Dynamic Analysis of weather and climate : atmospheric circulation, perturbations, climatic evolution [2. ed.] 9783642046797, 3642046797

Citation preview

DYNAMIC ANALYSIS OF WEATHER AND CLIMATE — Atmospheric Circulation, — Perturbations, Climatic Evolution

Marcel Leroux

he debate about 'global warming', 'climate change' and other environmental issues shows no sign of abating. In this updated and comprehensive second edition, the late Marcel Leroux challenges some of the 'scientific evidence' for global warming and shows how meteorology and climatology are natural sciences that rely on direct observation of phenomena. Through reconstructing thegeometry of atmospheric circulation, he demonstrates that very little is owed to hazard or chaos there is no 'unruly climate' but intensity shifts of the sum of weather patterns that constitute the climate - and he debunks the artificial separation between meteorology and climatology. In particular, he shows that solid observation provides more concrete results than abstract, computer-based predictions.

T

Dynamic Analysis of Weather and Climate • offers a perspective for the future of meteorology and climatology;

• makes clear how the Mobile Polar High (MPH) concept excels at explaining the meteorological facts and observations; • puts a meaningful and verifiable perspective on the future evolution of the climate;

• is fully illustrated with new figures and satellite images to give a better understanding of the concepts; • gives the most up-to-date, complete synthesis of the late author's research into weather and climate.

ISBN: 978-3-642-04679-7 springer.com www.praxis-publishing.co.uk

Marcel Leroux

Dynamic Analysis of Weather and Climate Atmospheric Circulation, Perturbations, Climatic Evolution (Second Edition)

Published in association with

Springer

Praxis Publishing Chichester. UK

Professor Marcel Leroux (deceased) Formerly Laboratoire de Climatologie, Risques et Environnement (LCRE) Lyon France

SPRINGER-PRAXIS BOOKS IN ENVIRONMENTAL SCIENCES SUBJECT ADVISORY EDITOR: John Mason, M.B.E., B.Sc., M.Sc., Ph.D.

ISBN 978-3-642-04679-7

Springer is part of Springer-Science + Business Media (springer.com)

Library of Congress Control Number: 2009936869

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction m accordance with the terms of licences issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside those terms should be sent to the publishers.

© Praxis Publishing Ltd, Chichester, UK, 2010 Based on the French second edition Published by Masson, Editeur, Paris © 2004 First English language edition published 1998

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Translator: Bob Mizon, 38 The Vineries, Colehill, Wimborne, Dorset, UK Project translation editor: Dr Marc Villeger Project copy editor: Rachael Wilkie Cover design: Jim Wilkie Typesetting: Aarontype Limited

Printed in Germany on acid-free paper

Contents

Preface.......................................................................................................................... xi

Acknowledgements....................................................................................................

xiii

Foreword......................................................................................................................... xv

Foreword to the English second edition..................................................................... xvii

Introduction: Perceptions of Weather and Climate and the Approach of this Volume.............................................................................................................. xxv Meteorology and/or climatology?.............................................................................xxv Perceptions of reality: schools of thought............................................................ xxvi Inadequacies in schools of thought, and associated problems............................... xxx The conceptual impasse........................................................................................ xxxi The approach of this book................................................................................. xxxiii

PART I GENERAL CIRCULATION IN THE TROPOSPHERE.................................

1

1 Radiation................................................................................................................. 3 1.1 Processes of radiation................................................................................ 3 1.2 The shape and motions of theEarth........................................................... 7 1.3 Greenhouse effect, water effect................................................................... 8 1.4 The geographical factor................................................................................. 12 1.5 Conclusion....................................................................................................... 13 Circulation in the lower layers of the troposphere....................................................... 14

vi Contents

2

Circulation in high and mid-latitudes: MPHs..................................................... 2.1 Perception of circulation in high and mid-latitudes............................... 2.2 The existence of mobile anticyclones....................................................... 2.3 Mobile Polar Anticyclones (Mobile Polar Highs (MPHs))................... 2.4 The polar thermal deficit............................................................................ 2.5 The birth of MPHs...................................................................................... 2.6 MPH trajectories........................................................................................ 2.7 The MPH-associated wind field.................................................................

3

Anticyclonic agglutinations.................................................................................... 45 3.1 A look at the so-called ‘subtropical’ high-pressure areas...................... 45 3.2 Meridional transport by MPHs and the formation of an.......................... anticyclonic agglutination (AA)................................................................. 48 3.2.1 The role of relief............................................................................ 48 3.2.2 The formation of an anticyclonic agglutination.......................... 53 3.3 Oceanic anticyclonic agglutinations......................................................... 55 3.3.1 Seasonal migration.......................................................................... 56 3.3.2 Migration in latitude..................................................................... 56 3.3.3 Migration in longitude................................................................... 56 3.4 Seasonal and/or temporary anticyclonic agglutinations in mid-latitudes................................................................................................ 57 3.4.1 Winter anticyclonic agglutinations................................................ 58 The ‘Siberian’ anticyclone.............................................................. 58 Duration of agglutinations............................................................ 60 3.4.2 Summer anticyclonic agglutinations............................................. 63 High pressure over the Mediterranean......................................... 63 The heatwave and drought of summer 2003 in Western Europe............................................................................... 64 The floods and the heatwave of summer 2007 .......................... 68 3.5 Conclusion.................................................................................................... 70

4 Tropical circulation................................................................................................ 4.1 A look at tropical circulation................................................................... 4.2 Pressure and wind fields over the tropics................................................ Trade circulation........................................................................................ Monsoon circulation.................................................................................... 4.3 The trade wind............................................................................................. 4.4 The trades.................................................................................................... 4.5 The monsoon................................................................................................ 4.6 Monsoons....................................................................................................

17 17 20 22 27 29 30 41

73 73 74 75 76 78 81 84 87

Circulation in the lower layers:conclusion................................................................... 91 5

General circulation............................................................................................... 5.1 General circulation: theevolution of ideas.............................................. 5.1.1 The birth of the tri-cellular model of circulation........................ 5.1.2 Improvements on the tri-cellular model of circulation..............

93 94 95 96

Contents

5.2 5.3

5.4

5.5

vii

Insufficiencies in the representation of circulation................................... 97 Units of circulation in the lower layers.......................................................99 5.3.1 The northern meteorological hemisphere....................................... 101 5.3.2 The southern meteorological hemisphere....................................... 102 5.3.3 Dynamical unity and climatic diversity......................................... 103 5.3.4 Fundamental questions................................................................... 104 General circulation in the troposphere........................................................105 5.4.1 The mean tropospheric picture........................................................105 5.4.2 Seasonal variation in circulation..................................................... 109 5.4.3 Partitioning and stratification in circulation.................................. Ill Conclusion: general circulation is perfectly organised............................. 115

PART II DYNAMICS OF THE WEATHER: DISTURBANCES .................................... 117

6

Pluviogenesis............................................................................................................ 119 6.1 Precipitable water......................................................................................... 119 6.2 Origin of an updraft.................................................................................... 121 6.2.1 The thermal factor...........................................................................122 6.2.2 The dynamical factor........................................................................ 123 6.3 Structural conditions.................................................................................... 125 6.4 Conclusion..................................................................................................... 126

7

Dynamics of weather in polar and temperate regions: MPHs........................... 129 7.1 Perception of the ‘disturbed field’ in high and mid-latitudes................. 129 7.1.1 The (impossible) control from above of phenomena lower down......................................................................................... 130 7.1.2 The FASTEX (Non-) Experiment................................................... 132 7.2 The MPH, the low-pressure corridor and the‘cyclone’........................... 136 7.2.1 Organisation of the pressure and wind fields................................137 7.2.2 The relationship between the MPH and the low-pressure area............................................................................. 139 7.2.3 Wind field above an MPH...............................................................143 7.2.4 MPHs and the ‘polar front’............................................................ 145 7.3 Weather associated with an MPH...............................................................145 7.4 Interactions between MPHs........................................................................ 148 7.5 Dynamics of weather in North America................................................... 151 7.5.1 The west coast.................................................................................. 154 7.5.2 East of the Rockies...........................................................................155 7.6 Dynamics of weather in France................................................................. 159 7.6.1 Relief and MPHs............................................................................. 159 7.6.2 Winter dynamics............................................................................... 162 7.6.3 Summer dynamics............................................................................. 164 7.7 Dynamics of weather in temperate and polar regions: conclusion ... 169

viii

Contents

8

MPHs at tropical margins............................................................................... 8.1 The temperate-tropical boundary......................................................... 8.2 North and Central America................................................................... 8.2.1 West of the Rockies................................................................... 8.2.2 East of the Rockies..................................................................... 8.3 South America........................................................................................ 8.3.1 West of the Andes..................................................................... 8.3.2 East of the Andes........................................................................ 8.4 The Mediterranean, North Africa, Arabia and the Indian Ocean . . 8.4.1 The Mediterranean..................................................................... 8.4.2 The Atlantic coastal area............................................................ 8.4.3 North Africa............................................................................... 8.4.4 Arabia, Indian Ocean................................................................ 8.5 Southern Africa...................................................................................... 8.5.1 The southern coastal area......................................................... 8.5.2 The Indian Ocean coastal area.................................................. 8.6 Eastern Asia............ •................................................................................ 8.7 Australia.................................................................................................. 8.7.1 “Migratory anticyclones”......................................................... 8.7.2 The southern coastal area......................................................... 8.8 Conclusion...............................................................................................

9

Pulses in trades and monsoons........................................................................ 195 9.1 Trade winds and ‘easterly waves’......................................................... 195 9.2 The vertical structure of the trades....................................................... 197 9.3 Pulses in the maritime trades................................................................. 198 9.4 Pulses in the continental trades............................................................ 204 9.5 Pulses in the monsoon............................................................................ 206 9.6 Conclusion............................................................................................... 208

171 171 174 174 175 177 177 178 180 180 181 182 185 186 186 187 188 190 190 192 193

10

The meteorological equator............................................................................ 10.1 The meteorological equator: the evolution of a concept................. 10.2 The inclined meteorological equator (IME)...................................... 10.2.1 The vertical structure of the IME......................................... 10.2.2 IME activity: squall lines (SL)............................................. 10.2.3 The active inclined meteorological equator........................ 10.3 The vertical meteorological equator (VME)...................................... 10.3.1 The VME over the oceans..................................................... 10.3.2 The ME over continents: IME and VME.......................... 10.4 Conclusion.............................................................................................

209 209 213 213 214 219 221 221 223 225

11

Tropical cyclones............................................................................................. 11.1 Cyclone structure andassociated weather.......................................... 11.2 Conditions forcyclogenesis................................................................. 11.3 The trajectories of cyclones.................................................................

227 227 229 234

Contents

11.4

11.5

ix

The geography of tropical cyclones...................................................... 238 11.4.1 The northern hemisphere..........................................................238 11.4.2 The southern hemisphere..........................................................243 Conclusion................................................................................................ 245

PART III DYNAMICS OF CLIMATE: CLIMATIC EVOLUTION THE “GLOBAL CLIMATIC SYSTEM”.............................................................. 247 12

Causes of climatic variations............................................................................... 251 12.1 Orbital parameters of radiation.............................................................. 251 12.1.1 Variation of the eccentricity of the Earth’sorbit..................... 252 12.1.2 Variation of the angle of inclination of the Earth’s polar axis....................................................................................253 12.1.3 Variation of the orientation of the polar axis........................ 255 12.1.4 Orbital parameters and climatic evolution............................. 256 12.2 Variations in solar activity..................................................................... 257 12.2.1 The sunspot cycle........................................................................ 257 12.2.2 Solar activity and climate......................................................... 257 12.2.3 New approaches.......................................................................... 261 12.3 Volcanism and climate............................................................................ 263 12.3.1 Volcanic emissions and ejecta: silicates and sulphates .... 263 12.3.2 Optical and radiative effects..................................................... 267 12.3.3 Varying, deferred and non-uniform thermal effects.............. 268 12.3.4 Climatic impact.......................................................................... 270 12.3.5 Volcanism: conclusion.............................................................. 272 12.4 The global warming myth........................ 273 12.4.1 Birth of a myth.......................................................................... 273 12.4.2 The reliability of estimates of the concentration of CO2 . . 275 12.4.3 Models and ‘global’ warming.................................................. 278 12.4.4 How representative is the ‘global’ thermal curve?................. 280 12.4.5 Thermal evolution at the poles................................................ 285 12.5 Conclusion on the causes of climatic variations..................................292

13

Palaeoclimatic variations and modes of general circulation............................. 293 13.1 Palaeoenvironments in Africa.................................................................293 13.1.1 Present-day dynamics of climate in Africa............................. 295 13.1.2 The palaeoenvironment of Africa at the time of the Last Glacial Maximum (18-15 kyr BP)........................................... 298 13.1.3 The palaeoenvironment of Africa at the time of the Holocene Climatic Optimum (9-6 kyr BP)............................. 303 13.1.4 Palaeometeorological interpretation......................................... 308

x

Contents

Variations in insolation and in modes of general circulation.......... 13.2.1 Variations in insolation......................................................... 13.2.2 Rapid general circulation (cold scenario)............................. 13.2.3 Slow general circulation (warm scenario)............................. Glaciation and deglaciation................................................................ 13.3.1 The onset of glaciation......................................................... 13.3.2 Dynamical processes of glaciation......................................... 13.3.3 Antarctic glaciation................................................................ 13.3.4 Glaciation in the north......................................................... Palaeocirculations over Africa............................................................ 13.4.1 Circulation at the time of the Last Glacial Maximum . . . 13.4.2 Circulation at the time of the Holocene Climatic Optimum................................................................... Conclusion.............................................................................................

310 310 312 315 318 318 321 322 324 329 329

Recent climatic evolution................................................................................. 14.1 Dynamics of the great Sahel drought................................................ 14.1.1 Sahelian pluviogenesis............................................................ 14.1.2 Presumed causes of the great drought................................. 14.1.3 The southward shift of pluviogenic structures.................... 14.2 The dynamics of Antarctica................................................................ 14.2.1 The Western Antarctic............................................................ 14.2.2 The recent dynamic of the Antarctic.................................... 14.3 Climatic evolution in the Arctic/North Atlantic/Europe/ Mediterranean space............................................................................ 14.3.1 Description of the North Atlantic aerological unit........... 14.3.2 The dynamic of the Arctic................................................... 14.3.3 Arctic ice................................................................................ 14.3.4 The evolution of weather in the North Atlantic space . . . 14.3.5 Increased pressure: a key element in climatic evolution. . . 14.3.6 Conclusion.............................................................................. 14.4 Climatic evolution in the North Pacific space................................. 14.4.1 The aerological dynamic of the North Pacific.................. 14.4.2 Recent climatic evolution on the eastern side of the North Pacific.......................................................................... 14.5 El Nino the ‘star’, and the real El Nino........................................... 14.5.1 The real Nino of the eastern Pacific.................................... 14.5.2 El Nino and the Southern Oscillation: ENSO................... 14.5.3 Cyclogenesis inthesouth-east Pacific.................................. 14.6 Conclusion on recentclimatic evolution.............................................

335 336 336 339 341 344 344 347

13.2

13.3

13.4

13.5 14

332 333

351 351 354 357 360 364 368 369 369 372 376 377 379 384 386

General conclusion...........................................................................................

391

Bibliography..............................................................................................................

397

15

Index.......................................................................................................................... 421 {Colour plate section positioned in the middle of the book.}

Preface

Praxis Publishing is to be praised for its initiative in taking on the second English edition of Dynamic Analysis of Weather and Climate, thereby bringing a major work to the world’s notice. It was Marcel Leroux’ lifelong aim, pursued with steely determination even in his last days, to understand, and thereby to increase the understanding of others. As both geographer and climatologist, he knew that true understanding goes beyond our comprehension of tables of statistics; it involves thorough research into the complex web of facts about the interactions of the atmosphere, the oceans and the features of the land. His respect for nature in its entirety was founded upon real observation, the indispensable basis of scientific research to any student of the natural world. Leroux fosters understanding through the clarity of his explanation, solidly supported and encompassing the whole of the domain under scrutiny: the mark of the effective teacher. I would like to recall here, having witnessed it at first hand, the constant care that Marcel Leroux lavished upon his students, in particular the many who were preparing their theses under his guidance. In his researches, as in his teaching, Marcel Leroux was a great man.

Charles Toupet Emeritus Professor, Universite Jean Moulin Lyon III

Acknowledgements

It is in Man’s nature to err, but only the fool persists in his fault. Cicero My father passed away two months after the completion of this book. He was well aware that his health was failing, and that there would be no more books to follow. This therefore marks the final stage of his work. Originally destined for the French-language readership, the current work represents the third edition of La Dynamique du Temps et du Climat as well as the second English-language edition of Dynamic Analysis of Weather and Climate. My particular thanks go to geologist Marc Villeger, whose support has been invaluable: he has given me the benefit not only of his experience, but also of his command of English. Without his contribution, this book would never have been produced. He has allowed me to keep a promise made. May I also thank: • •

• • • •

My mother, Madame Sylviane Guerin, and my aunt, Madame Ginette Leroux, for their assistance. Madame Veglia Leroux-Balzani, my father’s widow, for the trust she has shown in me. Monsieur Charles Toupet, Professor Emeritus, forty years a colleague of my father, who kindly consented to write the preface to this book. Mr Clive Horwood of Praxis Publishing and all his colleagues. Mr Bob Mizon for his excellent translation. Dr George Kukla of the Lamont Doherty Earth Observatory, Dr Gary Sharp of the Center for Climate/Ocean Resources Study, Dr Sonja A. Boehmer-

xiv

•

Acknowledgements

Christiansen of the Department of Geography, University of Hull, Dr Hans Jelbring and Dr Donald Rapp for their support. Monsieur Jacques Brache, geophysicist. May they all be assured of my sincere gratitude.

Sandrine Leroux

Foreword

If you want to be hated, tell a complacent man where he is wrong. And if you want to be hated unto death, show him where he is wrong.

James Mills in One Just Man, 1974 Professor Marcel Leroux opened the French second edition of La Dynamique Du Temps Et Du Climat with this distinctive quote. Since the Anthropogenic Global Warming theory was conveniently renamed ‘climate change’, climatologists such as Leroux have experienced Mills’ prescient omen. This choice fits the man, the scientist and, thus, the resulting loss those who learned from him felt when the announcement of his passing on 12 August 2008 reached us. I was supposed to meet with Marcel Leroux in the fall of 2008. His books The Meteorology and Climate of Tropical Africa (Springer-Praxis 2001) and Global Warming: Myth Or Reality? The Erring Ways Of Climatology, (Springer-Praxis 2005), explained the intricacies of atmospheric circulation in a luminous and logical way to this geologist. We corresponded for about a year and his natural modesty was added to the many qualities his books exude. Then across came the news that, despite a deep interest from former colleagues, his French publisher surprisingly declined publishing what would have been the third edition of this book. His daughter, Sandrine, was left bereft and the trusted recipient of her father’s final work. Meanwhile Praxis publisher Clive Horwood expressed his keen interest in seeing the work translated and published in English just as Global Warming: Myth Or Reality had been by Springer-Praxis in 2005. Loyalty to the memory of Professor Leroux, to his contribution to meteorology and climatology - arguably a dynamic understanding revolution as important as

xvi

Foreword

Plate Tectonics has been to Earth Sciences - and toward his family dictated my involvement, stepping in and coordinating the publication of the English language second edition of Dynamic Analysis of Weather and Climate: Atmospheric Circula­ tion, Perturbations, Climatic Evolution on Sandrine’s behalf. Sandrine Leroux and Clive Horwood kindly offered me to write this foreword, a distinct honor. Through this updated edition, you will learn about the General Atmospheric Circulation, Mobile Polar Highs (MPH), Rapid and Slow Modes of General Circu­ lation, meridional exchanges and aerological units and their role in climatic evolu­ tion during Glacial and Interglacial periods. Its fundamental methodology gives the primary role to the observed and measured facts and how coherently they are integrated into the MPH concept. In building his scientific case, Marcel is the proud heir of Cartesianism. Yet his style borrows from the enlightenment age, through Voltairian touches of irony and the luminous verve of Diderot. For Marcel Leroux, meteorology and climatology are Natural Sciences based on concrete elements. This well organized machine owes very little to hazard or chaos. There is no “unruly climate” but intensity shifts in the climate that is the sum of weathers, debunking the artificial separation between meteorology and climatology. This logical understanding offers a meaningful and verifiable perspective for the evolution of climate. Upon reading this book, you’ll be given the tools to decipher satellite images and understand weather, and to place in proper context the avalanche of more or less scientific news on the subject and appreciate their true significance. Indeed, Marcel Leroux shared his knowledge until the end in order to empower his audience, and that is the mark of a true humanist since, these days, sharing climatological knowledge requires courage. Leroux had it. Dr Marc Villeger Geologist

Foreword to the English Second Edition

xvii

I do not claim to be able to reveal all the sources of our errors concerning this limitless subject; when we think that we have found the truth in one place, it escapes us in a thousand others. However, I hope that, by exploring the principal areas of the mind, I may be able to observe essential differences, and dispel a good number of those imaginary contradictions which ignorance fosters. Luc de Clapiers, Marquis de Vauvenargues, Introduction a la Connaissance de I’Esprit Humain, Paris, 1746.

Foreword to the English Second Edition

The French first edition of this book appeared in December 1996. An English translation was published in 1998 by Praxis-Wiley, with a French second edition appearing in 2000, and a revised version in 2004. During this period, ‘climatology’, hitherto relatively low-key, had become very much a la mode, in particular in the media. Now not a day goes by without climate change being evoked, from the start equated to (alleged) ‘climatic warming’, confronting humanity with dire threats. Meanwhile, the number of ‘climatologists’, once quite small and usually geographer­ physicists and/or archivists of the observational data of the weather services, became considerably inflated. Most of them seem to be self-proclaimed, or created by political authority and therefore without scientific qualification. However, they are ‘recog­ nised’, and have even become the ‘darlings’ of the media (in particular if they are helicologistes or ‘eco-jetsetters’, flying here and there and owning several houses ...!). This ‘new’ climatology is orchestrated by the IPCC (Intergovernmental Panel on Climate Change), an organization created in 1988 by the World Meteorological Organization (WMO) and the United Nations Environment Programme (UNEP). It must be stressed that the ‘I’ of IPCC means ‘intergovernmental’, which implies that its members are named by governments and not by scientific authorities, meaning that the number of true climate ‘experts’ involved is very limited. For example, the French delegation of the IPCC comprises one ‘climatologist’, who is primarily a theoretical modeller (see below, in the General Introduction). The IPCC has published four reports (in 1990, 1995, 2001 and 2007), which are nearly always referred to as the Summaries for Policymakers (SP). The thousands of pages which present the Physical Bases for Climate Change (http://ipcc-wgl.ucar.edu/wgl/wglreport.html) are generally ignored, while the few pages of Summary for Policymakers are the references slavishly adhered to by politicians, and amplified by the media. The Summaries, built upon postulates laid down as ‘facts’, have become the basis for

xviii

Foreword to the English Second Edition

economic and political negotiations among IPCC members; climatology is reduced to mere simplicity, an ‘accessible’ science - because it now consists of a few slogans, or even the one leitmotiv, “humankind is responsible for the climate”! The omnipresence of this brand of climatology in the media, and the ‘assertions’ founded on an alleged (and illusory) scientific consensus, can lead some to suppose that meteorology/climatology has made considerable progress, to such a degree that it now represents a perfected science, able to answer all questions, and even to predict the climate of the next century! Unfortunately, we are soon brought down to Earth when we realize that the discipline of climatology itself has reaped no benefit from such exposure in the media, or from being bandied around by politicians. On the contrary, it has become simplistic, even hyper-simplified, and often puerile. The ‘greenhouse effect’ is evoked to ‘explain’ away, indiscriminately, ‘all’ weather phenomena: anything and everything, and often at the level of saloon­ bar debate. Heat or cold, flood or drought, storm or tempest, even cyclones: no phenomenon is ruled out, and the more catastrophic the better! It is thus essential to take stock, without fear or favour, and point out the rudiments essential for the comprehension of meteorological/climatological phe­ nomena, the fundamental knowledge that any climatologist worthy of the name must absolutely know before claiming to be an ‘expert’ on climate. Such is the goal of this work. This book is the fruit of long experience as a university teacher and researcher, and of long reflection, founded first and foremost upon direct observation of phenomena, initially in the Tropics, and later in temperate and polar areas. •

•

•

My first research, from 1967 onwards, was directed towards a subject which was then (and still is) a matter of debate: the distinction between tropical and extratropical weather. My analysis of Incursions of Air of Polar Origin across Western Africa, (cf. http://lcre.univ-lyon3.fr/climato/ and bibliography in fine) clearly showed that there is no ‘barrier’ between climatic zones, but then, I was still far from imagining the paramount importance of polar air in the dynamics of meridional exchanges. Within both the University of Dakar and the Bureau d’Etudes Meteorologiques (Weather Research Department) of ASECNA (Agence pour la Securite Aerienne en Afrique et a Madagascar - Agency for Air Security in Africa and Mada­ gascar, undertaking the work of Meteo France in Africa), I analyzed the Dynamics of Precipitations in West Africa as part of a third-year thesis. This was published by the ASECNA Weather Research Department in 1970. After completing studies of The Meteorological Equator in Africa (1973) and The High-Altitude Wind Field over Western and Central Africa (1974), analyses published by ASECNA, and then of Processes of Formation and Evolution of Squall Lines in Northern Tropical Africa (1976), I undertook a state thesis on climatology. The Climate of Tropical Africa, my 1980 thesis, was published in 1983 by the World Meteorological Organization (WMO/OMM, Geneva), and was distributed within all member states. The publication also enjoyed subsidies from the Agence de Cooperation Culturelie et Technique (ACCT), of the

Foreword to the English Second Edition

•

xix

Ministere de la Cooperation, and from the Centre National de la Recherche Scientifique (CNRS). This work was reprinted and found a new form as an English version entitled The Meteorology and Climate of Tropical Africa (Springer-Praxis, 2001), a particular feature being the 250 weather and climatic charts of the incorporated Atlas (in the form of a CD), fundamental to our understanding of the vertical structure of the African troposphere. From 1986, and now at the University of Lyon, I soon saw confirmation of the interaction of climatic zones, with particular reference to polar sources of circulation within the lower layers. I noted especially that a deep dichotomy existed in the analysis of temperate and polar phenomena, in the knowledge that the theoretical models described in the literature do not agree with observed reality, particularly on synoptic charts and satellite images. In 1986 I offered my thoughts on this in an article entitled ‘L’Anticyclone Mobile Polaire: Facteur Premier de la Climatologie Temperee’ (‘The Mobile Polar High: Principal Factor in Temperate-area Climatology’). My conviction was regularly con­ solidated, within the context of the Laboratoire de Climatologie, Risques, Environnement (LCRE) in the form of articles, directions of doctoral research undertaken at various latitudes, and in various thematic issues of the Revue de Geographic de Lyon (RGL: 1990, 1991, 1993, 1995, 1997, 2000) dealing especially with the climate of France.

It was vital to consider the evolution of climate on all timescales, by applying in particular the concept of the Mobile Polar High (MPH) on a palaeoenvironmental scale and by defining the variations in the intensity of general circulation (cf. The Mobile Polar High: A new concept explaining the actual mechanisms of meridional air-mass and energy exchanges, and the global propagation of palaeoclimatic changes (1993). After several articles devoted to criticism of the alleged part played by the greenhouse effect, and in particular its part in weakening the dis­ cipline of climatology, in 2005 I published, Global Warming: Myth or Reality? The Erring Ways of Climatology (Springer-Praxis), following which a second edition was requested. Dynamic Analysis of Weather and Climate thus represents the end result of a permanent and continuous process, not dictated by the literature but imposed by the facts of weather themselves, from the tropics to the mid- and high latitudes, and ever more widely until the scale of general circulation itself is reached, with discussion of the causes of its variations. This approach is underpinned by the concept of the Polar Mobile High (MPH): its origins, characteristics, trajectories, influence on the dynamics of temperate and tropical zones; its role as the engine of general circula­ tion; its power, and the respective extents of the meteorological hemispheres and their displacement across the meteorological equator; modes of general circulation explaining climatic variations, from the seasonal scale, with reference to recent evolution, to the palaeoenvironmental scale (fast and slow modes, with enhanced or reduced intensities in meridional exchanges). Reactions to the MPH concept range from its favourable and often enthusiastic reception by those who had not found explanations of weather and climate in

xx

Foreword to the English Second Edition

traditional theories, to categorical rejection on the part of those for whom ‘the climatic die is cast’ (and cast a long time ago!). The latter are indignant when their certainties, so well anchored in routine, are challenged; they see as iconoclastic any questioning of their firmly established dogmas. So several different attitudes become apparent: •

•

•

Complete and immediate agreement with the MPH concept because, in place of the esoteric nature of current ideas and especially the vain attempts at conciliation between them, it offers a more coherent mode of reasoning than in the past (Demangeot, Bull. AGF, 1999, pp. 662-664), as well as a concrete and directly observable vision of meteorological phenomena. There are those who particularly appreciate seeing geographical climatology become a major dis­ cipline, free at last from fruitless and ill-fitting meteorological concepts. A (limited) attempt at vulgarization (or even retrenchment): “we knew that already; we have already described mobile anticyclones ... ”. In Chapter 2.2 I underline the fact that I did not invent these anticyclones whose displacement has been observed for such , a long time, particularly in the southern hemisphere. Everyone has the chance to ‘observe the anticyclone as it passes’, but then comes the questions: Where did the MPH come from? Where is it going? What are its characteristics? Why is it passing through this particular place? At what speed? What is its importance in the determination of weather, and its role in general circulation ... ? These questions were never asked at the right time, and the observation of anticyclones has never pointed the way to a logical and global ‘concept’: a concept that explains the dynamic of weather and climate on all scales of phenomena, space and time. Thus, it is not enough just to watch the anticyclone go by! A categorical refusal, en bloc, without analysis or debate. On the scientific level, this is at the very least an odd attitude: when a new concept appears, it is immediately discussed, in writing, and rejected if judged erroneous. But ever since the first presentations of the MPH concept in 1983 and 1986, not one documented, reasoned - or signed - article has taken the concept to task. Generally speaking, this is the result of a certain embarrassment on the part of the whole ‘weather lobby’, and the silence is total. Labasse and Foechterle (1999) summarized the situation well: “... because there is a kind of icy shell beneath which the meteorological establishment shelters whenever the subject of the MPH is raised. A shell of silence: the Centre National de Recherches Meteorologiques, after some shilly-shallying, officially decided to refuse to comment on this subject, an attitude far removed from scientific usage, but confirmed by Meteo France” (Science et Vie 879, p. 72). What is more, a representative of the communication service of the Meteo France Research Centre (CNRM) announced: “Meteo France does not attach credence to the theory of the mobile polar high’’ (cf. http://www.provence-web.com/mistral/ ampet.htm), but did not refer to any official document presenting the scientific reasons for this rejection. The reaction of Meteo France and of the Societe Meteorologique de France (SMF) is surprising in its ambiguity and scientific

Foreword to the English Second Edition

xxi

insincerity. The SMF, it seems, considers that “the debate is now closed” and it even states, in order to avoid the publication of any article containing the abbreviation MPH, that its review is not intended “to retail ideas” (La Meteorologie 16, 1996, 49-52)! The situation remains unchanged. Is this because the MPH model is ‘too simple’? It is true, as Lenoir (1997) emphasized in Futuribles (p. 105) that “the extreme conceptual simplicity of the MPH model ” is not the least handicap, given that ‘all those who have racked their brains trying (in vain) to make the ‘official’ version of atmospheric circulation work and, in despair, have taken the trouble to familiarise themselves with the jargon of ‘chaos’ and the ‘butterfly effect’, from which proceeds all that is physically inconceivable (i.e. ‘physical miracles’), will not unreservedly agree to admit the inanity of such effort”.

And moreover, why is there no answer from the meteorological establishment in defence or justification of ‘physical miracles’, those scientific aberrations which are the implicit bases of attempts at synthesis between the classical theories? Is it because, as Dady (1995) points out in La Recherche (No 276, vol. 26, pp. 479-480), meteorology “has set itself against using any concept” and thus “thinks it is in possession of the truth” derived from the raw equations of fluid dynamics, even though ‘concept’ is inherent in any scientific process? Dady sees in this “a reduc­ tionist illusion, induced in many fields of research by the development of computing power”, an attitude which drives a constant headlong rush “in the hope of mega­ computers”: increasingly powerful and increasingly expensive machines, each one expected to come up with the ‘miracle’ (which, for intrinsic reasons, it never can) Is this because the principal concern of the weather service is to forecast, and in that context, as Joly (1995) puts it, “forecasting progresses, but not comprehension”? Joly insists on reminding us that “to predict is not to explain” (La Recherche, 276: 26, p. 480, 1995). But, if comprehension does not progress, is it quite certain that forecasting will progress? Meteo France would have us believe this by maintaining via the media the fiction of the ‘7-day forecast’: but the forecast for the third or fourth day is given with a degree of confidence of 3 or 2 out of 5! Is this quite serious? Three out of 5, or 2 out of 5: these are simply a little more or a little less than 1 out of 2, or a half, therefore 1 chance out of 2, in other words: might rain, might not rain! Is this really forecasting? Armed with the modern profusion of observational methods, theories, models and calculating power (and at what cost!), they arrive at such a conclusion! How can a mathematical and statistical approach which takes no heed of the real factors of the weather predict the trajectory and future activity of an MPH as yet unborn? It is incredible that received ideas and old concepts are repeated ad nauseam, without offering the slightest (true) explanation of phenomena, by the ‘weather’ service, with such amazing technical means at its disposal. It is true that the MPH concept has plenty not to like: •

it gives priority to direct observation, which is usually thought to be obsolete, and is even scorned - particularly by theorists - with its basic questioning of the dogmas and theories of the ‘holy books’;

xxii

•

• • • •

•

•

Foreword to the English Second Edition

it leads to an overall vision of climatology through the dynamics of meridional exchanges, while ‘traditional’ climatology divides meteorological space into distinct and dynamically separate cells; it recognizes the primary importance of the polar zones in the release and the maintenance of general circulation, a role usually allotted to the tropical zone; it underlines the primacy of the lower layers of the troposphere and its geography, while the paramount role is normally ascribed to the upper layers; it stresses the primary role of anticyclones, and particularly mobile ones, while traditional meteorology ‘sees’ only the depressions; it allots to these anticyclones (MPHs) the roles of engines of general circulation and triggers of disturbances, particularly at high- and mid-latitudes, but also to varying degrees in the tropical zone, which is traditionally separated from other climatic zones; it makes it possible to appreciate the nature of climatic variations on all scales of space, time, and intensity, from the synoptic (real-time) scale to the palaeoenvironmental scale, which traditional concepts are still unable to explain in a coherent way; it presents (unforgivably!) not only the functional simplicity of the climatic machine, but also its rigorous organization, which obliges us to locate each weather phenomenon in its correct place, whereas the absence in the models of any realistic diagram of general circulation allows ‘free rein’.

But the MPH concept also has plenty to recommend it. The extent to which it appeals is of course inversely proportional to the extent to which previous concepts have infiltrated - or even intoxicated - our minds; because, as Allegre (1982) emphasizes, ''‘'The more novel an idea is, the greater its power to shock, and the more it disturbs those whose reputations have been established elsewhere and whose intellectual comfort is troubled by its emergence”. But reality cannot be circumvented, and the view from the weather satellite is not subject to appeal: what it sees is indisputable, and a refusal to admit this is tantamount to blindness. The weather depends not on abstractions resulting from calculations provided by models which do not reflect the reality of the phenomena, but on real objects (MPHs), “concrete and personalised physical entities” as Demangeot (1999) puts it: entities that are born, move, change, interact and are affected by relief, and end up merging to give rise to tropical circulation, creating weather that is peculiar to each MPH and to each stage of its evolution. By studying a satellite photograph, any observer can easily see the birth of the factors affecting the weather, follow their trajectories, and predict their probable displacement for the following day and for the day after that, from the polar regions to the very heart of the tropical zone. If outmoded notions are not challenged; if there is a refusal neither to observe reality nor to let satellite images be the arbiter when there are conflicts of opinion; if absolute preference is doggedly given to theory and if there is blind confidence in climatic models - in brief, if there is a refusal to make progress in the face of repeated failures - then we are dealing with a discipline foreign to climatology rather

Foreword to the English Second Edition

xxiii

than with true science. But, as Galileo (condemned in 1633) stressed: “eppur si muove” - the MPH is on the move, and nobody can stop it ... Vauvenargues, 15 May 2008 Marcel Leroux Professor Emeritus of Universities; former Director of the Centre de Recherches de Climatogie Tropicale Africaine, CRCTA (Dakar); former Director of the Laboratoire de Climatologie-Risques-Environnement, LCRE (Lyon); member of the Societe Meteorologique de France (SMF); former Member of the American Meteorological Society (AMS), Chevalier dans l’Ordre des Palmes Academiques.

The more novel an idea is, the more its power to shock, and the more it upsets those whose reputations have been established elsewhere, and those whose intellectual comfort has been troubled by its emergence. Originality is a prized virtue, provided that it is not too disturbing. Beyond a certain threshold, any bold innovation will be met with marginalisation, or even sacrificial reaction Claude Allegre, L’Ecume de la Terre, 1982)

Introduction

Perceptions of Weather and Climate and the Approach of this Volume

Weather and climate, cause and effect, instantaneous or synoptic scales in weather, and differentiated or statistical scales (based on means) in climate, which is the out­ come of (the sum of) weather: in all these knowledge is fragmented, even closeted, according to scales of space and time (duration) within several disciplines; chief among these are meteorology, climatology and, in many aspects, oceanography. In each of these disciplines, knowledge either has been, or still is, apprehended through several schools of thought, often differing from each other according to the areas considered. It must be pointed out straight away that there is no chapter in this book dealing solely with marine hydrology, given that the surface movements of the oceans are principally dictated by surface temperatures (density currents in high latitudes) and especially by the aerological factor (drift currents). So it is on the atmospheric disciplines that we shall concentrate, without, however, forgetting those watery vastnesses which cover three quarters of the Earth’s surface. They represent a kind of surface with a specific thermal behaviour and constitute the main reservoir of potential heat energy: their influence is necessarily integrated into the analysis of the thermal and hygrometric characteristics of air movements in the lower layers of the atmosphere. Modes lay stress on the responsibility, in more or less durable fashion, of some causal factor, with varying degrees of justification. So, rightly or wrongly, the dynamics of weather and climate seem complex: differences in interpretation, or of temporary focus, serve only to increase the (artificial) difficulty of understanding the (natural) complexity of phenomena. We must begin by being aware of this. Meteorology and/or climatology?

Meteorology (from the Greek meteos, lofty) and climatology (from the Greek klima, inclination, referring to the angle of the Sun’s rays above the horizon) are, to most

xxvi

Introduction

people, normally two separate things. Meteorology is the study of weather and its causes, whilst climatology devotes itself to analysis of consequences for various milieux. The separation between the two disciplines is the fruit of circumstance. With the rise of aeronautics, meteorology became more and more preoccupied with current weather, and “recentred itself upon forecasting, which is its raison d’etre” (Rochas and Javelle, 1993); climatology was, in the language of the meteorologists, the collecting of observational data (and/or the basis for data used in models). Physical geography, on the other hand, concentrates on the aerological component in defining environments. The artificial dichotomy between meteorology-weather and geography-climate leads to a veritable mutilation of the logic of phenomena, involving aberrant approaches:

•

•

meteorologists consider that “charts corresponding to sufficiently high levels above the ground surface (particularly that of 500 hPa) are the means of understanding structures of synoptic movements, without complications intro­ duced by surface geographical characteristics” (Rochas and Javelle, 1993); climatologists blindly adopt the meteorologists’ concepts and, like them, very often look for the causes of interface phenomena (which occur essentially near the ground), in the barometric situation 5500 metres above. For this reason “we’ve always done it this way” certain climatological analyses systematically start at the 500 hPa level (though nobody really knows why!).

The meteorologists conveniently ignore the complexity of surface conditions, distancing themselves from them and stressing phenomena at altitude, sometimes to the point of neglecting geographical factors (especially relief). The climatologists, although well placed to be able to appreciate the importance of those same geo­ graphical factors, exclude them in their turn, either implicitly or deliberately.

Perceptions of reality: schools of thought Atmospheric phenomena have been, and still are, analyzed according to different schools of thought, which have succeeded each other through juxtaposition or superposition without really becoming integrated. Each school has offered its own logic and definitions, rejecting the principles of others without necessarily prevailing. As these schools still condition the ways in which meteorological phenomena are ‘seen’ and interpreted, it is most important to review their main ideas, in order to appreciate the real value of their influence, the multiplicity and validity of definitions translating physical reality, and the real complexity of natural phenomena, while estimating the contribution of artificially imported complication. The long era of empiricism, or popular meteorology (Dufour, 1973; Galtier, 1982; Chaboud, 1993; Victoire, 2001), is still with us in proverbs and sayings, in agri­ cultural precepts, in the lore of local microclimatic knowledge and forecasting, and in the designation of local winds. Maritime vocabulary, often dating back a very long time, is responsible for many appellations, especially of winds, and is a prime

Introduction

xxvii

source: the Roaring Forties or Brave West Winds, trades and monsoons, and the Mistral are examples. These legacies are often difficult to preserve unchanged, as meanings shift or become generalized and inappropriate, especially where terms initially maritime - become transposed to the land. The climatological or ‘statistical’ school is concerned with the analysis of mean data, used as a basis for reference. Description of the weather is derived from the mean meteorological situation, as shown on charts of mean pressure and resulting winds. Having arisen around the end of the 19th and the beginning of the 20th centuries, its influence is still with us through the names of centres of weather activity, isobaric entities constantly referred to: for example, in the North Atlantic, the well established formula runs thus: ‘the Icelandic low brings bad weather, whilst the Azores anticyclone is responsible for fine weather’. Owing their existence to statistics, these centres of action are considered to be permanent and ‘fixed’, or moving but little around a mean position. It was during this era (the 1920s), and principally because of the efforts of G. Walker, that it was proposed that there be indices to measure oscillations as a function of the respective strengths of the centres of action, between an anticyclone and a depression. Those proposed were: the Southern Oscillation (SO), the North Atlantic Oscillation (NAO), and the North Pacific Oscillation (NPO). These indices, simple differences in surface pressure between two weather stations, were merely ‘tricks of the trade’ to begin with (in the absence of a meteorological concept): artifices meant to help with forecasting. However, they explained nothing about the origin of centres of action, the variations in the strengths of anticyclones or the depths of depressions, and told us even less about the physical relationships between the two. For example, when discussing the NAO, we know nothing about why the Azores anticyclone expands or contracts, or why the Icelandic low moves, or why it deepens or fills. The corresponding geographical school of thought - known as analytical or separative - is founded on the same principles: it examines tightly overlapping natural parameters separately, elements of climate being translated into mean values. Essentially descriptive or based on appearances, and considered static and nonexplicative, it cannot (like the corresponding meteorological school) deal faithfully with the dynamic nature of phenomena. Then there is the air-mass, frontological or Norwegian schools of thought. In the preface to the French edition of articles setting out the theory of the polar front, Bjerknes (1923), denouncing criticisms of the climatological school, wrote: “Dynamical meteorology can make progress only from the detailed study of real and particular cases. The use of mean values often gives values which are difficult to interpret, even to the extent that important facts are often concealed”. Bjerknes and Solberg (1921) claim that contrasts in air masses of different densities, separated by fronts, form the basis for explanation of weather in temperate regions, and they propose the ‘Norwegian’ model of mid-latitude disturbance involving two fronts centred on a depression. As well as the polar front, representing the contact between polar and tropical air, other fronts are active everywhere, all the way from the vicinity of the pole, with its arctic front, to the tropics with their so-called TWF

xxviii

Introduction

(trade-windfront) and ITF, or intertropicalfront. This concept has entered the realms of the dogmatic, and the synoptic meteorology of polar and temperate latitudes is still to a very large extent based on the concepts of the air-mass/frontological school, especially for analysis of surface charts. The dynamical, kinematic or disturbance school does not evoke contrasts in density, or frontal discontinuities, but rather ‘discontinuities in speed and direction of lines of flux’, the construction of the kinematic field being based upon wind data. The increase in observations at altitude, with the invention of the radiosonde in 1927, led to the discovery of strong winds (jets), and as a consequence this school ascribed the role of prime mover to the upper layers of the atmosphere. The planetary wave, in the form of ‘Rossby waves’ (Rossby and Weightman, 1939), undulations in the highaltitude westerly jet stream, became the causal atmospheric phenomenon in temperate regions. Phenomena in the lower layers, and especially subtropical cyclones, were now relegated (like mobile highs, lows and associated bad weather, and in particular the Norwegian disturbance), to being consequences of what was happening above, although the hypothesis had never been formally proved. The climatological method of the two previous schools combines frontology and kinematics, in the synthetic method involving air masses and weather types (also known as the dynamical school). It thus applies the principles put forward by Julius Hann (1882), Wladimir Koppen (1936) and in particular Maximillien Sorre (1934), according to which climate is “the series of states of the atmosphere above a place in their habitual succession”. This method is, however, based upon an initial misap­ prehension: what is known as a ‘weather type’ is in fact only a ‘situation type’, recognized from the surface pressure field, and to each synoptic situation there corresponds an infinity of local weathers. The permanent confusion between scales of phenomena, i.e. between situation and weather, explains why this method has never managed to lead to a genetic classification of climates. Numerical forecasting and climatic modelling: in parallel with the increased interest shown in the upper layers, meteorological analysis has moved away from the concrete realities of the lower layers, and from a naturalistic approach. The initial project was described by Bjerknes (1904) in these terms: “z/, as everyone who reasons scientifically believes, atmospheric phenomena develop from those preceding them, following exact laws we ought to know with sufficient accuracy those laws through which one atmospheric state develops from the preceding one”. This postulate of the continuity of phenomena is the very basis of climatic modelling. Richardson put Bjerknes’ concepts into practice, and tried to resolve the equations of weather forecasting by means of numerical procedures. He had little success in 1922, principally (though not exclusively) because he had no access at the time to the necessary calculating power. The first electronic computer, ENIAC (1946), devel­ oped mainly by Von Neumann, led in 1950 to the first numerical weather forecast, and this approach to meteorology would henceforth be, according to the modellers, more of a ‘science’. Then came the era of the ‘technicians’ (Coiffier, 2000); as time went on, progress was made, but mainly on the numerical level rather than that of meteorological knowledge itself, thanks largely to extraordinary technical advances in computers. The architecture of the model is schematically as follows: space is

Introduction

xxix

divided into elementary boxes or cells within which, at a particular point or at a given instant, essential climatic variables are defined, transfers of mass or energy being effected from one cell to the next. This method of reducing space to a cell, or an elementary point within that cell (or to the scale of the ‘particle’), is known as the reductionist method. Based on ‘weather equations’ (equations of movement and continuity, and thermodynamic and ideal gas equations), numerical models represent, theoretically, the most objective approach (i.e. detached from direct observation and its interpretation): so, implicitly, concept-free. Since models cannot integrate all the components of phenomena, it becomes “necessary to ‘parameterize’ the phenomena, as the modellers’ jargon puts it, i.e. to take them into account without really describing them explicitly ... using only the parameters fed into the model” (Rochas and Javelle, 1993). In spite of its claim to freedom from concepts, the method is based upon the dynamical school, the “the doctrinal corps of numerical forecasting” (De Moor and Veyre, 1991) which promotes the undulatory model and the impor­ tance of upper layers, and from observational data takes only those which fit the model, even when dealing with satellite data considered ‘difficult to digitize’: the only aim is to predict (without necessarily explaining). The development of numerical models of general circulation and the use of powerful computers lead to climatic modelling. In the same way as with short-term weather forecasting, numerical climatic modelling tries to reconstitute past climates and/or predict possible climate change on a global scale. In fact, and “surprisingly for non-specialists: climate is simulated using the same models as for weather forecasting ... or at least by the same methods” (Rochas and Javelle, 1993), i.e. with the same approximative (parameterized) representation of phenomena; as Duplessy and Morel (1990) put it: “A climatic model is simply a coherent numerical framework, allowing us to keep an exact tally of the fluxes of matter, energy and quantity of movements within a complex system, even though the laws determining these fluxes may not be known exactly.” This schematization thus often comes down to simple and even simplistic relationships, and even ‘rule-of-three’ reasonings. For example, taking no account of actual processes of pluviogenesis or of their geographical distribution, an immediate relationship is suggested between water vapour and precipitation, and even between sea temperatures and rain, or the activity of tropical cyclones, despite the pertinent knowledge that there is no direct relationship between precipitable water potential and that water which is effectively precipitated. Modelling, i.e. the idea of calculating the evolution of the atmosphere using the equations of fluid mechanics and thermodynamics, dates from the early 20th century. Climatic modelling leads to diagnostic climatology where, here, ‘climatology’ most often has a ‘meteorologistic’, ‘statistical, or ‘data processing’ sense. This climat­ ology has been built up on the establishment of teleconnections (remote correlations) based on statistical studies developed through the intermediary of the model. But a regard for the physical analysis of phenomena and the geographical analysis of climate, a component of the natural environment, is absent or ill-defined. This essentially statistical trend in fact corresponds to a return to the climatological tendency, much amplified by current methods of data storage and treatment. The

xxx

Introduction

exact nature of phenomena is not observed directly, ‘reality’ being perceived by way of models which represent it but imperfectly. Teleconnections, which are in reality statistical relationships or covariations in parameters, are wrongly considered as being co-relationships (causal and/or physical) between apparently unrelated phenomena; any relationship is unrevealed, since it does not appear in the data used, and there is no direct analysis. The division between the schools of thought is further complicated as a function of the zones studied. Thus there is a division between polar meteorology and tropical meteorology, and it is supposed that there exists between them a ‘barrier’ of so-called ‘subtropical’ high pressure areas: thus some phenomena, though energetic, disappear at the edge of the temperate zone, whilst others appear ex nihilo at tropical margins. The absence of a perception of the whole picture of meteorological phenomena is implicitly expressed by the closed eddies (cells) in general atmospheric circulation, and especially by the tricellular scheme proposed in 1856 (and still in existence), masking the reality of meridional exchanges. It must be stressed that climatic models use exactly this tricellular scheme, which is inappropriate to describe exchanges of air and energy or the migration .of aerological and pluviogenic structures.

Inadequacies in schools of thought, and associated problems A diversity of perceptions is not of itself disadvantageous, since it offers several ways of approaching phenomena, except where methods or definitions maintain their continued existence only by dint of routine or inertia. There is, however, no real synthesis of different and irreconcilable perceptions, as each school of thought acts upon its own internal logic, in the absence of bridges between them, each having elaborated its own definitions and imposed its own dogmas. Ambiguities and deficiencies persist in many areas, and disagreements remain, for example in the explanation of circulation and its interconnections and variations on all timescales, the structure of the troposphere, or the genesis of disturbances. There exists no coherent model of general circulation which will both express the reality of meridional exchanges and integrate vertical and horizontal discontinuities as well as fundamental pluviogenic structures. So there is no overall scheme within whose context we might (and should) ‘frame’ phenomena taken in isolation, such as - and there are various examples - whether El Nino has a role at the origin or at the end of the process of general circulation. Moreover, certain concepts are founded upon ‘physical miracles’, postulates often hallowed by usage but going against the logic of phenomena and contradicting elementary physical principles. Each school of thought, firm in its immutable dogma, has bequeathed to us a way of thinking and its precepts, as well as its foci (dare we say ‘fixations’?), and a number of unanswered questions. The climatological school is still very much alive, with Meteo France claiming that “Z/ze Azores anticyclone is to blame” for poor summer weather, since that mythological entity “w not in its usualplace”\ (Carriere, in Reuters, 9 July 2007). This school would have us believe in the existence of permanent centres of action

Introduction

xxxi

(anticyclones and depressions), with names and locations, when all the while everything is moving. Above all, it imposes in some minds an incredible confusion of scales, which tainted climatology throughout the last century and still falsifies our perception of phenomena: entities defined by means do not exist on the synoptic scale; they may be invoked to explain the (mean) climate, but absolutely not to explain real weather! The frontological school’s method is still universally used in the analysis of surface charts. Apart from its imperfections (especially the absence of horizontal and vertical boundaries to its fronts), it has drawn attention to depressions while deliberately ignoring the associated anticyclones but has left ‘in the air’ any determination of the origin of the initial depression, the centre of the perturbed system. What is more, by applying to the explanation of weather the principle that ‘it’s the holes that constitute the cheese’, it has spread the habit of ‘seeking out the depression responsible’ (and ideally giving it a name: for example, the Icelandic or Aleutian Low, or even the Ligurian or Cypriot Low, though they are incapable of assuming such a responsibility since they are not causes, but statistical consequences revealed by mean pressure charts and not synoptic entities in themselves). The physical impossibility of synthesizing the two preceding schools (one dealing with the statistical scale and the other the synoptic) has led to the fixed idea that there exists a permanent field and a perturbed field, independent of each other, a dichotomy which is itself a considerable aberration; for, if everything is on the move, there can be only one single field! The dynamical school applies itself to the analysis of high-altitude charts. It gives the impression that it has resolved the problem of the origin of the ‘initial depression’ involved in mid-latitude disturbances, ascribing responsibility to undula­ tions in the high-altitude jet stream. However, this school ignores the existence of moving anticyclones in the lower levels of the atmosphere: the favoured (and undemonstrated) explanation of their origin consists in generalising it (falsely and counter to physical principles), in polar and temperate zones, to the ‘subtropical anticyclonic cells’ of the climatological school, i.e. subsident movements within the descending branch of Hadley cells! The numerical school, which declares itself free of concepts and, thereby, is the possessor of objective ‘truth’, is preoccupied with distilling virtual images and models, and has not accorded the importance they deserve to the real images pro­ vided by satellites. Within these images they seek primarily or exclusively confirma­ tion of outmoded perspectives, rather than allow reality to assert itself in their minds. Diagnostic climatology, ever more statistical and decreasingly meteorological, is no longer interested in the physical links between phenomena and cannot explain the (statistical) relationships with which it comes up.

The conceptual impasse So, for more than 50 years, meteorology has been in a veritable conceptual impasse. There is no longer any advance in our understanding. It might even be said that it

xxxii

Introduction

has regressed for we see in the literature at the beginning of this century yet more dei ex machincr. ‘oscillations’, the origin of which was not explained when they were proposed, and which remains unexplained in the context of traditional concepts! As Dady (1995), initiator of numerical forecasting in France, stressed: ‘'"there is a universal mental block in meteorology". And there is more than one mental block: •

• • •

•

•

Although the climatological school is over 100 years old, we still hear from Meteo France and from the media about that ‘star player’, the Azores anticyclone, brought out to explain away both good and bad weather. The Norwegian school, now nearly 90 years old, is still incapable of explaining the weather in temperate and polar regions. The dynamical school, arising in 1939 and so in its 70s, is still trying to show (in vain) that the causes of disturbances reside at high altitudes. The tricellular model of general circulation, proposed by Ferrel in 1856, and therefore over 150 years old, is still the reference in the dynamical context, especially among the modellers. The basic principle of these models proposed in 1904, more than a century ago, lays all at the door of the elementary cell, but cannot describe the overall functioning of the general circulation of the atmosphere. Satellite images, irreplaceable, are still widely ignored after 50 years, since the theorists decry practicalities and especially direct observation of meteorological phenomena.

So there is a considerable hiatus between, on the one hand, the functional simplicity revealed by satellite imaging and, on the other, the multiplicity and complexity of definitions which had been put forward during that period of relative ‘blindness’ when phenomena were partially seen from ground level only. Definitions persist essentially without basic modification, even if they are not confirmed by direct observation and satellite images: a case in point is that of the existence in the temperate zone of all the supposed ‘fronts’, and their presumed undulations; or, in the tropical zone, of a very extensive supposed ‘catalogue’ of disturbances. Hypotheses have not always been put to the test of reality provided by satellites; their images, which have now been available for nearly 50 years, have not always contributed to a decisive resolution of controversies. In spite of considerable technological progress in both the observation and handling of data, meteorology finds itself blocked, crystallized within its immutable dogmas, some of them a century old; any attempt at enlightenment is hastily condemned as a heresy. The intrusion of modes constitutes an additional artificial complication. For example, the explanation of the Sahel drought has varied according to a series of pet theories: Charney’s mechanism (albedo); sea surface temperatures (SSTs); jets; and El Nino. Phenomena attributed to some presumed factor are soon reassigned to whatever new factor is a la mode', for example, the very media-friendly El Nino and La Nina

Introduction

xxxiii

have jostled for position as factors in the greenhouse effect and the ‘climatic warming’ which now reign supreme. Afraid of being left behind as fashions come and go, the media (and not only the media), amplify the rate of change. The El Nino phenomenon, even before it has been analyzed in terms of its local components and placed within the context of general circulation, is associated with the southern oscillation and is sometimes considered (without the slightest evidence) as "the key to the vault of global climatology, of which diagnostic climatology is the expression" (Pagney, 1994): therefore, it can explain all climatic disorders! This is also the case with the so-called ‘Walker cells’, which multiply and gyrate this way and that as the whim takes them, according to circumstances and commentators. And what more can we say about ‘climate change’ which, though undefined, is still responsible for everything? Phenomena have not escaped being ascribed to chaos, or to a mysterious "attractor’ making atmospheric movements chaotic and unpredictable, or even to the "effects of chance and opportunity prevailing during encounters between eddies" (Joly, 1995), triggering stormy weather at mid-latitudes!

The approach of this book

In this situation, where the discipline of climatology is divided among many schools (and their confusion of scales), weakened by inadequacies precisely where it should be at its most solid (i.e. when dealing with the physics of phenomena), attracted by fleeting trends and crippled by intellectual blindness leading to conceptual impasses, it is absolutely necessary to:

• •

•

•

•

question everything at all times, without fear or favour, especially credos, received ideas and ephemeral tendencies; lend no credence to ‘physical miracles’, which fly in-the face of all logic and are to be denounced when and where necessary; test everything using real facts, stressing direct observation, using satellite information and synoptic charts as a reference; thereby dealing with evidence that really exists, underlining the hiatus between concepts and reality, and bringing real judgement to bear on questions involving conflict of opinion; stand resolutely outside arbitrary compartments of the disciplines of climatology and meteorology, and shun the idea of barriers between climatic zones, in order to be able to appreciate the total coherence of phenomena on a planetary scale, respecting the notion of a global system of climate, properly integrating merid­ ional exchanges into the context of the general circulation of the atmosphere; to consider phenomena, and verify their mechanisms, across all scales of intensity and all spatial scales, with continuity between one zone and the next, and across all temporal scales, from the instantaneous through the seasonal to the palaeoclimatic, in the knowledge that the climatic system can function only in the geographical conditions determined, according to identical modalities..

xxxiv

Introduction

Since the same reality has often been defined using different names, a pre­ liminary clarification may be necessary (at the beginning of the chapters), according to the subject to be treated, to focus our analysis on the genetic unity of phenomena rather than on diversity of characteristics and definitions, in order to obviate what is no longer acceptable, and to rule out the possibility that the proposed concept might be an extra, unjustified insertion. Our approach takes firmly into account the fact that meteorology-climatology is incontrovertibly a geographical discipline, a science of Nature: geographical factors (normally ignored) existing in the lower layers of the troposphere do in fact have a fundamental thermal and dynamic role. The essential aim of this book is to examine, in the simplest possible way and concentrating only on essential phenomena, the workings of meridional exchanges, variations in the intensity of general circulation, and the production and spatial distribution of weather with special reference to the migration of pluviogenic structures. This basic knowledge (with which all real climatologists ought to be thoroughly familiar) about the real mechanisms of meteorological phenomena, and about the processes whereby climatic modifications are transmitted, is necessary for the analysis and understanding of climatic evolution, across all scales of intensity, space and time. To sum up, our subject is the clarification end explanation of the dynamics of weather and climate, past and present, in order to be able to delineate the probable scenario for the near future.

PART I GENERAL CIRCULATION IN THE TROPOSPHERE

Circulation in the atmosphere or, more precisely, in its lowest stratum - the troposphere - is caused by the Sun’s radiation, the unequal distribution of which across the surface of the Earth brings about meridional exchanges. At their origins are two foci of cold - the polar regions, feeding air towards a focus of warmth at the centre of which is found the meteorological equator, the axis of symmetry of general circulation. General circulation is driven by thermal deficit at the poles, causing mobile polar highs (MPHs), and assigning them a principal role: that of feeding cold air towards the tropics, which in its turn (and this is a fundamental process) causes warm air to move up to the poles. The distribution of land masses, their relief, and the Earth’s rotation control the trajectories of MPHs, and the formation of anticyclonic agglutinations (AAs) which are the result of the merging of MPHs in subtropical latitudes. The latter determine the entities of tropical circulation, feeding trade winds and their possible pro­ longations as monsoons. Tropicalised air then returns to the poles, redressing the thermal balance of the Earth/atmosphere system.

1 Radiation

Atmospheric circulation, meteorological phenomena, and energy transfers with their associated energy transformations, are the consequences of a limited number of initial factors:

• • •

solar radiation, the primary source of energy for the Earth/atmosphere system; the shape and movements of the Earth, which are responsible for the differential radiative balance according to latitude, and for seasonal variations in this balance; lastly, geographical factors, especially relief and the nature of the substratum, given that radiation processes make the Earth’s surface the principal source of energy.

The Sun is the source of the energy received by the atmosphere and the Earth. The energy supplied by the Sun is 5700 times greater than the flux which emanates from the interior of the Earth as a result of the radioactive disintegration of uranium and thorium, a much weaker flux given its value of the order of 0.06 W/m2. The Earth’s thermal output is therefore negligible compared with that entering the Earth’s atmos­ phere from the Sun, with a daily value of 342 W/m2 (mean 24-hour value at the top of the atmosphere). Initially thermal, the energy is transformed by cascade into kinetic and/or potential energy, degenerative transformations which tend to dissipate it as motion (circulation and convection), heat or sound. The energy is continuously absorbed, transformed, redistributed, dissipated and renewed. All meteorological phenomena, energy transfers and transformations, the circulation of the atmosphere, weather and disturbances, and climates are consequences of processes involving the utilisation of this initial solar energy.

1.1 PROCESSES OF RADIATION

Solar radiation propagates in straight lines at a velocity of 300000km/s. It lies essentially in the visible (VIS), emitted in short waves, and the near infrared (thermal

4

Radiation

[Ch. 1

Figure 1.1 Wavelengths of solar radiation, and its absorption. Short-wave absorption involves mostly ozone; the ‘optical window’ represents the transparency of the atmosphere in short waves for incoming flux. In the infrared, absorption is mostly by water vapour and carbon dioxide. The ‘atmospheric window’ represents the near-transparency of the atmos­ phere in long waves for outgoing flux at this wavelength.

IR); the wavelength of the Sun’s maximum emission is of the order of 0.5 m (1 m, or micron, is equal to 1/1 000000 m or 1/1 000 mm), almost all the radiant energy (99%) being emitted as short waves (Figure 1.1). The radiation balance is the difference between, on the one hand, that part of incident solar radiation absorbed by the atmosphere and its constituents and/or the Earth’s land or ocean surface (flux received), and on the other, the direct (reflection) and infrared (IR) radiation sent by the surface and the atmosphere into space (flux emitted). Satellite-borne radiometers can make direct measurements of these components of the radiation balance to an accuracy of about 0.3%, at the upper limit of the atmosphere. The mean value over one year, and per 24 hours of incident solar radiation is 342 W/m2, and that of the planetary counter-radiation escaping into space is 240 W/m2 (Figure 1.2). •

The short-wave solar flux is likely to be reflected (as a function of albedo), refracted (by diffraction or scattering), or absorbed and transformed into heat. The estimates shown in Figure 1.2 indicate that about 30% (102 W/m2) of the incident flux is reflected or scattered into space, mainly by the tops of clouds and aerosols, and by the ground. The rest (70% =240 W/m2) is absorbed; partly by the atmosphere (20%), which is almost ‘transparent’ (low absorption) to visible short-wave radiation, with the exception of stratospheric ozone for ultraviolet radiation; but mostly by the Earth’s land and ocean surface (50%) in almost equivalent proportions between scattered radiation (28%) from clouds and aerosols, and direct radiation (26%).

Sec. 1.1]

Processes of radiation

5

planetary counter-radiation

flux into space = 70 = 240 w/m2 6

+

38

+

26

Z\

atmospheric emission absorption H2O CO2

< emission c from clouds

}

convection

-6

Earth-atmosphere budget: 20+50=70= 240 w/m2 a - short-wave radiation

- 109 + 95 -30 energy lost by surface = -5 0

b - long-wave radiation TCR : terrestrial, CCR : celestial

Figure 1.2 Radiation balance of the Earth/atmosphere system. The value of the flux received (342 W/m2) is represented by the figure 100; the elements of the balance are expressed as percentages of this (after Ramanathan et al. (1989), Kandel and Fouquart (1992), Kiehl and Trenberth (1997) in IPCC: Scientific Basis).

•

The Earth’s surface receives two and a half times as much incident energy (short­ wave) as does the atmosphere, and is therefore (paradoxically) the source of energy. This is why the Earth’s surface is of prime importance in climatology, especially where phenomena of thermal origin are concerned, and in particular in the lower layers of the atmosphere: depressions or anticyclones, convection or subsidence of the air, and meridional exchanges. The surface re-emits energy upwards (terrestrial counter-radiation) as long waves. Because of the opacity of the atmosphere to IR radiation (long-wave), most of the radiation from the Earth’s surface is absorbed, with the exception of that within specific radiation ‘window’, for example the band between 8 to 12 pm, the so-called ‘atmospheric window’ allowing direct re-radiation into space. The trace gases which absorb and re-emit infrared radiation (LW) are found in the lower layers, which are the densest. Heat is therefore conserved in the lowest 5000 metres of the atmosphere, temperature decreasing with altitude until a mean temperature of —58°C is reached at the tropopause. The particular feature of these trace gases is their transparency to visible light (SW), but they are opaque to most infrared radiation (LW) re-emitted by the Earth. Each of these emissive gases has a specific absorptive capacity, and they are warmed and re-emit in all directions. Absorption is effected by water vapour (H2O), carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), halocarbons (CFCs) and ozone (O3), whilst the major components of the air, nitrogen (78.08%) and oxygen (20.95%) play almost no part in thermal exchanges.

6

•

•

•

Radiation

[Ch. 1

Energy may also be transferred from ground to atmosphere by non-radiative processes (convection) in the form of perceptible (7%) and latent (23%) heat fluxes. Perceptible heat flux accompanies thermal convection. Vaporisation heat, held in reserve, is released over time during changes of state (condensation of water and cloud formation). The contribution of latent heat (23%) to the heating of the atmosphere is of the same order of importance as the direct absorption of short-wave solar radiation (20%). The atmosphere radiates in its turn; this ‘celestial re-radiation’ is emitted either into space or towards the Earth’s surface, involving warming of air near the ground. The notion of ‘celestial re-radiation’, comprehensible if it goes into space, is however a matter of great debate (even if it convenient for ‘pedagogical’ purposes) when we consider re-emission towards the ground. In fact, this idea contradicts the second law of thermodynamics, which tells us that, for all kinds of heat transfer, the emitter must be warmer than the absorber. Now, since tem­ perature decreases with altitude, a return transfer from emissive gases towards the Earth’s surface is impossible. It follows from this that the hypothetical ‘greenhouse effect’, i.e.. the return of heat downwards, is not happening, and is only possible through reflection: it ‘violates the laws of physics’ (Thieme, 2003; Gerlich and Tscheuchner, 2007). This scenario is therefore but a mere figment of the mind, and is not confirmed by observations. However, horizontal advections are triggered to compensate for updrafts (meridional exchanges are not taken into account in Figure 1.2 and by the models). The global radiation budget at the top of the atmosphere is thus (theoretically) in a state of equilibrium between incoming energy, i.e. the mean incident solar flux of 342 W/m2, and outgoing energy, i.e. solar flux (short-wave) of 102 W/m2 reflected as a result of the planetary albedo, and planetary re­ radiation (IR, long-wave) averaging 240 W/m2.

Figure 1.2 sums up the notions commonly involved when discussing the ‘greenhouse effect’. However, this is not to say that this concept (on which ‘global warming’ is based) is definitely established on firm foundations, since all the numbers put forward are more or less contentious; the level of scientific comprehension is generally held to be “very low”. Few values are in fact based upon real, direct measurements (especially satellite measurements) and most of them are the results of calculations that are often approximate and differ between analyses. Processes, and also the respective causes behind them, are open to debate, mainly because these estimates of the radiation budget represent phenomena that take place in a theoretical space that is vertical and stable. This virtual space is held to be free of any shear or vertical discontinuity, all the way from ground level to the upper layers of the atmosphere; numerical models create the “hypothesis of a mixed atmosphere” (IPCC, 2007); the virtual space has no horizontal exchanges of advected energy, i.e. the schema is not representative of reality. There are also other representations of the ‘greenhouse effect’, in particular that of Lindzen (2007), who allots a major role in the lower layers to water vapour (absorbing the radiation from CO2 molecules),

The shape and motions of the Earth

Sec. 1.2]

7

whilst convection raises IR radiation into the upper troposphere to be emitted into the stratosphere and outer space.

1.2 THE SHAPE AND MOTIONS OF THE EARTH

The sphericity of the Earth, its rotation and its orbital motions all affect the radiation budget in space and time. •

•

This budget is a function of the angle of incidence of solar radiation, at a maximum when the angle of incidence is 90° (vertical or zenithal rays); the energy received decreases with the angle. Since the mean annual solar flux is most intense around the Equator and decreases towards the poles, the annual radiation budget is positive in the tropics within latitudes 35°N and 35°S, and negative from these latitudes towards the poles. The Earth’s orbital motion around the Sun is responsible for the variation in day length, and seasonal variations in angle of incidence. Figure 1.3A shows diagrammatically the positions of the Earth at the solstices. The winter solstice (December 22) occurs near perihelion (January 6), which is when the Earth is nearest the Sun at a distance of 147 million km. The northern winter is therefore comparatively mild. The summer solstice (June 21) occurs near aphelion (July 6) when the Earth-Sun distance is greatest at about 152 million km, meaning that the southern winter is that much more severe. The difference (5 million km) caused by the eccentricity of the Earth’s orbit entails a variation in flux received, from 352.5 W/m2 to 330 W/m2, a difference of 22.5 W/m2, representing 6.8% of the least value of the mean solar flux (in southern winter).

Near the Equator, daylight always lasts about 12 hours, the period varying as one approaches the poles until, within the polar circles, there may in a 24-hour period be continuous daylight or continuous darkness according to the season. At latitude 45°N (Table 1.1), mean monthly day lengths vary by 6 hours 48 minutes between winter and summer. The tropical zone, cosmically defined, with a spread of 46°54' between the latitudes of the two tropics at 23°27' N and S, is the only area receiving the solar flux at angles between 90° (from the zenith) and 43°06' (angle of incidence at one tropic when the Sun is zenithal at the other). Figure 1.3B shows the (apparent) movement of the zenithal Sun between the two tropics, where it comes to a halt at the solstice Table 1.1

Variation of day length, in hours and minutes, at latitude 45 °N.

J

FMAMJ

9 18

10 24

11 48

13 36

14 54

15 30

J

ASOND

15 12

14 00

12 30

10 54

9 30

8 42

8

Radiation

[Ch. 1

equinox

B.

Tropic of Cancer

23 September—^-21 March

Tropic of Capricorn

Figure 1.3 A: Relative positions of the Earth and Sun at the solstices. B: Apparent movement of the zenithal Sun in the tropical zone.

(sol stat) and then returns; note that the figure also shows the twice-yearly passage of the zenithal Sun across the equator at the time of the equinoxes. With small variations in day length, and the greater altitude of the Sun above the horizon, the tropical zone is a source of heat experiencing only minor seasonal variations. Outside the tropical zone the Sun is never overhead, and there are large variations in the angle of incidence (from nearly 90° to 0°), with variations in day length increasing towards the poles, reaching 24 hours within the polar circles and, at 90°N and S, five months without direct sunlight. Variations in temperature are therefore great, and the notion of a ‘temperate climate’ in mid-latitudes is more often than not an artifice derived from thermal means. Polar regions, with their shallow angles of incidence, experience unfavourable amounts of radiation, with low temperatures; they therefore constitute two sources of cold, with very marked seasonal variations.

1.3 GREENHOUSE EFFECT, WATER EFFECT The notion of the ‘greenhouse effect’, now principally associated with carbon dioxide, has actually quite a long history, dating back to Fourier (1824), Tyndall (1861) and Arrhenius (1896). For a long time, the idea was sidelined owing to inherent physical contradictions, especially those involving the pre-eminence of

Sec. 1.3]

Greenhouse effect, water effect

9

water vapour and its partially overlapping absorption bands, originating in the molecular structure of gases, water vapour and carbon dioxide (Figure 1.1). It is easy to understand why scientists waited for more than a century to engage with this question, and it can also be asked why they embraced it with such assurance during the 1980s, not long after the hypothesis had been rehabilitated. This was probably because it lends itself to the reductionist method, which consigns phenomena to the level of the elementary cell in climate models. The ‘greenhouse effect’ or ‘hothouse effect’ was originally so called because the atmosphere was thought to act like glass of a garden greenhouse. The glass allows through the ‘luminous heat’ (solar radiation), but the ‘dark heat’ (infrared) cannot pass out through the glass and it builds up, raising the temperature within the greenhouse. However, as Wood pointed out in 1909, the atmosphere and the greenhouse are not directly comparable, as a glazed-in volume retains heat because of the absence of convection and advection, rather than through effects of absorption through glass, and re-emission. But the term ‘greenhouse effect’, accepted through usage, is now general, to the detriment of the more appropriate term ‘atmospheric effect’. Trace gases, known as greenhouse or emissive gases, absorb infrared radiation emitted by the surface of the Earth, by the atmosphere itself, and by clouds. They re-emit it into the surrounding gases. Therefore, these gases trap heat within the surface-troposphere system, and re-transmit it: the natural ‘greenhouse effect’ (cf. IPCC, 2001, Glossary). These emissive gases, which are essentially natural in origin, are:

•

•

•

•

•

•

water vapour (H2O). This is the principal greenhouse gas. Its distribution is extremely uneven, both geographically and as a function of altitude, and its lifetime in the atmosphere is brief. Through the mechanisms of evaporation and precipitation, the renewal of the global water potential takes less than two weeks; carbon dioxide (CO2). This gas originates from both natural and anthropic sources. Emissions due to human activities (combustion of fossil fuels, deforest­ ation and agricultural practices) represent only one-twentieth of the natural additions of carbon to the atmosphere; methane (CH4). Methane originates in the decomposition of organic matter, the combustion of fossil fuels and biomass, and fermentation (animals, rice-paddies, household waste tips); nitrogen protoxide (N2O). This is both natural and anthropic in origin, and is a by-product of nitrogen-based fertilisers, stock breeding, and industry; ozone (O3). This is formed by photochemical recomposition, mainly of pollutants found in urban atmospheres. It is a screen against both incident ultraviolet radiation and infrared radiation from the surface; CFCs (chlorofluorocarbons, or halocarbons). These are solely of human origin.

Aerosols, either of natural origin or associated with human activity, are responsible for a negative forcing, lessening the increase in temperature. Quantities vary constantly and regionally, and they react rapidly. Their lifetimes are relatively short. The amplitude of the forcing due to aerosols is very uncertain.

10

Radiation

[Ch. 1

Clouds affect the radiation budget in the same way as greenhouse gases, but their effect is complicated by the fact that they also reflect incident solar radiation, while their formation releases latent heat. Indeed, cloud may either reflect short-wave solar radiation from their upper surfaces, or absorb it; they also absorb and re-emit long­ wave radiation from the Earth’s surface. In the former case, they cool the surface by intercepting the solar flux, and in the latter they warm the lower layers. The relative contribution of these effects depends on the height, type, density and optical and radiative properties of the clouds. The term ‘atmosphere’ comes from the Greek atmos, signifying ‘vapour’. This etymology recalls the unique role of water vapour in the climatic system, linking surface and atmosphere in a water cycle. Water vapour has a triple action: as an absorbent gas (short waves and long waves), a major emissive gas, and a vector of latent heat through meridional transfers and updrafts. Moreover, its distribution is unequal at different altitudes: “nearly half the total water in the air is between sea level and about 1.5 km above sea level. Less than 5-6% of the water is above 5 km, and less than 1% is in the stratosphere, nominally above 12 km” (AGU, Special Report, 1995). Its concentration varies by several orders of magnitude in just a few kilometres. We should now re-estimate the respective importance of the different greenhouse gases contributing to the atmospheric effect.

•

•

As a preliminary estimate, leaving aside water vapour (as the IPCC regularly does), carbon dioxide is far and away the principal emissive gas (72.37%), the other gases contributing about 28% (Figure 1.4). The enormous ‘dominance’ of CO2 is obviously much diminished when the contribution of water vapour is taken into account. Now, the proportions are as shown (Figure 1.5):

water vapour (WV) carbon dioxide (CO2) methane (CH4) nitrous oxide (N2O) CFCs and other gases

95.00% 3.62% 00.36% 00.95% 00.07%

vs vs vs vs

72.37% 07.10% 19.00% 01.43%

So water vapour represents 95% of the greenhouse effect, and the possible influence of the other gases is down to 5%! If we consider the real atmosphere, carbon dioxide (from both natural and anthropic sources, and the gas upon which the ‘warming scenarios’ are based) represents no more than 3.62% of the greenhouse effect (i.e. 26 times less than water vapour!). The so-called ‘greenhouse effect’ is fundamentally a ‘water effect’, with other emissive gases possibly playing a part. This is a point which cannot be overemphasised: the equilibrium of the radiation budget depends principally on direct convective heat transfers from the surface towards higher levels, above all through the intermediary of the water cycle, which transfers the latent energy stored during evaporation and liberated in updrafts.

Greenhouse effect, water effect

Sec. 1.3]

11

Figure 1.4 Contribution to the ‘greenhouse effect’. Methane, NOx and various gases are adjusted for heat-retention characteristics relative to CO2 (natural and man-made sources, water vapour not included) (from Hieb, 2004).

100%

80%

'

60%

'

40%

20% 1

0%

1

• 3. SI 83;

Methane

Water Vapor CO2

Figure 1.5

0 3m G.g5QX

0.0723;

■

Wise, gases

N20

Contribution to the ‘greenhouse effect’ (natural and man-made sources, water

These reflections on the major role of water vapour should oblige us to reconsider the many aspects of the water cycle, especially where water vapour is concerned. What we have to do is to analyse most thoroughly the evolution of its concentration and spatial distribution, the significance of its transfers (particularly within the lower layers which concentrate those exchanges), its vertical divisions as a function of aerological stratification, the causes of and dynamical processes involved in its evolution, and its actual participation in any ‘warming’.

12

Radiation

[Ch. 1

1.4 THE GEOGRAPHICAL FACTOR The radiation budget is also influenced by geographical conditions, the Earth’s surface being the level of maximum short-wave absorption, and therefore the prin­ cipal heat source (in re-emitted long-wave radiation) for the lower atmospheric layers. Figure 1.6 shows diagrammatically the interactions between the ground and the lower atmosphere above surfaces of different types; for convenience, short-wave (solar radiation by day) and long-wave (terrestrial and celestial counter-radiation by day and by night) are shown separately. Where water is absent (and there is never a total absence), for example in the case of a continent over which the air is dry (Figure 1.6A), where the atmospheric effect is much reduced, continental-type thermal activity is characterised by strong warming of lower atmospheric layers during the day, encouraging thermal convection, low night-time temperatures due to considerable nocturnal radiation, and high thermal amplitude on both daily and annual scales. Inertia relating to the cosmic factor is comparatively low, and maximum monthly temperature is usually reached one month after the summer solstice. Absence of thermal inertia is observed (infre­ quently) when continental conditions are emphasised, as is the case in the mid-Sahara where the hottest month is June, with the Sun at the zenith. This type of continental activity is also encountered over cold, dry, glacial surfaces (inlandsis), and where the air is clean, away from built-up areas, and especially over mountainous areas. Over water (oceans, lakes and marshes), where the air is humid (Figure 1.6B), different processes of interception, especially absorption by water vapour, attenuate the incoming short-wave energy flux. Some is also consumed by evaporation, after reaching the ground, and the energy is taken up by water vapour. Long-wave

A- DRY AIR : continent mountain pure air

B- MOIST AIR : ocean

dense forest polluted/dust-laden air sr.tcr

TCR + CCR A

SC dif i i

SC I

dif evap i i I i i abs

I D :

T/

N : T

large thermal amplitude very limited greenhouse effect

I I I I

abs

0; T

dif evap

CCR

[TlilLUU^ N : T

small thermal amplitude intense greenhouse effect

SR : solar radiation (SW) TCR : terrestrial counter-radiation CCR : celestial counter-radiation LW abs : absorption ref: reflection SC : scattering evap : evaporation conv: convection, turbulence

Figure 1.6

Radiation budget and thermal activity. A: continental type; B: oceanic type.

Sec. 1.5]

Conclusion

13

emission is in its turn intercepted and reflected towards the ground. The temperature cannot therefore rise sharply, and any thermal convection is considerably reduced, if not prevented altogether, especially over oceans. As a consequence, thermal behaviour of the oceanic type is characterised by a moderate rise in daytime temperature, a small diminution in night-time temperature, and reduced thermal amplitude, on both daily and annual scales. Inertia as compared with the cosmic factor is comparatively high, maximum monthly temperature being reached (as a function of the amount of humidity) two or three months after the summer solstice. This oceanic type of thermal activity is also seen where forests are involved, especially the dense evergreen expanses of the Congo and Amazonia, where the activity is hyperoceanic, with practically no daily or annual amplitude. This type is also observed to a lesser extent when atmospheric dust concentrations are high, for example with the continental trade winds of Africa causing dust storms, or in dirty, polluted atmospheres over urban agglomerations where the atmospheric effect is increased. At greater altitudes, the thermal gradient is dictated by water (and by decreasing atmospheric pressure). When the atmosphere is dry, this decrease can attain 10°C/km (dry adiabatic), whilst in a moist atmosphere the value goes down to 5°C/km (moist or saturated adiabatic). Within an ascending volume of air there is cooling (lower pressure) and condensation, and therefore liberation of latent heat, which attenuates the cooling and encourages upward movement, without which the vertical develop­ ment of cloud formations could not proceed.

1.5 CONCLUSION To sum up, the nature of the radiation budget means that the Earth’s surface and the lower atmospheric layers are the fundamental source of energy for both vertical and horizontal atmospheric movements.

•

•

The Earth’s surface receives twice as much short-wave radiation from the Sun than is retained in passage through the atmosphere. It is therefore the prime agent for vertical atmospheric movements of thermal origin, either convective or downward and, on a completely different scale in space and time, may cause thermal low-pressure areas or maintain thermal high-pressure areas. This capacity is, however, a function of the thermal behaviour of the substratum, and is therefore marked over land masses, and insignificant or nonexistent over the surfaces of oceans. The Earth’s surface warms the lower layers of the atmosphere, by means of vertical transfer (convection) and long-wave radiation (terrestrial and celestial counter-radiation), thanks to absorption by greenhouse gases. These gases, and especially water, exist entirely within the lower layers, so the importance of these layers is considerable in accumulating and distributing perceptible and latent energy. The so-called ‘greenhouse effect’ is fundamentally a water effect.

14

Radiation

•

The unequal distribution of solar energy causes lateral movements; the existence of two sources of cold causes the atmosphere to be symmetrically divided between two meteorological hemispheres. Within each of these, circulation moves towards the meteorological equator, which lies in the middle of the central focus of warmth, and is then redirected out towards the sources of cold. At high latitudes the temperature of low-level air is observed to fall, as it receives less solar energy, and the surface supplies little while the air continues to dissipate energy. So, cooling, it becomes denser, and migrates towards tropical latitudes (in compact masses: MPHs) while its departure in the lower layers is compensated for by the arrival of warm air originating at lower latitudes. Seasonal variation, small within the reservoir of warmth but great within the cold sources, brings about cosmic modification in the characteristics of each meteorological hemisphere, and a migration in latitude of the meteorological equator. General circulation in the troposphere, and associated meteorological phenomena, result from the interac­ tion of these cosmic, radiative and geographical factors.

[Ch. 1

These meridional exchanges, which transfer colossal quantities of energy and in their turn bring about aerological stratification and disturbances (leading to updrafts and vertical exchanges of heat), are not correctly accounted for in the establishment of the radiation budget (Figure 1.2) and in the processes determining surface temperatures. So, radiation and its possible variations are constantly brought to the fore by the modellers, and treated at the level of the elementary cell. However, solar radiation can ‘explain’ only the mean conditions controlling the observed temperature, and its seasonal evolution and its variation across vast expanses such as climatic zones (cf. the heat source and the sources of cold). But on the local scale (exactly that of the models’ cells) of the weather station itself, the evolution of temperature depends above all on advections of air of varying origins and characters, on the intensity of radiation and the chorological conditions (relative to that location), jointly modifying (more or less rapidly) the advected thermal character­ istics. A temperature value, especially if it is a mean value, is not therefore in itself representative of the solar radiation reaching the ground. Radiation and temperature are not therefore necessarily related. For example, an equatorial forest receives large amounts of radiation, its interception by the vegetation and by water limits thermal variations, to the point where the seasonal rhythm vanishes. In temperate zones, where cold and warm air masses alternate, mean temperature masks the thermal extremes both daily and seasonally. The influence of radiation, and the radiation budget, like the evaluation of surface temperature, can be estimated only by taking into account the dynamic of meridional exchanges, in the context of general circulation, which is not integrated in any realistic way into the numerical models.

Circulation in the lower layers of the troposphere The troposphere is, by definition, the most turbulent part of the atmosphere. Within it, the lower atmospheric layers - especially the planetary boundary

Sec. 1.5]

Conclusion

15

layer which roughly corresponds to the first 1500 metres - exhibit the most complex and perturbed circulation. These lower layers deserve our special attention, for many reasons:

• •

•

•

they are the densest, and half the atmosphere is contained in the first 5500 metres (above which level, pressure falls below 500 hPa, i.e. half that at the surface); they contain nearly all the water vapour, which is involved in rainfall and supplies energy to disturbances, and the emissive gases (which include water vapour), the atmospheric effect being imperceptible above 5000 metres; paradoxically, the principal source of heat is not the Sun but the Earth’s surface, which warms the atmosphere. The interaction between surface and air is therefore of great importance, arising from differences in the substratum, thermal behaviour over continents and oceans, vertical upward or downward movements caused by contact with the surface (compression, anticyclonic stability, or turbulence and convection), and thermal gradients resulting from differential warming (transforming thermal energy into mechanical energy), the formation of thermal anticyclones above deficient regions, especially polar (MPHs), and/or the attraction exerted by deep thermal lows (especially in the tropical zone) with associated vast horizontal circulations; among geographical factors, and in conjunction with the distribution of oceans and continents, relief acts upon surface temperature through altitude, and because of its topography is a powerful aerological factor, especially in determining the paths of a great number of meridional exchanges.

All these surface-related factors contribute to the delimitation of vast spaces where air circulates in the lower layers of the troposphere. These units of the lower layers are topped by an essentially zonal circulation, whose simplicity increases as the influence of surface conditions wanes with altitude, and quantities of air involved decrease. It is therefore necessary to pay great attention to the lower layers, which are not only the most complex but also the most important where meridional exchanges are concerned. There is no hiatus in tropospheric circulation. It occurs without interruption in each meteorological hemisphere, but with variable modalities, from the poles to the tropical zone and back, in the lower layers as at altitude (with meridional transfer always more important in the lower layers). The division of the following sections in this chapter is therefore merely for convenience: 1

2 3

In the lower layers, polar and temperate circulations are controlled by Mobile Polar Highs (MPHs) issuing from the poles, carrying with them cold air but also bringing a compensatory counterflow of warm air towards high latitudes. MPHs form anticyclonic agglutinations (AAs), virtual buffer zones between temperate and tropical circulation; Air relayed by AAs maintains tropical circulation and the flow of trade winds, and their possible extensions into the transequatorial monsoon circula­ tions, moving towards the meteorological equator (ME) at the heart of the tropical zone.

2 Circulation in high and mid-latitudes: MPHs

Circulation in polar and temperate regions is controlled by Mobile Polar Highs (MPHs). Here, we examine only their contribution to circulation; the associated weather is analysed later (Part II, Chapter 7).

2.1 PERCEPTION OF CIRCULATION IN HIGH AND MID-LATITUDES

The climatological school sees circulation in polar and temperate zones through analysis of mean values: polar high pressure controls (theoretically) easterly polar winds, whilst subtropical high pressure controls temperate westerlies (‘Brave West Winds’). Between these highs, so-called subpolar lows form a long zonal corridor of depressions in the southern hemisphere, and in the northern hemisphere these are concentrated into two closed depressions known as the Aleutian and the Icelandic. These statistically defined centres of action are thought of as ‘permanent’, even ‘fixed’, or sometimes ‘semi-permanent’, and the circulation they control is also assumed to be just as permanent. Any changes (for, in reality, everything is mobile) in the ever-moving air masses are thus attributed to the alternation (or change in position) of strangely ‘fixed’ multiple centres of action: thus, in the North Atlantic, we see individually named anticyclones: the North Atlantic, Fenno-Scandian, East Atlantic (polar) and Azores (tropical), but also depressions such as the Icelandic, North Sea, Atlantic and Azores (Pedelaborde, 1982). The question of the origin of the mean values, through which definitions have been established, is never considered. The real phenomena concealed within these means have not been clearly identified. There now seems to be a distinction between a permanent and a perturbed field of wind and pressure, brought about by confusing (fixed) statistical visions with (mobile) synoptic reality. The origin of these centres of action is uncertain. It was attributed to thermal factors in the case of polar high pressure areas (PHPs), and a similar proposition was then mooted for the origin of subpolar lows, even though their maximum deepening

18

Circulation in high and mid-latitudes: MPHs

[Ch. 2

is seen in winter (which, as with the latitude involved, naturally refutes an origin of this type). A dynamical origin is also proposed: undulations in high-altitude jets, causing alternations in convergence and divergence in the upper layers, are said to produce subsidences and updrafts, creating even in the lowest layers ‘depressions and anticyclones in mid-latitudes, with a lifetime of the order of one week’ (Sadourny, 1994). The principle according to which altitude controls the pressure fields at low levels is unanimously accepted, and the usual presentation is that of a wave (at about 300 hPa) above a surface disturbance (depression) - and not a disturbance in the lower layers - delineated according to the Norwegian theory, with two fronts. But the space between the two levels is not complete and the frontal surfaces, though inseparable from a front, are not even represented at altitude. The ‘theory’ proposed by Meteo France (Joly, 1995, Figure 4), exactly illustrates this schematic vision. A high-altitude eddy (an anomaly of the jet current) is thought to induce the deepening of a surface depression, “the eddy and the depression being mutually reinforced”, and leading to “the most sudden and intense of storms”. This theo­ retical vision has never been demonstrated, because the intermediate space between the surface and the atmosphere at altitude is not (even partially) ‘filled’. The controlling role of altitude suggested by the dynamical school is still merely at the hypothetical stage (cf. Chapter 5). The downward influence of the upper layers in the creation of an anticyclone is also far from being proved. This has been proposed in order to explain the formation of ‘subtropical’ anticyclonic cells, but this is erroneous because subsident movements cannot reach the surface (cf. Chapter 3, Anticyclonic Agglutinations). The extension of this alleged ‘explanation’ to include the temperate zone is therefore wrong. It must be remembered that higher layers are always at very low pressures compared with lower ones; pressure at 9000 metres, of the order of 300 hPa, is lower (by more than 700 hPa) than surface pressure. In order to validate such a hypothesis, an extraordinary ‘physical miracle’ would be needed for levels of rarefied atmosphere to be able to engender low-level anticyclones with high pressures and low, even strongly negative, temperatures; movements of subsidence can create only warm anticyclones, of low energy, because of the low density of warm air. The air in this anticyclone would therefore be unable to cause the formation of a (relatively or absolutely) cold anticyclone, like an MPH. Yet another great ‘physical miracle’ would have to occur for the so-called high-altitude wave to persuade a lower-level anticyclone to follow it as it moves eastwards! Moreover, some anticyclones adopt an obviously meridional trajectory, and the wave certainly cannot explain that. Vertical velocities would also have to be considerable (to compensate for density differences), but nevertheless “one of the characteristic principles of large-scale atmospheric movements lies in the fact that their vertical velocities are extremely low, much lower in fact than their horizontal velocities, of the order of only centimetres per second” (Rochas and Javelle, 1993). High vertical velocities down­ wards are not observed above mobile or ‘fixed’ anticyclones in the lower layers, which are also capped by a marked discontinuity which shows that the supposed subsidence does not reach the surface. This presumed subsidence would in fact tend

Sec. 2.1]

Perception of circulation in high and mid-latitudes

19

to dissipate the often dense cloud formations which are frequently found above lower-layer anticyclones. This ‘explanation’ would also imply that lower level mobile anticyclones would be able to form only in mid-latitudes (i.e. on the tropical edge of the wave), in flagrant contradiction of reality, which sees them forming at high latitudes. Another of the dynamicists’ formulations compares the passage of the wave to “the crest of a swell moving through the sea without the water itself being displaced”, the wave thereby causing variations in pressure “without the displacement of air” (Laboratoire de Meteorologie Dynamique, LMD, personal communication). Such a mental construct cannot however explain meridional exchanges, temperature differ­ ences or, more especially, the origin of cold air, or of warm air to furnish the energy for disturbances; neither can it give the reason for the general west-east motion of an anticyclone-depression pairing and their associated weather and violent winds. This scenario would also involve another of our ‘physical miracles’ - the fact that accelerations in flow would have to happen with no increase in the volume of air displaced, a theoretical vision which would have us revise atmospheric physics ... but with no reference to observed reality! The exaggerated role ascribed to the upper layers appears to be based on a misapprehension: originally proposed to explain depressions, the explanation therefore took no account of the presence of the anticyclone which promotes the deepening of these depressions. In fact, anticyclone (cause) and depression (effect) are closely interdependent, and the dynamicists’ explanation involving only the depression is simply unacceptable in the case of the anticyclone, whose existence either is, or must be, deliberately ignored in order to validate the initial hypothesis. There are still other contradictions of physical principles: for example, in the proposition according to which “polar air (is) directed by the Icelandic depression zone” (Triplet and Roche, 1988) - but how can a depression, or cyclone, control circulating cold air, which is dense and by its very nature anticyclonic, and would quickly overcome it? Other unanswered questions include that of the origin of seasonal variations in the energy of centres of action, and in the intensity of meridional exchanges. It is claimed that there is a relationship between high-altitude jets and meridional circulation; in summer the jet is slow and performs considerable undulations (low index), but in winter it is fast, with slight undulations (high index). What then of the reality of the seasons, if we believe that strong undulations of the high-altitude westerly jet are vital for the existence of meridional exchanges, especially in the case of ‘cold flows’? Summer would then be the season of most active meridional exchanges! Also, what is the physical relationship, both synop­ tically and seasonally, between anticyclones and mobile cyclones, and (on the statistical scale of mean values) between subtropical anticyclones and subpolar lows? This relationship is expressed by an oscillation (in the North Atlantic: the NAO, or in the North Pacific, the NPO) represented by a difference in pressure between an anticyclone and a low, but this oscillation has still not been explained. The functional simplicity of climatic mechanisms is still masked by the persistence of concepts which remain unchallenged, or are sometimes simply juxta­ posed (see Introduction). Thus, for example, Sadourny (1994, p. 39): “... movements

20

Circulation in high and mid-latitudes: MPHs

[Ch. 2

produced by baroclinal instability take the form of waves and vortex-like structures. This is the reason for the development of depressions and anticyclones in mid­ latitudes ...”. Here we see the ‘wave’ concept ascribed to Rossby and Weightman (1939). However, later on the same page we read: “a classic situation of winter blockage is that involving a stable anticyclone anchoring itself in the area of Scotland; this results in a wave of cold across Western Europe, with a flux of polar air coming down from Scandinavia”. Whence comes this stable anticyclone (which is manifestly not stable, but moves), causing a cold spell; and if such anticyclones exist, why do they intervene so opportunely only in such a circumstance? What are they doing at other times? Why has this anticyclone anchored itself at this precise location? Is it really anchored at all? This anticyclone in the lower layers (is it really from Scandinavia?) cannot be explained by the theory previously mentioned, which proposes moreover a completely different explanation for blockage. What we have here is an MPH on the Scandinavian trajectory, which appears in the figure on the following page (Sadourny, 1994, p. 40). The juxtaposition of these two concepts, lacking analysis and any attempt at synthesis, is incoherent and needlessly com­ plicates the perception of weather. The many ambiguities which live on in perceptions of polar and temperate circulation, coupled with the recourse to various ‘physical miracles’ in order to get round inconsistencies, ought to be strongly challenged from a basis of direct observation.

2.2 THE EXISTENCE OF MOBILE ANTICYCLONES Under the influence of the Bergen school, which was well placed geographically to have an exaggerated view of their importance, low-pressure areas or cyclones were represented as the key factor in weather dynamics, doubtless because of the bad weather associated with them. Depressions take up markedly less space than highpressure areas. However, the clement weather which usually accompanies anti­ cyclones seems to make their passage (almost) unnoticed. Whether they are called post-cyclonic, post-frontal, or cold or polar discharges, mobile anticyclones have long been observed, in synoptic terms, on the surface and in the lower layers. Although they “evolve behind the cold low-pressure fronts of the polar front”, they are apparently independent of these fronts. Their presence in the southern hemisphere was noticed, and was at the time considered to be a specifically southern pheno­ menon: Duverge (1949) stressed “the unending circulation of anticyclones” around southern Africa and Madagascar; Taljaard (1967) and Zhdanov (1967) observed the succession of cyclones and anticyclones around Antarctica; and Gentilli (1971) noted the passage over Australia of “travelling anticyclones”. In the northern hemisphere, Pettersen (1956) and Klein (1957) analysed tracks of cyclones and anticyclones. Paul (1973) observed shallow thermal anticyclones over Canada, underlining their “extreme mobility”. But he did not ask the question: firstly, where do these anti­ cyclones really come from, and secondly, what becomes of them after they have moved out of the zone of study? In the case of North America, it was considered that

Sec. 2.2]

The existence of mobile anticyclones

21

“anticyclones tend to disappear along eastern coasts” (Zishka and Smith, 1980), and that “the development of superficial fronts” along the eastern seaboard was linked to a “sea-land interaction” (Bane et al., 1990; Nielsen and Neilly, 1990). The same concept has been applied to the north-west Pacific, where “the configuration of land and sea, the topography of continental land masses, and the positions of warm and cold currents” are deemed to be responsible for the distribution of depressions (Yarnal and Henderson, 1989). A study by Klein (1957) clearly revealed the existence of mobile anticyclones in the northern hemisphere. Twenty years of analyses of synoptic charts (US Weather Bureau, Historical Maps series) established, month by month, the frequencies and places of origin of anticyclones and depressions, as well as their principal trajectories. Figure 2.1 presents, by way of example, the author’s analysis for the month of October of the number of anticyclones, and Figure 2.2 the number of depressions, and also the displacement of these entities. This remarkable analysis, which revealed

Figure 2.1 The number of anticyclones per sector (5° of latitude and 5° of longitude), over 20 Octobers. Iso-frequencies are separated by tens, and zero-frequency regions are in grey. Principal trajectories of anticyclones follow the axes of maximum frequency (Klein, 1957).

22

Circulation in high and mid-latitudes: MPHs

[Ch. 2

Figure 2.2 The number of depressions par sector (5° of latitude and 5° of longitude), over 20 October. Iso-frequencies are separated by tens, and zero-frequency regions are in grey. Principal trajectories of depressions follow the axes of maximum frequency (Klein, 1957).

the great mobility of anticyclones and depressions without, however, underlining the relationships between these isobaric individuals, was not followed up in spite of its relevance: then, it was a la mode to examine synoptic surface charts! Moreover, satellite images would not be on the scene for some years ... Indeed, from the 1960s onwards, weather satellites should have renewed our perception of meteorological phenomena, now visible in their totality as viewed from above, and no longer only from the ground. 2.3 MOBILE POLAR ANTICYCLONES (MOBILE POLAR HIGHS OR MPHs)

Now, almost 50 years after the first satellite images, anticyclones are still not clearly individualised and recognised as primary players on the meteorological stage. NOAA satellite images (Photo 1A-D) show particularly well the organisation of cloud formations resulting from the penetration into the North Atlantic of flows

Sec. 2.3]

Mobile polar anticyclones (Mobile Polar Highs (MPHs))

A - 31 January 2008

B - 1 February 2008

C - 2 February 2008

D - 3 February 2008

23

Photo 1A-D Evolution from 31 January to 3 February 2008 over the North Atlantic and Europe. NOAA satellite images, visible, 12 h UTC (Satmos).

of cold air (MPHs) originating at high latitudes on both sides of Greenland on 31 January and 3 February 2008. The Norwegian school holds that only depressions are worthy of our interest, and mobile anticyclones are therefore generally ignored. The chart (Figure 2.3) from the Deutscher Wetterdienst (Berlin) is worth mentioning: it does not merely indicate the ‘warm’ and cold fronts, clearly separated and rapidly interrupted (strictly

24

Circulation in high and mid-latitudes: MPHs

[Ch. 2

A - 1 February 2008

B - 2 February 2008

C - 3 February 2008

D - 4 February 2008

Figure 2.3 Synoptic evolution from 1 January to 4 February 2008. Surface, 00 h UTC (Deutscher Wetterdienst).

according to the method, cf. Chapter 5), with depressions specifically named, but and this is new - certain anticyclones are shown. Also, the trace of the front partially ‘encircles’ the mobile high which is organising it, thereby marking it out. This represents undoubted progress towards coherence, but imperfections nevertheless remain - for example, the distinction between the cold front and the so-called ‘warm’ front (cf. Chapter 5), or the failure to take relief into account (especially the

Sec. 2.3]

Mobile polar anticyclones (Mobile Polar Highs (MPHs))

25

influence of the Cantabrian-Pyrenees chain and the Alps, which marshal the flow of advected cold air and determine the boundaries of the MPHs). The satellite images (Photo 1A-D) and synoptic charts (Figure 2.3) let us see the dynamics of MPHs (Figure 2.4) from 1 January (12 h UTC) to 4 February 2008 (00 h UTC):

•

•

•

•

On 31 January, an advection of cold air MPH Hl, descends to the east of Greenland into the Norwegian Sea, accompanied by intense cyclonic circulation (low-pressure area RESI, Figure 2.3A) across northern Britain (Photo 1A). On 1 February (Photo IB), the southern edge of this MPH (Hl) reaches northern Spain, while the associated low (RESI) causes a violent storm over Britain and the North Sea. MPH Hl is reinforced by newly-arrived cold air (MPH H2) over the Norwegian Sea (Figure 2.4A). On 2 February (Photo 1C), MPH H1+H2 (now labelled CHRISTFRIED, Figure 2.3B) invades Western Europe and flows southwards towards the eastern Atlantic and Mediterranean France. Low RESI has returned north, over Scandinavia. On the same day MPH H3, coming from Canada (Figure 2.4B), and passing west of Greenland (Photo IB), thrusts vigorously into the North Atlantic, causing a depression (STEFFI) to form (Figure 2.4B). On 3 February, MPH Hl + H2 (CHRISTFRIED) is crossing central Europe to the north of the Alps, drawing Low RESI eastwards, and flowing across the western Mediterranean between the Pyrenees and the Alps (Figure 2.4A). MPH H3, reinforced by a new southerly advection, MPH H4, moves rapidly south­ wards across the Atlantic, its leading edge the site of a vigorous cyclonic flow from the Iberian peninsula to the North Sea, accompanied by dense cloud formations; Low STEFFI is now centred over the British Isles.

Figure 2.4 Dynamic of MPHs: A MPH Hl and H2; B - MPH H3 and H4, from 1 February to 4 February 2008 (H: MPHs 1, 2 etc. in order of appearance; bracketed numbers are dates of observations).

26

Circulation in high and mid-latitudes: MPHs

[Ch. 2

Photos 2A and 2B An ‘exemplary’ MPH: 28 April 1986, over the North Atlantic, 12H UT, Meteosat. A: Visible light; B: IR (CMS-Lannion). An MPH on the American trajectory has passed into the Atlantic south of Greenland, and is very well delineated by cloud formations (moist surrounding air); the IR image shows that the lifting of the air around the perimeter of the MPH increases to the north, the IR revealing the vertical development of the clouds, with grey clouds massing in the lower layers, and cold-topped white clouds rising in the upper layers. A Scandinavian-trajectory MPH covers central Europe and, while moving towards Russia, flows from its southern edge between the Balkans and the high ground in Turkey, towards the eastern Mediterranean and northern Africa. It was on the southern and then western edges of this MPH, following its anticyclonic rotation, that radioactive pollution from the Chernobyl (Ukraine) nuclear plant spread over Western and Northern Europe. Between these two potent MPHs, over Western Europe, another MPH has ceased its eastward movement to the north of the Alps, and is flowing out towards the western Mediterranean between the Pyrenees and the Alps, and to the north of the Cantabrian Mountains towards the Atlantic.

Sec. 2.4]

•

The polar thermal deficit

27

On 4 February, MPH Hl +H2 (CHRISTFRIED) has reached the Black Sea, while the flow around its trailing edge towards the Mediterranean is progress­ ively weakening. MPH H3 + H4 is now far south, out across the Atlantic, heading for the tropics, while its leading edge has reached the continent of Europe; Low STEFFI is now between Iceland and Scotland.

The satellite evidence, backed up by synoptic charts, is irrefutable: anticyclonic air masses from the polar regions are certainly on the move! Nobody can now deny the existence of Mobile Polar Highs (MPHs), responsible for disturbances and the driving force behind general circulation. Mobile Polar Highs, or MPHs (Leroux, 1983, 1986, 1993), are huge discs of dense air (Photos 2A and B) which are mainly responsible in high and mid-latitudes for variations in pressure, the speed and direction of winds, temperature, humidity, cloudiness and rainfall: and it follows therefore that they control the perpetual variations of the weather, and climatic variability, on all timescales.

2.4 THE POLAR THERMAL DEFICIT

MPHs result from the basal cooling and subsidence of air over the polar regions, since the radiation budget at the surface is always negative. The low angle of incidence, the conditions at the interface (especially the high albedo of icefields and the inlandsis), and the absence in winter of any solar radiation, explain the negative thermal budget. Over the glacial Arctic Ocean and around its edges temperatures exhibit a wide seasonal amplitude, and are always negative (Table 2.1). The pole of cold in winter (especially January-February) lies in the Canadian Arctic archipelago (Putnins, 1974). This thermal pole is sometimes found over Greenland (especially in the summer, with a frequency of 24%), with its 2 186000 km2 of high icefields creating a permanent source of cold air (Table 2.2). Over the Antarctic continent, where winter occurs at aphelion, the altitude and thermal response of the ice explain the extremes of cold (Table 2.3). Table 2.1. Mean monthly temperatures (1) in °C; and extreme minimum temperatures (2) in °C, in the Arctic; P: North Pole, from observations carried out between 75° and 85°N, 19571959; B: Barrow, Alaska, 71° 18' N, 156°47' W, 7 metres; O: Ostrov Domashniy, Severnaya Zemlya, 79°30'N, 91°08'E, 3 metres (after Vowinckel and Orvig, 1970).

P-1 P-2 B-l B-2 0-1 O-2

J

F

M

A

-33,6 -48,7 -26,2 -47,2 -25,6 -46,7

-36,2 -50,2 -27,7 -48,9 -25,0 -47,2

-32,3 -45,7 -26,1 -46,7 -27,8 -45,6

-28,2 -40,8 -17,9 -41,1 -22,2 -40,0

M

J

J

A

S

O

N

-10,4 -1,8 -0,1 -1,5 -10,1 -19,4 -28,1 -30,2 -10,6 -4,5 -11,9 -30,8 -41,1 -39,0 -7,3 4,3 3,6 -0,8 -8,3 -17,4 1,1 -27,8 -13,3 -5,6 -6,7 -17,2 -28,3 -40,0 -10,0 -1,7 0,0 -3,3 -10,6 -18,9 1,1 -26,7 -10,6 -4,4 -8,3 -17,8 -31,1 -35,6

D

-33,1 -49,6 -23,6 -48,3 -25,0 -41,7

28

Circulation in high and mid-latitudes: MPHs

[Ch. 2

Table 2.2 Mean monthly temperatures, Central Station, Greenland, at 3000 metres (after Putnins, 1974).

A

M

F

J

M

A

J

J

O

S

N

D

Tmean —36,4 -39,9 -35,8 -34,3 -18,7 -12,7 -10,4 -11,2 -19,7 -31,9 -33,2 -37,5

Table 2.3 Mean monthly temperatures (1) in °C, and extreme minimum temperatures (2) in °C, in the Antarctic; P: South Pole, 90°S, 2800 m; V: Vostok, 78°28'S, 106°48'E, 3488 m. (from Schwerdtfeger, 1970).

P-1 P-2 V-l V-2

J

F

M

A

M

J

J

A

S

O

N

D

-28,8 -40,6 -33,4 -48,3

-40,1 -56,1 -44,2 -64,0

-54,4 -70,0 -57,4 -75,0

-58,5 -72,2 -65,7 -84,8

-57,4 -73,3 -66,2 -82,0

-56,5 -76,1 -66,0 -83,0

-59,2 -80,6 -66,7 -81,1

-58,9 -77,2 -68,4 -88,3

-59,0 -77,2 -65,6 -82,8

-51,3 -67,2 -57,4 -75,7

-38,9 -53,9 -43,6 -63,1

-28,1 -38,3 -32,7 -48,0

Cooling and compression of the air raise the pressure at the surface, but there is no permanent dome of high pressure to be observed over the poles, neither synoptically or on the scale of means, as a result of the continual departure of MPHs. Over the Arctic, average pressure reaches 1024hPa between Alaska and eastern Siberia in January, and 1017 hPa in July over the North Pole (Triplet and Roche, 1988). These average pressures are quite low compared with: • •

firstly, the average pressures actually observed in the centre of anticyclones by Serreze et al. (1993), north of latitude 65°N (Table 2.4); secondly, the real (synoptic-scale) strengths of anticyclones in the Arctic basin: - in winter, anticyclones are more potent, with pressures above 1035 hPa, and sometimes surpassing 1050 hPa, especially over Siberia north of the Verk­ hoyansk Mountains, and over Alaska/Yukon, with pressures also observed between 1025 and 1034hPa over the central Arctic, Greenland and western Siberia; - in summer, anticyclones are less potent than in winter: over northern Siberia and more especially the Beaufort Sea, the greatest pressures are recorded, exceeding 1025 hPa.

Table 2.4 Mean pressures at the centre of arctic anticyclones (1952-1988) (after Serreze et al., 1993).

Month

J

F

M

A

M

J

J

A

S

O

N

D

hPa

1033

1034

1033

1031

1029

1023

1022

1023

1024

1026

1030

1031

Sec. 2.5]

The birth of MPHs

29

2.5 THE BIRTH OF MPHs Cold polar air is regularly if discontinuously transported away, as vast dense masses detach themselves. This cold air, in a shallow layer under the thermal influence of the ground, and subject to subsidence and spreading through cooling, is ejected from the polar regions like a drop of water detaching itself when it reaches a critical mass, or like an iceberg breaking away from the inlandsis, in the form of mobile anticyclones under the combined impetus of the increase in the cooled mass, centrifugal force and the slope of the land (the catabatic winds of Antarctica and Greenland). The original cold flow, fast-moving, or a succession of merging flows, organise rapidly into an MPH under the influence of geostrophic force, which is strong at high latitudes and ensures the coherence of the MPH. As each MPH departs, it leaves behind a void, a short-lived low, attracting accelerating masses of warm air towards the polar zones; as this advected air cools, low-level high-pressure areas are regenerated by it, and a new MPH is born. Movements are in opposite directions as a function of layers: low down, cold air is transported en masse away from the pole, whilst above, a compensatory warm flow moves towards it. Between the two (i.e. above an MPH) an inversion level of wind and temperature occurs at about 1500 metres above the surface in the Arctic, where, as Vowinckel and Orvig (1967) emphasised: “the inversion is maintained in position and intensity both by surface cooling and simultaneously by subsidence, as well as by the advection of warm air above”. Over Greenland, and most notably over Antarctica, the outflow of cold air is accelerated by the slope of the land. Antarctica experiences “catabatic winds as a constant feature of the lower troposphere” (Parish and Bromwich, 1991), as relief disperses the cold air in all directions away from the inlandsis, often as winds of extreme violence. The same superposition of fluxes moving in contrary directions is observed at higher altitudes of about 3000 metres, as a result of the altitude of the surface (White and Bryson, 1967). MPH formation (Photos 2, 3, and 4) is continual over the Arctic (Table 2.5), with an average of 329 MPHs every year; thus one MPH is born almost every day, the five-year average having been established as one every 1.1 days, throughout the seasons, with slightly increased frequency during winter. In addition to their greater frequency, winter MPHs are characterised by their stronger nature and larger dimensions. Analysis of the dynamics of the North Atlantic aerological space between 1950 and 2000 has revealed that the mean annual number of MPHs ejected from the Table 2.5 Average number of MPHs formed over the Arctic basin from 1989 to 1993, based on the work of the Laboratoire de Climatologie, Risques, Environnement, LCRE (Guimard, Mollica, Moreau, P. de la Chapelle, Reynaud) using charts from the European Meteorological Bulletin. J FMAMJ J ASONDAn Frequency Mean 27.2 26.4 30.0 28.4 33.0 29.4 29.8 24.4 26.0 25.2 24.0 24.8 328.6 1/1.Id

30

Circulation in high and mid-latitudes: MPHs

Photo 3

A ‘Greenland’ MPH.

Photo 4

[Ch. 2

A ‘Scandinavian’ MPH.

Arctic northwest (to the north of the American continent) and encroaching on the Atlantic (cf. Figure 2.4B) is 166.5, or one MPH every 53 hours, i.e. 2 days and 5 hours (Pommier, 2005). In the North Atlantic, advections brought by MPHs on the ‘American’ tra­ jectory (Photo 2) may be reinforced by violent catabatic air masses from Greenland (Photo 3), and by MPHs on the ‘Scandinavian’ trajectory, i.e. entering the oceanic space via the corridor of the Norwegian Sea (Photo 4).

2.6 MPH TRAJECTORIES Ejected from the Arctic or the Antarctic, MPHs form lenticular, dense air masses, mobile and relatively homogeneous, or elastic, discoid masses. They are shallow (initially about 1500 metres in depth), but have enormous diameters which may be of the order of 2000-3000 km (Photo 2). They normally proceed from west to east, with a (variable) meridional component taking them gradually (sometimes rapidly) away from the poles. The dynamicists (who ascribe the pre-eminent role to the upper layers) offer no explanation for their motion, given the inconceivability of an undulatory phenomenon at altitude being able to draw along after itself an extensive, low-level anticyclonic mass of very dense air: and what of those MPHs (among the most powerful and rapid) that take a distinctly southward path (Figure 2.4A) instead

Sec. 2.6]

MPH trajectories

31

of moving from west to east as high-altitude waves most often do? Certain MPHs, steered by relief or by other, more powerful MPHs, may even go suddenly into reverse, and move from east to west. Anticyclones in the lower layers certainly do move, and this reality (the evidence for which resides in the satellite photos) must be taken into account. The progressive subsidence of air over the poles in the polar vortex, which turns from west to east, gradually brings the ever-colder, dense air into step with the Earth’s rotation. The same rotational direction, anti-clockwise, is adopted by the pack ice of the Arctic Ocean. The proximity of the Earth’s axis of rotation imparts to the polar air masses the maximum rotational rate, equal at the pole to the torque of the Earth. According to the principle of the conservation of rotational energy, and given that the local effect diminishes as one moves away from the poles, the torque proper to the moving air mass is greater than the local effect; the relative torque (difference between the planetary torque and the local torque) being positive, the displacement is generally in a west-east direction, which means that the mobile anticyclone moves faster than the surface of the Earth beneath it. This advantage over the rotation of the Earth eastwards is however transient, and gradually diminishes as the lines of latitude grow longer away from the poles. The relatively greater velocity of the MPH slowly falls off, finally equalling that of the rotational velocity and ‘immobilising’ the MPH; there may even be a reversal (westward) of motion, if no other factors (especially relief) have previously inter­ vened. At the same time, while the Coriolis force ensures the coherence of the MPH, in accordance with the equation of continuity, or conservation of mass (‘the mass of air contained within a volume limited by a closed fluid surface remains constant during the motion of that volume’ (Triplet and Roche, 1988)), the spreading out of the air (i.e. divergence and subsidence) within an MPH increases the area of the frictional surface below and contributes to a braking effect. The dynamical conditions of the displacement of MPHs vary with the Coriolis force (which diminishes at lower latitudes), with the seasons, and with the nature of the ground beneath. The seasonal variation is dependent on the polar thermal deficit: in winter, MPHs are stronger and their paths carry them further towards the tropics, while in summer their lesser strength means that they leave the polar regions more slowly. Unhindered as they cross oceans, MPHs there achieve their greatest velocities: in the North Atlantic, they last for about 6.5 days, from formation to their agglutination over the eastern Atlantic, covering a mean distance of 6754 km at an average speed of 43.5 km/h. The highest values are found in winter (Pommier, 2005). The most powerful MPHs (in terms of pressure) normally travel the furthest south (Figure 2.5). In the essentially oceanic latitudes of the Southern Ocean, including the ‘Roaring Forties’ of the mariners of olden days, MPHs emerge from the Antarctic as catabatic flows with great violence and regularity (Photo 5, and Photo 7-1). Continental land masses have a strongly perturbatory effect on dynamical conditions, through increased friction, and especially because of relief. Cold air (cold in relative or absolute terms) is dense and therefore not easily uplifted. The influence of mountain masses depends upon their altitude and orientation, and also on the depth, and direction of motion, of the MPH.

32

Circulation in high and mid-latitudes: MPHs

[Ch. 2

A

Figure 2.5 Trajectories and frequencies of ‘American-Atlantic’ MPHs as a function of their mean annual pressure. A: above 1020hPa (95.70%); B: above 1025 hPa (66.03%); C: above 1030hPa (32.43%); D : above 1035hPa (15.59%). Period 1950-2000 (Pommier, 2005).

Photo 5 GOES 7 image of the South Pacific, 14 December 1995, visible light (after NASA Explorer Arc). MPH ‘A’ is an example of the gigantic MPHs which roll off the Antarctic (catabatic winds). MPH 'B’, halted by the barrier of the Andes, flows out in a north-westerly direction, towards the tropical zone.

Sec. 2.6]

MPH trajectories

33

Photo 6A and B Effect of the Alps on MPHs (Meteosat, visible. 12H UT, CMS-Lannion). On 27 January 1986. an MPH moves down towards central Europe, and its southern edge is halted by the Alps, whose snow-covered summits are clearly seen. Another energetic MPH follows it across the Atlantic. On the trailing edge of the first MPH. an intense flow of cold air (mistral) moves off towards the western Mediterranean from between the Alps and the Pyrenees. The general eastward movement on 28 January 1986 causes this MPH to wheel around the barrier of the Alps to the north: a brisk cold outflow (bora) goes between the Alps and the Dinaric chain toward the Adriatic, and thence to the eastern Mediterranean. These two MPH fragments lift the warmer Mediterranean air before them, and propel it northwards.

34

Circulation in high and mid-latitudes: MPHs

[Ch. 2

Thus, relief acts on different scales: •

• •

by simply dividing an MPH, which flows around relief which is not too extensive, and reforms later; by fragmenting MPHs, and sending a part (of whatever size) off in another direction (Photo 6); or, as in the case of continuous chains such as the Rockies, the Andes or the Asiatic heights of Anatolia-Taurus which run into the Himalayas and Tibet, by blocking and channelling the entire MPH, thereby determining the behaviour of vast circulatory units.

A) In the southern hemisphere (Figure 2.6), rapid and violent ejection of cold air is facilitated by the slope of the Antarctic land mass, and dispersion (in the form of violent winds) occurs in all directions. The Andes, first encountered at 55 S, become impassable north of latitude 40°S; MPHs encounter southern Africa at 35 S, and, if

relief barring the passage of MPHs MPH trajectories

-------------MPHs ------- — flux lines (trades)

Figure 2.6 Trajectories of Mobile Polar Highs from the Antarctic and resultant circulation in the lower levels of the troposphere: southern hemisphere.

Sec. 2.6]

MPH trajectories

35

J

Photos 7 MPH divided and channelled by southern Africa. 1-5 August 1999 (Meteosat Image Bulletin. ESA. visible. 12H UT). 1. 2 etc.: dates; S: Summary chart: evolution 31 July 6 August 1999. The MPH is divided on 2 August as it reaches the mountains of the Cape, and then by the Great Escarpment. The main mass moves up through the South Atlantic towards the Gulf of Guinea, the flux (now a trade wind) being accelerated at the foot of the Namibian escarpment. The anticyclonic ‘Saint-Helena' cell is thereby reinforced. A smaller part of the MPH runs along the Drakensberg into the Mozambique Channel. The plateau of central southern Africa is not reached directly by the air of the MPH. Another MPH appears (2 and 3) to the south-west and approaches southern Africa.

36

Circulation in high and mid-latitudes: MPHs

[Ch. 2

they have crossed the vast spaces of the southern Indian Ocean, the eastern highlands of Australia at about 38°S. Orography does not have to rival the Andes to have its effect, however; for example, in southern Africa, the mean altitude of the southern African plateau is of the order of 1000 to 1500 metres, and the Great Escarpment at its edges reaches 2000 metres, and in the case of the Drakensberg, 3000 metres. Since the thickness of an MPH is certainly less than these values (of the order of 1000 m), an MPH approaching the Cape is neatly divided into two, one section flowing northwards along the foot of the Namibian escarpment, and the other following the coastline of Natal towards the Mozambique Channel (Photos 7), where Madagascar divides it further (Photo 8). The intervention of relief, early on in the case of the Andes, and later, at the margins of the tropical zone, creates three aerological spaces: Pacific, Atlantic and Indian/Australian.

Photo 8 MPH divided and channelled by Madagascar on 4 July 1992, Meteosat, visible (after IONIA, ESA/ESRIN). Anticyclonic rotation within the MPH brings the polar air towards the eastern escarpment (altitude of the order of 1500 metres), and causes it to spill around the south of the island. To the south of Madagascar there appears a line of confluence in the MPH between the air diverted by the escarpment and the more recently arrived polar air. The air of the MPH flows directly up the Mozambique Channel towards Tanzania and Kenya.

Sec. 2.6]

MPH trajectories

37

Figure 2.7 Trajectories of Mobile Polar Highs from the Arctic, and resultant circulation in the lower layers of the troposphere: northern hemisphere.

B) In the northern hemisphere (Figure 2.7), the highlands of Greenland divide trajectories from the outset. Greenland lies between latitudes 83°N and 60°N, and in its central region the land rises to 3230 metres, with a mean altitude of 2135 metres. Greenland interferes with the downward turning movement of the Arctic vortex and, together with the high ground of Ellesmere Island and Baffin Land, funnels MPHs preferentially towards North America and, with the exception of air descending directly from Greenland, feeds them into the Atlantic south of latitude 60°N around Cape Farewell. Three vast spaces are delineated, as a function of the trajectories imposed upon the MPHs:

•

North America (east of the Rockies)/Atlantic/western Europe and western Mediterranean: in this dynamic space circulation is directed by the immediate intervention of Greenland, sending a high number of MPHs towards Canada, at a rate of one Arctic MPH in every two (Table 2.6). These MPHs are often of gigantic size and may at a given moment cover the whole of North America.

38

Circulation in high and mid-latitudes: MPHs

[Ch. 2

Table 2.6 Mean number (1989-93) of MPHs following the ‘American’ trajectory: the flow is to the west of Greenland, onto North America and thence to the North Atlantic (LCRE).

J

F

M

A

M

J

J

A

S

O

N

D

Yr

Mn. 15.0 12.2 13.6 11.4 16.2 12.4 13.2 13.6 12.2 12.4 11.2 16.6 158

19 June

20 June

21 June

22 June

Frequency

1/2.3 d

Photos 9 Movement of MPHs over North America on 19-22 June 2003 (GOES E, visible, 18 h UTC). On 19 June, three successive MPHs are clearly visible over the North American continent, separated by their cloudy fringes (fronts). Another MPH has already reached the Atlantic. Born over the western Arctic (northern Canada), they move south, following two paths. One is more southerly, towards the Gulf of Mexico, the Caribbean and the Atlantic. The other, more zonal, and mainly across Canada, passes south of Greenland towards the northern part of the Atlantic, and thence towards Europe.

The Rockies are an impassable rampart to the cold air of MPHs, so the main trajectory towards the Atlantic lies between Baffin Land, Greenland and the northern Appalachians; these last channel part of the flow towards the Gulf of Mexico, and beyond the Isthmus into the Pacific, especially in winter (Photo 9). Most of the advected polar air remains in the Atlantic area, with

Sec. 2.6]

MPH trajectories

39

Table 2.7 Mean number (1989-1993) of MPHs following the ‘Scandinavian’ trajectory: the flow is to the east of Greenland, towards the north-east Atlantic and/or Europe (LCRE). J

Mn. 5,4

•

F

M

A

M

J

J

A

S

O

N

D

Yr

Frequency

5,2

6,6

6,6

9,4

8,4

6,8

4,0

6,8

8,6

6,4

4,8

79

1/4,6 d

possibilities for outflow, to the north of the Alps towards central Europe; towards the western Mediterranean; and towards North Africa south of the Atlas range, in winter when MPH tracks are further to the south. Central Europe/western Asia/eastern Mediterranean/northern Africa-Arabia: MPHs entering this space move in along the ‘Scandinavian’ trajectory (Table 2.7), and they may be joined by ‘Russo-Siberian’ MPHs from the east. Some Scandinavian MPHs descend directly upon Western Europe, in part channelled by the western slopes of the Scandinavian mountains. The Russian trajectories, however, direct MPHs mostly towards central Europe (Photo 10).

Photo 10 A powerful MPH on the ‘Scandinavian’ trajectory, from the Arctic to the Black Sea, flows across central Europe on 5 February 2008 (NOAA 18, visible, 11 h 15 UTC, www.hartmuts.de). It rounds and partially crosses the Scandinavian Alps and the southern section of the Norwegian relief (which is higher), and is divided into two tongues flowing directly southwards. The first of these reaches Romania, where the Carpathians (encircled), appear clearly above the cloud formations. The second reaches the Ukraine, and its leading edge spills into the Black Sea.

40

•

Circulation in high and mid-latitudes: MPHs

[Ch. 2

MPHs are directed either towards the Black Sea (channelled by the Urals) and the eastern Mediterranean basin, or otherwise towards the Turan basin occupied by the Caspian and Aral seas, from which it is difficult for them to escape, since this low-lying area is surrounded by high relief on three sides. This immense space is divided by the continuous line of relief stretching from the Anatolia-Taurus-Zagros range all the way to the Himalayas and Tibet. This orographical barrier is completed to the north by the Tien Shan, issuing from the Hindu Kush and Pamir ranges, then, north of Dzungaria, by the Altai massif, the Sayans and the highlands of Mongolia. North of latitude 55°N, outflow towards eastern Siberia can occur with some difficulty through Dzungaria in the direction of the Gobi Desert, or to the south by spilling over the Iranian plateau towards the Arabian Sea. In the southern part, MPHs arriving in the eastern Mediterranean basin proceed towards North Africa, and to the south of the Turkish and Iranian highlands (Taurus-Zagros) by way of the Arabian Peninsula on their way north to the Indian Ocean. Eastern Asia/North Pacific/western North America: potent winter Siberian MPHs flow across eastern Siberia through the Lena depression, between the Siberian-Mongolian highlands and the Verkhoyansk Mountains. These MPHs, reinforced by fragments of others arriving by way of Dzungaria and the Gobi desert, reach the Pacific across northern China and Japan (Table 2.8 and Photo 11). They gain reinforcement from MPHs coming out of the Arctic basin through the Bering Strait. Pacific MPHs encountering the western side of North America are blocked by the Rockies and turn southwards.

In these aerological spaces relief creates bottlenecks, through which the mass transportation of air by MPHs is relayed by the (concentrated) linear flow of often violent winds (compression and acceleration due to the Venturi effect), which then reconstitute the fragments of the MPHs. This is particularly true of the Mediterranean basin (Photo 6 and Chapter 6, Photo 11). These aerological spaces, enclosed by mountains, are often hermetic to cold air, but communication is possible, for example over Western Europe where Atlantic and Scandinavian MPHs may arrive. At the junction of two trajectories, the interaction of air masses is conditioned by their respective densities; the more recent and therefore cooler MPH takes advantage of its relatively higher density to divert, block or even lift and disperse the air of the less dense MPH (Chapter 5).

Table 2.8 Mean number (1989-1993) of MPHs following the ‘Pacific’ trajectory: the flow is across eastern Siberia and China, towards the Pacific (LCRE). J

FMAMJ

Mn. 10.6 7.8

11.8 11.6 10.4 7.8

J

ASONDYr

7.6

7.0

9.8

9.8

9.2

10.4 113

Frequency

1/3.2 d

Sec. 2.7]

11 April

The MPH-associated wind field

41

12 April

Photos 11 11 and 12 April 2007 (MTSAT01, Vis. 12 UTC). These images show the two trajectories followed by MPHs through Asia to the North Pacific: the northern, over eastern Siberia and the Sea of Okhotsk and/or the Bering Strait, and the southern, through China, where the southern MPHs are joined by those from Dzungaria and the Gobi Desert. On 11 April there are three successive MPHs, from western America towards Asia. On 12 April, the MPH leaving China, whose leading edge had been over Japan the previous day, has moved away eastwards, the Japanese relief ‘fanning out' the cold air flow.

2.7 THE MPH-ASSOCIATED WIND FIELD

As it moves, the relatively greater density of an MPH, with its cold air (cold in relative or absolute terms), allows it to push aside or lift before it warmer, less dense air. This (relatively or absolutely) warm flux may be the remnant of the trailing edge of the previous MPH which has been warmed to a greater or lesser extent. It may also have its origin in the deviation of a tropical flux which has been evolving for a longer time. It is observed that, all the while the MPH maintains the advantage of greater density, which is constantly under threat as it pursues its track, the process of creating the associated pressure and wind fields evolves, in broad outline, thus:

•

•

The advancing low-level anticyclone first causes the surrounding flux(es) to move away or deviate, then lift along the MPH’s leading edge, as determined by its direction of motion (the MPH is thus surrounded by a more or less active discontinuity/front). Since the MPH acts in a mechanical fashion, the extent of the lifting is a function of the strength and dynamism of the MPH, as well as of the thermal contrast between the MPH and the anterior flux, warm air having a natural tendency to rise. Figure 2.8 shows how this contrast, which is more marked on the southern edge of an MPH of northern origin, gives rise to a depression initiated by the MPH at the point where the contrast in density is at its greatest, on the axis of motion (where the directions of the MPH and the surrounding flux are diametrically opposed). The lifting of warm air by the MPH’s leading edge causes a drop in pressure at lower levels and the formation of a dynamic low, whose depth is a function of the intensity of the updraft.

42

Circulation in high and mid-latitudes: MPHs 2. extension of low-pressure corridor

[Ch. 2

3. deepening of closed depression

1. initial low “d" forms

Figure 2.8 Establishment of pressure and wind fields associated with an MPH (northern hemisphere).

•

•

•

The low then organises the low-level circulation into a cyclonic rotation. On the northern edge (Figure 2.8) of the depression, the surrounding air is directed towards the leading edge of the MPH (front). The intensity of the uplift, the resulting drop in pressure and the induced cyclonic circulation bring about the progressive northward drift of the low (or succession of lows); as long as the diverted air is not interrupted, the initial low becomes a peripheral low-pressure corridor, at its deepest along the leading edge of the MPH. During the formation of the low-pressure corridor, the diverted cyclonic northward flux (Figure 2.8 (2)) clings to the leading edge of the MPH, at first because of the general eastward motion, and later because of the geostrophic wind direction, whose resultant is in the opposite direction to the mass displace­ ment of the MPH. The cyclonic path of the diverted flux is directed westwards in both the northern and southern hemispheres (Figure 2.9). This dynamical component thus adds to the efficacy of the mechanical and thermal (densityrelated) uplift associated with the leading edge. A dense, powerful and rapidly moving MPH is therefore surrounded by very deep lows and strong updrafts, and an intense cyclonic circulation of warm air diverted towards the pole. After the diverted flux has passed around the aerological barrier presented by the MPH’s leading edge, the geostrophic force may act to its fullest extent. Rotational forces create a vortex, with low pressure at its centre, as a result of convergence and the resulting upward movement of air. As a function of the geostrophic force equal to: 2mQ.vsinL (m = mass, Q= angular velocity of the Earth’s rotation, v = velocity, L — latitude), the intensity of the vorticity and ascending movements, and therefore of the depth of the depression or cyclone, depend upon:

- m, which is the mass of the diverted flux, that quantity of diverted air itself being dependent upon the strength of the factor responsible for the diversion and therefore upon the strength of the MPH;

Sec. 2.7]

The MPH-associated wind field

43

r—i MPH ------ line of confluence C calm —► direction of motion---------- *flux line

Figure 2.9 an MPH.

Diagram of associated (anticyclonic and cyclonic) winds and displacement of

- v, the velocity of the diverted flux, itself dependent (unless local geographical factors or other MPHs intervene) upon the strength of the MPH and its velocity of displacement; - L, the latitude, zero at the equator (sin 0° = 0) increasing towards the poles where it reaches its maximum (sin 90° = 1).

•

•

Vorticity progressively increases with latitude, northwards and southwards, within the low-pressure corridor (Figure 2.9); deeper mobile lows may move within the corridor, merging into the main low. In the (‘Norwegian’) cyclone the intensity of vorticity is related to that of the diverted flux, and therefore to the strength of the MPH. The activity of the cyclone depends also on the intensity of the potential energy advected by the cyclonic flux, and the release of this energy intensifies updrafts. As a result of latitude, vorticity in the closed depression (vortex) is stronger on the polar edge, contributing, as it moves, to the distancing of the vortex from the MPH (cf. Chapter 5). The intensity of the diverted cyclonic flux also contributes

44

Circulation in high and mid-latitudes: MPHs

[Ch. 2

to this. The MPH moves south-east (Figure 2.8) while the associated cyclone moves off to the north-east. Simultaneously, the displacement of the MPH southwards improves the thermal qualities of the diverted flux, possibly enhancing its water content (and thereby energy), encouraging this progressive distancing while creating more updraft and lowering pressure further (through the further release of energy). The closed depression (cyclone) cannot therefore be dissociated from the anticyclone (MPH) which originally caused the rise of the initial depression, its extension, and the cyclonic diversion. Its characteristics (ignoring geographical effects) are closely tied to the strength and the velocity of the MPH and to the inherent qualities of the diverted flux. A powerful and fast-moving MPH is thus surrounded by very deep lows and strong updrafts, with an intense cyclonic circulation moving warm air towards the poles. The establishment, thus schematised, of the associated wind field around an MPH has not yet taken into account the inevitable interactions between MPHs, or the effect of relief. This wind field is also modified by the ever-changing ‘relationships of forces’ between the MPH and surrounding fluxes (cf. Chapter 5). A cyclonic circulation precedes the MPH, causing warm air to turn along its leading edge towards the poles (Figure 2.9). The whole assemblage of MPH, peripheral low, diverted flux and cyclone proceeds eastwards. The surface wind field associated with an MPH brings about the alternation, as it moves, for example in the northern hemisphere, of: • • •

• •

a diverted flux, in front of the MPH and moving towards the pole, with a strong southerly component (cyclonic); then an interlude of calm as the line of confluence or interaction (front) passes over, horizontal movements being replaced by vertical upward ones; then a (cold) flux with a strong northerly component (anticyclonic) as the MPH moves across; another calm period as the centre of the anticyclone passes (divergence and stability); then another southerly component at the trailing edge of the MPH, whereas within the vortex (cyclone) the wind may be blowing from any direction.

This instability of wind direction associated with the passage of an MPH shows that the calculation of a ‘mean’ or ‘resultant’ wind, deduced from its varying directions, is not representative, at these polar and temperate latitudes, of real low-level circulation. The latter is further complicated by local factors, especially relief and/ or breezes. The westerly resultant wind, supposed to be typical of mid-latitudes, in fact signifies the displacement of the MPH, but not the real winds associated with this displacement. MPHs gradually lose their dynamism as they travel, and encounter factors more potent than themselves, such as colder and more recent MPHs, or relief which leads to the formation of anticyclonic agglutinations (AAs).

3 Anticyclonic Agglutinations (AAs)

A ‘statistical’ perception of phenomena based on mean pressures is at the heart of the idea of the anticyclonic ‘centres of action’: subtropical high-pressure areas at the boundaries between the temperate and tropical zones, over oceans. To these is currently attributed responsibility for circulation in the tropics, with their easterly winds, and in temperate areas with their westerlies. Similarly, they are said to be responsible for the existence of tropical deserts, or for the separation between climatic zones according to whether or not they permit meridional exchanges. Within the temperate zone, there also exist vast individual temporary anticyclones, over land masses as well as over the sea. Ignorance of the reasons for their presence causes them to be too hastily attributed to some hypothetical displacement of ‘subtropical high-pressure areas’ that have sometimes strayed far from their usual locations. They are constantly referred to in order to explain the weather, especially in France where the Azores anticyclone - a fundamental player in the cast of anticyclones on the meteorological stage - brings fine weather as it expands', and then it withdraws, letting bad weather in, both synoptically and seasonally. The dynamical reality is very different. It requires some preliminary reflection upon the origin of these high-pressure areas, upon their characteristics and their role in circulation.

3.1 A LOOK AT THE SO-CALLED ‘SUBTROPICAL HIGH-PRESSURE AREAS’

First described by Maury (1855) from data collected by ships, ‘subtropical highs’ have been slotted into the scheme of general circulation, and ever since have been called “a key element of the surface pressure field” (Hastenrath, 1991). Discovered over the ocean, they have retained their ‘oceanic’ character and, in the words of Trewartha (1961), “their origin is not fully understood”. This is still true to a great extent; several questions remain to be answered. What is their origin? Their

46

Anticyclonic Agglutinations (AAs)

[Ch. 3

geographical situation? Their division into cells, their vertical structure, their presumed ‘migrations’ in latitude and longitude, variations in their strength (especially if they have ‘extended’ themselves across land masses), and their internal motions - all require explanation. These highs observed at the surface are normally attributed to permanent subsidences, roughly located above latitudes 30°N and S and associated with the descending sections of Hadley cells (as Maury (1855) and Ferrel (1856) suggested when discussing general circulation). Since compression leads to warming, these downward movements cannot be used to explain either the pressures achieved at the surface (warm air being light), or the temperature, which should be constantly warm but in fact is often observed to be low in high-pressure areas. Neither can such slow, descending movements be responsible, on any timescale, for the abrupt changes in weather seen at these latitudes, or for the vigour of the trade winds and their variations in speed, sometimes involving violent acceleration. There is certainly subsidence, but it is low-powered; given the great mass which moves through these anticyclonic cells to supply tropical circulation, extremely powerful and rapid subsidences - with, as a consequence, considerable warming - might be expected. Such intense phenomena are, however, not observed. It should be remembered that these subsident movements cannot normally reach the Earth’s surface. The geographical distribution of ‘subtropical’ highs, their fragmenting and the presence of maximum pressures to the east of the cells (statistically), invalidate the hypothesis of influence from above. If vertical movements of Hadley cells are really responsible, a continuous high-pressure zonal belt should exist permanently around latitude 20-30° in each hemisphere; but at the latitude where there is ‘permanent subsidence’ we see instead markedly individual anticyclonic cells, each one exhibiting different climatic characteristics at its edges. At the same latitude, we encounter the climates of the Sahara or the Cape Verde Islands, as well as of the Caribbean and the Yucatan. Reference to a hypothetical “descending eastern edge” (Pagney, 1994) of cells, supposed to account for these climatic features, does not explain why, in the free atmosphere and for no obvious (or demonstrated) reason, subsidence should be more intense at certain longitudes. This would also imply a 'non-subsident edge' on the western side, of equally mysterious origin. On the eastern sides of tropical oceans flow cold currents, and trade wind circulation causes upwelling from deep down, maintaining a supply of cool water near coastlines. The presence of this water is presumed to be responsible for the subsidence of the air above it. However, this simplistic ‘relationship’ carries its own disproof: if the subsidence were (thermally) reinforced in eastern parts of oceans, we should there encounter the highest temperatures, due to the effects of compression by subsident air. This is the opposite of the proposed relationship, and more especially of observed reality. Disproportion in the scales of the phenomena would moreover be considerable, since the thermal behaviour of the ocean is insufficient to cause the presumed vertical movements (cf. Figure 1.6). Also, again, it would be necessary for the subsidence to reach the ocean surface, and this is not observed. The constant west-east displacement, experienced as a pressure wave and observed within oceanic or continental anticyclonic cells on their temperate edges, is

Sec. 3.1]

A look at the so-called ‘subtropical high-pressure areas’

47

variously interpreted, as are the changes in strength and extent of highs. Gentilli (1971) explained their great mobility in the southern hemisphere thus: “subsident air, engendering the belt of high-pressure areas and the tropical divergence, is subdivided, as a result of the Coriolis effect, into a series of migratory anticyclones”. This unverified hypothesis would not explain the migrations of ‘travelling anticyclones’, and could only be applied to warm anticyclones (which is not the case). Migrations of subtropical cells in latitude and longitude throughout the year, observed by using mean values of pressure, “are still not well understood", as Hastenrath (1991) stressed. Centres of action are often thought of as independent entities, as actual meteorological individuals, capable of ‘swelling’ and ‘shrinking’. Such personifica­ tion reduces our explanation of the weather to a kind of ‘meteorological animism’ (Leroux, 1992a); thus, Choisnel (1991) had the Azores anticyclone out of its normal seasonal position (it was at the time over the British Isles) and apparently "drifting as it were north-eastwards, which is particularly out of character in winter ...” - which is the least one can say! In the preceding case the ‘presence of the anticyclone’ was thought to be responsible for ‘fine weather’ in the winter but, in July 2007, because it was "not in its normal position" (according to Meteo France), it was also - and officially (cf. Le Figaro, 13 July) - "the fault of the Azores anticyclone" that the weather was so bad that summer! Let it rain or shine: it’s always the ‘fault’ of that anticyclone! They bemoan the "departure of the anticyclone, which has returned to its traditional home", and announce "the return of the anticyclone", but nobody, not least the forecasters, can point it out on a synoptic pressure chart or a satellite image - simply because, on the scale of present time (i.e. the current weather), this famous anticyclone does not exist. If someone happens to mention ‘an anticyclone’ (i.e. an MPH), it should be pointed out that this is a mere snapshot, and that a few hours before (and after), the anticyclone was (and will be), in a different location: for as time passes, everything moves, both anticyclones and depressions. Tropical highs are still seen as a virtual barrier, more or less impassable: a sort of anticyclonic rampart forming a fundamental boundary between the circulations of temperate and tropical zones, which therefore seem almost (and erroneously) independent of each other. This presumed barrier, implicit in various descriptions of general circulation, and in particular the tri-cellular schema and palaeoclimatic concepts (notably by COHMAP, 1988), is even deemed to control and/or prevent meridional exchanges: thus, “frequent meridional exchanges imply a weakening of the anticyclones”, whilst, conversely, “a lessening in the frequency of meridional exchanges implies a strengthening of tropical anticyclonic cells”. This concept, left over from traditional cellular constructs of general circulation, is a misguided one on any timescale, from the synoptic to the palaeoclimatic, since observations of reality show the opposite to be true. The origin of these anticyclonic centres of action is not really known, because a statistical view of phenomena, initially defined using mean pressures, does not ask the essential question: how do these anticyclones (‘subtropical’ or formed at higher latitudes) actually come into being? The confusion between the statistical scale

48

Anticyclonic Agglutinations (AAs)

[Ch. 3

(of means) and the (real-time) synoptic scale is thus, inexplicably, still very much alive. Tentative explanations have been put forward: for example, Viaut (1942) stated that “this kind of ‘renewal’ of the subtropical high-pressure belt by mobile polar anticyclones which become stationary and warm up is a regular occurrence”. The problem might well have been resolved at that time, if MPHs had been seen as the individual entities that they are, and recognised as the (permanent, not occasional) vehicles for meridional exchanges. 3.2 MERIDIONAL TRANSPORT BY MPHs AND THE FORMATION OF AN ANTICYCLONIC AGGLUTINATION (AA)

What is required for an anticyclonic agglutination (AA) to be able to form? The movement of MPHs must be slowed, or even halted, by ‘fixed’ relief or by ‘mobile’ relief (i.e. a denser MPH). MPHs cannot keep moving indefinitely eastwards: as they spread, frictional forces increase, and their relative excess of speed compared with that of the surface of the Earth diminishes; movement is progressively slowed. Moreover, because of the progressive weakening of the geostrophic force, the MPH gradually loses coherence, and finally (as a result of its own dynamic only) ceases to advance; it flows outwards into areas of lower pressure, rather like a drop of water spreading, and gives rise directly to a trade flux. As a result of differences in the strength of MPHs, one MPH may be overhauled by the MPH following it, resulting in total or partial merger if their densities are similar (cf. Photo 1, Figure 2.4, for example the merger of H1+H2). Differences in strength, speed and trajectory between MPHs, and/or the action of land masses in increasing friction, encourage this telescoping effect. As a function of the direction of travel of MPHs and of the alternation between continental and oceanic circumstances, conditions differ according to which side of oceans is involved:

•

•

on the western edges of oceans, near land masses, immediate passages into tropical circulation are constantly observed. The MPH directly fuels the trade­ wind circulation. Photos 12 illustrate this transformation from MPH into trade flux above the southern Indian Ocean; At the eastern edges, the transition towards tropical trade circulation occurs via an AA formed by the merging of successive MPHs.

3.2.1 The role of relief The most telling factor, however, is relief which, as a function of its altitude and of the depth of the MPH, may slow, channel, or halt it, and interlock successive MPHs. The most vigorous (and permanent) effects are those of relief which, as a function of its altitude and of the depths of MPHs, decelerates, channels, blocks and causes the merger of successive MPHs to form an AA.

Sec. 3.2]

3 June 2007

Meridional transport by MPHs

49

4 June 2007

Photos 12 Between 1 and 4 June 2007 (Meteosat 07, Visible, 06 h UTC). Example of the direct transformation of an MPH into a maritime trade flux, and thence into a monsoon. An MPH from the Antarctic moves north-east across the south Indian Ocean and pro­ gressively spreads. Its leading edge reaches southern Madagascar on 1 June. The air that it transports moves northwards along the Mozambique Channel, whilst the eastern Escarpment of Madagascar is still blocking the flow on 2 June. The reinforced trade rounds the Escarp­ ment on 3 June and, crossing the geographical Equator, becomes a monsoon (the Indian monsoon, kusi) on 4 June.

50

[Ch. 3

Anticyclonic Agglutinations (AAs)

14 June 2007

15 June 2007

16 June 2007

17 June 2007

Photos 13 Channelling by the Andes of air advected by an MPH, 14-17 June 2007 (GOES 12, Visible, 18 h UTC). On 14 June an MPH, divided by the snow-capped southern Andes, which are impassable beyond 40°S, is moving northwards across the Pacific and Argentina. On 15 June, over the Pacific, the air of the MPH is integrated into the maritime trade circulation and spreads north-westwards across the ocean to form the south Pacific AA known as the "Easter Island cell’. Meanwhile, over the continent, the leading edge of the MPH reaches the Gran Chaco and the southern part of the Amazon Forest. On 16 June a new MPH is in its turn divided as it moves northwards, and on 17 June it too is integrated, to the west of the Andes, into the maritime trade flux; over the continent the eastern MPH proceeds northwards across Argentina.

Sec. 3.2]

28 March 2006

Meridional transport by MPHs

51

29 March 2006

Photos 14 The zonal mountain barrier of Asia, from the Pontic-Zagros ranges to the Himalayas and Tibet (Meteosat 07, Visible, 06 h UTC). On 28 March 2006 a ‘Siberian’ MPH proceeds southwards across the plains of western Siberia, towards the Turan depression (which contains the Caspian and Aral Seas). To the east, it reaches the Tien Shan Mountains and the Hindu Kush. On 29 March a large fragment of this MPH infiltrates the eastern part of the Mongolian Plateau (Gobi Desert) via the Dzugarian corridor. To the south, it remains ‘trapped’ by the Turan depression, with only a small part escaping towards the Indian Ocean through the lowlands of Iran. Tibet and the Himalayas form an impassable barrier, protecting India from advections of cold air.

Continuous north-south barriers such as the Rockies or the Andes (Photos 13), or east-west barriers such as those stretching from Turkey to China (Photos 14), stop and divert almost the whole mass of MPHs (Figures 2.6 and 2.7). It is not, however, indispensable that the mountain barrier should be as formidable as in America and Asia. Interaction with continents may be a more complex affair. Such is the case when MPHs encounter Western Europe; at first, they move easily inland, but the Vosges, the Jura and, to a larger extent, the Alps and the Cantabrian-Pyrenees system make France a veritable orographical ‘funnel’. The resulting compression, caused by the blocking action- of the relief, the anticyclonic rotation within the MPH, and the orientation of the Pyrenees and Cantabrian mountains providing a virtual ‘springboard’ towards the west, throw most of the associated air vigorously back towards the Atlantic, in the opposite direction to the normal mass motion. Another section of the air may skirt around the north of the Alps into central Europe, and yet another, under pressure deep within the funnel and partly blocked by the Massif Central, may escape between the Alps and the Pyrenees towards the basin of the western Mediterranean, occasioning violent winds such as the cierzo of the Ebro valley, the tramontane at the foot of the Pyrenees, the cers of the Narbonnaise and the bise and mistral of the Rhone valley (Photos 15). The situation of 27 April—1 May 1986 (Photos 15 and Figure 3.1) illustrates the genesis of the north-east Atlantic ‘Azores’ AA. That fraction of the MPH returned to the Atlantic from 29 April, in this case the largest part, still contained and diverted southwards by the barrier of the Cantabrian Mountains and the Pyrenees, the Iberian plateau and its sierras, in conjunction with the Moroccan Atlas, joins the mass transported by MPHs from the west (the ‘American-Atlantic’ trajectory),

52

[Ch. 3

Anticyclonic Agglutinations (AAs)

28 April 1986, 15 h 30

29 April 1986, 09 h 30

29 April 1986, 15 h 30

30 April 1986, 09 h 30

Photos 15 Evolving situation between 28 April 1986 (15 h 30 UTC) and 30 April 1986 (09 h 30 UTC). Photo 2 shows this MPH on 28 April at 12 h UTC. (Meteosat, visible, CMS Lannion).

Figure 3.1 Genesis of the North Atlantic ‘Azores’ AA. Evolution between 27 April and 1 May 1986, based on Meteosat.

Sec. 3.2]

Meridional transport by MPHs

53

Table 3.1 Mean number (1989-1993) of ‘American-Atlantic’ MPHs entering the ‘Azores’ AA (LCRE). J

F

M

A

M

J

J

A

S

O

N

D

Yr

Mean 14,6 13,4 13,2 12,4 13,4 12,2 11,6 11,4 11,6 11,8 12,2 13,8 151

Frequency

1/2.4 d

and possibly from the north (meridional ‘Scandinavian’ trajectory). The north-east Atlantic AA receives in this way an average of 12.4 ‘American’ MPHs every month, the frequency being greater in winter than in summer. The average renewal period is 1 MPH every 2.4 days (Table 3.1). 3.2.2 The formation of an anticyclonic agglutination

Figure 3.2 shows diagrammatically the formation of an anticyclonic agglutination, encouraged by relief with a north-south orientation. •

•

•

•

• •

•

On the extratropical edge, a low-pressure corridor may still remain open along the leading edge of a more recent and/or more rapid MPH. On the synoptic scale, a depression may occupy the statistical location of an anticyclonic cell from time to time. The narrowing of the corridor between the leading edge of the MPH and the relief accelerates the intensity of the cyclonic flux (diverted towards the pole). As an MPH gradually merges into the previous one, and thence into the agglutination of the MPHs which preceded them, the peripheral depression weakens and fills, and the cyclonic circulation on the leading edge of the invading MPH (with flux diverted towards the pole) gradually dies away; anticyclonic stability asserts itself, and both updrafts and the likelihood of rain fade away. Progressively, in a field of high pressure areas formed by the amalgamation of MPHs, there will eventually remain only anticyclonic rotation, especially where relief provides compression and channelling of the eastern edge of the AA. By contrast, absence of wind is a dominant feature at the centre of the agglutination; hence, the existence of calms at the ‘horse latitudes’, where rainfall is markedly low (on becalmed sailing ships, it was the custom to slaughter horses, as they consumed a lot of water). Tropical (trade) circulation begins where alternate meridional components of MPH-associated winds are replaced by one unique anticyclonic flow. A succession of MPHs, i.e. ‘gusts’ of polar air, causes the trade wind to pulsate; each ‘pulse line’ (PL) is evidence of newly-arrived air. The air transported by MPHs in the lower levels now slides beneath the air moving downwards at latitudes 30°N and°S (the descending branches of the Hadley cells). The upper boundary of the air of polar origin now forms the trade inversion (TI) (Figure 3.2b).

54

Anticyclonic Agglutinations (AAs)

[Ch. 3

Figure 3.2 Formation of an AA (encouraged by relief with a north-south orientation) as successive MPHs merge, and tropical trade-wind circulation results (northern hemisphere). Left: surface wind and pressure field (al, a2, a3, AMP4 are successive MPHs). Right: vertical meridional section corresponding to the surface field (PL: pulse line in trade, TI: trade inversion).

As MPHs move, decelerating towards the tropics and meeting continents whose relief causes the air masses to turn, slow down or locally accelerate, interlock, halt or be compressed - AAs are formed and are ceaselessly sustained by new MPHs. Anticyclonic agglutinations, both oceanic and continental, are characterised by low-level anticyclonic stability and divergence (spreading), subsidence in upper layers (mostly between latitudes 20-30°N and S) atop low-level AAs, stratification between layers with different origins and characters, and sometimes a lack of precipitable water.

Sec. 3.3]

Oceanic anticyclonic agglutinations

55

3.3 OCEANIC ANTICYCLONIC AGGLUTINATIONS

Discovered originally on marine pressure charts, these have been ‘statistically’ defined and allotted names: the ‘Azores’, ‘Saint Helena’, ‘Hawaii’, ‘Easter Island’ and the ‘Mascarene’ Highs/Anticyclones. They are located on the eastern edges of oceans, as a result of MPHs encountering continental obstacles in their paths. The AA of the southern Pacific (‘Easter Island’ High) is highly representative. Not only does the Andean cordillera form a barrier from 50°S onwards, becoming impassable at 40°S, but as it curves westwards in Peru it increases compression and further slows the northward flow (Figure 2.6 and Photos 13). Similarly, the Rockies present a formidable barrier to cold air, from Alaska to southern Mexico, halting any eastward movement towards the Hawaiian AA, channelling air towards the Equator. This is also the case with the less challenging Great Escarpment in Namibia and the ‘Saint Helena’ cell (Photos 7). In the case of Australia, relief (for example, the Great Dividing Range in the east), is more modest; so the Indian Ocean ‘Mascarene’ cell is comparatively less well defined. In the North Atlantic, the mountain barrier is less efficient, with possibilities for outflow north of the Alps, towards the Mediterranean, and across Africa south of the Atlas Mountains, especially in winter when MPH trajectories are found further south (Figure 2.7). The movements of sea surface waters are also traditionally associated with anticyclonic rotation. Thus, the oceanographer Voituriez (2007) states that the Gulf Stream of the North Atlantic “is a marine current driven by winds associated with the Azores anticyclone, which blow clockwise”, just as the Kuroshio of the North Pacific is associated with the Hawaiian cell. This ignores the fact that, if we retain the traditional definition of this “centre of action”, a subsidence coming from the upper layers in the troposphere would be incapable (if it reached the surface of the ocean) of pushing along these immense marine circulations at the surface, and neither could it explain the direction of the rotation. And the main thing it ignores is exactly that most powerful of phenomena, the phenomenon at the origin of the AA: the movement of MPHs. It is these that, arriving at the eastern edges of the continents from across the oceans, exert the most powerful motive forces (through tension or stress) on the air above the sea water: a force much more considerable than that exerted by the already warmed (less dense) trades. Also, it is well known that these currents move, not continuously, but in successive pulses, corresponding to the impulses exerted by MPHs. Thus, the Gulf Stream, impelled by MPHs from the American continent, crosses the ocean “almost in the middle of the North Atlantic” (Voituriez, 2007), its encounter with Europe dividing it into two branches, one heading north (North Atlantic Drift and Norwegian Current), and the other south (Canaries Current) and then east (North Equatorial Current). Compression resulting from encounters with relief and the consequent rise in pressure on eastern margins, the large volume of air transported, the channelling and acceleration (Venturi effect) - all these maintain the force that draws the air across the water, and ensures the movement of the surface water towards the Tropics. So the Rockies and the Andes serve to greatly strengthen air currents in the lower levels, and to give sea surface currents (Californian and Humboldt) in the Pacific

56

Anticyclonic Agglutinations (AAs)

[Ch. 3

unequal amplitude, especially in the South Pacific. The Eckman diversion thus draws surface coastal waters out into the open ocean, and encourages the upwelling of waters from the deep. The vast gyres of marine drift currents (first from west to east, then meridionally, with associated cold currents and upwellings, and then from east to west in the tropical zone) are thus organised, the initial impulsion coming from the MPHs as they cross the oceans.

3.3.1

Seasonal migration

Oceanic AAs exhibit a seasonal ‘migration’, real in latitude and apparent in longitude, reflecting variations in intensity of MPHs.

3.3.2

Migration in latitude

Migration in latitude is a real phenomenon, and is the distant consequence of radiation, conferring variable strength upon MPHs as the seasons pass (note that a powerful MPH has a more meridional trajectory - the less powerful, the less meridional).

•

•

In winter, the greater dynamism of MPHs is accompanied by an increase in pressure and a general displacement of MPH and AA trajectories towards the tropics, along with a general acceleration of trade wind circulation in the relevant meteorological hemisphere. Meridional trajectories of winter MPHs create ‘additional’ AAs, as revealed by pressure data. In the North Atlantic, the statistical ‘Bermuda cell’ stems from the direct descent of MPHs through North America between the Rockies and the Appalachians into the Gulf of Mexico. The huge Pacific space is actually shared by two AAs, even in summer, but with a marked reduction from June to September (Terada and Hanzawa, 1984): to the east, the ‘Hawaiian’, and to the west the ‘Philippine’, evidence of the vigorous nature of Asian MPHs descending, especially in winter, into the South China Sea. In summer, the strength of MPHs is less, and their trajectories, now not so meridional, move the agglutinations away from the tropics, lessening the dynamism of tropical circulation. Now anticyclonic cells appear generally more individualised on the eastern rims of oceans. This is a result, synoptically, of diminution or lack of direct MPH trajectories on western rims.

3.3.3

Migration in longitude

Longitudinal ‘migration’ of statistical anticyclones is only apparent, and is merely an artifice of mean pressure measurements: the ‘disappearance’ during the summer of the westward ‘Bermuda cell’ extension of the ‘Azores’ AA might be interpreted as a

Sec. 3.4]

Seasonal and/or temporary anticyclonic agglutinations in mid-latitudes

57

slippage eastwards. Over the Indian Ocean, in the absence of any telling orographical feature, the translation of the so-called ‘Mascarene’ cell (in reality the core of the most powerful MPH, since observation is based upon synoptic-scale charts) is the most striking, moving from longitude 60°E, near Reunion, in the southern winter, across to 89°E in summer (U.S. Navy, 1976; Taljaard and Van Loon, 1984). This variation is not to be taken as proof that a real migration of the (statistical) cell has taken place, however. What it does reveal is the vigorous nature of winter MPHs in the southern hemisphere, moving up the eastern flanks of Africa and Madagascar to supply the maritime trades, some of which will contribute to the energetic Indian monsoon (Photos 12). In summary, ‘migrations’ of oceanic anticyclonic cells are normally artefacts of pressure charts, and their ‘motion’, meridional or zonal, is in fact that of the MPHs crossing and constructing them, and causing their possible extension. In any case, these cells, designated on the basis of mean values, cannot migrate far from their statistical ‘locations’, since their synoptic position is determined by dynamical conditions (the movement of MPHs) and geographical facts (notably relief).

3.4 SEASONAL AND/OR TEMPORARY ANTICYCLONIC AGGLUTINATIONS IN MID-LATITUDES Over oceans, AAs have a ‘permanent’ character, meaning that they are constantly regenerated by the arrival of new MPHs and by stable geographical conditions which cause them to merge. AAs are not only found over oceans, since they owe their existence to the likelihood of permanent or seasonal displacement or blockage of MPHs. According to the dynamical school, which accords a pre-eminent role to the upper levels, such situations result from a “blockage in the upper layers caused by a warm ridge at altitude”, considered to be an “essential condition for the maintenance of the high-pressure cell” (Besleaga, 1990). A structure within which warm air lies above a layer of cold air evidently does not favour updrafts, and therefore hinders rain. But the explanation is incoherent: by what physical process (or - again miracle) can warm air constitute a condition, when it is located in the upper layers where the density is much lower, and how can it explain the presence and renewal of cold air, which is much denser, lying beneath it at lower levels? This ‘explanation’, like that put forward by the climatological school with its ‘travelling’ Azores cell, is similarly incoherent. Attentive analysis of MPH trajectories clearly explains the genesis not only of winter agglutinations but also of their summer counterparts; such phenomena can occur anywhere, whatever the nature of the substratum, because their origin is dynamical. As well as with geographical conditions, their formation and main­ tenance are associated with very powerful and/or very frequent MPHs, and/or with blocking by MPHs on north-south trajectories, the latter being denser and responsible for waves of cold (absolute or relative).

58

Anticyclonic Agglutinations (AAs)

[Ch. 3

3.4.1 Winter anticyclonic agglutinations In winter, the greater strength of MPHs and the thermal contribution of colder land masses continents are added to the effects of relief, encouraging the formation of temporary Aas of variable duration. Therefore, anticyclonic stability brings with it, for varying amounts of time, clear skies, contrasting temperatures and an absence of rain or snow.

The ‘Siberian’ anticyclone The high mountains of Asia, stretching latitudinally from Anatolia to Tibet, and longitudinally from the Urals to the Tien Shan and Verkhoyansk Range, stop - or considerably hinder - any southward or eastward flow. So the movement of MPHs coming directly from the Arctic on the Russian and Siberian trajectories is hindered, while the continental substratum experiences (very) low temperatures in winter. Thus is formed, in winter, the famous ‘Siberian anticyclone’, or ‘Sibero-Mongolian anti­ cyclone’ seen on pressure charts, with synoptic values often greater than 1050 hPa. In the same way, over North America, the charts show the ‘Manitoba anticyclone’. This powerful Asiatic agglutination, maintained by newly-arrived MPHs and by the coldness of the continent, spreads considerably, spilling southwards. South of 55°N, in the Turan basin, which contains the Caspian and Aral Seas, there is a veritable cul-de-sac for MPHs, and the area is anticyclonic in both winter and summer. To the east, a narrow outlet leads through Dzungaria towards Mongolia and the Gobi Desert (Photos 14). This is the exit taken by the ‘yellow wind’, which has carried eastwards the loess from the moraines of the Last Glacial Maximum, and still brings dust clouds (‘dry fog’) episodically across China. An easier exit for the overspill of the local AA, though in the opposite direction to that of MPHs, is westwards, north of the Caucasus and towards the Black Sea and the eastern Mediterranean. In winter, the MPHs’ greater depth permits overspill of the upper part of the AA between the Elburz Mountains and the Afghan heights, across the Iranian plateau by way of the least elevated Dasht-e-Kavir, Dasht-e-Lut and Seistan into the Indian Ocean. This is the time of the sansar, sarsar or shamsir, the ‘wind of deathly cold’. In summer, the lesser depth is compensated for by the heat, and by the attraction exerted by the depressions in the south of Arabia and Pakistan. The seistan, named after the province it crosses in south-east Iran, blows strongly southwards; it is sometimes known as the ‘120-day wind’, active from the end of May or beginning of June, and dying away at the end of September, and contemporaneous with the Indian monsoon above which it rises (i.e. above the Inclined Meteorological Equator, IME). Traditionally, the theory holds that Eastern France can be covered in winter by an extension of this anticyclone ‘bringing air from Siberia’, spreading across central Europe and reaching the Atlantic. In fact, it is the arrival and contribution of new MPHs on the ‘Scandinavian’ or ‘Russian’ trajectories that cause this westward extension: the density of the cold air over Siberia and European Russia does not allow the less cold and therefore less dense MPHs to pass eastwards.

Sec. 3.4]

Seasonal and/or temporary anticyclonic agglutinations in mid-latitudes

59

Figure 3.3 shows the pressure field on 10 February 2008, indicating four MPHs on the western side of the agglutination over Western Europe and the near Atlantic, where polar air is channelled towards the tropical zone. Anticyclonic stability in this immense agglutination can manifest itself as low clouds or radiation fog (here, over Russia). Clear slots and the absence of wind nevertheless allow diurnal temperatures

6 February 2008

7 February 2008

8 February 2008

9 February 2008

Photos 16 Formation of a Eurasian anticyclonic agglutination by successive arrivals of new MPHs, from 6-9 February 2008 (NOAA, Visible, 12h UTC, Satmos). On 6 February, pressure is already high over Russia (higher-pressure AAc, at 1050 hPa, see Figure 3.3)), and an MPH on the 'Scandinavian' trajectory (AMP1), blocked at its eastern edge, moves down across Western Europe. Re-supplied on 7 February, it will extend during the following days into the Mediterranean and the Atlantic. On the same day, a new MPH (2) arrives from southern Greenland, is unable to move east and will be progressively integrated on 8 February into the Atlantic edge of the agglutination. On 8 February a new 'American' MPH (3) arrives in its turn, and spreads southwards, blocked by high pressures. Yet another MPH (4) comes to feed the AA.

60

Anticyclonic Agglutinations (AAs)

[Ch. 3

Figure 3.3 Synoptic situation on 10 February 2008 at 00 h UTC (from Deutscher Wetterdienst). Added to the diagram are: AAc = continental anticyclonic agglutination, western extension of the ‘Siberian’ anticyclone - AMP1 ... AMP2 = MPHs entering the agglutination (date of the observation in brackets: (8) = February 8 (cf. Photos 16). This agglutination will be constantly supplied until 21 February 2008.

to rise rapidly, while the nights are seasonally cool or cold, with the chance of heavy frosts as the ground radiates heat away.

Duration of agglutinations

The duration of AAs in winter is variable. They can last for a few days or persist for over a month. This was particularly well demonstrated during winters in Western Europe from 1988 to 1993, when an anticyclonic cap extended across the Atlantic, the Mediterranean and Western Europe (Leroux et al., 1992b, 1994d). By way of example, we see in Figure 3.4, for 15 December 1988, over the eastern Atlantic and Europe and stretching beyond right across Asia to the Pacific, the presence of four MPHs, conjoined, but still obvious individuals, on the configuration of the pressure field. Two of these are of the American-Atlantic trajectory (with A2 preceding A4), and the other two are cooler MPHs coming straight down from east of Greenland on the Scandinavian trajectory. Al is the oldest, and has been reinvigorated several times, especially by A3 which is younger and faster. The continental anticyclonic

Sec. 3.4]

Figure 3.4

Seasonal and/or temporary anticyclonic agglutinations in mid-latitudes

61

Surface chart for 15 December 1988, 00 h UT (after EMB).

agglutination (AAc), which is itself continually crossed by Russo-Siberian MPHs, and is coldest at its centre, prevents the eastward zonal displacement of Scandinavian MPHs, which then stand in the way of others coming from the west. Succes­ sive MPHs interlock and slip slowly southwards over the Atlantic and the Mediter­ ranean, and thence to North Africa, reinforcing the agglutinations over them (the ‘Azores’ and ‘Saharan’ AAs). Table 3.2 Location of the centre of the anticyclonic agglutination from 1 December 1992 to 31 March 1993: number of times (days) and frequency in % (121 days).

Europe

December January February March Total

1992 1993 1993 1993

23 9 19 15 66

74% 29% 68% 48% 55%

Atlantic

Mediterranean

3 19 0 8 30

10% 61% 0% 25% 25%

5 3 9 8 25

16% 10% 32% 25% 20%

62

Anticyclonic Agglutinations (AAs)

[Ch. 3

Table 3.3 Dynamics of long-lasting winter agglutinations over western Europe, 1989 to 1993. The table shows dates, duration, total of MPHs merging with them, number of MPHs from the American and Scandinavian trajectories, MPH frequencies (in the form of 1 MPH/n days), and extreme observed values for pressures at the centre of MPHs. Period

Duration (days)

Total MPHs frequency

Amer, traj.

Scand. traj.

Pressure hPa

5 Jan-22 Feb 1989 19 Dec 198925 Jan 1990 12 Jan-7 Feb 1991 11 Jan-9 Feb 1992 10 Dec 199210 Jan 1993 13 Jan-20 Feb 1993

49

30 1/1.6 23 1/1.7 15 1/1.7 17 1/1.8 26 1/2.6 30 1/1.3

23 1/2.1 17 1/2.3 13 1/2.0 10 1/3.0 16 1/2.1 20 1/2.0

07 1/7.0 08 1/4.8 03 1/8.6 07 1/4.3 10 1/3.3 10 1/3.9

1025 to 1045 1025 to 1040 1030 to 1050 1025 to 1045 1030 to 1050 1030 to 1050

39 26

30 33

39

d d d d d

d

d d d

d

d d

d d d

d d d

During the winter of 1992-1993, the centre of a very large AA formed in this way was to be found preferentially over Europe (Table 3.2), at the junction of the Atlantic and the Scandinavian MPH trajectories. Such vast anticyclonic areas have occasioned near-desert conditions, and can be extremely stable, one result being heavy urban pollution, and another the absence of snowfall on mountains. Durations of such long, rainless periods of anticyclonic stability have been (Table 3.3): - 77 days, in three periods, the longest being 49 days, in the winter of 1988-1989; - 86 days, in 4 periods, the longest being 39 days, in the winter of 1989-1990; - 52 days, in 4 periods, the longest being 26 days, in the winter of 1990-1991; -111 days, in 5 periods, the longest being 30 days, in the winter of 1991-1992; - 105 days, in 5 periods, the longest being 39 days, in the winter of 1992-1993.

Climatic consequences often show contrasts. In January 1995, for example, a large number of very strong MPHs caused floods from Brittany to north Germany. In northern France, between 17 and 29 January, 313.6mm of rain were measured in Brittany, 343.8 mm in the Ardennes, and 406.4 mm on the Ballon d’Alsace mountain; this was 5 to 9 times greater than normal, with record monthly rainfall (since 1949) for January being recorded at 21 weather stations (Coudert, 1995). The MPHs, responsible for intense carriage at their leading edges of humid Atlantic air, later agglutinated further south over the Atlantic (in the ‘Azores’ AA) and the western Mediterranean, bringing a severe and long-lived drought to the Iberian peninsula, Morocco and the Canary Islands (Table 3.4).

Sec. 3.4]

Seasonal and/or temporary anticyclonic agglutinations in mid-latitudes

63

Table 3.4 Winter rainfall 1994-1995 for the Canaries, at Mazo (La Palma) and Los Rodeos (Tenerife): monthly totals.

Stations

November 94

December 94

January 95

February 95

Mazo Los Rodeos

8.6 mm 11.8 mm

5.4 mm 4.2 mm

4.5 mm 10.6mm

5.3 mm 8.8 mm

3.4.2 Summer anticyclonic agglutinations

In summer, less powerful MPHs proceed freely eastwards, unless relief hinders them. But an anticyclonic agglutination may also form under particular circumstances, for example in the Mediterranean with its surrounding relief, or above a meeting point of trajectories, or even if the pressures and/or frequencies of one of more MPHs are abnormally high.

High pressures over the Mediterranean In winter, the Mediterranean area experiences its rainy season, as MPHs or fragments of MPHs cross the basin. In summer, however, precipitation is rare, with relatively high temperatures associated with high pressure. This anticyclonic situation is not due, as the traditional theory holds, to some hypothetical extension of a “ridge from the Azores”. Over the Mediterranean, air riding in on the MPHs is under pressure as a result of the blockade it has already experienced when encountering the relief which determines the pathways open to the lower layers. In the case of the western basin, channelling occurs between the Pyrenees and the Alps, between the Alps and the Dinaric Alps, and through the breaks in the Spanish sierras, spilling as a sheet over the Iberian plateau, especially in winter when MPHs are deeper; the zonal barrier of the Atlas Mountains prevents any direct low-altitude flow into the Sahara west of Tunisia. In the eastern Mediterranean basin, MPHs coming between the Balkans and the Anatolia-Taurus mountains have a straighter track and are therefore energetic, especially in winter when, further to the north, continental agglutinations dilate and ‘overflow’. Pressure is thus generally high across the Mediterranean and its environs and, as a result of the origins and trajectories of the MPHs replenishing them, the high-pressure areas encountered here form a specific anticyclonic entity. These Mediterranean highs flow out across North Africa, away from the Atlas Mountains and, according to mean data, form a so-called ‘Mediterranean-Saharan’ cell in summer, becoming more ‘Saharan’ in winter because of the reinforcement of the MPHs and the general shift southwards (Leroux, 1983). Further east in the winter, the flow of MPHs between the high ground around the Red Sea and the Taurus-Zagros chain creates the seasonal synoptic ‘Arabian’ AA, which sustains the trade-wind circulation across the northern Indian Ocean.

64

[Ch. 3

Anticyclonic Agglutinations (AAs)

The heatwave and drought of summer 2003 in Western Europe

The heatwave of summer 2003, when little rain fell, was a remarkable event for western Europe, not least because of the many people who succumbed to the heat. The partisans of the ‘global warming’ scenario hastened (automatically) to attribute the heatwave, drought and forest fires near the Mediterranean to the greenhouse effect (i.e. to the human factor). In reply to the question “What mechanisms lie at the origin of this heatwave?” Meteo France offered: "The presence of the Azores anticyclone ... with an extended ridge ... and above it, a layer of unusually warm air from the south".

1 August 2003

5 August 2003

3 August 2003

7 August 2003

Photos 17 Formation and maintenance of the anticyclonic agglutination over Western Europe, 1-15 August 2003 (NOAA, Visible, 12 h UTC, Satmos). For convenience, images are shown for two-day intervals, since the situation evolves relatively slowly during the period in question

Sec. 3.4]

Seasonal and/or temporary anticyclonic agglutinations in mid-latitudes

9 August 2003

11 August 2003

13 August 2003

14 August 2003

15 August 2003

Photos 17

(Continued).

65

[Ch. 3

66

Anticyclonic Agglutinations (AAs)

•

Bringing in the ‘Azores anticyclone’ immediately puts us back into the realms of magic: more meteorological animism, incomprehensible when it comes from an official meteorological institution! The anticyclone in question has an ‘extended ridge’ which seems to have appeared out of nowhere and we are told nothing of its origin, or why it extends as far as about 60°N, an amazingly high latitude for a ‘subtropical’ anticyclone from the Azores! This anticyclone has above it (indicating its discoid nature, which is correct) “a mass of very warm, very dry air from the south, coming in through the Pyrenees ... both near the surface and at altitude" (cf. Meteo France website). For what physical reason could this flux have established itself, and survived so long? And what is the significance of the (never demonstrated) presence of warm air advected at altitude, given that warm air cannot reach the ground, as the inversion above the low-level anticyclone shows? To the questions: “Why did this cap of warm air move so far to the north, and why did it remain fixed there for so long?” Meteo France replied: "There is no ready answer ... many mysteries remain" (!). Surprising ...

•

•

•

What really happened, then, during the summer of 2003 or, to be more precise, during the first two weeks of August when the weather was at its hottest? To answer this question, we need only study the satellite images (Photos 17) and synoptic charts (Figure 3.5), where the unusual frequency (for the season), of meridional-trajectory MPHs is evident. From 1-17 August, 12 MPHs (Photos 17 A to L), exhibiting fairly high pressures for this time of year (1020-1025 hPa), made their contributions to the maintenance of the AA: seven of them followed the ‘American-Atlantic’ trajectory, and five the ‘Scandinavian’, with some ‘American’ MPHs being reinforced en route by cold air from Greenland, an example being MPH D on August 4 (D’4). The Scandinavian MPHs are colder, because Scandinavia experiences lower temperatures at this time than are normal in the rest of Europe (Pielke Sr. and Chase, 2004), and they form a barrier to MPHs coming from the west, for example the enormous MPH B blocks the eastward path of MPH A from 1 August. Another example is that of MPH H hindering the progress of MPH F, the latter having merged with MPH E on 8 August. MPHs of various origins are merging over Europe, which sits squarely beneath the meeting point of the two trajectories, and their encounter maintains high pressures on land as well as at sea. The agglutination was particularly homogeneous from 3 August until 12 August, and maximum temperatures occurred during this period of great anticyclonic stab­ ility. MPH I, passing south of Greenland on 9 August, was considerably reinforced on its southward trajectory, moving directly over Great Britain on the 13th, and covering Northern Europe on the 14th, from France to Denmark. On the 15th it was moving mainly eastwards, and on the 17th it stretched from the Pyrenees to the Black Sea. The temperature then fell in France by more than 10 degrees C and precipitations occurred. The heatwave was now ended, and the period that followed it, like the autumn, was relatively cool.

Sec. 3.4]

A

Seasonal and/or temporary anticyclonic agglutinations in mid-latitudes

1-5 August 2003, surface

67

B 5-9 August 2003, surface

D

12-17August 2003, surface

Figure 3.5 Surface evolution of MPHs from 1-17 August 2003 over the North Atlantic and Western Europe, (a) 1-5 August, (b) 5-9 August, (c) 9-12 August, (d) 12-17 August. From European Meteorological Bulletin surface charts. (Note: A , B , C ...: leading edge of an MPH.)

There is nothing exceptional about this type of situation, since it occurs regularly at lower latitudes during this season in the eastern Atlantic and over the Mediterranean. There is nothing really out of the ordinary here, the peculiarity of this situation being the unusual extension of this anticyclonic dome, in both space and time. Also, barometric situations like this, involving such temperatures, are not rare. The summers of 1998, 1995, 1994, 1985, 1983, 1976, 1964, 1947, 1921 (with only a quarter of the usual rainfall), 1901, 1900 etc. were equally hot, or locally even hotter. In France in 2003, 70 records - out of 180-were broken, but all-time records remained unchallenged. The national record is still held by Toulouse where,

68

Anticyclonic Agglutinations (AAs)

[Ch. 3

on 8 August 1923, a temperature of 44 C was recorded. The (drier) summer of 1976 is remembered everywhere in France as the symbolic ‘summer of drought’. The summer of 2003 created new national records in Portugal, Germany and Switzer­ land. In Britain, that summer is still outranked by those of 1976 and 1995, when the very hot spells lasted longer. Consequently, and indisputably (as long as we observe actual phenomena), the cause of the heatwave was the presence of an anticyclonic agglutination. It was absolutely not the result of an extension of the ‘Azores’ anticyclone, nor of air coming up from the south, either at altitude or at the surface. On the contrary, it was caused by: • •

the concentration and deceleration of low-level anticyclonic air from the north, i.e. coming from the Arctic and transported in the form of an MPH; and the rapid diurnal warming of that air, at high pressure.

The floods and the heatwave of summer 2007 In Europe the summer of 2007 was marked by strong thermal and pluviometric contrasts, notably in July. From 15 to 21 July, a period representative of thermal contrasts across Europe, the temperature was below normal by from 1° to 3°C in Scandinavia, Great Britain and the west of France, with the western part of the Iberian Peninsula experiencing a thermal deficit of 5°C. However, central Europe and the Mediterranean recorded higher values than normal (by 7°C) (NOAA, CPC), exceeding 40°C, or even 45°C in Italy, Greece, Bulgaria, Romania, and Hungary; this heatwave was accompanied by a severe drought and forest fires. In Western Europe, rainfall was higher than normal: in two-thirds of northern France there was more than 40% more rain than usual, and in the departement of the Manche from 2-3 times more, while there was a marked lack of rain in the south. Great Britain, where rainfall in May had been twice as heavy as normal, received up to four times the normal in June. The result, from the end of June and through July, was epic flooding. It was no surprise that Meteo France attributed the situation to “the whims of the Azores anticyclone ... located far from Europe, in mid-Atlantic around 35°N” (cf. www. meteofrance/dossiers). On the other side of the Channel, the culprit was the atmospheric jet stream, responsible for “torrential rainfall over the United Kingdom and torrid temperatures in the Mediterranean” (cf. Vaughan, President of the Royal Meteorological Society, Guardian, 2 September 2007). The same old mystical recipes, explaining nothing! Observation shows that bad weather is associated with the intervention of MPHs on meridional trajectories, and fine weather to the consequent agglomeration of those same MPHs over central Europe and the Mediterranean. Photos 18 and Figure 3.6 illustrate a situation that was repeated during the period June-August 2007. The AA over central Europe, spilling onto the Mediterranean (Figure 3.6) was formed by the deceleration and integration (especially because of the presence of the Alps) of five MPHs: MPH1 (Al), established by 13 and 14 July and reaching the Black Sea by the 15th; MPH 2 (A2), a powerful mass with a meridional trajectory

Sec. 3.4]

16 July 2007

20 July 2007

Seasonal and/or temporary anticyclonic agglutinations in mid-latitudes

69

18 July 2007

Figure 3.6 Dynamics, 15-20 July 2007

Photos 18 The formation of an AA over central Europe and the Mediterranean by MPHs on the meridional trajectory, 16 20 July 2007 (NOAA, Visible, 12 UTC). The anticyclonic stability, clearly visible as the absence of cloud formations, is responsible for the heatwave in central Europe and the Mediterranean.

Figure 3.6 Formation of the AA over central Europe and the Mediterranean, 15-20 July 2007 (based on the synoptic charts of the Deutscher Wetterdienst, 00 UTC).

70

Anticyclonic Agglutinations (AAs)

[Ch. 3

(with pressure greater than 1020, even 1025 hPa on the 18th), and clearly visible on the photograph for July 16: descending southwards over the Atlantic and eastwards across Europe to join the agglutination on 19 and 20 July; two MPHs (A3 and A5) descend directly upon the AA via Scandinavia, and one MPH (A4) reinforces the eastern flank of the AA. MPHs 2, 3 and 5 contribute to a concentration of bad weather over the North Sea, the British Isles (D) and western France, before integrating with and reinforcing the AA. On 20 July, a new MPH arrives in Western Europe.

3.5 CONCLUSION

MPHs are responsible for ‘bad weather’ all the time they are moving and creating cyclonic circulation at their leading edges (cf. Part II). However, as soon as they have been slowed, or halted, and when they merge, they impose their anticyclonic stability. This consequence has long been recognised by sailors in the so-called subtropical AAs, feared because of their calms and lack of rain. These oceanic AAs have a ‘permanent’ character, meaning that they are in a constant state of formation, because of the dynamic of MPHs and particular geographical conditions. However, AAs can form anywhere, over both land and sea, as a result of the coalescence of several MPHs, merging for variable lengths of time, seasonally or temporarily. Weather associated with agglutinations of MPHs is controlled by raised pressure and anticyclonic stability, which bring heat, drought and pollution. The higher the pressure, the greater the molecular conduction and infrared absorption; air cannot rise and, at near-ground levels, becomes overheated (for the same quantity of solar energy received), especially when winds are light or non-existent. The heat brings about a marked diminution in relative humidity, i.e. the air becomes very dry, and more so if water vapour does not penetrate the anticyclonic area; the natural greenhouse effect, principally associated with water vapour, is considerably reduced, allowing more energy to reach the ground during the daytime. The absence of clouds also offers optimal insolation. So there are wide temperature contrasts (with much increased diurnal thermal amplitude), and the days become abnormally hot, or ‘spring-like’ in winter. Nights are cold, with frequent frosts and radiation fogs. In summer, the heat builds up to gradually to create what are traditionally known as ‘dog days’, especially in urban areas which are less ‘ventilated’, hotter, dryer and polluted: emissive gases are an aggravating factor. At the same time, the anticyclonic character limited to the lower layers and the absence of horizontal and vertical air movements concentrate pollution near the ground (below an inversion layer at about 1000 metres), while the strong insolation accelerates photodissociation (the production of ozone, encouraging the temperature to rise in the lower layers). Precipitation is hindered, or even halted, and drought results in winter as in summer. In winter, there is no more snow at altitude when the AA covers the mountain massifs. In summer, the increasingly dry conditions caused by the heat bring widespread forest fires (cf: as illustrated during Australia’s dramatic forest fires in

Sec. 3.5]

Conclusion

71

2009, Ed. Note). Drought can also cause the raising of fine particles from the ground, as occurs in North America in the ‘Dust Bowl’ of the Great Plains of the Midwest. Vast clouds of dust are borne along at the leading edges of MPHs crossing the agglutination, travelling as far as the east coast. The ceaseless procession of MPHs causes permanent, seasonal or temporary anticyclonic agglutinations (key elements in the climate of temperate and subtropical regions) to be buffer zones for meridional exchanges, marshalling extratropical circulation before feeding, in a diluted manner, air brought by MPHs to circulate in the trade winds of the tropics.

4 Tropical circulation

Tropical circulation is the uninterrupted prolongation of circulation at mid-latitudes. The passage from one to the other occurs in the lower layers, either directly by way of an MPH, or more slowly through the intermediary of a ‘subtropical’ anticyclonic agglutination (AA). The mass transported by MPHs is transformed into the trade flux, and thereafter into the monsoon flux. 4.1 A LOOK AT TROPICAL CIRCULATION

Tropical circulation has been known about, and defined to a greater or lesser extent, for a long time. In ancient times, much was known empirically about the coastlines of the Indian Ocean, from eastern Africa to Malaysia, with their respective mon­ soons; and more scientifically, knowledge of it predated that of temperate circulation, as the first chart showing oceanic winds, prepared by R. Halley (1686) from observations by navigators, dealt with the trades of the tropical zone. This antiquity doubtless explains why definitions of tropical fluxes, through superposition and blending of interpretations, have become particularly ambiguous. When navigators who were accustomed to the (MPH-associated) variability in direction of the winds of temperate zones approached the Tropics, they were struck by the ‘regularity’ of the winds in those parts. Indeed, the French term for the trade wind, alize, is derived from Old French alis, meaning smooth or even. Having been long thought of as a uniquely oceanic phenomenon (a view influenced by sailors), the trades are the ‘winds of passage’ to America: passato to Iberians, whence the German Passatwind, normally shortened to Passat. The commercial overtones live on in the English term ‘trade’. Although there is little ambiguity as far as the trades are concerned, they are described in a simplistic and erroneous way as north-east winds in the northern hemisphere and south-east winds in the southern, except when the term is used in a purely oceanic sense; though it is certainly wrong to call them the winter monsoon. The term monsoon (mousson, monpao, monsun) is rich in meaning, being used either in the singular or the plural as a designation for a season, a flux or a wind system:

74

Tropical circulation

•

a (navigational) season, based on wind direction, from the original Arabic mausim or Malay monsin, and meaning just that; a ‘rain-bearing’ flux, especially in India or in Africa, where they speak of monsoon rains (though the experience of Somalia shows us that the presence of the monsoon is not necessarily synonymous with rain); a wind system, showing a reversal in direction between winter and summer; in this hypothesis, the aerological year consists of two monsoons, the summer monsoon and the winter monsoon, the latter being in reality, i.e. genetically, a trade wind.

•

•

[Ch. 4

This multiplicity of concepts (cf. Leroux, 1975, 1983) is responsible for the incorrect usage of the term ‘monsoon’. ‘True’ and ‘pseudo-’ monsoons are evoked in the lower layers and also at altitude (each time there is a seasonal reversal in circulation, or even a marked change in direction), both in the Tropics and also elsewhere, and even into the Arctic. Many variable indices have been thought up to characterise mon­ soon regions (e.g. monsoon index, in this case limited to the borders of the Indian Ocean: eastern Africa, south Asia, and Australia (Ramage, 1971; Das, 1986). As well as these Asian and Australian monsoons, the World Meteorological Organisation (WMO, 1998) has added what it terms a ‘North American monsoon' formed by the advection in summer of the Atlantic trade from the Gulf of Mexico towards the interior of the continent, at that time partly free of MPHs coming south directly through the Great Plains (cf. Part II). Explanations are as varied as they are debatable. The most widely heard is that the monsoon(s), especially in Asia, is like a large-scale breeze mechanism, with two alternating air currents: the continent ‘breathes out’ in winter, and ‘breathes in’ in summer. This thermal concept, put forward long ago by Halley (1686), is still much subscribed to in spite of its simplistic nature: “in winter, the thermal low is ‘replaced’ by the Siberian anticyclone... the air expelled by this anticyclonic cell is very cold and dry. Having undergone the foehn effect on the slopes of the Himalayas, its humidity is lessened and at the same time it is considerably warmed, finally resulting in a practically cloud-free air mass” (Triplet and Roche, 1988). So the Indian ‘winter monsoon’ (in reality a trade wind) has its origins in the Siberian anticyclone, meaning that the Tibetan-Himalayan barrier is crossed by cold Siberian air, with a. foehn effect over India! The multiplicity of ideas and the insufficiency of explanations (or even gross errors as above) require us to define tropical fluxes according to uniquely aerological criteria, i.e. as a function of the relationship between the pressure field and the wind field.

4.2 PRESSURE AND WIND FIELDS OVER THE TROPICS The tropical zone is flanked to the north and south by ‘subtropical’ anticyclonic agglutinations (AAs), between which stretches the zonal corridor of intertropical

Sec. 4.2]

Pressure and wind fields over the tropics

75

lows (ITL) occupied by the meteorological equator (ME), the line of symmetry towards which converge the contents of northern and southern MPHs. This zone experiences two types of flux unique to the tropics: the trades and the monsoon. The fundamental distinction between these is one of trajectories: the trades, flowing predominantly westwards, do not cross the meteorological equator; the monsoon, with its strongly eastward tendency, does. These flows occupy specific pressure fields, marshalled in the lower layers by tropical and extratropical factors. As a function of the conditions relating to their supply of air from outside the tropics (seasonal variation and flow modalities), AAs determine the initial circulation of the trades, separated by trade discontinuities. Factors within the tropical zone result from the Sun’s yearly motion, which displaces the ITL and the meteorological equator, bringing about the veering motion when a trade wind crosses the geographical equator to become a monsoon. Schematically seen, the tropical pressure field shows two typical surface topographies, symmetrical and asymmetrical in relation to the equator, combining dynamical and geographical factors (Figure 4.1):

• Trade circulation On either side of the Equator, two AAs stand almost opposite each other, more or less staggered in longitude according to geographical conditions. Between the AAs, pressure is lower, and at the bottom of the equatorial thalweg (the axis of the ITL) runs the meteorological equator. The characteristic low pressures of the medial corridor, and its seasonal migration, depend closely upon the dynamics of the agglutinations: the ME shows only a slight annual migration, not moving far from the geographical equator. This topography is found mainly over oceans where

Figure 4.1

Surface pressure and wind fields in the tropical zone: trades and monsoons.

76

Tropical circulation

[Ch. 4

‘permanent’ AAs form, as the thermal behaviour of the water does not permit any very marked deepening of the low-pressure corridor, whose character is essentially relative and dynamic. This kind of pressure field, where lines of flux are parallel to the isobars, is typical of trade-wind circulations on either side of the equator; • Monsoon circulation The intervention of the continental thermal factor alters this symmetrical distribu­ tion. The Sun’s yearly motion displaces maximum warming from one tropic to the other, the theoretical maximum displacement being 46°54' (cf. Figure 3B). The thermal factor can certainly affect the pressure field, though this is not possible over the oceans where AAs dominate and warming is never excessive. Over continents, however, the ITL corridor is locally very deep, and displaced by moving thermal lows, which distort the meteorological equator, lagging more or less behind the Sun’s zenithal motion. The trade wind coming from an AA thus crosses the geographical equator by way of a transequatorial pressure declivity between high pressure in one hemisphere and low pressure in the other. The Coriolis or geostrophic force, a function of the sine of the latitude, is nil at the Equator, and the initial trade wind can cross the Equator under the effects of inertia and of the pressure gradient. Progressive penetration into the opposite hemisphere, towards the axis of the thermal low which is moving away from the equator, increases the geostrophic force,

Photo 19 This visible-light image from one of the first satellites (ESSA 8, ASECNA, Dakar), taken over the Atlantic on 11 November 1964 at 12H 00 UTC, shows very clearly the change in direction over the Atlantic of the trade wind from the south, becoming the Atlantic monsoon as it crosses the equator in the direction of the south coast of West Africa.

Sec. 4.2]

Pressure and wind fields over the tropics

77

now of opposite sign: the flux is diverted eastwards as a monsoon (Photo 19), and the lines of flux are now perpendicular to the isobars (Figure 4.1). Genetically, a monsoon is therefore the extension into one hemisphere of a trade wind originating from an AA in the opposite hemisphere, or directly from an MPH, drawn across by a thermal low in the summer hemisphere. The term summer monsoon is thus tautological. Along the ME, the monsoon meets a (continental) trade wind from the opposite hemisphere, and the area of extension of this trade is considerably reduced. Satellite images (Photos 20 and 21) taken on 18-20 May 1995 underline the uninterrupted nature of the flow (Figure 4.2): MPHs agglutinate (note especially MPH 3 which merges on 20 May into MPH 1), and the pulse lines which mark out the progression of the MPHs across the tropics progressively become the trade wind circulation, which changes over the equatorial Atlantic into the monsoon circulation heading for western and central Africa, in the lower layers. Tropical circulation in the lower layers is organised into individualised pressure and wind units on a topo-geographical basis. The arrangement of these units depends on the distribution of oceans and continents, relief, and the thermal behaviour of the

Photo 20

Figure 4.2

Photo 20 19 May 1995, over the Atlantic, at 12H UT, Meteosat, visible (UTIS, CRODTORSTOM, Dakar). Figure 4.2

Meteosat image interpreted: 19 May 1995, Atlantic and Africa.

78

Tropical circulation

[Ch. 4

Photos 21 Southern MPHs feed the ‘St. Helena' cell, 18-20 May 1995, over the South Atlantic, and the southern maritime trade is born (Meteosat, Visible, 12 UTC, UTIS, CRODTORSTOM, Dakar).

surface below (for example, oceans, dense forests, or land masses), which give the trades and monsoons their particular characteristics. 4.3 THE TRADE WIND

A trade wind circulation occurs in a distinctly individualised unit, entirely within the same geographical hemisphere, and involves only the lower layers (Figures 4.3, 4.4, photos 22). In Figures 4.3 and 4.4, mass air transportation by MPHs and in AAs is shown differently from the lines of flux, as the resultant wind vector (mean wind) does not represent reality in the temperate and polar zones, where MPH-associated winds are very variable (cf. Chapter 2). Only in the tropical zone is this the case, as actual wind directions there are quite close to the mean directions. We observe as many trade winds as there are permanent or seasonal subtropical AAs, the circulations being separated (at the surface) by trade discontinuities (TD). TDs are often confused with relief, especially if the relief presents a north-south orientation as in the case of the Rockies and the Andes. Across southern Africa the Great Escarpment fixes the Cape TD just to the north of Angola; over West Africa, the TD, blocked for a short distance by the Atlas range, is finally freed to fluctuate around the coastline or occasionally to push further in across the continent, especially in winter. Over the eastern Pacific, the TD is first of all fixed by the relief of western Mexico, notably by the Sierra Madre del Sur; its displacement over the ocean is a function of the respective strengths of the Pacific and Atlantic trades. The Isthmus belongs in winter to the Atlantic trade unit.

Sec. 4.3]

Figure 4.3

9-

The trade wind

Tropical circulation and zonal vertical structure of winds in January.

,________________________________________ „_________________________________________________________ _____________ ».

w

w

w

c • zonal section of winds at 15°S

Figure 4.4

Tropical circulation and vertical zonal structure of winds in July.

79

80

Tropical circulation

[Ch. 4

15 August 1986 " Photos 22 Evolving situation over the North Atlantic, 13-16 August 1986 (after Meteosat Image Bulletin, ESA, visible, 12H UTC). MPH 'a’ spreads gradually south-eastwards, engendering the lower stratum of a new maritime trade flux. MPH 'b’, hindered on its eastward track, flows out over the Atlantic and agglutinates with the northern part of MPH 'a’. MPH 'c’ merges into this AA. On this synoptic scale, the outline of 'a’, or of 'a’ 4- 'b’, can in no way be identified with the 'Azores anticyclone’, whose definition is 'statistical’. This evolution of the situation over the North Atlantic shows the difference between the mean perception, expressed by Figures 4.3 and 4.4, and synoptic reality, which affirms that tropical circulation is far from uniform: it consists of successive and interlinked units of circulation, variable in both intensity and extent.

In each unit of circulation, the anticyclonic rotation of the AA determines the direction of flow, with a dominant easterly component, but from various directions; in the northern hemisphere, for example, the trades swing from northerly to north­ easterly, then easterly, south-easterly and finally southerly, in the direction of the temperate zone. This rotation is relatively constant, any shift from the mean depending upon the arrival of fresh MPHs in the AAs. Charts showing stability in wind directions (Mintz and Dean, 1952) show that the frequency of real directions deviating by less than 45° from the normal attains 60%, or even 80%, only in trade wind circulation. The anticyclonic agglutination, a buffer zone for meridional

Sec. 4.4]

The trades

81

exchanges, acts as a reservoir of air for redistribution, and regularises the tropical flow by checking its speed relative to extratropical circulation. Paradoxically, on the statistical scale, the average resulting wind speed exhibits high values only in the tropical zone (Mintz and Dean, 1952). These mean values are in fact a result of the great variability in wind direction in the extratropical zones, due to the constant effects of the endless procession of MPHs; the real fluxes may sometimes be very violent, whilst tropical fluxes have lesser average speeds, but considerable stability in their direction after the anticyclonic rotation asserts itself and the trade wind is born (Figure 4.2). The most rapid flow is observed to the east of agglutinations, and especially when the proximity of an orographical obstacle accelerates the flux (the Venturi effect), and the average speed progressively drops in the direction of the meteorological equator or towards the western edge of the agglutination. Over the tropical North Atlantic, Delourme (1956) observed a mean displacement in 24 hours of 700 to 1200 kilometres (30 to 50 km/h) along the eastern edge, and of 500 to 600 kilometres (20 to 25 km/h) to the south of the agglutination, with speeds further reduced towards the western edge whilst the frequency of possible trade-wind breakdowns increases.

4.4 THE TRADES Air fed in from outside the tropics is initially cold (either relatively or absolutely), and as it gradually moves in to become ‘tropicalised’, the original geographical situation of the AA and the nature of the surface below determine the thermal and hygrometric characteristics of the trades. According to the trajectory followed we can distinguish:

•

•

the maritime trades (MT), which spread across most of the Tropics by reason of the great expanse of ocean surfaces and the permanent nature of the oceanic AAs. These trades warm slowly, their oceanic diurnal amplitude being small (Figure 6B), but they constantly gain moisture and may reach saturation at the end of their courses. Maximum heat and humidity (and therefore precipitable water) are experienced at the end of summer, as a result of the thermal inertia of the ocean environment; the continental trades (CT), which warm more rapidly and have much greater diurnal amplitude (Figure 6.A); their saturation deficit increases in a similar way, and their large evaporation capacity is never appeased. The harmattan, which blows across tropical northern Africa, is a characteristic example of a trade wind (Figure 4.5).

When the nature of the surface below the trajectory changes, the characteristics of these winds are modified. A continental trade may gain moisture: an example is the Arabian CT (shamal) as it passes from the Peninsula onto the northern Indian Ocean in northern winter; the same occurs with the cold, dry continental flux moving out of Asian MPHs as it reaches the Pacific from China. A maritime trade may be

82

[Ch. 4

Tropical circulation

continentalised, as in the case of the etesian winds or meltemi (sources of an MT over the eastern Mediterranean), which become the harmattan, a dry and torrid CT in North Africa; or the southern Indian Ocean MT, which moves across the South African plateau, first depositing some of its rain on the Great Escarpment in Zimbabwe and Natal, and later being warmed in the Kalahari basin to become a CT over the plateaux of Damaraland.

[ ~ relief over 1000m ....... .. escarpment Fi i i I i dense forest -------------- boundaries of MPHs *-■ direction of motion of MPHs-------- *■ trade/monsoon ----------- trade discontinuity (TD) and/or Interoceanic Confluence (IOC) Ec Meteorological equator : — — “ inclined (surface line) —— vertical (middle layers) ------ ► ocean current coastal upwelling C Canaries current G Guinea current

B Benguela current

Figure 4.5

A Aghulas current

M Mozambique current

S Somali current

Surface circulation over Africa in January and February (Leroux, 1983).

Sec. 4.4]

The trades

11 June 2006

12 June 20062

13 June 2006

14 June 2006

83

Photos 23 Direct supply of a maritime trade over the South Pacific, east of Australia, 11-14 June 2006 (Metsat 01, Visible, 00 UTC). In spite of its modest character, the meridional alignment of the eastern mountains seems to be an effective barrier to the progress of the maritime trade into the interior of Australia. The incessant procession of MPHs ‘builds' an anticyclonic cell, known - in terms of mean pressures - as the ‘Kermadec anticyclone'.

If no orographical factor intervenes to encourage the formation of an AA, the transition from extratropical to tropical circulation may be much more rapid. The tropical face of an MPH, where cohesion is reduced because of the diminution in the geostrophic effect, opens out (the ‘water-drop’ effect), and air transported by an MPH becomes a trade wind flux directly. Photos 12 show that the idea of a ‘Mascarene cell’ is meaningful only in terms of mean pressure values, as MPHs

84

Tropical circulation

[Ch. 4

march rapidly across the southern Indian Ocean to sustain successive maritime trade wind circulations. This also happens in the North Pacific, east of Australia, where MPHs directly supply a maritime trade, to form (statistically) an ‘anticyclonic cell’, the ‘Kermadec cell’ (Photos 23). This maritime trade (originally from the Antarctic) is generally cooler than the one that passes through the ‘Easter Island’ cell. Such is also the case in the North Pacific, across which Asian MPHs spill, or the North Atlantic, where there are American MPHs with meridional trajectories; similarly, the (Pacific) ‘Philippine’ and (Atlantic) ‘Bermudan’ cells have only statistical reality. The replenishment of AAs by MPHs, with division into circulatory units, takes place only in the lower layers. As a result of the juxtaposition of circulatory units, the trade discontinuity (TD) at the surface is overlain in the west by the higher eastern flux, and becomes the trade inversion (TI). The zonal vertical sections in Figures 4.3 and 4.4, at 15°N, along the equator and at 15°S, underline the complexity of the circulation, organised in units (or cells) in the lower layers when affected by geographical factors, and the relative simplicity above them, with essentially zonal circulation which climbs above the trade inversion of each unit. The lower stratum of . a ‘late-stage’ trade (warmed and turbulent) therefore passes over a ‘nascent’ trade of more recent origin, cooler and denser. It merges into the upper stratum of the new trade, possibly modifying its initial characteristics. Trade wind circulation therefore consists of two distinct strata, which extend the vertical structure (Figure 3.2) of the original A A (or isolated MPH); these strata, with their different origins and characteristics, are separated by a TI, between the turbulent lower stratum and the subsident upper stratum:

•

•

the ‘lower stratum’, initially shallow and cool, increases in depth towards the west and towards the ME, while progressively moving towards the tropics (tropicalisation), where warming and turbulence will occur as a function of the surface crossed, to a lesser extent over oceans but markedly over the land; the ‘upper stratum’, flowing predominantly from the east, becomes less and less perturbed by relief as it gains altitude. Its air is subsident, warm and dry (through adiabatic compression).

The trade inversion is therefore lower on the eastern edge of an AA, and higher on the equatorial and western edges, depending always on the nature of the surface traversed (cf. Part II).

4.5 THE MONSOON The term monsoon means specifically a tropical flux (not a season), with a trajectory across a transequatorial declivity. This configuration of the pressure field has an anticyclonic cell in one hemisphere and an ITL-type low in the other. A monsoon circulation may therefore arise anywhere in the tropics, over oceans or continents. However, over the oceans, with their vast AAs, the ME deviates but little from the Equator, the low-pressure corridor is lacking in depth, declivities are not very steep

The monsoon

Sec. 4.5]

85

and deviation is slight. It is over continents that monsoon fluxes are at their most powerful as a result of the attraction exerted by thermal lows a long way from the Equator; the effect of this is to move moisture far inland from the distant oceans. The formation of a pressure field amenable to monsoon circulation is often a complex matter. Over Africa conditions differ with longitude, and there are three different possible situations to be considered (Figures 4.5 and 4.6):

U— ... J relief over 1000m ■■ i > x escarpment r~~rrn dense forest ----------- boundaries of MPHs direction of motion of MPHs-------* trade/monsoon ------------- trade discontinuity (TD) and/or Interoceanic Confluence (IOC) Meteorological equator: — “ inclined (surface line) ■' - — vertical (middle layers) ■—-» ocean current coastal upwelling C Canaries current G Guinea current

B Benguela current

Figure 4.6

A Aghulas current

M Mozambique current

S Somali current

Surface circulation over Africa in July and August (Leroux, 1983).

86

Tropical circulation

•

Eastern Africa is the only place in the tropical zone in which the uninterrupted seasonal migration of the line of the meteorological equator, nearing a theoretical maximum dictated by the overhead position of the Sun, may be traced (cf. Figure 3B): in northern summer (June to September) the thermal low sits over northern Sudan/Ethiopia and southern Arabia, migrating rapidly southwards from October onwards, and in December reaching northern Mozambique and north-west Madagascar, where it remains during the southern summer. It moves back northwards, at first slowly in March, and then more rapidly in April and May. In central Africa, at the longitude of the Congo basin, the substratum is continental on both sides of the Equator, but the dense forests act as a virtual ‘radiation trap’ with hyperoceanic thermal qualities (cf. Figure 6B), preventing any substantial warming above and keeping thermal lows outside their borders. There are two types of surface lows, from direct or indirect thermal effects, depending on whether or not the Sun is overhead in the hemisphere in question. To the north of the forests the thermal low is of indirect character in northern winter, and of direct character from March/April to September/October. To the south of the forests the direct thermal low leaves Angola-Zambia in February/ March, and from April onwards is halted by the southern boundary of the forested region, becoming indirect in character until September. In western Africa, due to the presence of the South Atlantic, the influence of the Sun’s position is felt only in northern summer. The Tanezrouft thermal low is very much a fixed entity from June until September, sliding south gradually in October and accelerating in November; from December to February, a shallow low-pressure corridor (of the indirect thermal type) lies along the ‘effective’ boundary, i.e. the northern edge of the dense forests which bring oceanic aspects onto the continent. Northward migration begins in March, with warming encouraged by increased dryness over the savannahs, and accelerates in April when the Sun’s more elevated position re-establishes a direct thermal character.

•

•

[Ch. 4

Discontinuities (as surface lines: ME and Interoceanic Confluence or IOC) are therefore brought to a halt at the edges of the forests. During the southward migration, the meteorological equator holds firm to the west and centre in the northern hemisphere, but in the east it dips towards Mozambique, its meridional branch ‘fixed’ to the east of the dense forests by the highlands along the western Rift Valley. As northward motion occurs, the IOC, which is a TD, but also becomes a discontinuity between monsoons, is maintained to the south of the forest mass, before it changes direction towards Ethiopia. In the Asia-India Ocean complex, (indirect) thermal lows slowly deepen until April over the Deccan and Thailand. These lows first of all draw in neighbouring maritime trades, a process known in India as the ‘little monsoon’, which in the strictest sense is not a monsoon as it does not come from the southern hemisphere. Then in May, as a result of direct lows exerting a greater attraction, of the ME migrating over eastern Africa and especially of the now sustained replenishment by southern hemisphere winter MPHs, the Indian monsoon abruptly surges onto the

Sec. 4.5]

Monsoons

87

Indian mainland, a phenomenon known as the ‘burst of the monsoon’. The monsoon flux meets the thermal lows of Rub-al-Khali (southern Arabia) and Sind (Pakistan) and then falls upon the foothills of the Himalayas, which ‘fix’ the meteorological equator in the lower layers. In the same way, an abrupt turn southwards will reverse the direction of the circulation, in the west towards the Mozambique-Madagascar low and in the east towards southern Australian low. The transformation of a trade wind into a monsoon may modify the direction, but does not for the time being change the thermal and hydric qualities of the original flux. However, such a transformation always involves considerable structural modifications. Attraction by thermal lows acts only in the lower layers, and therefore the translation into a monsoon affects only the lower stratum of a trade wind, which progressively moves beneath the trade circulation in the opposite hemisphere. This initial lower stratum, at first topped by a trade inversion (TI) in one hemisphere, crosses into the other hemisphere beneath the vertical structure of the meteorological equator (VME), and is later surmounted by the meteorological equator in its inclined structure (IME). This latter structure is increasingly unproductive along the direction of the line on the map of the ME, and the moisture-laden lower-level monsoon, shallow and flowing predominantly from the west, is surmounted by a warm, dry continental trade, mostly from the east (cf. Chapter 10).

4.6 MONSOONS •

The Indian and Chinese monsoons; the Madagascar and Australian monsoons The greatest monsoon complex is found over the Indian Ocean, the western Pacific and their rims. In northern summer, the maritime trades, replenished by strong southern MPHs from Madagascar to Australia, move towards Asia in two main currents, one heading for India (Indian monsoon, Photo 24), and the other for southern China (Chinese monsoon) across the peninsula of Indo-China. The western flux, blowing from the south-west off Africa, is called the kusi (Kenya) or kharif (Somalia), and comes from the south-west, as a low-level jet along the coasts of Kenya and Somalia (Findlater, 1971). In August this ‘Somali jet’ exceeds 50 km/h at an altitude of about 1500 to 2000 metres (Leroux, 1983). In winter, the dynamic nature of northern MPHs reinforces the trades which, having become monsoons, push the ME southwards. To the west, the kaskasi (or, in Swahili, kasikazi), a north-easterly trade and extension of the belat or shamal from across the Arabian peninsula, and of the sansar or shamsir which spreads out across the Iranian plateau, becomes, south of the Equator, the Madagascar monsoon, reaching the north-west of the island and northern Mozambique. To the east, the potent continental flux coming out of China becomes a maritime trade across the western Pacific, and, drawn in by the North Australian thermal low, becomes the Australian monsoon (Photo 25). In the middle of the Indian Ocean, between these two monsoons, the ME is found near the geographical Equator (Figure 4.3).

88

Tropical circulation

[Ch. 4

Photo 24 17 June 17 2007, 06H UT, visible, Meteosat 07. The Indian monsoon, from the south-west and west, fed by the southern trade coming directly from two MPHs, is surmounted from Arabia to NW India by the structure of the inclined meteorological equator. Cloud-formation development is thwarted; except in the case of infiltration into valleys, mostly in Tibet, the Indian monsoon flux is halted to the north of the sub-continent by the barrier of the Himalayas.

•

Africa: the Atlantic monsoon, the kusi and the Madagascar monsoon Over Africa (Figures 4.5 and 4.6), the Atlantic monsoon, issuing from the South Atlantic trade, moves up in summer towards the Saharan lows stretching from the Atlantic to Ethiopia, and after crossing the Abyssinian highlands joins forces with the Indian monsoon. Across the south of West Africa and the Congo basin, this flux is permanent. The branch turning towards the south of the basin in northern winter (‘Congo air’) is not strictly speaking a monsoon as it does not cross the Equator, but is a diverted trade, ‘fanned out' by thermal lows towards the north of the forests and the western Rift, and most preferentially towards the Angola-Zambia low (Leroux, 1983). So, in Africa, we can identify three monsoon fluxes: the kusi (Indian monsoon), the Madagascar monsoon and the Atlantic monsoon.

Sec. 4.5]

Monsoons

89

Photo 25 25 January 1996, visible, GMS 5, Explorer Arc NASA: two vast MPHs from Asia proceed across the Pacific. The leading edge of the preceding MPH, ‘a’, seals off almost the whole of the North Pacific from the Philippines eastwards, with tropical air (Hawaiian AA) being transferred directly towards the Aleutians. In the MPH, the south-westward flow of warmed air (as the trade assumes a maritime character) is checked by the mountains of Annam, to flow out over the South China Sea, where two pulse lines are clearly visible, and become monsoon 'm' across Indonesia towards northern Australia, pushing the VME southwards. MPH ‘b’ moves across the northern part of the ocean more rapidly, near Japan and south of Kamchatka, with accelerated motion through low-lying areas in the Japanese highlands acting as ‘nozzles’; air at ‘bl’ reinforcing MPH ‘a’. The southern part of MPH ‘b’ is blocked to the east by relief (Yunnan and Szechwan), causing the lifting, over southern China, of warmed and humidified air (trade) from part of the southern edge of MPH ‘a’.

•

The Amazonian monsoon During southern summer,in South America, the maritime trade wind originating from the winter Bermuda-Azores AA is drawn in by the thermal low located at about 20°S, to the south of the dense forests, in January, into the Gran Chaco low, to become the Amazonian monsoon (Photo 26). The ME is then ‘fixed’ in the lower layers by the eastern slopes of the Andes that form a solid barrier upon which the monsoon releases heavy rains, which drain away into the Amazon basin to be recycled by evapotranspiration from the dense forests. The ME then turns upwards towards the noreste along the western slopes of the Planalto Brasilero and its sierras, while the eastern slopes remain within the Atlantic maritime trade (from the ‘Saint Helena’ AA).

90

Tropical circulation

[Ch. 4

Photo 26 The Amazonian monsoon, 14 January 2007 (GOES 12, Visible, 18H UTC). The maritime trade flux over the North Atlantic is diverted southwards, towards the Amazon basin, and the monsoon is halted by the Andes. The Cordillera forms a barrier between the monsoon and the maritime trade flowing along its western foothills. The Amazonian monsoon flux is progressively diverted eastwards, with the arrival of the low-pressure corridor on the leading edge of an MPH over Argentina and southern Brazil contributing to its southward motion, and accentuating its eastward diversion; part of the flux contributes to the creation of MPH-associated cloud formations.

•

The Panamanian monsoon Over the Pacific, the maritime trade, channelled northwards by the western slopes of the Andes, crosses the Equator and, during northern summer, curves in towards the Isthmus of Panama to form the Panamanian monsoon from Colombia to south-western Mexico as far as 15 N. Here, there is no deep thermal low but rather a weakening of the dynamism of the northern trade (and especially a diminution in the meridional trajectories of American MPHs through the Gulf of Mexico), as a result of the comparative shelter offered by the western flank of the Sierras Madre (both Mexican and Guatemalan); a major feature is the reinforcement of the initial trade by southern MPHs (Photo 27). A recent way of thinking (cf. WMO, 1998) has tended to make a ‘monsoon’ out of the maritime trade moving into North America from the Gulf of Mexico. This flux, drawn towards the interior of the continent in both winter and summer by the low-pressure corridors on the leading edge of MPHs (cf. Chapter 7),

Sec. 4.5]

Circulation in the lower layers: conclusion

91

Photo 27 The Panamanian monsoon, 14 August 2007 (GOES 12, Visible, 18H UTC). The southern maritime trade crosses the geographical Equator, while progressively diverted eastwards as a monsoon flux. To the west, the Panamanian monsoon is halted by the longitudinal relief of Mexico, and to the east it crosses the Isthmus towards the Gulf of Mexico in the direction of a tropical low. Note that, over the Pacific, the Panamanian monsoon passes beneath the Vertical Meteorological Equator (VME), marked by a zonal alignment of discontinuous cloud formations, just north of the Equator.

and reaching maximum intensity in summer, has nothing genetically in common with a monsoon. It does not cross the geographical Equator, and therefore is not involved in a trcmsequatorial declivity, and neither is it drawn in by a tropical thermal low over the land.

Monsoon circulations are therefore multiple, formed under different geographical conditions: • •

in southern summer, the Amazonian, Madagascar and Australian monsoons blow; and in northern summer, the Panamanian, Atlantic and Indian/Chinese. The most intense transequatorial currents are observed in northern summer as a result of the attractive effect of the Saharan and Asian thermal lows, mainly because they are sustained by powerful southern MPHs, given the comparatively more severe southern winter. The pressure field that favours the establishment of a monsoon flux, which (usually) opposes an AA in one hemisphere to a continental thermal low in the other, has the effect of encouraging the most intense transfers of oceanic precipitable water towards the interiors of land masses.

Circulation in the lower layers: conclusion In summary, circulation continues uninterruptedly in the lower layers, from the pole in the direction of the Meteorological Equator and back again. At polar and

92

Tropical circulation

[Ch. 4

temperate latitudes it is impelled by MPHs, which organise the transportation en masse of polar air towards the Tropics and, within the low-pressure corridors on their leading edges cause, warm air to return polewards. The passage to the tropical zone is effected either directly by the ‘emptying’ of an MPH, or through the intermediary of AAs which create vast units of trade-wind circulation, particularly over oceans. The trade fluxes thus supplied then become monsoons which penetrate deeply into the continents. In its turn, a major part of the tropicalised air transfers tropical energy towards the polar and temperate zones, again through the intermediary of the low-pressure corridors associated with MPHs. This circulation restricts itself to the lowest 1500 metres, because of the initial depth of MPHs and geographical conditions, especially relief.

5 General circulation

In its Glossary (Physical Basis, Appendix I, 2001), the IPCC reminds us that: "the climate is determined by atmospheric circulation and its large-scale interactions with ocean currents and the land with its features'' - an idea long agreed upon by all. The general circulation of the atmosphere is considered to be the vehicle for climatic variations, and all known (and all predicted) variations are therefore meant to be analysed, particularly by models, within the framework of general circulation. This means (or should mean) that climatology is (or should be) already in possession of a coherent schema of the general circulation of the atmosphere, applicable on all spatial and temporal scales. That schema explains (or should explain) how the ‘atmospheric system’ works, why and how it changes, and how the climatic consequences of previously established causes are transmitted through the links of general circulation. Also, such a schema ought to enable us to put each element into its correct place in the logical chain of phenomena, since each element, individually identified, is necessarily involved in the ensemble of mechanisms through links of causality. Logically, this is the way things should be, but is it the way they are? Our thoughts turn immediately to models, and especially to 3-D numerical models, more specifically known as General Circulation Models (GCMs). Are they really an indispensable tool? The answer, unfortunately, is no. In the words of Lindzen, “models do not begin with a scheme of the general circulation; the general circulation is supposed to emerge as part of the solution” (communication to the author, 2004). So the circulation is ‘supposed to emerge’. It is not even a certainty! Surprising, but there it is. The numerical model, established on the basis of elementary cells, seems therefore to have no continuity through the various elements, except through ‘contagion’ by neighbouring cells. The general layout and logic of the whole will emerge by themselves (it is supposed), from the equations linking the basic cells. The only framework is that suggested by the points on the grid, but how each point, or set of points, relates to the wider entity is unspecified.

94

General circulation

[Ch. 5

This shows the pointlessness of interrogating such a model as to how meteoro­ logical phenomena are governed by some initial cause, since the main thrusts and chains of activity are not identified. Without a general coherent schema, it is impossible to appreciate the real importance of any one element in the context of the whole, and allot to it an exact place in the unravelling processes. Unfortunately, climatology uses similarly futile ‘reasoning’ when it suggests, for example - without establishing causal links - that droughts, heatwaves, floods and extreme meteorological events are the ‘results’ of the greenhouse effect; or claims that the Sahel drought or the Dust Bowl are caused by temperature changes in the oceans even though the root cause of the changes is neither known nor investigated. At the level of the elementary cell, a ‘simplistic equation’ is thought to establish this relationship between “rain and temperature, air and water”! The same caricature of reasoning occurs when there are discussions of the ‘causal role’ of El Nino, which is in reality only a consequence of a whole chain of processes. How is such a situation still possible today? Dady (1979) underlined that the advent of modelling had spelt the end of working with concepts beyond the level of the particle, in an effort to “understand atmospheric evolution”. Ever since, for 50 years, climatology has been in a real conceptual impasse as far as general circulation is concerned.

5.1 GENERAL CIRCULATION: THE EVOLUTION OF IDEAS

A description of general circulation has been sought since the earliest days of meteorology, but this fundamental aim remains to be achieved. Let us examine the principal efforts. The great voyages of the 15th century led to a progressive description of the ‘Brave West Winds’ of temperate latitudes, which brought Columbus back from America, and the tropical trades and monsoons. The first chart of winds, by Edmond Halley in 1686, showing the trades between 30°N and 30°S, was used to elaborate the first theory of the dynamics of wind: the so-called ‘equatorial chimney’. Because of the differential distribution of solar radiation, general circulation may be likened to a thermal motor, with two sources of cold and one of heat, between which exchanges take place. Cold air falls in high latitudes, warm air rises in the tropics, and thermal energy is converted mechanically into kinetic energy by pressure gradients whilst vertical movements liberate potential energy. High- and lowpressure areas at the surface are surmounted by inverse pressures, and meridional exchanges take place in opposite directions at low level and at high altitude. The unicellular ‘equatorial chimney’ concept, for both hemispheres, was intro­ duced by Hadley in 1735. This first theory of general circulation asserted the primacy of tropical heating: warm air rose at the Equator, streamed at altitude towards the poles and returned to the tropics via the lower layers, drawn in by the relative vacuum of the ‘chimney’: general circulation comprised two convection cells, one in each hemisphere. Hadley claimed that the direction in which the trades blew was the

Sec. 5.1]

General circulation: the evolution of ideas

95

result of the difference in the speeds of rotation of the Equator and the poles: air coming from the pole is slowed by the Earth’s surface, which moves fastest at the Equator, and therefore seems to come from the east. By the same reasoning, air moving towards the pole travels eastward faster than the Earth’s surface, hence the existence of westerly winds in mid-latitudes. This schema was progressively amended, since the turning Earth adds mechanical energy to the initial thermal impulse. So the idea arises of individualised wind belts: those blowing from east to west, retarding the Earth’s rotation, in the tropics and near the poles (though the presumed polar easterlies have yet to be observed); and westerlies, in mid-latitudes, accelerating the Earth’s rotation (the east/west balance should attain equilibrium). This concept underpins outlines of general circulation (especially at altitude), but as Rochas and Javelle (1993) point out: “easterly winds in the tropics, and westerlies in temperate latitudes, are the two most important characteristics of atmospheric circulation which require explanation: but it must be admitted that there is as yet no ready answer to these simple questions’’.

5.1.1 The birth of the tri-celiular model of circulation

A century later, in 1835, Coriolis showed that the trajectory of any object moving across a rotating body will describe a curve at all points on that body’s surface, independently of the starting point and of the direction of the object, once set in motion. The oceanographer Maury carefully collated a considerable amount of documentation concerning sea and air currents from ships’ logs, and revealed the previously unsuspected existence of subtropical zones where mean atmospheric pressure remained relatively constant. In 1855 he put forward a plan of general circulation incorporating these pressure zones, with two cells in each hemisphere, and air rising at the Equator and subsiding at latitude 30° in each hemisphere (already called the ‘Hadley cells’). But he also showed a current from the tropics to the poles in the lower layers, rising at the poles, which were calmer zones, like the Equator. Ferrel drew upon the principle set out by Coriolis and the observations of Maury, and in 1856 proposed a schema of general circulation marshalled by the geostrophic force, based on three circuits. This tri-cellular model would influence meteorology for a long time to come. In the subtropical (now ‘Hadley’) cell, equatorial air rises, then descends at about 30°N and 30°S to return to the Equator. However, not all the air returns towards the Equator: some travels towards the poles at low level, but at about latitude 60° it encounters cold air leaving the poles, and rises, moving back in upper levels towards 30° and falling again. This forms the second circulatory cell (later known as the ‘Ferrel cell’). The third cell is the polar one: cold air moves away from the pole, warms up, and at about latitude 60° it rises and returns towards the pole. Notice that, at the extremity of the temperate cell, we have a meeting of ‘warm’ air from the south and ‘cold’ air from the north: cold air which, in the polar cell, is nevertheless capable of rising on its journey back to the pole! Ferrel’s tri-cellular schema gave (provisional) answers to questions of the day, such as the existence of the equatorial calms feared by sailors at the junction of two

96

General circulation

[Ch. 5

tropical cells. Subsidence of the air at latitudes 30°N and 30°S, with high-pressure belts, explained the great continental deserts and the oceanic tropical calms known as the horse latitudes. At polar latitudes, the Coriolis (or geostrophic) effect caused prevailing easterlies, and for the same reason the equatorial (trade) winds were westerlies. The convergence, at about 60°, of polar air and warm air from the temperate zones formed the great depressions and anticyclones of middle latitudes. Ferrel’s model was hailed in his day as a great breakthrough, and the call of the tri-cellular model of circulation still echoes loudly in today’s meteorology. However, it has its shortcomings, and many have tried to improve upon it. For example, Thomson’s schema of 1857 caused Ferrel to modify his own in 1889. The cells became far less individualised, and the polar easterlies disappeared to be replaced by the ‘polar calm’, but the ‘tropical calms and dry belt’ and the ‘doldrums and equatorial rain’ were retained.

5.1.2 Improvements on the tri-cellular model of circulation The end of the 19th and the early 20th centuries saw the advent of many new schemes of circulation. Bjerknes (1923) kept the trade wind, the counter-trade and subtropical subsidence, but introduced the Polar Front between latitude 30° and the pole, so symbolic of the Norwegian school, with families of disturbances and sporadic winter irruptions of polar air into lower latitudes (foreshadowing Mobile Polar Highs, MPHs). Rossby, another member of the Bergen School, did not follow Bjerknes’ schema, and proposed a tri-cellular circulation model (1941) with the dynamical factor taking precedence over the thermal factor. This model is actually quite close to that of Ferrel in its main features, with three cells in each hemisphere: the Hadley cell, the Ferrel (with exchanges between these two cells) and the polar. A new feature was the Polar Front, a continuous separation between polar air and air flowing from subtropical high-pressure areas of essentially dynamical origin. The Ferrel cell, also essentially dynamical, is supplied with energy mobilised by enormous Norwegian cyclones. Radio-sondes have revealed the exist­ ence of a jet stream, a high speed ring in the upper troposphere, while experiments have shown (Rossby and Weightman, 1939) that large-scale turbulence within dis­ turbances (polar front) creates the westerly jet, determining its intensity and position. In 1947, Rossby overturned concepts when he stated that the origin of temperatelatitude disturbances (the Polar Front) lay in the high-altitude jet stream. In 1921, Defant had explained that exchanges took place in steps rather than in a continuous current from the pole towards the Equator, and that these exchanges were mechanical and carried out within large-scale disturbances (anticyclones and moving lows), which create widespread turbulence. The progress of families of cyclones brought about step-by-step meridional exchanges. Palmen (1951) took up this idea, and attempted to reconcile matters: he considered the polar zone to be a zone of mixing, and he introduced jets and connected the temperate and tropical cells. Palmen’s model (which nevertheless owes much to Ferrel’s tri-cellular model),

Sec. 5.2]

Insufficiencies in the representation of circulation

97

with later slight modifications (Palmen and Newton, 1969), is considered to be the best elaborated of the schemas of general circulation. It must be pointed out that, without special instructions, models based on radiation can only deal with the more or less direct consequences of temperature differences (on pressure gradients or circulation) from cell to cell. So they can only describe a unicellular circulation in each hemisphere, from the pole towards the Equator in the lower layers, and in the opposite direction at altitude, in fact reproducing Hadley’s initial ‘equatorial chimney’ schema of 1735. Representations of general circulation therefore remained at a certain point, and for about 50 years Ferrel’s tri-cellular model (1856), as revised by Rossby (1941) and Palmen (1951, 1969) has been the preferred doctrine. Although this model is now seen as “a vast oversimplification, it still provides a useful conceptual tool” (Henderson-Sellers and Robinson, 1986). In France, it continues to be reproduced and taught, especially in universities (cf. Beltrando and Chemery, 1995; Vigneau, 2005). Meteo France, the French Meteorological Society (SMF) and the Dynamical Meteorology Laboratory (LMD) still consider that the general circulation of the atmosphere comprises ‘‘three meridional cells in each hemisphere, from the Equator to each pole. These are the Hadley cells, by far the largest and most active, the Ferrel cell and the Polar cell” (De Felice, 1999, cf. site www.smf). LeTreut (2007), a modeller and the director of the LMD, still refers to this cellular schema, with its three cells in the northern hemisphere, but only two (with no reason given) in the southern hemisphere.

5.2 INSUFFICIENCIES IN THE REPRESENTATION OF CIRCULATION

Although Ferrel’s legacy has been amended, it does not even approach an explanation of the true nature of meridional exchanges. The principal shortcomings are as follows:

•

•

•

in the polar cell (direct sense) ‘warm’ air rises and ‘cold’ air falls, conforming to thermodynamics. But it is difficult to understand how air flowing from the pole might have been ‘warmed’ on reaching latitude 60°, and be warm enough to rise and return to the pole! This polar cell is improbable. Also, polar easterly winds are not observed. What is more, if this happened, it would never be cold outside the polar regions! We know that this cell has been re-named, but to now call it a ‘mixing zone’ does not tell us much more about the nature of meridional exchanges. In the temperate or Ferrel cell (indirect sense) the air which is moving back towards higher latitudes is progressively cooled, but it still rises at around 60° latitude! It is supposed that the movement is brought about by the other cells: a rather simplistic viewpoint. This cell, which works contrary to physical principles, is also therefore quite improbable. In the Hadley cell (direct sense) equatorial updrafts are certainly observed, but the subsident aspect (which is supposed to create deserts) does not involve the

98

•

General circulation

[Ch. 5

surface, because at the latitude where subsident movements happen, we find not only the Sahara, but also the West Indies and the Yucatan (cf. Chapter 3). the tri-cellular schema segments the exchanges and thereby does not envisage long-distance transfers, nor can it explain waves of heat and cold; also, it does not allow for thermal equilibrium in the Earth-atmosphere system. The ‘cellular’ concept might well have been credible in the days when distances seemed so great and zones so well separated, when tropical weather was ‘independent’ and red rains across Europe were known as ‘rains of blood’ ... though nowadays we all know that the red dust comes from the Sahara, just as those living around the Gulf of Mexico know that the nortes, like the winter rains of the Cape Verde Islands, are of far-away ‘polar’ origin.

Ferrel’s schema and those that followed it were based on observations: initially those of Maury, and mainly of the mean pressure field: • •

with low pressure corresponding to convergence and upward movement; with high pressure corresponding to divergence and downward movement.

With these principles in mind, it was easy to imagine the three cells (and ‘imagine’ only, since at the time the conditions at altitude were unknown); but these schemas established “truths”, often still considered to be incontestable postulates. For example: •

•

the meteorological equator, axis of the so-called ‘equatorial’ lows, is a location of updrafts, a notion implied within the generic term ‘Intertropical Convergence Zone’ (ITCZ). This is not always true, especially in the case of the Inclined Meteorological Equator (IME), which is not necessarily characterised by vertically ascending air; anticyclonic areas are created by air descending through the whole depth of the troposphere. Chen et al. (2001) mention the ambiguities which surround this idea: “The low-level tropical anticyclones in particular continue to be an open question”. So ‘tropical high-pressure areas’ (e.g. the ‘Azores’ or ‘Hawaiian’ anticyclones), defined using mean pressures, are generally considered to be ‘permanent centres of action’ (cf. Chapter 3). This way of thinking has caused an incredible confusion of scales, and still falsifies the perception of phenomena.

Let us also mention that disturbances, and especially those of middle latitudes, are not integrated into the dynamic of circulation. For example, “the most significant feature of the general circulation of the atmosphere in temperate latitudes is the existence of prevailing westerly winds. Superimposed upon these are disturbances which can mask the dominant character of these winds” (Rochas and Javelle, 1993), meaning that disturbances in temperate areas, appearing ex nihilo, are not integrated into the westerly flow but ‘graft’ themselves onto the circulation without really becoming an integral part of it, even if they are moving in the same direction! Stress is often laid upon different levels of altitude. In the mid-troposphere over the northern hemisphere, three long waves, with a wavelength of about 8000 km, are

Sec. 5.3]

Units of circulation in the lower layers

99

attributed to the presence of north-south oriented relief (a hypothesis which could however apply only to the Rockies), and to the distribution of continents (but not really in the case of Eurasia). Three identical waves are posited for the southern hemisphere, but only the one associated with the Andes matches the amplitude of the northern waves. Cold air is said to occupy the thalwegs, and warm air the ridges. Shorter undulations of the order of 2000 km (about the diameter of an MPH) are superimposed upon the large waves, and are associated with temperate cyclonic disturbances. The zonal movement of the free atmosphere (the atmosphere above the influence of the substratum) thus effects step-by-step exchanges, most often during the passage of disturbances, of closed eddies of warm or cold air breaking away northwards and southwards from sizeable undulations. According to Rochas and Javelle (1993): “the fundamental question is: by what mechanism is kinetic momen­ tum transported from the tropics towards temperate latitudes? The answer has been long in coming, but it is now clear: disturbances transport kinetic momentum, and also heat and water vapour”. Consequently, meteorology, and therefore the models, still offers no coherent schema of general circulation. This may seem incredible to the non-climatologist, but it is a fact that cannot be ignored, especially when we are trying to predict the evolution of the climate: because no parameter can evolve separately, just as no climatic region evolves separately. Everything is more or less closely linked. It is often the case that statistical concepts become confused with synoptic, changing reality; for example, westerly winds are deemed to be characteristic of mid­ latitudes, but this is normally true only of upper levels since, lower down, MPHassociated wind directions are essentially variable. In general, explanations of circulation do not bear comparison with the reality of meridional exchanges, which are disguised by the idea of division into cells, especially when circulation is put into reverse (as with Ferrel cells), and when curving systems close up. These explanations mix observed facts and theoretical concepts, and exaggerate the importance of the upper levels. Above all, because of its perceived complexity, models of general circulation considerably minimise circulation in the lower layers, although they are the densest; but it is precisely the Earth’s surface which is the source of the thermal impulse which marshals and maintains general circulation, via the regular injection in the lower layers of MPHs from high latitudes.

5.3 UNITS OF CIRCULATION IN THE LOWER LAYERS The lower layers of the troposphere are of primary importance in climatology. They are the densest layers, and contain nearly all the water vapour and other greenhouse gases. Also, circulation in the lower layers is of a particular kind, and much more complex than that of the upper layers, because of:

• •

the Earth’s surface warming the atmosphere; differences in the substratum;

100

• • • • •

General circulation

[Ch. 5

the thermal behaviour of oceans and land masses; vertical upward and downward movements caused by the surface (compression, turbulence, convection); thermal gradients arising from differential warming; effects of attraction caused by deep tropical thermal lows; differences in altitude (especially the presence of high mountain masses).

The spaces within which circulation occurs are fairly distinct, if extendable, and there may be ‘traffic’ to varying degrees between these spaces. Figure 5.1 offers a schematic overview of circulation in the lower layers. Relief is represented if it has a strong aerological influence, capable of forming a boundary of a unit of circulation by channelling mass air transfers. The distinction is made between:

• •

barrier relief, impassable to dense MPH air except through lower-lying sections. Warm air is not thus affected; and more modest relief if it determines circulation over large areas. This kind of influence must usually be a localised one: for example the Ghats, and especially

■mm

relief barring passage of dense MPH air (limit of circulation) aerologically significant high ground AA anticyclonic agglutination • MPH area of origin MPH trajectory — -*• warm flux diverted towards pole ——• surface line of meteorological equator, seasonal positions (DJF, JAS) ......... - trade/monsoon discontinuity ------------ *• trade-------------- ► monsoon

■

i• ■ ■

Figure 5.1

Areas of circulation in the lower layers (diagrammatic).

Sec. 5.3]

Units of circulation in the lower layers

101

the Western Ghats, are not included, despite their major role in the climatology of India, because they do not affect circulation as a whole over the Indian Ocean; while the high plateaux of Africa are shown, because of their role in the delimitation of the Atlantic and Indian aerological spaces. The notion of the transportation of air en masse by MPHs signifies simultaneous movements of cold air (i.e. anticyclonic, remaining in the lower layers) and the returning warm flux (cyclonic, and able to rise). The extreme seasonal positions of the surface line of the meteorological equator (mean position December-February, and mean position July-September) outline the zone of alternation of the trade and monsoon fluxes. The geographical factor determines well individualised areas of circulation in the lower layers, and these spaces may ‘communicate’, with ‘overflows’ between spaces occurring permanently or seasonally. Six aerological units are recognised, each fairly distinct, if extendable: three in each meteorological hemisphere.

5.3.1 The northern meteorological hemisphere The highlands of Greenland divide MPH trajectories from the outset. Three spaces are relatively well delineated by relief:

North and Central America/North Atlantic/Western Europe: MPHs from the western Arctic preferentially follow the American trajectory, to the Atlantic. The Atlantic ‘Azores’ AA sustains a maritime trade wind which, in northern winter, becomes the Amazonian monsoon. Winter meridional trajectories of MPHs, at the root of the seasonal westward extension of the Atlantic (‘Bermuda’) AA, facilitate the flow of Atlantic trade winds across the Isthmus south of the Mexican highlands and into the eastern Pacific. The main return of air towards the north pole occurs across the Gulf of Mexico, up the eastern seaboard of America towards the Norwegian Sea, mainly between Greenland and Scandinavia. ‘Outflows’ towards neighbouring units occur in the direction of central Europe north of the Alps, into the western Mediterranean, and in winter towards northern Africa to the south of the Atlas Mountains (where the trade discontinuity is not ‘fixed’. Northern and central Europe/western Asia/southern Africa/Arabia: MPHs have Scandinavian and Russo-Siberian trajectories, more or less channelled southwards by the Scandinavian Alps, the Urals and the Asian highlands. Dense air, channelled by the northern edge of the Alps and skirting around the Carpathians (Ion-Bordei, 1988; Aubert, 2008), flows away mostly towards the eastern Mediterranean, North Africa and Arabia, and further east, towards the Turan basin where a vigorous AA is formed. ‘Outflows’ occur north of 55°N from the Turan AA, in the direction of eastern Siberia, through the Dzungarian sill towards Mongolia and the Gobi desert, and towards the Sea of Oman through lower-lying parts of Iran. In winter, con­ tinental cold air, relief and MPHs maintain a powerful and highly stable ‘Siberian’

102

General circulation

[Ch. 5

AA; this aerological space overflows westwards in the direction of the Black Sea and the Mediterranean, and southwards across the Iranian plateau into the western part of the Indian Ocean, to sustain the maritime trade wind (kaskasi in Africa) and then the Madagascar monsoon. In summer, less evident and/or warmer sorties take part in establishing the Atlantic and Indian monsoons, blowing from the south. East Asia/North Pacific: MPHs pass between the Sayans and Verkhoyansk moun­ tains into Manchuria, strengthened by fragments from other MPHs which have crossed the Gobi desert (the yellow wind, laden with loess). They also reach the North Pacific via the Bering Sea. Air returns towards the pole partly out of Asia across coastal regions, and partly through strong channelling west of the Rockies towards Alaska. MPHs with meridional trajectories, which are energetic in winter (engendering the statistical ‘Philippines’ outcrop), feed a trade wind which develops a maritime character. Having crossed the peninsula of Indo-China, this joins up with the northern Indian Ocean trade wind and, like the trade, it becomes a monsoon, tending towards Australia. The maritime trade wind from the Pacific (‘Hawaiian’) AA also joins this flow. Limited ‘outflows’ occur across the Rockies (by way of three high, narrow passes) onto the plains of North America, which belong to the North Atlantic aerological space.

5.3.2 The southern meteorological hemisphere

MPHs move out in all directions from the Antarctic, and the three southern aerological spaces intercommunicate easily before coming under the influence of the Andes, which have a marked effect north of 40°S. South America/South Atlantic/western and central Africa: MPHs are diverted towards the Atlantic and agglutinated by the great Namibian Escarpment (the ‘Saint Helena’ AA). The return flow of air towards the Antarctic occurs mainly to the west, nearer to South America. The maritime trade wind (or ‘monsoon’ ‘Congo Air’) is permanently over the Congolese forests, and/or heads off towards eastern South America. In northern summer, this space expands considerably, over southern Africa (Atlantic monsoon), the ocean and the Amazon Basin.

Southern and eastern Africa/Indian Ocean/Australia: This Indian Ocean space is relatively modest in size in northern winter, when the Madagascan and Australian monsoons encroach from the north, albeit extending over southern Africa (up to the great Western Escarpment reached by the continentalised trade wind), and most of Australia west of the Great Dividing Range in the east. In southern winter it covers a vast space, spilling over from the Indian Ocean onto eastern Africa (as the kusi), southern Asia and the north-west Pacific, by means of the Indian and Chinese monsoons. Eastern Australia/South Pacific/western South America: Closed off to the east by the Andes, which create the powerful ‘Easter Island’ AA, this immense aerological space

Sec. 5.3]

Units of circulation in the lower layers

103

exhibits only slight seasonal variability, and here the ME migrates but little. The vigorous trade wind channelled by the Andes shifts the ME almost permanently into the northern hemisphere, and in northern summer develops into the Panamanian monsoon. Further west, to the north of Australia, some of the trade flux is diverted northwards with the current of the Chinese monsoon. The main features of circulation in the lower levels underline the fact that there is no hiatus between the temperate and tropical zones. The transfer of air from pole to Equator and back carries on uninterruptedly. In each circulatory space immense ‘figures of eight’ are created, with air leaving the poles and returning after passing through the tropical zones where it takes on heat and water. At the crossover point, the warmed air passes before and above cold air, with conflict occurring as an MPH migrates (cf. Part II). It is important to stress the similarity between the circulation of air, and circulation in the upper layers of the oceans, the latter being principally driven (as drift currents) by the former, with warm water also passing above cold water (density currents). Ocean currents participate to a proportion of about 20% to 25% in meridional transfers, with three-quarters of the exchanges being effected by the atmosphere at a much greater comparative velocity.

5.3.3 Dynamical unity and climatic diversity

Within a unit of circulation, regional diversity is wide and, as a function of dynamics, various regions exist. These may: • •

• • •

be regions of preferential departure for cold air, near the pole (with little warm air coming in from the south near the surface; lie directly on the trajectories of MPHs, with alternating cyclonic circulations (to the fore of MPHs) and anticyclonic circulations beneath the MPH. The western flanks of the units transfer mostly cold air, while the eastern flanks experience intense arrivals of warm air; lie outside the preferred path of MPHs, i.e. in the domain of associated lows and warm cyclonic fluxes (though the passage there of an MPH cannot be ruled out); be beneath an AA, with its remarkable ‘stability’ (or relative stability, since variations in strength and north-south migrations are nevertheless observed); or they may lie beneath a trade wind prolonging the AA, itself eventually becoming a monsoon; now, in the tropics, it is the eastern flank which has the cooler air (and water), while the western flank is warm.

Some areas experience constant or near-constant conditions (especially in the presence of mighty barriers), while others see continual daily and seasonal, even interannual, variations. Between the two extreme seasonal positions of the ME (Figure 5.1), the trades and the monsoon blow in turn. Still others lie at the junction of two aerological spaces: for example, Western Europe may be alternately influ­ enced by either the Atlantic unit of circulation or the Eurasian, the latter being responsible, especially in winter, for the bitterest cold spells. In the case of Tunisia, at

104

General circulation

[Ch. 5

the meeting-place of the dynamics of the eastern and western Mediterranean basins, which are involved in two different aerological spaces, there may sometimes be sudden violent confrontations, leading to dramatic flooding (Leroux, 1991b). This geographical diversity should not, however, be allowed to hide the fact that, within an aerological space, the initial dynamic is the same, controlled to various degrees by MPHs entering that space; all the parameters of the unit are inter­ dependent. Thus, within a given space, even if the logic of processes seems to differ, and climatic consequences are diverse in nature, there exists a general covariation of climatic parameters. A different aerological space will follow its own dynamic, con­ trolled by its own MPHs and its own geographical conditions affecting circulation. The possible associations between units may arise, in the same hemisphere, either from particular communications between more or less compartmentalised units, or from the conditions prevailing at the point of departure of the circulation, i.e. at the pole concerned. This is a reality which must be strictly borne in mind before imagining - as diagnostic climatology does not hesitate to do (without any previous meteorological analysis) - ‘connections’ between regions with no aerological com­ monality (unless it is the polar origin of MPHs).

5.3.4 Fundamental questions

The way in which circulation in the lower layers is organised calls into question many points which had been thought (and which are still thought by some) to be solidly established. •

•

•

•

The origin of the circulation does not lie in updrafts resulting from heat in the tropics, but rather is caused by the thermal deficit, and the variations in that deficit, over the polar regions. Because of this deficit, new ‘discs’ of cold air are constantly being injected into the circulation, with varying degrees of energy, and thereby encouraging the return of warm air towards the poles. Circulation is not a continuous process (i.e. flowing step by step as in models), but is constantly renewed as the injection of cold, lenticular air masses (MPHs), transporting their inherent thermal characteristics (and others picked up along their paths), perturb the circulation of higher and middle latitudes. Then, calmed, they contribute to tropical fluxes. The arrival of these enormous vol­ umes of ‘new’ air (or, more correctly, recycled above the poles), orchestrates the intensity of meridional exchanges, both warm and cold. The polar/temperate zone is not, strictly speaking, a ‘mixing zone’, but a zone of rapid meridional transfers (and confrontations), where the notion of ‘mean winds’ has no climatic meaning. Cyclonic circulation to the fore of MPHs effects an intense transfer of water vapour and, consequently, of energy, most but not all of which originates in the tropics. Most of the water vapour is transported in the lower layers (i.e. the lowest 1500 metres), as Peixoto and Oort (1983) underlined, remarking especially that “the transport of water vapour is clearly influenced by topography”.

Sec. 5.4]

•

•

•

•

General circulation in the troposphere

105

There is no break in the circulation within each aerological space, and no inter­ ruption between temperate and tropical circulation, from the pole to the Meteorological Equator, and therefore there are no closed cells in the lower layers. There are definite hiatuses between different units of circulation, and in the lower layers, within which most air moves. So there is actually no ‘single’ general circulation, but specific circulations integrated into general circulation. The notion of globality and the interdependence of all phenomena, as presented by the models, is a long way from being a fact. Anticyclonic agglutinations (AAs), or ‘subtropical highs’, are present in the lower layers, with no intervention from upper levels. Now, subsident air is still almost universally claimed to be the factor causing the presence of deserts. This is wrong for a number of reasons, not least of which is that the air subsidence in question occurs only above AAs, and cannot therefore reach the ground! The geographical factor is normally not much taken into account, especially by models. Its importance is considerable as far as circulation is concerned, however, with relief being a main factor. For example, the Rockies, stretching from Alaska to southern Mexico, divide North America into two practically distinct units as far as cold air in the lower layers is concerned. Similarly, to the east of this barrier, water vapour is advected essentially from the Atlantic. The Andes do the same with the cold air and water vapour of South America. Again, the Himalayan-Tibetan mountains form a fundamental climatic bound­ ary, denying MPHs access to the Indian subcontinent and depriving central Asia of oceanic air. The role played by relief is immediately obvious in the case of the high mountain chains, but more modest relief can also have a comparable effect.

5.4 GENERAL CIRCULATION IN THE TROPOSPHERE It should be remembered that the degree to which mean winds are representative is not the same in the tropical zone as in the extratropical zones:

•

•

in the tropics, mean values are representative of the real wind field, i.e. the direction of the synoptic wind varies but little from the vector of the mean resultant wind; outside the tropics, because of the continuous passage of MPHs, the wind varies constantly, and in diametrically opposite directions. A ‘mean wind direction’ therefore has no synoptic relevance.

So it is impossible to represent tropical and extratropical circulations in the same way in a schema of general circulation (which is necessarily ‘mean’) by lines of flux, which are representative of reality only in the case of tropical fluxes. 5.4.1 The mean tropospheric picture The complexity of circulation in the lower levels is connected with both the geographical factor (especially relief), and the thermal factor, causing the formation

106

General circulation

[Ch. 5

of lenticular MPHs or tropical thermal lows over land masses. The influence of these factors can only fall off, and eventually disappear, with increasing altitude: circula­ tion then becomes progressively simpler, while at the same time the air becomes less dense. Only large-scale currents are now apparent, moving from the west outside the tropics and from the east within them, organised in jets near the tropopause. Figures 5.2, 5.3 and 5.4, which limit themselves to the troposphere, show this progressive simplification diagrammatically. Synoptically, observation shows a succession of

FFTFl

Polar highs

BP

Polar lows

Tropical ighs

D

Intertropical lows

Polar front Trade inversion

Figure 5.2

Meteorological lows

E. W.

Zonal wind component

Diagram of general circulation in the troposphere (Leroux, 1983).

of circulation, taught at the Ecole Nationale de la Meteorologie since 1992 (cf. Bonnissent, 1992), and published in Meteorologie Generale et Maritime, Cours et Manuels No 14, Figs. 9.7, p. 83 (Ecole Nationale de la Meteorologie, Meteo France, Toulouse, 2001).

Sec. 5.4]

General circulation in the troposphere

107

Key for Figs. 18 and 19 : : MPH : AA__________ Flux diverted back towards pole (warm air) Meteorological equator: ME and ITL I VME Vertical ME------- 1 IME inclined ME ------- trade and/or monsoon (lower layers) --------- trade inversion (Tl) W westerly jet E tropical easterly jet \ \ Tropical highs at altitude

Figure 5.4

General circulation in the troposphere (Leroux, 1993).

nuclei of acceleration above zones of intense convective activity, forming on the one hand a tube of high-speed air at a fairly high altitude in the general westerly flux (the westerly polar jet), and on the other hand nearer the Tropics a subtropical jet moving out of the tropical zone from the south-west in the northern hemisphere, accompany­ ing the deep penetration of an MPH into the tropical zone, and reinforcing the westerly high-altitude current. However, working from mean values, as in the figures mentioned above, differences in velocity in these wind nuclei and variations in direction are erased, leaving only a uniform circumpolar current, in each hemisphere, whose flow is continuous when looked at in this statistical way. Figure 5.2 (Leroux, 1983) represents a step in the direction of Figure 5.4 (Leroux, 1993a). It still refers to the ‘polar front’, but shows the continuity of circulation in the lower layers from the poles to the meteorological equator. Figure 5.3 underlines the ambiguity of the position taken by Meteo France, which advocates the tricellular schema, but copies, teaches and publishes my schema from 1983 ... which calls into question the schema of 1856. Meridional exchanges follow various routes. Departing from the poles, cold air is exported by MPHs. The polar thermal deficit therefore impels general circulation incessantly. The passage of MPHs though mid-latitudes acts like a ‘snow-plough’, causing two vast movements of air, horizontally and vertically: •

•

the return of warm air towards the poles in the lower layers, to the fore of MPHs, and above them, supplying air for future MPHs; the lifting of warm air at the leading edges of MPHs (front and closed low) and the release of latent energy.

At these favoured latitudes, where intense vertical transfers occur, accelerations in westerly circulation (jets) are also observed. These accelerations are strongest in winter and weakest in summer, because of the variations in the strengths of MPHs and in the intensity of updrafts, which shift nearer to the tropics in winter and are further from them in the summer, as MPHs move in latitude. In spite of their primal

108

General circulation

[Ch. 5

importance, these transfers appear only very imperfectly on a vertical section (Figure 5.4), since they occur horizontally and mainly in the lower layers, before rising (above MPHs) as they approach the poles (Figure 5.5). Areas where the circulatory ‘figures of eight’ cross over exhibit intense vertical exchanges, driven by the dense air of MPHs forcing warm air upwards. These crossovers manifest themselves in either hemisphere by three high-altitude thalwegs (connected with the fact that there are three aerological units in the lower layers (see Figure 5.1). In the northern hemisphere, one of these thalwegs is found to the east of north America, one over east Asia, and the third, which is weaker, over central Europe, i.e. precisely above the areas where MPHs find their outlets. The jets (as Rossby supposed in 1939) are the consequences of phenomena in the lower layers, and especially of considerable vertical transfers of air and energy triggered by MPHs. The reverse is not the case.

Key for Figs. 18 and 19: MPH ~ aax AA__________ Flux diverted back towards pole (warm air) Meteorological equator: ME and ITL I VME Vertical ME ------- 1 IME inclined ME ------- trade and/or monsoon (lower layers) --------- trade inversion (TI) W westerly jet E tropical easterly jet \ \ Tropical highs at altitude

Figure 5.5 General circulation in the troposphere (Leroux, 1993). Plan and vertical section (according to mean values).

Sec. 5.4]

General circulation in the troposphere

109

Anticyclonic agglutinations (AAs), formed by lenticular MPHs, are found only in the lower layers. From them springs the trade circulation or, more properly, the lower stratum of the trade, originally (within the AA) of the order of 1000 metres deep. The lower stratum of the trade becomes progressively warmer, and slowly spreads and deepens, becoming laden with water vapour. It meets the trade or the resulting monsoon (Figure 5.3) coming from the opposing hemisphere, along the meteorological equator (ME). Along the ME, updrafts, with deep convection, become general, for thermal but especially dynamic reasons, connected with the confluence/convergence of the circulatory hemispheres, impelled initially from the poles. The upward movement of air at the heart of the tropical zone has two major consequences:

1

2

It supplies the Tropical Easterly Jet (TEJ), the vigour of which varies with the intensity of the updraft. It raises pressure at higher levels, forming Tropical High Pressure areas (THPs) which enclose the tropical zone in an inverted ‘V’ configuration (Leroux, 1983). These highs drive circulation in the direction of the poles, but the geostrophic force will not permit meridional exchanges across such a distance. Obeying mechanical laws, air at higher levels is rapidly drawn down towards the surface, the descents tending to close off the Hadley cells at around latitudes 30°N and S.

The downward movements do not, however, reach the surface, as the lower layers are already occupied by the AAs and/or the lower stratum of the trade wind issuing from these same AAs. A fundamental discontinuity is thereby created between, on the one hand, subsident warm, dry air above and, on the other hand, the AA, or the lower stratum of the trade wind below, as it warms, spreads and possibly gains moisture. This discontinuity is the Trade Inversion (TI), the climatic consequences of which are essential within the tropical zone as it is a horizontal, unproductive discontinuity, i.e. discouraging the vertical development of cloud formations. This is also the case in extratropical zones, as this discontinuity firmly dictates how water vapour is utilised, hindering its upward dispersion and concentrating it in the lower stratum of the trade. Then, the remaining subsident (Hadley cell) fluxes may pursue two possible directions, one back towards the meteorological equator above the lower stratum of the trade in the middle layers, and the other towards the temperate and then polar zones above the MPHs, thus closing a circuit initiated at the pole (Figure 5.4).

5.4.2 Seasonal variation in circulation The Sun’s yearly motion causes seasonal variation of meridional exchanges (Figure 5.6). In terms of radiation received, all latitudes are affected to varying degrees, but variations in the polar thermal deficit have the greatest effect upon circulation, because of the dynamics of MPHs. The polar thermal deficit strengthens MPHs and

110

General circulation

[Ch. 5

Key for Figs. 18 and 19 : : ~ MPH 7■■"■vxs AA__________ Flux diverted back towards pole (warm air) Meteorological equator: ME and ITL I VME Vertical ME ------ j IME inclined ME ------- trade andfor monsoon (lower layers) --------- trade inversion (Tl) W westerly jet E tropical easterly jet \ \ Tropical highs at altitude

Figure 5.6

General circulation in the troposphere (vertical sections: according to season).

AAs in winter, intensifies confrontations at mid-latitudes and strengthens the trades. Throughout the winter hemisphere, circulation is accelerated, both away from and towards the pole. The more meridional trajectories of MPHs shift the most intense vertical transfers associated with disturbances, and the westerly jet, more plentifully supplied by these enhanced transfers, is also shifted and exhibits maximum velocity, double that observed in the summer. This helps to shift it towards the tropic, moving it away from the axis of rotation. The increase in dynamism enlarges the hemisphere, at the expense of the summer hemisphere. The latter experiences a weakening of its MPHs, AAs and tropical fluxes, a slowing of exchanges, and less intense disturbances, as a result of the less pronounced polar thermal deficit. This hemisphere also sees a lesser shift towards the tropics of the trajectories of MPHs, of AAs and of the slowed westerly jet. Around the tropic of this hemisphere, deepening lows facilitate, through attraction, overspill from the opposite hemisphere in the form of transequatorial fluxes (monsoons). Since the thermal factor has influence only on layers near the surface, the ME, pushed by a strengthened flux and drawn by a deepened depression, is markedly displaced towards the summer hemisphere only in the lower layers, over land masses. The uninfluenced part of the ME, in the middle layers, and the tropical easterly jet high above in the upper layers, merely share in the relatively weaker overall shift in general circulation, which is the same as that of the AAs. So the ME has two vertical structures:

•

an inclined structure (IME) in the lower layers and encroaching deeply into the opposite hemisphere, superimposing a trade wind upon the transequatorial monsoon flux (Figure 5.6). This superposition (trade above monsoon) causes the

Sec. 5.4]

•

General circulation in the troposphere

111

IME to be a fundamental discontinuity, making the two fluxes unproductive because they have different origins, characters and directions; and a vertical structure (VME) in the middle layers, uninfluenced or not much influ­ enced by phenomena in the lower layers. Above it flows the high-level Tropical Easterly Jet (TEJ). The VME and the TEJ (like the AAs) follow the overall shift in general circulation. The VME is the axis towards which the mid-layer easterly fluxes of the middle layers (the upper strata of the trades) are directed.

As the seasons change, displacement is general, southwards in northern winter and northwards in southern winter. Figure 5.4 therefore shows transitory situations during which there is momentary equilibrium between the dynamics of the two meteorological hemispheres. In southern winter, at Earth’s aphelion, the southern polar thermal deficit is accentuated, encouraging a greater expansion of the southern meteorological hemisphere, a process reinforced by northern thermal lows deepened over the land (Figure 5.1): well-fed monsoons penetrate far into the hemisphere and the surface line of the meteorological equator migrates northwards.

5.4.3 Partitioning and stratification in circulation As we have already stressed, modellers see the troposphere as uniform, homogeneous, and smooth, with neither partitioning nor discontinuity. Lindzen (communication to the author, 2004) wrote: “there is a difference between sharp gradients and discontinuities, and it is the latter that do not exist [in the models]”. This ‘ideal’, but hypothetical, atmosphere has little in common with the real thing which, on the contrary, is rigorously organised: it is separated into near-autonomous units of circu­ lation and possesses vertical and horizontal discontinuities with distinct identities. Circulation is, first of all, separated out in the lower layers (Figure 5.1). One of the primary discontinuities is relief, forming barriers at various altitudes and blocking the cold, dense air of MPHs. Consider, for ex-ample, both North and South America, where the west coasts and the eastern areas have distinctly different climatic characteristics, are visited by MPHs of different origins and paths and their precipitable water is of different origins. The Great Escarpment around the southern African plateau prevents the maritime trade, blocked at the Namibian coast, from penetrating inland until it passes beyond the highlands of Angola, to flow into the lower-lying Congo Basin (Figures 4.5 and 4.6, Leroux, 1983). Acknowledging the role of relief would preclude inanities like this one from Meteo France (Cours et Manuels No. 14, 2001, p. 152), on the subject of the Asian ‘winter monsoon’: “the masses of cold air expelled from this thermal anticyclone experience considerable compression (a foehn effect) beneath the wind of the Himalayan and Tibetan relief”. This seems to suggest that, contrary to all that is known about density, the very cold, dense and thin layer of air of the MPHs moving across China (Figure 4.5) manages to rise several thousand metres in order to cross the Himalayas and descend into the Indian subcontinent! This kind of ‘explanation’ is manifestly absurd.

112

General circulation

[Ch. 5

The partitioned circulation thus created will possibly flow further into the tropical zone via a Trade Discontinuity (TD), a division between two trade-wind circulations. Thus, in southern Mexico, the evolved Atlantic trade, rounding the Sierra Madre mountains, meets the nascent Pacific trade, which is denser, and rises above it (Figures 4.3 and 4.4). Similarly, south of the discontinuous barrier of the Cantabrian Mountains, the Pyrenees, the Meseta, the Iberian Sierras and the Atlas range in Morocco, a Trade Discontinuity (TD) separates the Atlantic maritime trade from the continental Saharan trade which is energised by MPHs moving into northern Africa via the eastern Mediterranean basin. The warmer and lighter continental trade, with its load of dust, then passes above the cooler and denser maritime trade. In extratropical zones, the passage of MPHs sets up, around and above the MPHs, a synoptic and discontinuous stratification. The mobile inversion of wind, temperature and humidity marking the top of the MPH separates fluxes of different origins, directions, vertical movement and characteristics. Situated at an altitude of about 1500 metres in the vicinity of the pole, it progressively sinks as the MPH moves towards the tropics. The low-pressure corridors between MPHs possess no such inversion, and here updrafts dominate in the cyclonic circulation. The stratification becomes permanent and continuous within the AAs. An inversion at about 1000 metres separates air advected by MPHs from the subsident air above. This inversion, in concert with the anticyclonic character of the AA, is particularly unproductive (i.e. it inhibits the vertical development of cloud formations). It extends from the AA to which it belongs further into the trade circulation, becoming the (equally unproductive) Trade Inversion (TI). Beneath this horizontal discontinuity, the turbulent lower stratum warms up and may become moist (over the ocean) or drier (over land). Above, the subsident upper stratum is warm and dry. Between the two, the Trade Inversion marks the ceiling for thin stratiform clouds when the trade is a maritime one, or the upper limit of the concen­ trations of dust lifted by turbulence within the lower stratum of the continental trade. The Inclined Meteorological Equator is another stratified, unproductive struc­ ture. The superposed fluxes (easterly trade above westerly monsoon) are of different origins, directions and characters. Beneath such a structure, which may extend for several hundred (or even a thousand) kilometres (cf. Figure 4.5), precipitable water advected by the lower-layer monsoon is scarcely exploited, with rain normally absent (Leroux, 1983). This is a fact that cannot be ignored when the question of the ‘Sahel drought’ arises. These elements of stratification differentiate the temperate zone, in which the lack of permanent inversion permits upward movement of air (if driven vigorously from lower layers), from the tropical zone, where the vertical structure is a deter­ mining factor in the development and distribution of meteorological phenomena.

Zonal ‘Walker' circulation: myth or reality? Should the meridional circulation illustrated in Figures 5.2 to 5.6 be complemented in the tropical zone by a schema of zonal circulation based on closed cells? In the

Sec. 5.4]

General circulation in the troposphere

113

lower layers, geographical factors are responsible for circulation which is at once meridional and zonal. However, in the upper layers, and increasingly with altitude, circulation is remarkably zonal. The Earth’s rotation means that absolute meridional circulation is ruled out, and it is inevitable that there will be some zonal element in every displacement. But what of the existence of uniquely zonal cellular circulations strung out along the Equator? Bjerknes (1969) supposed this to be true for the Pacific. Considered to be a component of the Southern Oscillation (Walker, 1923-24), this presumed circulation became known as the Walker cell, with subsidence to the east of the Pacific and updraft over Indonesia, easterlies in the lower troposphere and westerlies in the high troposphere. Flohn (1971), and subsequently others, extended this concept to the whole tropical belt. Four cells were posited, along the Equator (some commentators had more: no less than seven in the case of Krishnamurti et al., 1973): the Pacific, Atlantic, Congo and Indian cells. Later these cells were extended well away from the Equator, covering almost the whole tropical area. Progressive simplification led to a supposed relationship which may be summed up thus: “updrafts are found over the warmest regions, and subsidences over the coolest” (Fontaine, 1990). It should be noted that we are dealing here with near-Equatorial ocean areas where tem­ perature differences are slight, and the labels ‘warm’ and (especially!) ‘cool’ have very limited meaning. However, as stressed by Hastenrath (1991), “considerable controversy remains”. This presumed ‘relationship’ over tropical oceans, which seems to be a new incarnation of the “cold water, no rain/warm water, rain” hypothesis, carries its own disproof: if subsidence were reinforced on the eastern side of oceans, there should be associated high temperatures due to compression, which is the opposite of what is observed in reality, and of the proffered hypothesis. If the relatively small differences in oceanic temperature were really capable of causing such movements throughout the troposphere, even to the extent of reversing circulation, what would happen over continents, and especially over the Sahara? Even in summer, with temperatures at ground level touching 50°C, the tropical structure (modified in the lower levels by a deepening thermal low) is not greatly transformed (the depressionary status existing only below 1500-2000 metres), and it is certainly not reversed! The presentation of such an unproven hypothesis seems all the more simplistic when we consider that the thermal behaviour of the ocean cannot be responsible for the presumed vertical movements (neither upwards nor downwards), and that it has moreover not been verified (neither in fact nor in analysis) in the high-altitude wind field, or in its horizontal and vertical movements. In this scheme of things, Africa would be under the influence of two cells:

•

•

one over the Congo basin, with (rainless) subsidence to the west over Gabon or Cameroon-Nigeria or in an extreme case over Liberia, and updraft (with rain) over the highlands bordering the western Rift Valley on the frontier between Zaire and Uganda; the other, the descending (rainless) branch of the Indian cell over East Africa (Kenya). Thus, we have a presumed lack of rainfall over the west coast - but this

114

General circulation

[Ch. 5

is, as is well known, well watered in the area around the Bight of Biafra (where record rainfall totals for Africa are recorded, sometimes exceeding 10 metres); further west, along the Liberian coast, rainfall exceeds 4000 mm annually! Analysis of the climatic facts of tropical Africa leaves one in no doubt that the presumed effects of alleged ‘Walker cells’ are never observed, whether it be on the Equator or towards the Tropics (Leroux, 1983, 1994a).

Observation shows that, in Africa, as elsewhere in the tropical zone, this concept of zonal circulation, originally a simple hypothesis, has become (through abandon­ ment of early reservations, and systematisation) an amalgam, bringing together, simplistically: • •

•

facts scattered in space and time, brought to the vicinity of the Equator and viewed in a limited way; unproven relationships, such as those linking cold water with lack of rainfall and/or warm water with its presence or even subsidence/updraft above the open ocean, whose thermal behaviour (in absolute values and variations) cannot cause the presumed vertical movements (cf. Figure 1.6); the hypothetical existence of a “subsident edge of subtropical cells”, an (erroneous) idea put forward to justify higher pressures on eastern edges of AAs (which ought therefore to be the warmest, as a result of the presumed com­ pression). AAs would have to be of a rather different nature, in that the associated subsidence of the descending branch of the Hadley cell would actually have to reach the surface (and this does not happen). The concept of zonal circulation must in fact deliberately take no account of tropical aerological stratification and the vertical structure of the trades as evidenced by the trade inversion, which is but an inversion of subsidence.

So there are no Walker cells, i.e. cellular circulation in the more or less immediate vicinity of the Equator. Tropical circulation is zonal and continuous above the complexity of the lower layers. The presumed existence of such cells would split up this zonal circulation (which would by turns be from the east and from the west) and thus Saharan dust would not cross the Atlantic towards America, the Tropical Easterly Jet would not be continuous above the VME and would not be reinforced and doubled in summer above Africa by air brought in from Asia. The meteoro­ logical realities which this concept is said to explain, in a simplistic and erroneous manner, have causes obvious enough to allow their recognition, always on the understanding that this is based on direct observation of real events:

•

•

in the lower layers, the appearance of a westerly flux is evidence of the replace­ ment of a trade flux by a monsoon flux (with displacement of the ME), but it is certainly not evidence of a reversal of the direction of the trade; in the upper layers, the appearance of a westerly flux is evidence that the westerly extratropical flux (via the translation of the westerly jet, cf. Figure 5.6), or the deep penetration of a thalweg) is temporarily replacing the easterly tropical circulation (TEJ), but it is certainly not evidence of a reversal of the tropical jet.

Sec. 5.5]

Conclusion: general circulation is perfectly organised

115

5.5 CONCLUSION: GENERAL CIRCULATION IS PERFECTLY ORGANISED

General circulation is complex, partitioned, stratified ... but perfectly organised. When meteorological phenomena are labelled “chaotic” (a comforting assertion, for peace of mind), it is normally a sign of a (deliberate?) lack of understanding of this rigorous organisation. The fact is that ‘chance’ plays a small part, and introducing it is often the resort of unavowed ignorance. So, when the supposedly ‘unruly’ climate is discussed, it is usually because those discussing it do not appreciate the rules, i.e. the rigorous mechanisms determin­ ing, localising and characterising climates. Questions such as “Can the climate be turning on us?” (Bard, 2004) are absurd, echoing the catastrophism of ‘media­ weather’, and they serve only to point up ignorance of the way in which climate works. ‘'Turning' suggests that the causes, mechanisms or even the direction of general circulation really could, for some reason or other, abruptly change, or ‘turn’ into reverse! The genera] circulation of the atmosphere is rigorously organised, is always subject to the same physical principles and always functions according to the same mechanisms (in well-defined geographical conditions). Its variations are therefore not variations in its nature, but are the result of variations in its intensity. Before we claim that there is some relationship between two parameters or, worse still, that one is the cause of the other, we must understand and recognise the respective places of each of these parameters, and the sequences linking them within the context of general circulation. We cannot say that something is tied to something else without having first established the reality of the physical link between them. No matter how sophisticated the statistical analysis, it is completely worthless climatologically if it is performed ‘blind’, with no basis in any previous and solid meteorological analysis; observation and, above all, proof of the reality of physical causal links are indispensable in the study of climate. Even if the schemas presented above, involving MPHs, are as yet incomplete, they represent the most realistic version (i.e. the version closest to meteorological reality), because they are based faithfully on direct observation. Let those who contest them, or query (as does Meteo France), the role of MPHs at the origin of general circulation, supply proof that this concept does not conform to observed reality. Let them indeed suggest an alternative, indisputable concept, equally well supported and based on the facts as they are observed. The concept of the MPH as applied to the general circulation of the atmosphere has the advantage of representing, in the field of current research on this subject, the only schema embracing the initial cause of circulation, and the cause of its daily, seasonal and indeed palaeoclimatic variations. It offers a complete and coherent perception of the dynamic of meteorological phenomena, encompassing all events, normal or extreme. It is applicable to all scales of intensity, time and space. This is why we shall constantly come back to it in the course of this book.

116

General circulation

[Ch. 5

For ease of presentation, the characteristics of circulation discussed above are essentially averages. This initial approach must be completed and corrected through an analysis of the disturbances which are an integral part of that circulation, and constitute the only way by which weather truly manifests itself, and, by repetition, becomes climate.

PART II DYNAMICS OF THE WEATHER: DISTURBANCES

We have fallen into the bad habit of making a distinction, often cut-and-dried, between the permanent field and the perturbed field. Symptomatic of this way of seeing things is an observation by Rochas and Javelle (1993): “the defining trait of general atmospheric circulation at temperate latitudes is the existence of prevailing westerly winds upon which are superimposed disturbances, which succeed in concealing the dominant character of those winds”. We cannot, however, confuse the artificially stable statistical vision with synoptic reality which is necessarily mobile, nor liken the disturbances to transient ‘grafts’, thereby dissociating them from the circulation of which they are an integral part (cf. Figures 5.4 and 5.6). The so-called permanent field is in reality but a fiction built upon mean values of pressure and wind, a convenient statistical procedure to establish dominant features, but it does not really exist since it does not correspond to real weather, based on actual experience. The so-called perturbed field is the only true one, representing circulation in all its vibrant, dynamic reality. It is therefore vital to take account of the dynamics of circulation and weather, which is the only route to the analysis of climatic dynamics resulting from the repetition of weather aspects.

6 Pluviogenesis

Precipitation, which is of primary importance in climate, is a quite random parameter, and very discontinuous in time and space. All meteorological parameters are affected by perturbed weather, but precipitation occurs only in such circumstances, and is indeed the event most associated with them, resulting as they do from a breakdown in the equilibrium of the concentration of water in the air. Before analysing disturbances, it is necessary to remind ourselves of the conditions necessary for precipitation to form. Often associated with the formation of precipitation are condensation nuclei, such as dust, volcanic ash, pollen grains, sea-salt crystals, ice crystals and charged particles (ions). These nuclei are much in evidence in air and, although they stimulate the formation of water droplets, they cannot by themselves be the explanation for rain. In fact, three conditions need to be present simultaneously: 1 2 3

the existence of precipitable water, and its rapid renewal; the unleashing of a vertical ascending movement; the presence of an aerological structure which will not hinder ascent.

6.1 PRECIPITABLE WATER At any given moment, the water reserve in the atmosphere is equal to 13 000.109 (13 000 billion) tonnes. Water is lost in the form of precipitation (rain, snow, hail), or of condensation (fog, dew). The total lost from the atmosphere annually is of the order of 450 000.109 tonnes (or m3), a rate of 15 million tonnes every second. Precipitation is compensated for by evaporation. Under normal conditions of pressure, vaporisation heat of the order of 600 calories (or more precisely, the process being temperature-dependent, 597 calories at 0°C, 580 calories at 30°C) is needed to transform 1 -gramme of water into water vapour. This expenditure of energy by air and evaporating water explains the thermal behaviour of water surfaces, and the low

120

Pluviogenesis

[Ch. 6

Table 6.1 Variation of potential humidity as a function of temperature (pressure not taken into account). The maximum saturated mass is expressed in grammes per cubic metre of air.

T °C

-20

-10

-5

0°

+5

+10

+15

+20

+25

+30

+35

+40

temperatures observed above them. The quantity of water which a volume of air is capable of evaporating and containing (potential humidity) is a function of its temperature (Table 6.1). Energy supplied during evaporation is conserved in the form of latent heat, which may be transported over thousands of kilometres without being utilised. This energy is also called heat of condensation, because it is liberated during a change of state, for example on return to liquid. Therefore 600 calories per gramme of water condensed are available for heating the atmosphere. This process is of great importance for the atmospheric energy budget, as about 85% of the energy comes from the release of latent heat during pluviogenesis. The distribution of mean precipitable water, or of real water vapour content (specific humidity) is greatest over oceans and in tropical regions, and least over continents (with the exception of tropical forests) and mountains, and at high latitudes, in the direction of which values decrease. Tropical disturbances therefore have a ready supply to draw upon, while in extratropical regions disturbances have to rely on considerable horizontal transfer (advection). Even in well-supplied regions, there is no direct link in any given place between precipitable water and actual rainfall. Heavy rain is not necessarily associated with high humidity: for example, the areas around the Red Sea are very humid, but rainfall is absent; the mighty precipitable water of the Indian monsoon passes above Somalia in summer, and above that the Atlantic monsoon - yet the area benefits little from this, and has humid deserts where contact condensation is much more likely than precipitation. Precipitable water is advected over differing distances and, especially outside the tropics, paths may be very long. Horizontal transfers take place between areas of surplus and of deficiency, the main exporting regions in the tropics being between latitudes 12°N and 35°N, and 10°S and 35°S. Near the Equator, between 12°N and 10°S, part of this potential is used for the activity of the meteorological equator and the provision of monsoon circulations. Peixoto and Oort (1983) stressed the fact that water vapour circulates in the lower troposphere and most is transported in the planetary boundary layer, i.e. the lowest 1500 metres, whilst “water vapour trans­ portation is clearly influenced by topography” (as are the trajectories of vector fluxes: cf. Figure 5.1). The circumstances of water vapour transfer therefore confirm the importance of aerological circulation spaces in the lower layers: thus, in North and South America, east of the Rockies and the Andes, nearly all atmospheric water originates from the Atlantic Ocean. Transfers of precipitable water highlight the importance, in pluviogenesis, of the factor capable of promoting advection across distances of hundreds or thousands of kilometres. Also evident is the importance of renewal of the water stock and

Sec. 6.2]

Origin of an updraft

121

therefore of energy. An intense disturbance will, of course, need to be abundantly supplied, with rapid renewal to ward off degeneration. Let us consider some very basic examples of water transfer. A flux of air at 15 °C, at the surface, and if saturated, contains 12.85 g of water per m3 (Table 6.1); cooled to 0°C (which would require an ascent of 3000 metres), this m3 would give 8 g of water. To generate 1 mm of rain, which is 1 litre per m2, advection and uplift of 125 m3 of air are necessary over that square metre. Air at 10°C (ascent reduced to 2000 m) requires an advection of 219 m3 to give 1 mm of rain, the figures for air at 20°C being 4000 m and 80 m3. A downpour of 300 mm, like the one which burst the banks of the River Ouveze on 22 September 1992, causing the Vaison-la-Romaine disaster, therefore requires, per m2, an advection of 24000 m3 of saturated Mediter­ ranean air at an initial temperature of 20cC. For an area of 1 km2, the necessary advection in this case was therefore of the order of 24.109m3 of air; over the 90 km2 of the lower basin of the Ouveze, where the torrential rain fell, 2160.109m3 of air passed in about 3 hours between 13H 30 and 15H 30. This represents an advection of at least 12.109m3 of warm, moisture-laden Mediterranean air per minute in the lower layers! This situation lasted between 21-24 September 1992, from the Pyrenees to Italy, with heavy downpours; the departement of the Herault saw 448 mm of rain, the Lozere and the Gard 310 mm, and there were floods in Genoa. Transfers of water of such amplitude and and intensity can be caused only by a factor involving con­ siderable dimensions and energy, ruling out merely local factors of thermal convec­ tion or relief, hypothetical high-altitude “suspected thalwegs”, or even mysterious “mesoscale regenerative convective systems” (Giorgetti et al., 1994; Leroux, 1995b). This energetic factor must obviously be identified and analysed if we are to be able to forecast the risks associated with it (cf. Chapter 7).

6.2 ORIGIN OF AN UPDRAFT

Cooling is necessary to bring about condensation of water vapour, by contact or in mass. As warm, moist air passes over a cold surface, cooling takes place, together with saturation of the air in contact, and a deposition of dew (the so-called ‘hidden precipitation’ is not precipitation in the normal sense, nor is it hidden, as this condensation is very real). A more intense cooling, for example through night-time radiation, lowers the temperature of the layer of air near the ground and results in the formation of so-called ‘radiation’ fog in the air mass. Cooling is, however, most frequently caused by the upward motion of air, which involves: decompression (expansion), cooling (of the order of 5-6°C per kilometre in moist air), saturation (or even supersaturation of prolonged duration) and condensation into droplets (clouds). In moist air, the release of latent heat during ascent slows cooling. The decrease, following the saturated adiabatic, is of the order of 0.5°C every 100 m, whilst in dry air it may be greater than 1°C per 100 m (following the dry adiabatic). Clouds thus formed are short-lived, and may disappear by evaporation if there is warming during the hotter part of the day or through the foehn effect. However, they may also persist for long periods without any rain falling,

122

Pluviogenesis

[Ch. 6

since each water droplet remains isolated from its neighbours by its own magnetic field, its tendency to fall being balanced out by air resistance. The type of cloud formation depends on the energy of the updraft, the aerological stratification and the amount of water supplied (i.e. amount of energy, which is indispensable for the vertical development of clouds). The triggering of an updraft, with resultant cooling, may occur because of the intervention of either the thermal or the dynamical factors. The thermal factor (perceptible heat and latent heat) and the dynamical factor (involving relief and the aerological aspect) are closely associated in nature. Their separation below is merely for the sake of clarity.

6.2.1 The thermal factor

When air is warmed through contact with the ground it tends to rise (though the ocean - cf. Figure 1.6 - cannot achieve sufficient warmth to cause updrafts). This is known as thermal (or natural) convection. Above bare earth, more intense warming may bring about the formation of a twisting column of warm, dry air, a familiar sight in desert areas during the hottest part of the day. Names for this include dancing djinn, whirlwind dervish, zobaa (zobayya in Egypt and Mauretania), paraca

300-

£

E

fks. (i

500 •

• 6

700 -

850-

Clouds

Phenomei

clear sky

na I ------------

sparse cloud isolated cu

dfy f°9 heug

summer (non-rainy)

300-500 km

800-1200 km

Width Depth of monsoon 0-1000 m

JTT'

Tl

rainy: wintering

transitional

dry

illl

VME

IME

Season

- 3

i

_______________ —X-*,'

Al

- 9

1000 to 3000 m

1500 m

squall lines cumulonimbus

overcast altocumulus WllvwUI V IUIUU altostratus

overcast stratus

windstorms thunderstorms

continuous rain

light rain, drizzle

Meteorological equator------------southern trade inversion

----------- harmattan Tl

Figure 10.1 Diagrammatic cross-section of the troposphere in August along the 0° meridian over western Africa and the Atlantic (after Leroux, 1970, 1974).

Sec. 10.1]

The meteorological equator: the evolution of a concept

211

Figure 10.1 deals with this stage in the idea, showing the weather zones associated with the furthest advance of the monsoon over western Africa in northern summer. The ITF (IME) structure is identified with discontinuous rain, thunderstorms and stormy squall lines. The ITCZ (VME) structure is identified with continuous, abundant and non-thundery rain, while beneath the southern trade inversion rain is light, tending towards drizzle. Also shown are the nuclei of strong winds, African easterly jets (AEJ), in the middle layers, and tropical easterly jets (TEJ) in the upper layers (Leroux, 1974). The vertical structure of the meteorological equator can be best described by the elaboration of mean monthly charts of the overall circulation of tropical Africa, at various levels: surface, 1000 m, 1500 m (850 hPa), 2000 m, 3000 m (700 hPa), and 5500 m (500 hPa), with a view to constructing meridional vertical and zonal sections of the African troposphere (Leroux, Atlas, 1980: 168 vertical sections, 108 zonal sections). Figure 10.2, based on these results, shows mean monthly positions of the meteorological equator (ME) over western Africa at about 0° longitude, and at 20°E (Congo basin) and 40°E (Kenya-Ethiopia-Mozambique Channel), for JanuaryFebruary, April, July-August and October (Leroux, 1983). According to the level considered, the ME possesses two structures and two different behaviours: •

•

In middle layers the ME is vertical (VME), representing the ‘planetary’ structure of the ME, which is the prolongation onto land of the oceanic structure (cf. Figures 4.3 and 4.4). This structure slowly moves around throughout the year in a restricted zonal band. Its position, annual migration and characteristics are almost independent of conditions at the surface, its dynamics emerging from general circulation. In lower layers the ME has an inclined structure (IME), riding above a shallow monsoon circulation. The annual migration of its surface line is associated with the thermal behaviour of the substratum: if the substratum is entirely continental, then migration is rapid and covers some distance,- as over eastern Africa; over oceans and forests, however, it is curtailed and halted (Figures 4.5 and 4.6).

Nowadays the term ‘intertropical front’ is restricted to French-speaking Africa, although identical structures are found elsewhere. The term ITCZ is current almost everywhere. The abbreviation is used loosely to describe either the surface line of the ITF, inactive and invisible because of its lack of cloud in this type of ME, or the zonal alignment of dense middle-layer clouds shown on satellite images. This visual approach to the term is an inappropriate one, since any vaguely zonally-aligned cloud belt can then be labelled an ITCZ, without any structural analysis whatsoever. We hear of ‘doubling of the ITC’, or of ‘doubling of the ITCZ’, designations of convenience representing a backward step to the days when the idea of two ITFs (northern and southern) was mooted, as if there might exist at one and the same time two equators for general circulation, which is a genetically absurd hypothesis. Such a (temporary) configuration is observed when a pulse line nears the ME, ready to penetrate it and intensify its activity (cf. Photo 20).

212

The meteorological equator

[Ch. 10

Figure 10.2 Monthly positions of the meteorological equator over Africa (north-south sections at 0°, 20° and 40° longitude, after Leroux, 1983).

Figure 10.3

Vertical structure of the meteorological equator: (a) ocean; (b) continent.

Sec. 10.2]

The inclined meteorological equator (IME)

213

The next step, then, would be to throw off the old terminology of ITFs and ITCZs, unsuitable labels, but hallowed by usage (Leroux, 1992c). The notion of the meteorological equator (ME) should be to the fore, its structure specified in terms of the inclined meteorological equator (IME), or the vertical meteorological equator (VME) (Figure 10.3).

10.2 THE INCLINED METEOROLOGICAL EQUATOR (IME) 10.2.1 The vertical structure of the IME

The surface line of the IME runs along the axis of the thermal lows associated with the north-south movement of the Sun. This thermal character limits active migration to the surfaces of land masses, and to the lower layers by dissociating them from the middle layers where this thermal influence is not felt. The ME is therefore inclined in the lower layers (above the monsoon flux), with the surface shared by the ground­ level thermal low and the ME in the middle layers almost horizontal whilst these entities are hundreds of kilometres apart. The IME therefore becomes another component of tropical stratification. In this inclined structure, there are significant discontinuities in characteristics between the adjacent flux layers (the dew point temperature being the traditional criterion for tracing the ‘ITF’): the moist monsoon, more or less continentalised, is overlaid by a warm, dry continental trade. The IME is also a wind discontinuity. Confluence occurs in different ways, as in Figure 10.4: the fluxes may be fairly parallel, or come together in a marked confluence, or flow in diametrically opposed directions. They may even have opposing directions but parallel trajectories, which involves a strong shear effect, or the ME may represent an axis of diffluence, as happens at the western edges of AAs (where trade ‘breakdowns’ are frequent). The type of confluence affects the nature of the associated convergence, with conditions ranging from those possibly most encouraging to updraft (opposition of fluxes) to those discouraging it, i.e. diffluence and wind shear. There is pronounced wind shear between a monsoon (lower stratum) and an easterly trade (upper stratum). As a result of discontinuities in characteristics (especially humidity) and wind, the IME structure is particularly unproductive. Cloud formations - possibly associated in the monsoon with pulses whose energy gradually diminishes towards the IME, or with thermal convection which increases as the thermal low is approached and the monsoon is progressively warmer - are decapi­ tated, evaporated or sheared off as soon as they rise into the trade. The IME structure is not then particularly associated with cloud formations (and therefore cannot be called the ITCZ, as it is not the seat of convergence). In spite of the large amount of precipitable water in the monsoon, with high tem­ peratures favouring instability, and in spite of the presence of mountain ranges, there is practically no rain. The southern Sahara, into which the Atlantic monsoon irrupts in northern summer, is a telling example of this. The presence of a monsoon is not always a guarantee of clouds or rain, and the normal weather brought by a

214

The meteorological equator

[Ch. 10

f2---------------fl________________________________________

to

f0--------------------------------------------------------------------

— fo

fO--------------------------------------------------------------------

CONV fl---------------------------------------------------------------------

f2---------------------------------------------------------------------

------------- isotach

f3>f2>fl>fo

Figure 10.4 Types of confluence along the meteorological equator.

continentalised Atlantic monsoon is dry and humid, or even stifling, but rainless. It cannot rain because of the unproductive character of the stratification, unless through the intervention of some foreign perturbing element able rapidly to modify those structural conditions discouraging vertical cloud development. A squall line, the typical disturbance of the IME structure, will then result. 10.2.2 IME activity: squall lines (SL) Rain falls from squall lines, during events such as windstorms, thunderstorms and showers, all associated with tall cloud masses (cumulonimbus) forming a curve with north-south orientation (Figure 10.5). This definition may apply to other phenomena (particularly active pulse lines) but, especially in Africa, it has become reserved for this mobile disturbance in the monsoon flux. This disturbance was at first attributed to (essentially thermal) ‘local storms’, and then to a kind of‘monsoon front’. Later, it was called a ‘tornado’ because of the eddying wind that precedes it, although it has nothing structurally in common with the grimly famous tornadoes of North America (which in fact propagate on the leading edges of powerful MPHs, where strong thermal contrasts develop from stormy supercells). The origin of a factor from outside the IME structure which

Sec. 10.2]

The inclined meteorological equator (IME)

215

Figure 10.5 Vertical zonal structure of an active squall line. (cb: cumulonimbus; Ac, As: altocumulus, altostratus; $: gusting wind)

changes its unproductive character is still a matter of controversy, as recent labels suggest: ‘easterly unstable storm line; easterly disturbance; African wave’. Its liken­ ing (through similarity) to an easterly wave has led to a largely factitious debate: •

•

the trajectories are fundamentally different, since in the IME only the upper flux is from the east, whilst the squall line moves from east to west, i.e. running counter to the direction of the westerly/south-westerly monsoon of the lower layers; structural conditions are also different, with a trade inversion in the case of the easterly wave (without shear) and the IME in the other, with the trade above the monsoon.

Moreover, the cloud formations are specific to squall lines, which move in an arc generally aligned north-south (Photos 49 (colour section) and 50). Figures 10.6 and 10.7 show in diagrammatic form the process of formation of a squall line (SL) over northern Africa, after the penetration of a powerful trade wind pulse into the Atlantic monsoon blowing in the opposite direction. The continental trade or harmattan is warmed, and normally passes over the monsoon flux (Figures 10.1), but a denser pulse cannot rise and clashes with the south-westerly flux in the

216

The meteorological equator

[Ch. 10

VME: Vertical Meteorological Equator EMI: Inclined Meteorological Equator SL: Squall Line

Photos 50 9 June 1986, Meteosat (after Meteosat Image Bulletin, ESA, 12H UT, visible and infrared - the IR brings out the tall clouds with their cold summits). These images show: the absence of clouds in the unproductive IME structure, except on the squall line; the zonal alignment of clouds associated with the VME clearly visible to the south, below the south­ facing coast of West Africa; and the disposition of clouds in the squall line (SL) welling up to the fore of a knot of easterly wind (pulse) entering the monsoon. The oval outline is plainly seen.

lower layers. The pulse in the continental trade, carrying dust particles as evidence of its dynamism, is fragmented by a relief between the Atlas Mountains and the Hoggar, the Hoggar and the Tibesti, or the Tibesti and Darfur (cf. Photo 45). This channelling by the relief causes the pulse to be strengthened locally (Venturi effect). The two latter corridors open onto the very places where SL formations are most frequently observed. The anticyclonic nucleus causing the trade pulse is recent in origin, and given the fact that even at equal temperatures moist air is lighter than dry air, the monsoon is forced to give way. At first, the monsoon escapes laterally owing to the compression generated by the pulse, and then the contact surface between the fluxes is stretched (Figure 10.7). Then the monsoon rises and a great deal of energy is released with updrafts and cloud formations above the pulse line, giving it its characteristic bowed form. Cumuliform clouds on the advancing front spread out in the middle layers above the pulse.

Sec. 10.2]

The inclined meteorological equator (IME)

SURFACE

W Figure 10.6

E

217

VERTICAL SECTIONS

Process of formation of a squall line in tropical Africa.

Weather associated with squall lines A squall line, which moves at an average speed of 40-60 km/h from east to west, usually brings with it the following weather pattern:

•

In front of the SL there is calm and relative warming; heat, humidity and lack of wind combine to create uncomfortable conditions, with updrafts damping the westerly wind near the ground.

218

The meteorological equator

[Ch. 10 ZONAL CROSS SECTIONS

z

L .

- 1 Westerly winds

Meteorological Equator

i—______ ———---------------------------------------------------- ——I

3 Monsoon flattening phase, formation of pockets

4 Blocking phase, extension ofcontact suface

5 Monsoon/easterly wind pockets burst

Figure 10.7

•

•

Phases in the formation of a squall line: zonal vertical sections.

Violent gusts of wind occur as the cumulonimbus (Cb) line passes over, causing great material damage and raising sand and dust. These events are popularly but erroneously called ‘tornadoes’. Sometimes, especially in the vicinity of the surface line of the IME, the sudden winds, accompanied by whirling dust, are the only manifestation of the SL, which may then be known as a ‘dry tornado’. Later, wind-driven rain falls with great force, accompanied by thunder and lightning, and the temperature falls rapidly by about 10°C. After the storms have passed, continuous rain falls from dense cloud masses in the middle layers (Ac and As), while the surface wind blows from the east. Then the cloud thin, the rain stops and the wind turns westerly with the re-established monsoon.

The next SL will arrive between one and three days later, though a delay of a week or more is known; like the heavy rains, such irregularity is characteristic of the regions beneath the IME structure.

Sec. 10.2]

The inclined meteorological equator (IME)

219

Photo 51 2 October 1988, Meteosat 3, visible, 12 h UTC. A meridional pulse, moving down across the Sudan along the western foothills of the Ethiopian highlands, has caused the formation of a squall line over the eastern part of the Congo basin. In its southern section, the SL is intensifying the activity of the EMV structure.

When displacement is zonally westward, the easterly pulse line will easily lift a tenuous, warm monsoon, and rainfall from the SL is intense and stormy. If, however, the movement is in a more meridional direction, the pulse burrows beneath the IME structure (in the direction of the VME), opposing an ever denser monsoon, which leads to thickening clouds and heavier rain from the SL, but brings it to a halt before finally absorbing it. The SL, reaching into the heart of the tropical zone, therefore becomes a factor in the activity of the VME (Photo 51). Though characteristic of an IME-type stratification, SLs are also met in Africa in the inclined structure of the Interoceanic Confluence (IOC) to the south of the Congo basin (Figure 10.2b): the low-level westerly Atlantic flux, humidified by the forest, is ridden over by the trade which, to a lesser or greater degree continentalised, is coming from the Indian Ocean. An IME structure is also found in summer over Pakistan and north-west India, where the shallow Indian monsoon (about 1000 metres deep) is capped by subsidence and by the westerly continental trade off the Arabian peninsula (belat) and the north-westerly seistan from the Iranian plateau. The pulses driving these dry fluxes cause SLs, here andhis, or northwesters after their motion, bringing wind-, sand- and thunderstorms, but not much rain in the west in the Thar Desert. This type of inclined structure is also known in the Amazon Basin, especially in the south around the exit of the Gran Chaco where the upper flux is continentalised, but also in the eastern and nordeste regions of the Basin, where the upper trade from the Atlantic has more or less spent itself passing over relief. In northern Australia, in southern summer, the IME structure is also present. The depth of the monsoon is about 1000 metres in November and March, and 2000 metres over Darwin in January-February (Gentilli, 1971). 10.2.3 The active inclined meteorological equator The IME structure is not as unproductive in some other places, because the superposition of fluxes may bring accumulated humid air of both trade and monsoon

220

The meteorological equator

[Ch. 10

fluxes. The Indian monsoon is overridden in the west (in the Arabian Sea) by a continental flux, but in the east by the maritime trade from the Pacific, which has previously passed over the Chinese monsoon (cf. Figure 4.4). Over Australia, only the central part of the IME is overridden by a continental trade; to the east and the west, the Australian monsoon is covered by maritime trades. The IME structure may moreover cease to be unproductive, and may become the seat of intense activity (the active IME), as contrasts between the fluxes are eradicated. This occurs in the Mozambique Channel in southern summer, when the Madagascar monsoon and the Indian Ocean trade, both approaching saturation, meet almost head-on along a practically vertical IME structure over 2000 metres deep (Figures 4.5 and 10.2c, Photos 52 and 54). This also happens in the case of the North Pacific trade when it rises above the Chinese monsoon, and likewise when the South Pacific trade meets the Australian monsoon (in the Queensland area) and passes above it, and when the North Atlantic trade passes over the Panamanian monsoon at the latitude of the Isthmus (Figures 4.3 and 4.4). The IME, like the VME, may now harbour intense activity, with

Photo 52 9 January 1990, 12H UT, visible, Meteosat. Over northern Africa, the IME, whose surface line runs along the edge of the forests (Figures 4.5), has no cloud. Those which have developed are along the VME which is in the middle layers along the line of the south-facing coast of West Africa. The line of the IME runs around the dense forests of the Congo basin and along the western Rift Valley. The IME becomes active over Mozambique, turning northwards to the east of Madagascar and reaching the VME over the central Indian Ocean. Air advected by the southern MPH1 (east of South Africa) encounters the IME directly over Mozambique and its Channel. The Atlantic AA supplies the monsoon near the southern part of West Africa and the ‘monsoon’ (or diverted trade) near the western Rift Valley, as well as the Angola-Zambia thermal low near the IOC.

Sec. 10.3]

The vertical meteorological equator (VME)

221

abundant clouds and heavy rainfall; its structural limitations can be overcome by the rich potential advected by monsoon and maritime trade fluxes. 10.3 THE VERTICAL METEOROLOGICAL EQUATOR (VME)

The Intertropical Convergence, or ITCZ, has long been known by mariners by its original names of ‘equatorial chimney’ and 'pot au noir', and is defined as a region of ‘equatorial calms’ because of the updrafts that characterise it. However, the structure of the ITCZ over land masses has long been ignored, since the ‘convergence’ is there no longer directly perceptible, another structure having interposed itself in the lower layers: the inclined EMI structure crossed by the monsoon flux. The ‘intertropical convergence’ is now concealed within the middle layers of the atmosphere, where the planetary vertical structure of the equator of general circulation lies. 10.3.1 The VME over the oceans

Over the oceans and away from continental influence, the ME is vertical throughout its approximately 6000-metre deep structure (Figures 10.3a). Shallow intertropical lows, dynamic rather than thermal in origin, form a continuous corridor whose position is adjusted by the dynamics of AAs, themselves conditioned by the varying strengths of MPHs. The VME migrates but little, and in the course of a year may drift by 10° to 15° in latitude, its structure moving en bloc and retaining its vertical nature (cf. Figures 4.3 and 4.4, showing the differences between the IME and the VME). Its constant nearness to the geographical Equator means that it is the axis for encounters between trade fluxes. As trade inversions progressively rise and finally disappear, the VME represents the axis for the concentration of energy carried beneath the trade inversions, and for upward movements maintained by the energy of the trades. Its position at any one time - and likewise its yearly motion - are determined by the respective power of pulses coming from north and south. The VME therefore combines, in terms of structure and energy, the most favourable conditions in the tropical zone for rainfall. Updrafts are unleashed with great power, with thermal origin ruled out over an ocean surface. Lower-layer pulses in the trades, ending up in a weakened state in the VME, find there a new surge of energy from the tropics. The VME thus acts as an axis for dense cloud formations (whence the mariners’ traditional unfailingly ability to recognise it) and, in satellite images, its general modern appellation, the ITCZ (although this is not always true over land, where its structure is double). The VME’s structure, which is thermally homogeneous and uniformly humid, concentrates cloud masses in a zonally-oriented band about 200-300 kilometres wide (Photos 53 and 54). The VME is also an axis of much fine, regular and continuous rain, normally unaccompanied by thunder. Clouds here are typically in the middle layers, and often contiguous, though breaks in the clouds appear frequently; the VME is potentially active, but not always activated. Activity in the VME is orchestrated by the arrival of pulses, either

222

The meteorological equator

[Ch. 10

Photo 53 5 April 1986, 17H UTC, GOES W (after Guillot and Trigaux, Veille Climatique Satellitaire 12 (5), 1986). The VME runs noticeably unbroken and straight for more than 5 000 kilometres in longitude, from 85°W to 135°W. In the northern meteorological hemisphere appear six MPHs: ‘a’, ‘b’ and ‘c’ are over North America to the east of the Rockies, and kd’ nears the tropics, spreading across the Atlantic and beyond to the Pacific over the Isthmus, north of the VME; MPH ke’ causes a northward flow west of the Rockies, and poorly-defined T merges into the North Pacific AA. In the southern meteorological hemisphere there are three distinct MPHs, kg’ and kh’ in the Pacific and ki’ east of the Andes. Air from another MPH, kj’, passes into the ‘Easter Island’ AA which spreads westwards to the south of the VME, here with a more southerly position (cf. Chapter 14, El Nino).

simultaneously or individually, from trades to north and south, and as each new­ comer appears there is a temporary intensification of convergence, perhaps with mid­ layer cumulonimbus cloud development, showers and thundery episodes. When a pulse line nears the VME, the increase in energy and favourable structural conditions encourages clouds to form, and at this time one may visually have the idea of a "doubling of the intertropical convergence’. This simple snapshot does not, however, depict a particular ME type (the two lines of clouds having different origins) since, a few hours later, the pulse line in question has merged into the VME.

Sec. 10.3]

The vertical meteorological equator (VME)

223

Photo 54 4 January 2007, Meteosat 05. visible, 06 UTC. The VME stretches across the Indian Ocean from Somalia to Indonesia, in an almost continuous zonal alignment just south of the Equator, towards which pulse lines (PL) are moving from both north and south. In the western part of the Ocean, off Africa, the kaskasi (northern trade), which is being transformed into the Madagascar monsoon, is holding off the lower-layer structure of the IME near the Mozambique Channel and Madagascar, the activity of the IME manifesting itself as dense cloud formations.

The small width of the VME, and its limited amplitude in migration, con­ siderably restrict the area affected by rain associated with this structure. It plays its part, along with the AA and trade inversion, in maintaining the large extent of what might be termed ‘ocean deserts’.

10.3.2 The ME over continents: IME and VME Over continents the depth and migration of thermal lows modify ME structure in the lower layers (Figure 10.3b), and the effects of this spill over for some distance onto neighbouring oceans. The IME structure, an axis of confluence, is no longer the scene of convergence because of shear (Figure 10.4). In the middle layers, uninfluenced by the substratum, the VME structure holds firm throughout a depth of 3000-4000 metres, beginning at altitude of about 2000 metres and ending at about 6000 metres. Here the VME is the axis of confluence for the upper layers, above the lower fluxes, i.e. trades and

224

The meteorological equator

[Ch. 10

Figure 10.8 Structure of the troposphere and components of pluviogenesis over western and central Africa. TEJ: tropical easterly jet; nEAJ: northern easterly African jet; sEAJ: southern easterly African jet.

monsoons (Figure 10.8). Advection contributes humidity, evaporated from the lowerlayer monsoon flux, on the one hand above the trade inversion, and on the other above the structure of the IME. The passage of pulse lines in the lower layers also causes updrafts of potential, transported by the monsoon, directly into this structure. In spite of its being moved into the middle layers, the VME over continents (or over oceans covered with this double IME/VME structure) retains considerable amounts of precipitable water, and still acts as the axis of concentration for this water, for clouds and for rain. The density of zonal (sometimes separated) cloud formations, and its activity (not always constant), are affected by the arrival of pulse lines moving in the monsoon, or of squall lines. The resulting temporary intensifica­ tion engenders cumulonimbus formations, and stormy characteristics in the middle­ layer clouds with their continuous rain. The range of the yearly migration of the continental VME is enlarged in comparison with that of the oceanic VME (Figures 4.3 and 4.4). The VME shifts between about 10°N and 10°S, a 20° amplitude, double that of the oceanic VME but less than half of the free displacement of the IME (for example, over eastern Africa, Figures 10.2), which displays an amplitude of more than 40° in latitude (from north-west Madagascar to the south of the Arabian peninsula). The VME produces in its path abundant and regular rains, of moderate intensity but long-lasting, across an area which is greater on continents than on oceans. Monthly rainfall charts for Africa (Leroux, 1983) show that rainfall of or exceeding

Sec. 10.4]

Conclusion

225

Table 10.1 Monthly rainfall readings (RR) in mm at: Abidjan (Ivory Coast, 5°10'N); Libreville (Gabon, 0°20'N); Kisangani (Zaire, 0°30'N); Mogadishu (Somalia, 2°10'N); Gallacaio (Somalia, 6°40'N). In bold: rain associated with the passage of the VME.

RR mm

J

F

M

A

M

J

J

A

S

O

N

D

Year

Abidjan Libreville Kisangani Mogadishu Gallacaio

29 331 83 1 0

48 305 104 0 3

109 410 146 9 1

156 363 172 58 24

344 291 181 56 60

605 18 112 82 2

238 1 116 58 0

34 9 179 40 2

52 113 178 23 1

183 384 215 27 41

185 506 161 36 14

104 389 114 9 1

2087 3120 1761 399 149

200 mm per month occurs beneath the VME structure. To either side of the VME, beyond latitudes 10-12°N and S, rain is essentially a product of squall lines characteristic of the IME to the north and of the IOC to the south. The same slow, regular migration of rainfall is seen in Amazonia and (excluding effects of relief) in Indonesia. The double passage of the VME asserts the bimodal character of precipitation, with two rainfall maxima whose separation in time is a function of latitude. This pattern of two rainy seasons interspersed with seasons of less rain (rather than completely dry) is not a product of some hypothetical process of ‘equinoctial rains’, nor are ‘solstitial rains’ involved. The annual migration of the VME structure is at its origin. The examples in Table 8.3 of rainfall figures in equatorial Africa show (Table 10.1) that, in the area swept by the VME, the situation involving bimodal heavy rains is not always the case, and local conditions may be more influential. Annual totals for Abidjan, Libreville and Kisangani illustrate the ‘pluviometric security’ ensured by the two passages of the VME. In Abidjan and Libreville, from June-July to September, the dearth of rain is a result of the presence of the maritime trade, capped by its inversion (Figures 10.1). At Kisangani, squall lines in the IME (NovemberMarch) and the IOC (June-September) sustain the figures in the period between the passages of the VME in April-May and October. In Somalia, opposing factors are at work: in winter, the northern trade carries with it an unproductive inversion, and in summer the Indian monsoon is affected by divergence, the flow rate is rapid (Somali jet) and the ME sheared. Such conditions produce little rain, in spite of rich precipit­ able water, and the two passages of the VME are firmly isolated to May and October at Gallacaio, whilst further to the south at Mogadishu the VME’s passages are masked by the arrival in the southern winter of pulse lines in the trade/Indian mon­ soon (Figures 4.6). These characteristics naturally do not arise from the intervention of some hypothetical ‘Walker cell’ (Chapter 5).

10.4 CONCLUSION The meteorological equator thus constitutes the fundamental structure bringing rain to the tropical zone. Long recognised over the oceans, if somewhat hidden over land,

226

The meteorological equator

[Ch. 10

its vertical structure is generally ignored, and erroneously labelled the ‘ITCZ’. A planet-wide discontinuity, the meteorological equator exists in its two forms of the unproductive IME, bringing storms and occasional, haphazard rain, and the VME with its abundant and regular precipitation. As an energetic equator, the ME possesses the optimum conditions for the development of disturbances, and, naturally, of paroxysmal tropical cyclones, principally within the structure of the VME, though they may occur in the IME structure too.

Photo 29. Alpine valley filled with cold air in winter (photo M.L.). The top of the MPH is decked with stratified cloud, and radiation fog covers damp surfaces (River Dranse de Morzine and forest). The cloud cover prevents insolation, and in the course of the day there is little rise in temperature in the valley, which is then colder than the summits. Aerological stratification of this kind plays a major part in concentrating pollution, especially when there is no opportunity for the outflow of air from the valley through a low-lying area. MPH air will be only slowly brought out by any breezes blowing in the direction of high ground.

Photo 30. The Alps beneath a sea of clouds (photo M.L.). The summits appear at the same level from the top of the cloud deck. Above the sea of clouds, the Sun shines, and during the day the valleys are colder than the summits. Aerological stratification imposed by MPHs determines the altitude of villages and resorts wishing to enjoy the best of the sunshine, above the MPH and the cloud layer which is normally at about 1500-1600 metres.

Photo 42. Beneath the trade inversion above the island of Sal, in the Cape Verde group, May 1995 (photo M.L.). Note the cloud shadows on the bare landscape of the island.

Photo 43. Above the trade inversion over the island of Sal, in the Cape Verde group, May 1995 (photo M.L.). The clouds are immature and mostly separate from each other, and lie at about 800 metres, below the inversion layer; the upper stratum here is dust-free.

Photo 44. The trade inversion along the coast of Mauritania, July 1993 (photo M.L.). The inland penetration of the maritime trade progressively dissipates the cloud formations at the inversion level.

Photo 46. Above the continental trade inversion over the Sahara (photo M.L.). Dust of yellowish and reddish hues reflects the Sun’s radiation and hides, either partially or totally, the ground below.

Photo 47. The trade inversion layer photographed from Mount Mali (Fouta Djalon, Guinea), looking north-east, at an altitude of 1537 metres, January 1988 (photo M. Carn, ORSTOM). It is possible to see the inversion layer only when the lens is at the same altitude, as is the case with this photo. The inversion is still relatively low at this time of year (winter) and at this latitude (12 N). The concentrated dust layer is seen as a grey band across the top of the picture.

Photo 48.

A haboob in Mauritania (photo supplied by Z. Nouaceur).

Photo 49. A squall line over western Africa, 23 May 1992, Meteosat, visible (after IONIA, ESA/ESRIN). The development of cumulonimbus cloud is due to an easterly pulse (issuing from the continental trade and surrounded by the monsoon) whose contours may be easily seen.

Photo 58.

A tropical low over southern Arabia, 5 October 1992. Meteosat, visible (after

11 Tropical cyclones

The term cyclone (Greek kuklos, reminiscent of the coils of a serpent), was first used in 1845 by Piddington in Calcutta (Leborgne, 1986), to describe the tropical storm, and later passed into general use for any depression, and especially the Norwegian cyclone. For the Far East the term typhoon is used (Chinese thaifong, Japanese tai fu, in India toofan and in Arabia tufari), and in the Philippines baguio. In Australia it is called willy-willy (a term also used for tornadoes and whirlwinds), whilst the West Indian phrase hu ra kan gives in Spanish huracan, in English hurricane and in French ouragam, the most powerful examples merit the title super cyclones. Awesome in their powers of destruction, cyclones have stimulated research into tropical matters and are thus well understood, flights having been made into their ‘eyes’ and images having been captured by satellites. Nevertheless, there is still some disagreement as to their origins, notably about the presumed role of thermal conditions over the ocean, and high-altitude divergence; what causes them to take the whimsical and unpredict­ able paths they follow is also a subject of debate.

11.1 CYCLONE STRUCTURE AND ASSOCIATED WEATHER

The spiral of cloud constituting a cyclone is several hundred kilometres across (from 500 to 1000 kilometres, or even double the latter for ‘supertyphoons’). Separated by clear lanes, arms of cloud converge towards a compact ring around the depression proper (Figure 11.1). The ‘eye’ forms an immense amphitheatre surrounded by walls of cumulonimbus towering to the tropopause; measuring some 10 to 50 km across, sometimes more, it is generally calm within. In the case of a well-formed example at a mature stage, the eye is an indicator of the intensity of the phenomenon: compact eyes are a feature of violent cyclones, and larger eyes are evidence of less violence; as its eye expands, the cyclone will often be in decline. The eye is a focus of low pressure, and the record stands at 870 hPa, a measurement taken in the north-west Pacific. At the heart of the disturbance, air is warm, as intense convergence into the

228

Tropical cyclones

[Ch. 11

hurricane wave

Figure 11.1

The tropical cyclone: plan and vertical section.

girdle of cumulonimbus releases vast quantities of energy, causing violent updrafts. The whole spiral travels along at a very variable speed of the order of 30 knots (50-60 km/h), with sharp accelerations, and the higher the latitude the greater the speed will be. Cyclones are associated with great damage wrought by rain, wind and rough seas. The most frequent criterion used to differentiate types of disturbances, progressively more intense, is the wind: in tropical lows winds usually travel at below 34 knots (60 km/h); tropical storms reach speeds of between 34 and 63 knots, and beyond 64 knots (>119 km/h) the cyclone can be said to have reached maturity, a stage fortunately not attained by all storms. The Saffir-Simpson scale lists five categories of cyclonic activity as a function of wind speed:

• • • • •

category category category category category

1: winds of between 120 and 155 km/h; 2: winds of between 155 and 180 km/h; 3: winds of between 180 and 210 km/h; 4: winds of between 210 and 250 km/h; and 5: winds stronger than 250 km/h.

Sec. 11.2]

Conditions for cyclogenesis

229

Mature cyclones may contain winds of more than 200 km/h, at category 5 gusting to as much as 300 km/h, with acceleration factors reinforced by the re-use of part of the rising air, which descends outside the disturbance. The force of the wind at its worst has been likened to ‘battering-ram’ gusts: pressure exerted at 240 km/h is of the order of 600kg/m2. Torrential rain falls beneath the walls of cumulonimbus, and cyclones hold records for rainfall measured; the single example of Cilaos (Reunion) will suffice, with 1870 mm in 24 hours. On land, floods are aggravated by rising sea levels {hurricane tides and hurricane waves).

•

•

Hurricane tides are formed by winds at the polar edge of the cyclone, where they are at their strongest (the geostrophic force being greater here). They propagate in the form of a heavy swell to distances of more than 1000-1500 kilometres in front of the cyclone, with a rise in sea level of about 1 metre. The hurricane wave is a more dangerous entity as it is more directly a part of the cyclone, moving with the low-pressure area as a dome of water. Its elevation due to pressure is 1 cm per hPa: a drop in pressure of 100 hPa will therefore cause a rise in sea level of 1 metre, but the effect is greatly enhanced by the inrushing wind pressing in upon the water towards the centre of the cyclone. The hurricane wave raises river levels at the same time as heavy rain and flooding occur, and its effects are amplified by the configuration of coastlines, as illustrated by a tragic example: that of the River Ganges, flowing into the Bay of Bengal. Here, on 12 November 1970, the joint effects of the hurricane tide and the astronomical high tide (1.7 metres) combining with the funnelling effect of the shape of the Bay caused the normal level of the rivers Ganges and Brahmaputra, already in flood, to rise by 7 metres. At least 300000 people perished in the area now known as Bangladesh.

11.2 CONDITIONS FOR CYCLOGENESIS The simplistic catastrophism propounded by the IPCC and the media, as applied to cyclones, leads many to believe that all that is needed for the number and intensity of cyclones to increase is a rise in sea temperature of a few tenths of a degree. This is fortunately not the case, and the conditions for cyclogenesis are in fact somewhat more complicated. Five simultaneous conditions are necessary to engender and main­ tain cyclones: the preliminary existence of a depressionary field, the unleashing of convection, a supply of energy, development in altitude and the formation of an eddy.

The existence of a depressionary field in the lower layers Low pressures encourage the initial deepening of the depression. This condition, which cannot apply in regions (by definition anticyclonic) covered by AAs which are by definition anticyclonic (Fig. 89), is met by intertropical lows and more especially by their axis, the ME, in its vertical and inclined structures (when the latter is active, i.e. not unproductive). The pre-existing low-pressure field may also be the low on the

230

Tropical cyclones

[Ch. 11

leading edge of an MPH, where cyclones also form, more rapidly because of the latitude at which they are born (i.e. with greater vorticity). Once in existence, the cyclone tends to seek out the areas of least pressure, both within the tropics and without: the low-pressure corridors of intertropical lows or those on the leading edges of MPHs.

The unleashing of convection The unleashing of upward air movement is dynamic: the ME (with the exception of the continental IME) is activated by the arrival of pulse lines (PL), leading to the formation of zonal cloud concentrations (cf. Chapter 10), which may release intense precipitation but rarely turn into cyclones. Powerful pulses are more likely to start the formation of tropical lows by causing penetration of the ME, to the north or to the south, and by releasing the beginnings of an eddy (a stage in the development of the low) which needs reinforcement to attain the status of a tropical storm. Regions prone to energetic pulse lines (PL), such as those on eastern sides of oceans influenced by anticyclonic agglutinations (AAs), are more likely to be sources of tropical lows that evolve into cyclones. An example is the North Atlantic off" West Africa, where the cyclones form which will eventually reach the Caribbean.

The supply of energy Once updrafts are established, the cyclone becomes self-perpetuating, the vast energy involved simultaneously encouraging updrafts and the maintenance of the depressionary character; this continues all the while that the energy is constantly and regularly renewed. The oft-quoted necessity for high sea temperatures (above 26°27°C) is an example of a covariation considered as a condition. The oceanic equator, where warmer water congregates, is in fact confused with the ME, since it is air currents which displace surface water. So the cyclone is supplied by fluxes which need a long path over the ocean to store up enormous quantities of perceptible and latent heat. Latent heat is supplied in only minimal amounts by evaporation in situ, as the fluxes come already near-saturated into the vicinity of the ME. Moreover, charts of evaporation clearly show that the highest values are recorded for tropical, nonequatorial waters. Therefore, cyclones form and persist if they are fed by tropicalised fluxes which are warm and, most importantly, possessed of abundant energy (i.e. very humid). These conditions are met, either separately or simultaneously, by monsoon and tropicalised maritime (unstable) trade fluxes. Regions covered by continental trades are therefore cyclone-free, as are those where (stable) nascent maritime trades blow. As a result of the thermal inertia of the ocean, the humidity of fluxes is greater at the end of summer and in autumn, when cyclones are well nourished and at their most damaging. Energy supply must be rapid and uninterrupted, both at the begin­ ning of the process where powerful vector fluxes (strongly sustained by pulses) are needed, and later to drive accelerating winds; the cyclone sucks into itself all the energy from a space around it of more than 1000 kilometres in all directions.

Sec. 11.2]

Conditions for cyclogenesis

231

Development in altitude The cyclone must be able to develop right through the troposphere, so there must be no subsidence or shear in the initial stages. Cyclogenesis is therefore not a feature of stratified structures such as the IME and trade inversions (TI), both unproductive. The most favourable conditions are those realised within two structures: the VME and the low-pressure corridor of an MPH. The best conditions are found within the VME, which presents no structural obstacles and offers a rich potential energy supply; here, 80% of cyclones are born. When the VME is transformed into the IME on approaching land, structural conditions become more and more unfavourable, and some other factor (e.g. rich energy supply or increased vorticity) must come into play to compensate for this, or degeneration of the disturbance will follow. A cyclone will find excellent conditions to stimulate its vertical development on the leading edge of an MPH, but conditions relating to the supply of energy are usually less favourable unless compensated for by the mechanical energy of the MPH and the considerable vorticity associated with the latitude. It should be noted that divergence at altitude (in an anticyclonic direction), sometimes proposed as a condition, is in fact a consequence of strong convergence in the lower and middle layers; considerable divergence in the upper layers (with the formation of an ‘anvil’ at the summit of the cumulonimbus clouds) is a sign of intense activity and powerful updrafts.

The formation of an eddy

The formation or strengthening of an eddy (vortex), a prerequisite for drawing in and accelerating air and for concentrating convergence, depends upon the geo­ strophic force. This force, a function of latitude, is almost zero near the Equator, and cyclogenesis is therefore not possible between latitudes 4°/5°N and 4°/5°S. So the VME, harbouring the most favourable structural and energy conditions for cyclogenesis, but near the Equator, must become sufficiently distant from it (through the action of pulse lines or seasonal translation) to act as an ideal starting point for cyclones. Extension of the VME into the IME is therefore a positive factor in increasing vorticity and in transforming zonal perturbations in the VME into cyclones. The Coriolis force is a function of the speed of the fluxes and of the mass transported; as a result, the most rapid and largest cyclones are also those which rotate fastest. At the leading edge of an MPH that has driven deep into the tropics, structural conditions are not optimal, but the latitude (= increased vorticity) promotes the rapid evolution from depression to cyclone. Vorticity is a function of latitude, and is therefore stronger on the polar edge of a cyclone than on its equatorial edge. This relative difference in strength slowly shifts the cyclone, once rotation is established, northwards or southwards away from the structure in which it originated, bringing it ever nearer to the temperate zone. The resulting increase in vorticity makes up for decreasing energy supply and increasing structural constraints.

232

Tropical cyclones

[Ch. 11

Favourable Factors: Vertical meteorological equator (VME) Active inclined meteorological equator warm, very humid monsoon warm, very humid trade MPH leading edge : low-pressure corridor

Figure 11.2

---------- ////// 11 i j ; i

\ \ '• \ \

Unfavourable Factors: AA anticyclonic agglutinations unproductive inclined meteorological equator unproductive trade inversion (stable m t) latitude within 5°N and S continental trade

Conditions for tropical cyclogenesis.

The prerequisite conditions for cyclogenesis are, fortunately, of an extreme nature, unlikely to occur often at one and the same time, or to be sustained severally as the cyclone pursues its path. The number of cyclones is therefore relatively small. These conditions mean that cyclones are essentially oceanic phenomena, given the localisation of the VME, and the large energy requirement, which hastens degeneration as the disturbance moves over land and the supply of very moist air becomes limited. The VME is the most likely birthplace of tropical cyclones. However, a vigorous MPH may conspire with abundant tropical energy to produce cyclones {subtropical cyclones} too (cf. Chapter 8), and this is most likely either at the beginning of the Photos 55 Path of a cyclone across the North Atlantic, 4-17 September, 1986 (after Meteosat Image Bulletin, ESA, visible, 12H UT). On 4 September cyclonic circulation appears over Senegal, near the well-marked VME. This eddy, so far not very intense, moves off over the ocean on a northward track. The depression draws in energy from 6 September onwards, putting the VME in disarray. By 8 September, the tropical depression is very much an individual entity, and on 9 September the eye appears. The VME re-forms far behind near Africa, and at the same time MPH ‘a’ is moving towards the cyclone. On 10 and 11 September the cyclone, having gathered strength within the low-pressure corridor, moves northwards around the leading edge of MPH ‘a’; but this MPH is proceeding rapidly eastwards, and creates a barrier on 12 September to the north of the cyclone, weakening it. On the 13th, MPH ‘a”s eastward motion leaves a path open behind it, and the cyclone picks up again. However, MPH ‘b’ comes upon the scene to reinforce ‘a’, and then MPH ‘c’ arrives. The cyclone is trapped between the rear edge of ‘b’ and the leading edge of ‘c’, taking advantage at first of the increased advection of potential energy funnelled towards it; but as MPH ‘a + c’ spreads out, the cyclone is doomed, merging on September 18 into the pulse lines of the vast anticyclone and closing with the cloud formations on the leading edge of MPH ‘c’. The track of the cyclone only seems capricious; tropical dynamics direct it at first, handing the task over to MPHs from 10-11 September.

Sec. 11.2]

Conditions for cyclogenesis

Evaluiinn du 4 au

233

234

Tropical cyclones

[Ch. 11

cyclone season or, more especially, at the end (in conjunction with the energy of an MPH), with optimum conditions in autumn when potential energy is at a maximum. These cyclones, which intensify very rapidly as a result of the latitude at which they are formed, with strong vorticity, may also occur outside the cyclone season.

11.3 THE TRAJECTORIES OF CYCLONES The trajectories of cyclones are determined by both tropical and extratropical factors (Photos 55). The transformation of the VME into the IME on approaching land, and/or the increase in vorticity with latitude, shift cyclones slowly away from the Equator. These two factors, either together or separately, conspire with the inherent tendency of the disturbance to move away from the Equator (because of the stronger vorticity on its polar edge) to shift cyclones towards the low-pressure corridors in front of MPHs reaching tropical margins. The cyclone, tending to follow the easiest downward pressure slope, is attracted towards these lows, and will be re-energised and integrated into the extratropical weather dynamic, with changed direction and character. This meeting of the dynamics of tropical and temperate areas may thus shield the land by causing cyclones to change direction; they move westwards at first (in concert with the tropical dynamic) towards land, and then away from it (temperate dynamic) if they are intercepted and pushed eastwards on the leading edge of an MPH. This is usually the case, for example, along the eastern coasts of North America and the Far East. Unfortunately, this same dynamical pattern at the leading edge of

D Tropical low S Tropical storm H Hurricane 5 stage 5 on Saffir-Simpson scale (peaked on 26-27 Oct.)

Figure 11.3 Trajectory of hurricane Mitch, 21 October-5 November 1998.

Sec. 11.3]

The trajectories of cyclones

235

Photos 56 Mobile Polar Highs and hurricane Mitch, 26-31 October 1998 (GOES-E, Dundee University). On October 26 and 27 Mitch was at its greatest intensity (category 5), but was blocked by MPH ‘Al’. The anticyclonic barrier was reinforced by ‘A2’, which very soon merged with ‘Al’, and then by the more powerful ‘A3’. Mitch degenerated rapidly when pushed southwards, crossing the Isthmus and becoming a tropical low, sheltered from anticyclone ‘Al + A2’ reinforced by A3, to the south of the mountain ridge.

236

Tropical cyclones

[Ch. 11

an MPH can also bring ashore a cyclone moving westwards across the ocean, as happens on the west coast of Mexico. It is said that the trajectories of cyclones are ‘capricious’ and unpredictable. In fact, they owe very little to mere chance, but are strictly directed, first of all by the tropical dynamic and then by the temperate, with a possible contribution from relief. A memorable example is that of Hurricane Mitch, infamous for the vast toll in human lives (an estimated 24000 fatalities) and enormous damage to property. During the day of 21 October 1998, a pulse line in the maritime trade approached the meteorological equator and caused the formation of a tropical low in the south­ western part of the Caribbean Sea, at around 12°N (Barbier, 1999). Its vast energy (associated with the maritime trade and the Panamanian monsoon), the favourable vertical structure of the active IME and the nearness of the VME, and vorticity, transformed this low into a tropical storm on 23 October, and it rapidly devel­ oped into a category 1 cyclone, becoming category 2 by the evening of the 24th (Figure 11.3). By 25 October, Mitch had become a category 3 hurricane, was soon category 4 and was at its most formidable on October 26 and 27 (Photos 56). On the 28th, it eased slightly to category 4, and then 3, and on the 29th through 2 to 1, while its path turned southwards, across Honduras and then El Salvador and Guatemala, where it was again a ‘mere’ tropical storm by the 29th and 30 h. The north-westward trajec­ tory turned southwards on the 27th, towards the Isthmus. According to Barbier (1999), “All the American forecasting models announced that the cyclone was going to move towards the north and north-east... in Honduras and Nicaragua, television stations broadcast messages reassuring the population and claiming a true ‘divine miracle’ ... but careful study of the satellite photos would soon have revealed that Mitch could only move over the land”. What happened was that, on 24 October (Figure 11.4A) a powerful MPH (Al), having travelled in previous days across the Atlantic and the Gulf of Mexico, at first intensified the cyclone, its leading edge causing a concentration of energy in the low-pressure corridor, but the Atlantic offshoot of Al blocked its passage northwards. MPH1 was reinforced by MPH2 (A2) with which it rapidly merged, and then by MPH3 (A3) which again increased pressure to the north of Mitch. Activity was progressively reduced while Mitch, blocked to the north and west, moved onto the land and became downgraded to a tropical storm. The depression then ‘took refuge’ against the southern slopes of the mountains of the Isthmus (with the northern flank of the ridge halting the anticyclonic thrust). The route westwards was also barred by an intense northerly circulation (a norther) across the lower part of southern Mexico (the tehuantepecer) onto the Pacific. Finally, as MPH3 (A3) moved off, the way eastwards was again open and Mitch, now a tropical storm, was able to proceed northwards. This move­ ment took place along the trailing edge of A3 and the low-pressure corridor on the leading edge of A4. On November 5 the storm was over southern Florida, and on the 6th, off the Carolinas, filling on the 7th (Figure 11.4B). So, in spite of its considerable store of energy, Mitch was, like all other disturbances, subject to strict dynamical conditions governing its motion. Its trajec­ tory was organised, first of all by the geostrophic force taking it north-westwards,

Sec. 11.3]

A

B

The trajectories of cyclones

------------ . surface |jne of ime

237

te tehuantepecer

■——— Leading edge of MPH4 (A4) ===== Surface line of IME PM Panamanian Monsoon ------ -—- 1015 hPa isobar (trailing edge of MPH3 (A3)

Figure 11.4 Dynamics of hurricane Mitch: A: 24-31 October 1998; B: 1-6 November 1998 (based on synoptic charts from the European Meteorological Bulletin, EMB).

238

Tropical cyclones

[Ch. 11

and then by the leading edge of MPH1 opposing this motion and turning Mitch southwards; then, the relief of the Isthmus took over, and finally the trailing edge of MPH3 and low-pressure corridor on the leading edge of MPH4. So the trajectory was predictable, all the while the reality of the situation was observed rather than the response of models based on statistics and outmoded concepts.

11.4 THE GEOGRAPHY OF TROPICAL CYCLONES Every year, on average, 85 to 90 tropical storms, ranging from intense tropical depressions to cyclones, form across the tropical zone, though with very unequal distribution (Leborgne, 1986; Hastenrath, 1991).

11.4.1 The northern hemisphere

Two-thirds of these occur in two clusters in the northern hemisphere: in the north­ west Pacific/Indian Ocean, and the North Atlantic/north-east Pacific. The north-west Pacific In the north-west Pacific one-third of all tropical storms occur, about 28 per year of which 18 are typhoons. This is the region of cyclones par excellence, given the number of events, the duration (from May to November) of the cyclone season, and the size attained by super-typhoons, of which there are three or four a year. Several factors combine to create favourable conditions: the Chinese monsoon and the Pacific trade, both saturated, come together at the VME within longitudes 140° to 170°E, and thence along the active IME towards China, with the trade riding above the monsoon. MPHs also enter the equation, their power setting off cyclogenesis outside the summer season: it is a peculiarity of this region that a cyclone may arise at any time of year. Most of the trajectories curve inwards to the north-east before they reach land, but typhoons strike China four or five times a year, Japan twice, the Philippines four or five times and Indo-China once or twice. These storms cross the peninsula of Indo-China in the direction of the Bay of Bengal only very infrequently.

The northern Indian Ocean

In the northern Indian Ocean there are about six cyclones every year. The Bay of Bengal sees about four or five annually, these being of minor intensity. The Indian monsoon is a causal factor, but the IME rapidly pushes up against the Himalayas, whilst the VME is moved into the middle layers, along the latitude of the south Deccan (Figure 4.4). The most likely periods for cyclone formation are therefore those when the ME moves north, in May-June, and back south, in OctoberNovember. These events, during which the structure is near to a VME, coincide with peaks in cyclone activity, the autumn one being the higher. The summer respite is

Sec. 11.4]

The geography of tropical cyclones

11 November

12 November

13 November

14 November

239

Photos 57 11-14 November 2007, Meteosat 07, visible, 06 UTC. This sequence shows two simultaneous disturbances. One, moving meridionally within the VME structure, forms on 12 November south of the Equator, but remains only slightly active (its category still that of a low). The second, further to the north, forms on 10 November in the IME structure, and passes into the Bay of Bengal as a cyclone (Cyclone Sidr), moving up across Myanmar on 15 November and 16, where it finally dissipated.

240

[Ch. 11

Tropical cyclones

1 June

2 June

3 June

4 June

5 June

6 June

Photos 59 1-6 June 2007. Meteosat 07. visible. 06 UTC. A depression forms on 31 May and 1 June in the IME structure. The vorticity becomes apparent on 2 and 3 June, and the cyclone develops an 'eye' on 4 June. Then, intensity diminishes (as the clouds widen out) and the cyclone breaks down on June 6 over the land (Oman).

Sec. 11.4]

The geography of tropical cyclones

241

very marked, especially in August. A small number of cyclones seen in the Indian Ocean originate in the north-west Pacific, having made their way with some difficulty across the Indo-China peninsula. In the Arabian Sea, along the western coasts of India, frequency is low, with only one or two cyclones in a year. Periods of possible activity match those for the Bay of Bengal, from which cyclones may come, after crossing the southern Deccan, but the unproductive nature of the IME (with the continental flux above it) soon stifles activity as the cyclone proceeds westwards. Only rarely do such weak affairs reach Oman (Photos 58 (colour section) and 59).

The North Atlantic, the Caribbean and the Gulf of Mexico Here, there are on average nine cyclones a year, five of which reach hurricane strength. The season runs from May to November, with peak activity in September/ October. The two most influential factors generating lows over the central Atlantic are the presence of the VME, into which flows the evolved maritime trade, and the contribution of the Atlantic monsoon with its principal flux moving towards Africa. The vast majority of squall lines moving onto the ocean from the coast of Africa disappear into the unproductive trade inversion zone, but any managing to overcome this obstacle are involved in the creation of half of the cyclones. The latter then leave the structure of the VME, whilst MPHs descending vigorously across North America bend their tracks north-eastwards. Cyclones may reach the coast of North America from Texas to Florida at the rate of two or three a year, moving up the eastern seaboard, or they may hit Mexico at the western end of the Gulf, or even make the difficult crossing of the Isthmus to enter the Pacific (cf. Photos 55 and 56).

The north-east Pacific

This area extends the Atlantic zone westwards, constituting the second most active region of cyclones with a total of 15 a year. This is a triangular space, home of the Panamanian monsoon (Figure 4.4); the triangle is based on the Isthmus, from which come weakened Atlantic disturbances, and on two trade fluxes (to north and south) topped by a strong inversion. The cyclone season lasts from May to November, with both the Panamanian monsoon (with the Atlantic trade above it) and the presence of the ME (the active IME and the VME being close) being positive factors. Temporals, which are intense tropical lows associated with the VME, “move slowly and bring torrential rain” (WMO, 1992). These cyclonic systems have low rotational velocity because of their latitude, and only a third of them ever reach the cyclone stage (locally, cordonazo)', they move westwards across the ocean and normally decline fairly quickly. Here however, MPHs, which elsewhere will protect continents, come down and divert the short-lived cordonazos along their leading edges towards the western coasts of Mexico at the rate of two or three a year. These coasts are more open than those in the east, all the way up to southern California, where the Mojave and Sonora Deserts may experience exceptional rainfall (Photos 60).

242

[Ch. 11

Tropical cyclones

31 August

1 September

2 September

3 September

Photos 60 Dynamic of a cordonazo off the coast of Mexico, 30 August to 5 September 2004. GOES 10, visible, 18 h UTC, NOAA. Continuous line: leading edge of an MPH; dashed line: meteorological equator; circled date; tropical low/cordonazo', flux line; Mexican monsoon. A depression appears on 30 August in the VME structure off the Acapulco area, and begins a northward migration. At the same time, an MPH arrives from the north at the Californian shores. Their encounter occurs on 3 September and the cordonazo integrates itself into the peripheral low-pressure corridor of the MPH, and, like all lows in the same circumstances, migrates northwards, towards California, which it reaches on 4 September. On 5 September, the low fills and follows the MPH on its journey west.

The geography of tropical cyclones

Sec. 11.4]

Synthesis: 30 August-5 September

4 September

Photos 60

243

(Continued).

11.4.2 The southern hemisphere Notwithstanding its oceanic character, the southern hemisphere generates only onethird of these disturbances, along a continuous band stretching from eastern Africa to the western Pacific. The eastern Pacific and the South Atlantic are covered by an AA and an unproductive trade inversion and, what is more, are not visited by the ME: thus, even when the maritime trade becomes unstable, as occurs on the coasts of Brazil, cyclogenesis is fortunately absent.

The south-west Pacific

In the south-west Pacific the cyclone season lasts from November to April, with 12 storms a year, of which only three qualify as cyclones. Causal conditions are associated with the meeting of the Australian monsoon and the maritime trade along the VME, which becomes the active IME in the west before the continent is reached. Most tracks are diverted eastwards, but if the intervention of MPHs is not timely, then northern Australia may suffer, Queensland being most at risk with an average of two cyclones every year. New Caledonia is affected at around 21 S, every three or four years, and the Society Islands (Tahiti), 18°S and 150°W, are normally spared, all the while the VME remains in the northern hemisphere or near the equator (cf. Chapter 14: El Nino).

244

Tropical cyclones

[Ch. 11

The south-eastern Indian Ocean, or north-western Australia

These areas are an extension of the previous domain and receive its storms, passing them on to the Pacific by way of the Gulf of Carpentaria and Queensland. Thus the cyclone season lasts from November to April, with about seven events a year, of which two or three may reach the status of willy-willy. Again the Australian monsoon and the VME, with favourable conditions, are positively involved. The

14 February

15 February

16 February

17 February

Photos 61 14-17 February 2008. Meteosat 07, visible, 06 h UTC. A depression forms on 13 and 14 February in the active IME structure east of Madagascar. It rapidly evolves into a cyclone, named Ivan, and on the 16th a well formed ‘eye' is seen as it nears the east coast of Madagascar, where it causes considerable damage (especially on the lie Sainte-Marie). From 18 February, cyclonic activity declines and, as it crosses Madagascar, the cloud mass (in the active IME), slipping southwards across the island towards an MPH, meets it on 21 February south of Madagascar. Ivan fills on 24 and 25 February in the Mozambique Channel.

Sec. 11.5]

Conclusion

245

transformation to the south-east of the ME into the IME topped by a continental flux, and the usual westward movement of cyclones, ought to shield the continent from their attentions. But MPHs intervene to push them in the opposite direction and across western Australia on a NW-SE track. Once over the continent, they always degenerate rapidly, sometimes in the space of just a few hundred kilometres, but coastal areas are particularly likely to suffer their worst ravages. The south-western Indian Ocean

Here, the cyclone season runs from October to April, with 11 storms a year, half of which become cyclones, especially from January to March. Cyclones form in the VME structure where the Madagascar monsoon and the maritime trade meet; some form north of Australia, having propagated along the VME. Cyclones follow the IME southwards towards Madagascar, and are diverted south-eastwards by MPHs (Photos 61), or run across the island into the Mozambique Channel. The Channel is a meeting-place for certain conditions (Leroux, 1983): the IME is active, as a result of the Madagascar monsoon and the maritime trade, both saturated; MPHs constantly affect the south of the Channel, deepening the Mozambique low; and the latitude (20°S) favours a strong vortex. These favourable conditions combine to generate modest-sized tropical storms or cyclones which rapidly take off" eastwards, with the MPH dynamic; they peter out over Madagascar or, if they manage to cross it, regain strength over the ocean, descending south-eastwards (with the leading edge of the MPH); they may leave the area of influence of the MPH and turn back across Madagascar towards the Channel, following the IME. Every 15 years or so the island of Reunion is visited by a devastating cyclone, with damaging cyclones arriving every five years, and sometimes two in the same year.

11.5 CONCLUSION Cyclones, tropical phenomena par excellence, are by their nature destructive. Any optimistic hopes (cf. Project Stormfury) about combating their intensity, their heavy rains and the strength of their winds have been rapidly dashed. The only really effective way to protect oneself from the worst ravages of these hugely powerful phenomena is to flee, if possible, as happens in the southern United States where mass evacuations are organised at their approach. The tropical cyclone represents the paroxysmal stage in the activity of the meteorological equator, in its most specifically tropical VME structure. However, when a tropical cyclone is overhauled by the leading edge of an MPH, or even born of an MPH {subtropical cyclone), is its dynamic of a tropical or an extratropical nature? As it merges into the extratropical circulation, the tropical cyclone is evidence of the continuous major process of meridional exchanges: swept from the poles by MPHs, and back towards them to the fore of MPHs.

PART III DYNAMICS OF CLIMATE: CLIMATIC EVOLUTION

THE ‘GLOBAL CLIMATIC SYSTEM’

Parts I and II have dealt with the basics of climate and weather dynamics, in terms of the present day, and on both the synoptic and seasonal scales. They have offered a framework - general circulation and its workings - within which isolated climatic ‘events’, both in the past and more recent times, must be allotted their correct places, respecting the order in which phenomena occur and are linked. It is absolutely necessary (and it is the least we can do) to know whether, for example, the Sahel drought is a result of local conditions or more remote ones, or if El Nino lies at the start (as a cause) or at the end (as a consequence) of a physical process. Part III aims to examine how aerological structures and weather and climate mechanisms may be modified, intensified or attenuated, and/or displaced. The reasons for longer-term dynamical change in climate will be dealt with, beyond the synoptic scale, in recent times and on the palaeoclimatic scale. This necessarily and immediately assumes that these same processes, with changing intensities, occur through all timescales, from the instantaneous to that of palaeoclimates. The dynamics of general circulation obey the same laws, in comparable circumstances, and do not adjust themselves in one or another direction, or even obey contrary mechanisms, according to questions put and/or regions involved, in a multiplicity of opportunist ways. The ‘global climatic system’ (Figure III. 1) is composed of:

•

the atmosphere, the most unstable and changeable part of the system. The modification of its constituents is considered (by the IPCC) as the essential phenomenon involved in the greenhouse effect, thanks to the properties of emissive gases, mainly water vapour, to which are added solid and liquid

Dynamics of climate: climatic evolution

248

•

solar radiation

S I

p

(0 C0

4

I

4

I

I

I

OC O

S' .

Wr - tropopause -

2 0)

| '

volcanism

o S ** evaporation biosphere

noosphere

lithosphere

Figure III.l

•

• •

•

continental hydrosphere

The global climatic system (general circulation omitted).

particles in suspension (aerosols), and clouds. However, the most changeable aspect the mechanism of meridional exchanges is not, in the absence of a representative schema, correctly integrated into the system; the hydrosphere, in the form of all liquid and underground water, the fresh water of rivers, lakes and aquifers, the salt water of seas and oceans (the reservoirs and sources of carbon dioxide), not forgetting the water (in vapour or liquid form) in suspension in the air; the lithosphere: land masses and their distribution, and relief (altitude and disposition), soils, and volcanic or surface dust in the form of aerosols; the cryosphere', sea ice (icefields), ice on land (the inlandsis of Greenland and Antarctica, glaciers on mountains, permafrost), snowfields and ice crystals in high-altitude clouds; the biosphere, on land and in the seas, and especially vegetation, with special reference to large areas of forest, and plankton fields.

To these components we may add the noosphere (Greek noos, intelligence; Demangeot, 1994), representing the actions of the human race (though this does not always correspond to a definition of those actions). All these elements interact in complex ways, and on different scales of time and space. The atmosphere is the most changeable element, but exchanges with other elements occur over various amounts of time. In the case of oceans, for example, exchanges of heat and water, and also of gases like carbon dioxide, occur both instantaneously around the surface and over millennia in abyssal regions. Factors favouring modifications of the global climatic system arise from two types of process: internal processes, interactions affecting climate and dependent on it; and external ones, arising from factors which affect climate but are independent of it, and which thereby exert a 'forcing' effect on climate.

The absence of a global schema of the climatic system

249

THE ABSENCE OF A GLOBAL SCHEMA OF THE CLIMATIC SYSTEM The concept of a global climatic system does not automatically assume the existence of a global climate, a practical but reductionist notion which might be inferred from Figure III. 1. A ‘global’ climate does not exist; it is of course only a mental construct. The definition of ‘climate’ - or more precisely, ‘climates’ - is geographically based, in well defined areas. Also, given its static character, this schema (Figure III.l) cannot take into account the integral nature of the climatic system. What it does represent is, in a manner of speaking, the ‘Earth system’, but it does not encompass the climatic system as a whole, because that system is essentially one of movement. Alongside stable or slowly-modified elements (easily represented in such a schema) there exist changing elements, evolving more or less rapidly: in the oceans, movement at the surface is impelled by winds (drift currents) or by density contrasts caused by thermal gradients (density currents), or salinity (thermohaline circulation). The most changeable element is that of the atmosphere, which is much more reactive and fast­ moving, ensuring meridional transfers of air and energy (cf. Figure 5.4). These extremely mobile elements (especially air) cannot be represented in the schema of Figure III.l. In spite of their essential role in determining the climate, they are indeed often forgotten, particularly in numerical models, which ‘see’ the relationships between the different spheres on the scale of the elementary cell. The ‘climatic’ system therefore suffers from one fundamental omission: that of the idea of general circulation, which determines and distributes the consequences of interactions. We cannot and must not proceed directly from some internal or external factor or process to some global or climatic consequence, but rather through links controlling weather dynamics. These links transform, amplify or attenuate climatic effects and, from the same initial cause, may even lead to different or even completely opposite outcomes in different regions. Statistical relationships may be established between the variation of a certain cause and the variation of some parameter (which is what happens on the scale of the modellers’ elementary cell), but this will only have any real meaning when the real physical mechanism has been identified. A statistical relationship (teleconnectiori) is evidence merely of a co-variation, with no attempt to determine the link between cause and effect. Relationships are seldom binary (i.e. involving only two parameters isolated from their context). For example, it is not necessarily the case that the temperature must drop dramatically in high latitudes (as a result of orbital parameters) for inlandsis to form: the accumulation of thousands of metres’ thickness of ice is not dependent on local conditions, but stems from enormous transfers of water, involving enormously potent factors, over thousands of years. Similarly, the building of great systems of dunes, such as the longitudinal Ogolian dunes of the Sahara, is not merely a function of local shortage or absence of rain, but also depends on an intensification of meridional exchanges (a consequence of which is the acceleration of winds in the lower layers) whose beginnings lie far away, and on a modification of general circulation. On a different scale, it is not necessary for evaporation to increase with a rise in temperature for rainfall to increase in its turn, as it might be calculated by a basic climate-model cell: there is no direct relationship

250

Dynamics of climate: climatic evolution

between these two parameters, and conditions for pluviogenesis are far from being so simplistic. Need we say that the fact that a warm sea current appears north of Peru is not a sufficient explanation for a severe freeze in Quebec! What a bizarre inversion of meteorological processes! One element, plucked for convenience from the ensemble, cannot change on its own, and we must therefore keep in mind that phenomena to be analysed have their place in general circulation and in the hierarchy of phenomena, and are themselves a part of the global climatic system.

Causes of climatic variations

All the components of the climatic system are constantly changing, and it is impossible to apply present mechanisms to all timescales. If we based our reasoning on the present-day functioning of general circulation, then the nature of the litho­ sphere, and especially the distribution of land masses and relief, so vitally important in the marshalling of ocean current and air circulation, would have to be unchanging. This would rule out the geological aspect, which requires knowledge of the exact geography (and most importantly topography) of the planet. Only recent climatic variations (until the Upper Pleistocene), to which the logic of present-day phenomena may be readily applied, will be included. Both external and internal causes must be considered for the origin of these variations in the intensity, if not in the character, of the dynamics of weather in geographical conditions comparable to today’s. Varia­ tions in solar activity and in the Earth’s orbital characteristics are likely to modify the intensity of the radiation falling upon the planet. Factors affecting the quality of solar radiation and the amount of terrestrial counter-radiation can be derived from both natural processes due to volcanism, and from processes due to human activity, namely the increase in quantities of greenhouse gases, a presumed cause now con­ sidered by the IPCC as the ‘only’ factor responsible for ‘climatic warming’.

12.1

ORBITAL PARAMETERS OF RADIATION

Currently, seasonal and latitudinal variation in radiation is a function of the positions of the Earth with respect to the Sun (Figure 1.3). The polar axis, presently aligned with the star Alpha Ursae Minoris (the Pole Star), forms an angle of almost 66°33' to the plane of the ecliptic. The Earth-Sun distance changes by about 5 million kilometres between Earth’s aphelion (when it is furthest from the Sun, at the summer solstice in early July) and its perihelion (when it is nearest the Sun, at the winter solstice in early January, cf. Chapter 1). But the orbital parameters of radiation are constantly changing, with variations taking place in the Earth-Sun

252

Causes of climatic variations

[Ch. 12

distance and in the inclination to the ecliptic and orientation in space of the polar axis. These three parameters follow overlapping cycles of different lengths, and slowly modify conditions affecting the arrival on Earth of the Sun’s radiation. The astronomical theory of palaeoclimates, involving variations in the three aforemen­ tioned astronomical parameters, and the outcome of lengthy reflection about the genesis of glaciations discovered by Agassiz (1837), saw decisive progress with studies by Adhemar (1842) and Croll (1875) at the end of the nineteenth century, and thereafter especially through the work of Milankovitch (1924). Glacial periods had originally been associated with a decrease in energy received at high latitudes in winter, with corresponding accumulation of snow (necessitating vast transfers of precipitable water, a factor not then emphasised). Milankovitch, however, thought that conditions promoting the formation of ice corresponded to a minimum of summertime solar radiation in the northern hemisphere, insufficient to melt the snow from the previous winter (a concept embracing only a part of the problem). So Milankovitch deserves recognition not for an explanation of the mechanism of glaciations (which cannot be explained without recourse to the idea of general circulation), but for having revealed the causes of the long-term variations in insolation experienced by the Earth. The astronomical theory (sometimes known simply as the "Milankovitch parameter') has been alternately recognised and dis­ credited, as geological debates have continued. Then, in the 1970s, long chrono­ logical series based on the study of marine sediments and ice cores gave more credence to the theory: more recently, the Vostok cores from Antarctica (Petit et al., 1999) have revealed four cycles, lasting more than 400 000 years, of eccentricity in the Earth’s orbit, and lesser variations within those principal cycles. Periodicities of about 100000, 41 000 and 22000 years have therefore emerged. So, the astronomical theory is established as the fundamental explanation of long-term climatic modifications.

12.1.1 Variation of the eccentricity of the Earth’s orbit The Earth’s orbit is not circular, but elliptical, its revolution around the Sun being perturbed by the gravitational attraction of the other orbiting planets. The Earth is therefore not at a constant distance from the Sun, its orbital path varying from almost a circle to a more or less flattened ellipse. The term eccentricity describes the degree of deviation from a circular path, and is expressed as a percentage relating the minor axis to the major axis of the ellipse: the smaller the percentage, the more circular the shape becomes. Variation in the eccentricity, which has been relatively small since it has not yet exceeded 7% (Berger, 1992), shows a near-periodicity of about 100000 years (or 100 kyr, kiloyears), with variation between 90000 and 120000 years (90 kyr and 120 kyr). The last maximum of eccentricity, which was small (of the order of 2%), happened about 16 kyr BP (Before Present: according to convention, before 1950), and the previous one dates back to about 125 kyr BP, with an eccentricity of about 4% (Figure 12.1). The relatively slight eccentricity has little effect on the total energy received through the year, the difference amounting to only 0.2%. However, seasonal

Sec. 12.1]

Orbital parameters of radiation

253

Figure 12.1 Variation in orbital parameters of radiation from 150 000 BP (150kyr) to + 20000 (20kyr) (from Berger, 1992).

contrasts are important, and if we add to these those based on other parameters, they are involved in a major cycle of climatic variations of about lOOkyr, which constitutes the “principal cause of the quasi-periodic succession of glaciations on Earth” (Duplessy and Morel, 1990), as curves based upon studies of Antarctic ice cores confirm (Vostok >400kyr, 1999, and Epica > 800 kyr, 2007).

12.1.2 Variation of the angle of inclination of the Earth’s polar axis

The angle between the polar axis and the perpendicular to the plane of the ecliptic is changing very slowly away from 2°27'. This angle varies through 3°, between 22° and 25°, over a quasi-period of 41 kyr. The last maximum of obliquity (involving an angle of 24°30'), occurred about 9 kyr BP (Figure 7.20), and the polar axis has been slowly moving back ever since, at the present time towards a value of 23°26'.

254

Causes of climatic variations

[Ch. 12

A decrease in the angle, with the axis at its most upright position, reduces seasonal contrasts, with milder winters and cooler summers. Conversely, when the angle is larger and the axis more tilted, contrasts are more marked, and the winter hemisphere is relatively further from the Sun and is therefore colder, the sum­ mer hemisphere being warmer as it is ‘nearer’ the Sun. Variations are identical in the two hemispheres, but consequences differ according to latitude, especially in the tropics and polar regions.

•

•

In the tropical zone, variations through axial inclination are comparatively weaker, but the latitude of the two tropics moves from 22°N and S to 25°N and S, bringing about an extension or reduction of the astronomically defined tropical zone by 6 of latitude, from a width of 44° to 50°. This difference means an extension in latitude of roughly 667 extra kilometres above which the Sun may pass at the zenith. This may seem a small distance, but it is by no means inconsiderable if we consider the total surface in the tropics capable of receiving optimal (zenithal) solar radiation, and if we also consider the dynamic of the meteorological equator and the resulting reduction or amplification in monsoon circulation. About 9 kyr BP (Figure 12.1), the tropical zone stretched across 49° of latitude (24°30'x 2), but at our epoch this is reduced to 46°54', and will continue to shrink. The effects of obliquity are amplified in polar regions, where the latitude of the Polar Circles (today at 66°33', and moving towards 66°34') oscillates between 65° and 68°, reducing or extending in equal measure the area where 24-hour darkness in winter may occur; the (astronomical) tropical zone also expanding and contracting by as many degrees. During summer, a maximal tilt will increase polar warming, and the difference between the two extreme positions of obliquity means a difference of 14% in energy received in high latitudes. Conversely, when the axis is at its most upright and the tilt minimal, the energy received by the poles is at its least, a situation pertaining at about 25 kyr BP, when the inclination was close to 22°10' (Figure 12.1).

The cumulative effect of the total quantity of energy received simultaneously by the tropical and polar zones is particularly striking:

•

•

When the axis is more upright (i.e. not much inclined) the tropical zone is smaller and the zone of the ‘polar night’ correspondingly larger. The total energy deficit is considerable, especially near the poles where the winter deficit is greater and summer insolation correspondingly reduced. This was the situation at about 28 kyr BP. When the axis is more tilted, the tropical zone is larger and the zone of the ‘polar night’ correspondingly smaller. The total energy surplus is consider­ able, and the polar thermal deficit is less in the winter, with summer insolation correspondingly increased. This was the situation in the period centred on 9 kyr BP.

Sec. 12.1]

Orbital parameters of radiation

255

Remember that the axis is currently moving to a more ‘upright’ position, i.e. there is a very gradual reduction in the tropical zone and a no less gradual extension of the ‘polar night’ zone. But the timescale is very long (estimated at about 10 000 years, according to Figure 12.1).

12.1.3 Variation of the orientation of the polar axis

The Earth is not a sphere, but an oblate spheroid, with a slight bulge at the Equator due to rotation. The gravitational attraction of the other bodies in the solar system causes a slow gyration of the Earth’s axis, an oscillation like that of a spinning top. The polar axis does not therefore always point towards the same place among the stars, and the Pole Star (on the line projected from the polar axis northwards) is at present Alpha Ursae Minoris; 4000 years ago the Pole Star was Alpha Draconis, and in 12 000 years’ time it will be Vega, in the constellation of Lyra. This oscillation gradually displaces the positions of the solstices and equinoxes on the ellipse travelled by the Earth, and changes that moment in the year when it reaches its

-22000 BP

aphelion

solstice

equinox

Figure 12.2 Precession of the equinoxes or solstices. Precession may be roughly summed up thus: for the position of the solstice to be reversed from aphelion to perihelion (i.e. by 6 months), it takes 11 000 years (for the reversal of the direction of the polar axis); 6 months = 180 days = 4320 hours = 259 200 minutes; so in one year precession is 259200 minutes/11 000 = 24 minutes.

256

Causes of climatic variations

[Ch. 12

furthest point (perihelion) and its nearest point (aphelion) with respect to the Sun. The combination of axial precession, with a period of 26 kyr, and other orbital parameters, make the mean value of the periodicity of the precession of the equinoxes equal to 22 kyr; there are two dominant periods, one of 23 kyr, and another of 19 kyr. The outcome of this periodicity is a swing, in opposite directions, of the polar axis: thus, 11 kyr BP (Figures 12.1 and 12.2), at the June solstice, the Earth was at perihelion (unlike today when it is at aphelion), whilst the December solstice coincided with aphelion (perihelion today). In the northern hemisphere therefore, summers were warmer but winters colder; in the southern hemisphere summers were cooler, as a result of the increased solar distance, but winters were less cold, the distance being shorter. Seasonal contrasts were therefore less pronounced in the southern hemisphere, but very marked in the northern hemisphere.

12.1.4 Orbital parameters and climatic evolution The role of the astronomical factor in long-term climatic variations is undoubted, particularly in its determination of glacial-interglacial cycles. The 100-kyr cycle is obvious, and is apparent in all long-term curves, especially those based on ice core samples from the inlandsis. Since it is seen to be the same all over the planet, this cycle explains the synchronous character of major changes in the two hemispheres. Thus, the period centred around 125 kyr BP (Figure 12.1), when eccentricity was 4%, obliquity nearly 24° and summer insolation in high latitudes more than 13% greater than at present, constituted the interglacial Eemian period, which was some­ what warmer than the present. The most recent equivalent to the Eemian was about 6 kyr BP, and is known as the Holocene Climatic Optimum (HCO). About 115 kyr BP, by contrast, with eccentricity still pronounced, but with low obliquity (22°24z), and summer at aphelion, insolation in northern high latitudes was 9% less than at present. The cooling marked the beginning of the last (Wurm) glacial period, which through successive stages led to the Last Glacial Maximum (LGM) about 15 kyr BP. As a function of orbital parameters, “the cooling which began 6000 years ago will continue for another 5000 years” (Berger, 1992), then evolving into glacial conditions. These orbital parameters, fundamental to our comprehension of palaeoclimates, are too slow to be of import in the case of current evolution. However, they enable us to see recent millennia, and our own epoch, in the context of a slow evolution with a general trend of reducing insolation. Different interactions with the two other parameters introduce shorter variations within the major cycles. Consequences vary according to latitude, being slight near the Equator and increasing in magnitude towards the poles. Obliquity and preces­ sion have a great influence on insolation in polar latitudes (insolation varying little at the tropical latitudes), which are most affected by the 41-kyr cycle; through these latitudes, and the intermediary of MPHs, they also influence circulation in general (cf. Chapter 13), and circulation is accelerated (rapid mode) or decelerated (slow mode) as a function of thermal deficit at high latitudes (cf. Chapter 13).

Sec. 12.2]

Variations in solar activity

257

12.2 VARIATIONS IN SOLAR ACTIVITY The circulation of the atmosphere and the weather resulting from it are consequences of solar radiation (Chapter 1). It seems logical, therefore, to state that variations in the activity of the Sun are the primary cause of climatic variations. This has been a subject of debate for a very long time, and there is an abundant scientific literature (cf., for example, Waple, 1999). These variations crop up from time to time in the media to ‘explain’ meteorological anomalies. The Sun “goes crazy” and its “wrath” is responsible for, individually or severally, heatwaves, droughts and even floods. There is endless fascination with the subject, even though “in spite of the massive literature, there is little or no convincing evidence of statistically significant cor­ relations” (Pittock, 1983) between solar activity and weather or climate (with just a few exceptions). Interest has recently been rekindled following new observational and analytical capabilities, new methods of investigating past solar activity, and new theories moving beyond the Sun itself to take account of variations in the flux of galactic cosmic rays (GCR).

12.2.1 The sunspot cycle

The amount of energy supplied by the Sun has long been thought invariable: hence the notion of the solar constant, used to designate the quantity of energies arriving at the top of our atmosphere. However, this notion of a constant must be revised, as it varies - if only slightly - with solar activity. Irregularities in the Sun’s activity manifest themselves through the appearance of darker patches on the photosphere (the visible luminous surface of the Sun), linked to the Sun’s magnetic field. Very active zones in the solar magnetic field lead to the formation of these ‘spots’ on the surface of the Sun, and they appear at first at solar mid-latitudes (at around 40°). Then, the sites of fresh sunspots migrate towards the Sun’s equator, each large spot lasting for about a month. The spots appear dark because they are at a lower temperature (about 4500°C) than the rest of the photosphere (at about 6000°C). Observations of sunspot activity since the beginning of the 17th century have revealed that the number of spots varies between sunspot minimum and sunspot maximum, following an average cycle of 11 years (the ‘Schwabe cycle’), with variations around this mean value of between 9 and 13 years (Figure 2.8). Also, cycles occur in pairs, making the periodicity 22 years, involving the reversal of the Sun’s magnetic field. The minimum of cycle 22 occurred in 1986 and the maximum in 1991, the minimum of cycle 22-23 was centred on 1997, and cycle 23 saw a maximum in 2001 and lasted until 2007-2008, when cycle 24 began (Figure 12.3). These 11 -year cycles are of unequal amplitude (following the so-called ‘Glessberg cycle’ of the order of 80-100 years). The values of the minima remain very similar to each other, but the maxima exhibit considerable variations. The latter were very high around 1760-80, 1840-70 and 1940-60, and comparatively low around 1740-60, 1790-1820 and 1880-1910. There is a relationship between the number of spots and solar activity: maximum solar radiation (the active Sun) occurs at times

258

Causes of climatic variations

Figure 12.3

[Ch. 12

Evolution of solar cycles: from cycle 12 to cycle 23 (from Benestad, 2002).

of spot maxima, and spot-free zones are then brighter and emit more radiation. In the course of one cycle, variation in the intensity of radiation is never very great. Since 1978, variations in the intensity of solar radiation have been measured directly by satellites (Nimbus 7, SMM, ERBS), and results have shown that the amplitude of variation in an 11-year cycle is only 0.1% (Foukal, 1994, 2003, Figure 12.4). The amplitude of variations within a cycle attains 0.3% over short periods during phases of increased activity (around maximum), and is very small at times of least activity (around minimum). Cycles involve all solar activity, with maxima corresponding to periods of intense solar wind and greatly increased ultraviolet radiation. It is possible to infer past solar magnetic activity (and associated illumination) by analysing isotopes formed by the interaction of radiation and atmospheric molecules. Production of carbon-14 (14C) from nitrogen-14 (14N) in the upper atmo­ sphere varies with the intensity of cosmic radiation. Small variations in amounts of

i

i

1368 1367 1366

1365 1364 1363

------ 100-day smooth dally doto

: J—L-I

1980

1

I—L—I

I

1985

I

I

I

I«

I

1990

I

I

I

«


4) through contact with atmo­ spheric water. The size of these sulphuric acid droplets (less than 1 pm) allows them to remain in suspension in the stratosphere for several years, and there is a

Volcanism and climate

265

TROPOSPHERE

STRATOSPHERE

Sec. 12.3]

Figure 12.6

Effects of volcanic eruptions on the atmosphere (diagrammatic).

permanent aerosol layer at 20-25 km altitude. Very thin clouds of ultracirrus sometimes form in the mesosphere at an altitude of about 80 km. We now know that these sulphate aerosols, by reason of their high reflectivity as droplets, have a greater thermal impact than do ash and dust (Devine et al., 1984). Satellite observations of the eruption of El Chichon in 1982 supported this view. The way in which aerosols are dispersed depends upon the location at which they are ejected.

•

In the tropical zone, where the tropopause is higher (Figure 5.4), and where ejecta can reach a height of more than 20 km, dust is distributed both north­ wards and southwards and the spread can be planet-wide. Dispersed by highlevel winds, material takes from two to six weeks to spread around the planet in lower and mid-latitudes, and thence towards the poles. Material from the eruption on 14-15 June 1991 of Mount Pinatubo in the Philippines, at 15°N, reached a height of up to 35 km. It took 21 days to travel round the globe, according to satellite observations (Figure 12.7). Two months after the event, there were aerosols above 42% of the Earth’s surface between 30°N and 20°S (Bluth et al., 1992). These aerosols later gradually drifted northwards. The aerosols from Pinatubo were only slightly diverted towards the southern

266

Causes of climatic variations

[Ch. 12

N

0.5

■O 3

0.0

(Q

-0.5

Mt Hudson S

i

i

r


— = = = = =

[Ch. 13

limits of westerly (W) and easterly (E) winds== Meteorological equator (VME and IME) trade inversion trade-wind pulse — = = Interoceanic Confluence (IOC) water potential (water vapour content) --------------- ► flux line (trade - monsoon)

Figure 13.1 The meteorological equator and pluviogenesis in Africa.

Over eastern Africa (Figure 13.1c), the IME is not unproductive in southern summer when it lies over Mozambique and its Channel, and northern Madagascar. The Indian Ocean trade and the Madagascar monsoon, both very moisture-laden, meet here within a depth of the order of 2500 metres; the resulting intense activity means heavy rainfall and possible cyclogenesis. The lack of rain along the western edge of Africa is linked with the stability of lower-level AAs and with the unproductive structure of the maritime trades. The cool, shallow lower stratum is overridden by a dry flux which discourages any cloud formation above the trade inversion (Tl), and lower-layer humidity may condense out as dew. Briefly viewed, such are the main factors of Africa’s climate: factors which may all evolve as current seasonal and climatic variations indicate. Very few have their origins in conditions which are strictly local (i.e. limited to Africa), but are rather part of the dynamic of meridional exchanges. Figure 13.2 shows the main presentday climatic domains of Africa (Leroux, 1983), as reference for an analysis of palaeo­ climatic evolution in Africa over the last 18000 years. Africa’s palaeometeorology is approached through the palaeoenvironments of the two extreme situations pertaining during the Last Glacial Maximum (LGM) and the Holocene Climatic Optimum (HCO). The curve of Figure 13.3 follows the main climatic changes in tropical Africa during the last 18 000 years, at the same time showing extratropical evolution, including only general evolution. Below the curve are mentioned tropical African events; and above it extratropical events, with names

Sec. 13.1]

Palaeoenvironments in Africa

297

--------- 400 mean annual isohyet in mm ----------------southern limit of Mediterranean domain southern limit of Sahelian domain (between 100 and 500 mm) - Sudan : between 500 and 1500 mm -------- eastern limit of domains of stable maritime trade (with unproductive inversion) -------- limit of permanent Atlantic monsoon domain (Guinea-Congo forest) --------- limit of bimodal pluviometric regimes (associated with VME and, in the east, with high ground in Ethiopia)

Figure 13.2

The main climatic domains of Africa.

of periods of cooling, derived from fluctuations in the limits of Alpine snow and trees, borrowed from Furrer et al. (1987). Temperature change is not calibrated owing to the small amount of variation in the tropics and to the considerable increase in differences nearer the poles. Variations are noticeably synchronous, but the waning of the LGM is more rapid for tropical regions, as evidenced by the doubling of the curve, with the dashed line more closely following tropical evolution. The reconstruction of African palaeoenvironments is based on a very large number of studies, mostly those published in the volumes of Palaeoecology of Africa,

298

Palaeoclimatic variations and modes of general circulation

[Ch. 13

Figure 13.3 Climatic evolution in Africa during the last 18000 years.

and in Faure et al. (1986) and Miskovsky et al. (1987); also on regional or all-African syntheses, e.g., for northern Africa and the Sahara, those of Williams and Faure (1980), Williams and Adamson (1982), Rognon (1989) and Petit-Maire et al. (1991); for central and eastern Africa, Hamilton (1982), Roche (1989), Lanfranchi and Schwartz (1990); and for southern Africa, Van Zinderen Bakker (1982), Heine (1982) and Tyson (1986). Only general traits and main facts have been retained. Non­ fundamental changes have, if necessary, been omitted, especially where dates are concerned: more recent ones have been corrected. The main aim here is to create a picture of mean African palaeoclimates, based on the general coherence of the periods centred on the LGM and the HCO, with a view to describing firstly the dynamics of the African climate at the time of these characteristic meteorological situations, and secondly the progressive meteorolocgical evolution between these extreme situations (Leroux, 1994c).

13.1.2 The palaeoenvironment of Africa at the time of the Last Glacial Maximum (18-15 kyr BP) The Last Glacial Maximum (Figure 13.4) is roughly centred, in Africa, on the period 18-15 kyr BP. In the northern hemisphere, the period 20-18 kyr BP is considered to be the time of “maximum cooling during the last glaciation” (Frenzel et al., 1992), and is more precisely centred around 18 kyr by the CLIMAP Project (1976). For certain regions in Africa this date simply marks the beginning of the LGM, which may even last until as late as 13 kyr BP. Cold dominated, especially at the northern and southern margins of Africa, with the overall mean temperature estimated at 8°C less than that of the present day for the whole of northern Africa (Frenzel et al., 1992). Ocean levels were lower (Faure and Elouard, 1967; Giresse and Lanfranchi, 1984) by about 130 m. The Mediterranean occupied two basins joined by the Tunis Strait, the temperature in the western basin being from 5° to 9°C lower than today’s. To the north of the Red Sea, this figure is 4°C.

Sec. 13.1]

Palaeoenvironments in Africa

299

coastal upwelling : reinforced ==■= diminished ~ present limit of active dunes . • dune formations (Sahara, Namib, Kalahari) Southern limit (Ogolian-Kanemian) miiiiuiiiiij forest 'refuge* (sheltered by topography, or river 'refuge* - Congo Basin) "'!!!'!' low-altitude heterogeneous woodland -------------------- ---- • and its northern and southern li lake levels: high • low ® dried-up o precipitation: greater than present day + less than present day “ n6v6, glacier * temperature : -5 average number of degrees cooler (°C) than present day

Figure 13.4 The palaeoenvironment of Africa at the time of the Last Glacial Maximum (LMG).

The Maghbreb and Mediterranean

In the Maghreb and in the vicinity of the Mediterranean, rain was at first more plentiful at both the beginning and the end of the LGM, but the glacial maximum had a dry, cold climate in an environment resembling that of the Steppes. Atlantic Morocco was humid and misty (Soltanian period) but not necessarily rainy, the onset of semi-arid conditions being “in disagreement with the idea of a recent humid

300

Palaeoclimatic variations and modes of general circulation

[Ch. 13

Soltanian period in Atlantic Morocco” (Dutour and Miskovsky, 1991). The end of the wet period was characterised by pronounced aridification (Weisrock et al., 1985) and by considerable development of coastal, longitudinal dunes after 17 kyr BP. An ancient shoreline (17 kyr) has been found at 130 m on the Moroccan shelf (Pomel, 1986), and the Canaries experienced arid conditions after 20-17 kyr BP (Petit-Maire et al., 1986). Low-lying ground was arid, but the summits of the Atlas Mountains saw rain, and were partially covered by glaciers (Messerli and Winiger, 1980) in Morocco (in the Rif and High Atlas) and western Algeria (the Jurjura). Outflow towards the Atlantic created the Soltanian terrace, and towards the Sahara, the Saoura terrace; melting snow fed lakes at the foot of the Atlas Mountains, most notably in the Saoura valley. In the south of Tunisia, at the height of the glacial period, it was cold and arid with keen winds, the main effect of which was the building up of the third version of the Great Eastern Erg (Ballais and Heddouche, 1991). The Sahara In the Sahara, there were relatively high levels of precipitation on the central massifs, the Hoggar and Tibesti. In the Hoggar, and more especially in the Atakor, the permanent snowline came down to an altitude of 2500 m, and the lower limit of frost fracturing was at 1000-1400 m (Rognon, 1989). In the Tibesti, the cold, wet period lasted from 16 kyr to 14.5 kyr, and (non-glacial) snow formations existed above 3000 m in the north of the massif, with lakes occurring above 1000 m (Jakel, 1979). Vegetation and soil were of Mediterranean type and temperatures were more than 10°C lower in winter and 7°C lower in summer. In the Jebel Marra (Darfur), the caldera lake of Deriba was at its highest level between 19 and 14 kyr BP (Williams et al., 1980). Lower-lying parts of the Sahara and south Saharan latitudes became dry, with little water which was sporadically supplied to rivers by tropical rain, and water­ courses mostly did not reach the sea. At the foot of the Adrar in Mauretania, the Chinchane sebkha disappeared at 21 kyr BP, and the Senegal River was barred by Ogolian dunes and terminated 500 km from the coast (Michel, 1973); the Niger was halted by dunes in its inner delta (Macina), and Lake Chad also disappeared 15 kyr BP, though small rain-fed lakes persisted among the dunes, and there was water between 18 kyr and 13 kyr BP in depressions in the ancient Erg (Servant and Servant, 1983). The lower Nile Valley was invaded by dunes after 17 kyr (Butzer, 1980), and to the west of the White Nile, sand was transported as far as latitude 13°N (Wickens, 1982). The two Niles and the Atbara became seasonal wadis, and their Ethiopian tributaries may have had short-lived increases in flow in summer. The White Nile probably stopped flowing into the main river, which was very low, and the Nile Valley in Egypt and Nubia was hyperarid. Ogolian-Kanemian longitudinal dune formations stretched southwards for 500 km, as far as latitude 13°-14°N. Their advance did not, however, represent the furthest extension of dune formations during the Quaternary; they reached as far as 1° to 2° from the Equator (Nichol, 1999). Off the west coast, the Canaries Current and upwelling were greatly reinforced (Sarntheim et al., 1981; Hooghiemstra, 1986), and the temperature was lower by

Sec. 13.1]

Palaeoenvironments in Africa

301

between 4° and 10°C. Near the coast of Senegal, high levels of pollen from the Sahara and the arid coastal areas are evidence of the strengthening of the trade wind, whose maximum velocity occurred around 17 kyr BP (Lezine, 1991). On the southern margin of the Sahara, zones of vegetation were displaced south­ wards for great distances, for example, by 500 km in the Sudan (Wickens, 1982). The forest was very patchy and was replaced by savannahs, even in the Congo basin; the only remaining areas of trees were in topographical ‘refuges’, for example, on the coast of Liberia, the southern slopes of the Adamaoua, the Chaillu Mountains, and the western side of the western Rift. In Ghana, around Lake Bosumtwi, where an estimated 50% less rain fell and it was about 3°C cooler than today, there was a marked arid phase with two peaks of aridity at 18.5 kyr and 14.7 kyr BP (Talbot and Johannesson, 1992). Around Lake Barombi-Mbo (Cameroon), patches of forest remained in a slightly less arid environment than that of the lower west African coast of between 20 kyr and 15 kyr BP (Maley, 1987). Dense forests were replaced by mountain prairies and mountain vegetation appeared at lower altitudes, descending by around 800-1000 m on the Bateke plateaux, corresponding to a drop in tempera­ ture of about 5° to 6°C (Elenga and Vincens, 1990). The coasts of Gabon and the Congo, where water from Guinea would not again flow until about 13 kyr BP, became much colder (Giresse and Lanfranchi, 1984). In the Congo basin, savannahs were taking over from forests, but the hypothesis supposing the disappearance of dense, humid forests in the central Basin is, accord­ ing to Roche (1989), “implausible”. A heterogeneous forest mosaic appeared, with isolated low-altitude mountain forests replacing the more continuous canopy, and orographic or fluvial forest ‘refuges’ existed in marshy zones at the centre of the Basin, where the Congo River flowed in a random and attenuated way.

Eastern Africa In Eastern Africa, the Ethiopian Highlands were colder and drier, with very poor vegetation. In the Ras Dashan (Simen mountains), the lowest glacial tongue reached down to 3760 m, but most of the terminal moraines are at about 4100 m (Messerli and Winiger, 1980). Glaciers were at their greatest extent in southern Ethiopia, covering 600 km2 on the Bale mountains, with the high plateau at 4000-4200 m completely covered with ice; on Mount Badda there were 140 km2 of ice, and the terminal moraine came as far down as 3700 m. The Dankali depression was very arid between 17 kyr and 12 kyr BP, with a surface dated at 17-16 kyr at the bottom of the driedout Lake Abhe, which had been 150 m deep around 22 kyr BP. In the Ethiopian Rift, with its tendency towards aridity, lakes were at nearly their present levels, or were at lower altitudes (Gasse et al., 1980). Off Somalia, the summer Somali upwelling was much reduced, whilst in the equatorial Indian Ocean there were no great shifts in temperature, the range being of the order of 2°C (Prell and Kutzbach, 1987). In the East African highlands ice covered about 800 km2 (Hastenrath, 1991), with glaciation on Mount Elgon (today ice-free), Mount Kenya, where glaciers came down to an altitude of 3200 m between 21 kyr and 15 kyr BP (Mahaney, 1989), Mount Kilimanjaro, with ice at between 3300 and 3600 m (Coetzee, 1967), and the

302

Palaeoclimatic variations and modes of general circulation

[Ch. 13

Ruwenzori massif, where ice and vegetation types occurred at levels about 900 to 1000 m lower than at present. Rainfall was 25% lower, and temperature between 8.7° and 6.3°C lower (Livingstone, 1980). Lake Rudolf (Turkana), whose outflow did not reach the sea, was low, with dunes partially filling it (Mackel and Walther, 1984). Lake Victoria (Nyanza) was at its lowest (75 m below present levels) between 14.7 kyr and 13.7 kyr BP (Stager et al., 1986), with no outflow, and was salty and surrounded by areas of savannah. In the Kenya-Tanzania Rift, Lake Bogoria under­ went a sharp fall in level about 15 kyr BP (Tiercelin et al., 1982), as did Lake Nakuru and Lake Manyara, which was a salty lake until about 12.5 kyr BP. On the heights of the western Rift, a noticeable retreat of forests has been identified, with larger areas of open country, at both high and low altitudes, from 22 kyr BP (Roche, 1989). In Burundi, where conditions were cold and arid, mountain forests survived in very restricted ‘refuges’, with estimated temperatures at levels 3°_4°C lower than at present (Bonnefille et al., 1992), with 25-30% less rain. Lake Kivu was well below its present level about 14 kyr BP, in an enclosed basin and at a high level of salinity. Lake Tanganyika shrank between 17 kyr and 12.5 kyr BP, losing almost the whole of its southern part (Tiercelin et al., 1988) as water levels fell by about 300 m. The period of maximum aridity at the southern end of Lake Tanganyika occurred about 18 kyr BP (Van Zinderen Bakker, 1982), with the thin forests of the Zambezi reduced in area between 22 kyr and 12 kyr BP, a time of the coldest and driest climatic conditions (Vincens, 1991).

Southern Africa

In southern Africa, the western coastal strip, along the foot of the Great Escarp­ ment, was distinctly different from the Kalahari. Alongside the Namib, the Benguela Current was colder around 19-16 kyr BP (Diester-Haass et al., 1988), with a more powerful flow reaching further to the north. Upwelling was more pronounced, and temperatures lower by about 5°C (Morley and Hays, 1979), or even 8°-9°C. In the southern part of the Namib, south of the Kuiseb River, winter rains came north­ wards, with violent winds moving the longitudinal dunes, which are nowadays static. In the central and northern parts of the Namib it was cold, with little rain, and strong winds drove large formations of dunes northwards. South of latitude 25°S, conditions in southern Africa were generally humid between 17 and 15 kyr BP (Tyson, 1986). In the area around present-day Kimberley, rainfall was at only half its present value at 16 kyr BP, whilst in north-western Cape Province an ancient lake reached its greatest size between 16 and 14 kyr BP. Temperatures were from 6° to 8° lower than today’s, with minimum values occurring around 16 kyr BP. In the Kalahari basin between 19 kyr and 13 kyr BP, there was a distinct difference between the arid north and west, and the more humid east and south (Heine, 1982). The northern Kalahari had insufficient rainfall to submerge the Makgadikgadi depres­ sion, and only marshes and shallow lakes were present. Stronger winds meant that sand was moved towards the southern Congo basin, though, according to Alexandre et al., (1994), “the age of these phenomena is the object of some controversy”: the most ancient dunes of the north-west Kalahari seem to pre-date the humid period of

Sec. 13.1]

Palaeoenvironments in Africa

303

30-20 kyr BP, whilst dunes contemporaneous with the LGM would be those in north-eastern Botswana. In the southern Kalahari a semi-humid climate prevailed, with the Molopo River, a tributary of the Orange, then flowing continuously. Winds were vigorous, following much the same direction as they do today, southwards behind the Damara-Namara escarpment, with a secondary flow eastwards along the lower land of the Orange valley (Heine, 1982). There were great frosts on the high ground, at their most severe about 16 kyr BP, and on the plateau in the middle of the Transvaal, temperatures were 8°-9° lower than at present. On the high Drakensberg there are signs of periglacial activity, but no indication of permanent ice. Vegetation types were at levels about 1000 m below present ones, with a near-absence of forests, especially (as a result of strong winds) on the plains opening out onto the eastern coastline. The Agulhas Current was colder, and the temperature 3°-4°C lower off Mozambique (Van Campo et al., 1990). On Madagascar, savannahs increased in area at the expense of forests, and wind mobilisation was a feature of the south-east of the island (Salomon, 1987). In summary, the environment of Africa (compared with today’s) was char­ acterised at the time of the Last Glacial Maximum by: • •

•

•

general, if unequal, cooling, with temperatures lowest on the northern and southern margins and the western coasts; generally less rainfall, with consecutive lowering of lake levels, though there were marked differences; some sectors enjoyed a net increase in rainfall, and on the high mountains, with their greater precipitation, snow built up neves and glaciers; distinctly more rapid circulation of both air and sea currents, with strong upwellings except in the northern part of the Indian Ocean where the summer Somali upwelling was on the contrary very weak or absent. far less vegetation, with near-disappearance of dense forests (except at some sheltered sites), to the advantage of savannahs and steppes, with considerable extension of dune formations especially (if not uniquely) in northern Africa.

13.1.3 The palaeoenvironment of Africa at the time of the Holocene Climatic Optimum (9-6 kyr BP) The Holocene Climatic Optimum (Figure 13.5) lasted in Africa between about 9 kyr and 6 kyr BP, with marked differences but a generally warm climate. Seasonal variations were at their most marked around 9 kyr BP, and mean temperatures more constant around 6 kyr BP. During this period, at around 9.5 kyr BP, there occurred the hottest period known for southern latitudes, with sub-Antarctic surface waters reaching their highest temperatures at about 9.4 kyr BP (Hays, 1978). Europe had its highest temperatures around 6 kyr BP (the so-called ‘Atlantic’ maximum), with summer temperatures in mid-latitudes 2° to 4°C higher than at present (COHMAP, 1988). Around 8 kyr BP, the ocean around Africa passed through the peak of the Nouakchottian encroachment phase, which lasted until 5.5 kyr BP (Faure and Elouard, 1967). In the middle of this period, at around 7.5 kyr BP, a brief but intense

304

Palaeoclimatic variations and modes of general circulation

[Ch. 13

coastal upwelling : reinforced------- diminished 1111 111 11 QI dense forest (maximum extent) Sahara, Namib .............................. current limit of active dunes Sahel, Sudan : southern limit of Ogolian-Kanemian dunes, now fixed migration of steppe-type vegetation. ------- and its northern and southern limits------------precipitation : greater than present day 4* less than present day — lake levels: high, absolutely or relatively •

Figure 13.5 The palaeoenvironment of Africa at the time of the Holocene Climatic Optimum (HCO).

hiatus occurred, dividing the HCO into two often quite distinct parts, especially in Africa north of the Equator; the HCO finally came to a relatively abrupt end around 5 kyr BP.

Northern Africa In northern Africa, the northern and southern borders of the Sahara saw a simultaneous return to more humid conditions after 14.5 kyr BP (Gasse et al., 1990).

Sec. 13.1]

Palaeoenvironments in Africa

305

Precipitation was more abundant over the Maghreb and near the Mediterranean, especially at the beginning of the period, and on the Atlas Mountains where cedar forests gave way to deciduous forests. Further towards the Sahara, however, drier conditions prevailed, and a cold, humid climate became warm and humid (Ballouche et al., 1987). Regional peculiarities still existed, however, and in the Aures Moun­ tains before 6.3 kyr BP the climate was arid, and sand accumulated until the wetter period of Neolithic times set in, from 6.3 to 4.3 kyr BP (Ballais, 1985). Ancient lakes and marshes beneath the Atlas mountains on the Saharan side, for example to the north of the Great Western Erg (Algeria), “confirm the existence of a climatic phase more humid than today’s, beginning in 9.3 kyr BP and lasting until 3 kyr BP”; a permanent lake soon formed, and later declined, with the eventual drying out of the marshes occurring around 7.2 kyr BP; a new lake, of slight salinity, appeared around 6.2 kyr BP (Gasse et al., 1987). Mediterranean-type vegetation spread southwards by about 250-300 km, and chotts became lakes, notably in southern Tunisia. In the western part of the Libyan Desert, between 10 kyr and at least 7 kyr BP (Pachur and Braun, 1980), there were lakes, and extensive shallow marshlands with a scattering of vegetation, with rainfall estimated at between 200 and 300 mm. On the slopes of the high ground near the Red Sea, wadis were very active in winter, especially between 11 kyr and 8 kyr BP, adding to the Nile flooding (Butzer, 1980). On the other side of the continent, the Canaries experienced a humid period peaking between 9 kyr and 7 kyr BP (PetitMaire et al., 1986). The Sahara The Sahara certainly saw more rain, with moisture brought in from both north and south in a markedly seasonal way (depending on latitude). Arid regions were therefore of considerably smaller extent. The central massifs of the Hoggar and Tibesti were particularly well supplied, with increased amounts of rain during the first part of the period. In the Serir Tibesti, molluscs and calcareous muds have been dated at between 8.9 and 7.5 kyr (Pachur and Braun, 1980), and depositions in an ancient wadi suggest that material was transported from the Tibesti over a distance of 800 km towards the Serir Calanscio. This Tchadian Sahara (cf. the Tchadian period) was a place of scattered lakes and many marshes, the optimum of this lake­ forming period occurring around 8 kyr BP. In the Chinchane area of Mauretania at 21 °N, the era of lakes lasted from 8.3 kyr to 6.8 kyr BP, with the Saharan vegetation being replaced around 8.3 kyr by a pseudosteppe like that of the Sudan-Sahel (Lezine, 1989). Even in the Tanezrouft hyperdesert there were small freshwater lakes between 10.2 and 5.5 kyr; Lake Toudenni (9.5-6.5 kyr BP), in the north of Mali, was at its largest around 8 kyr BP (Petit-Maire et al., 1991). The Niger’s vast inland delta extended to the north of present-day Timbuktu, into the Araouane basin between 8.5 kyr and 3.5 kyr BP, to cover an area of more than 60000 km2. Where Lake Chad now lies, lake deposits linked to ground water occur for the period from 12 kyr BP onwards (Durand et al., 1984), and around 9-8 kyr, lakes were supplied simultaneously from the north (with water flowing in from the Tibesti)

306

Palaeoclimatic variations and modes of general circulation

[Ch. 13

and from the south; the flow from the Tibesti ceased around 7-6 kyr BP (Servant and Servant, 1983). Huge areas lay under water, with the ancient Lake Chad (never attaining, it seems, the dimensions of an inland sea) surrounded by savannahs; lakes were fed by rivers, flowing in as they do today from the south, the Logone river perhaps flowing for a time into the Benoue. Between 11 kyr and 8 kyr BP, the level of the White Nile was 3 m above its (unregulated) modern value (Williams and Adamson, 1982), and the volume of water carried in the Nile was at least three times greater than in normal times today. From the Sahel and possibly also from the highlands of the Sahara, tributaries of the great water systems (Senegal, Niger and Nile rivers) flowed almost continuously during the early phase of the HCO, and after the hiatus of about 7.5 kyr BP they flowed in a less regular and more seasonal fashion, with the maximum in summer (Servant and Servant, 1983; Talbot et al., 1984). In the southern Sahara, zones of vegetation were displaced far to the north, at distances of 500-1000 km from those at present, depending upon the region and the type of vegetation. Along the western coasts, where the Canaries Current was weaker and upwelling less intense, in the Niayes (interdunal depressions found in Senegal), mesophilic Guinean forests reached latitude 16°N between 9 kyr and 7.5 kyr BP. Elements of Sudanese and Sahelian type were present on today’s southern Saharan margin (at 21°N), spreading furthest around 8.5 kyr BP (Lezine, 1989). In the central Tenere, Sudanese-type savannah, with trees, stretched more than 400 km north­ wards (Schulz, 1987), and in the eastern Sahara where the episode of greatest rainfall occurred between 9.5 kyr and 4.5 kyr BP, zones of vegetation were displaced north­ wards by 400-500 km (Ritchie and Haynes, 1987), with the extent of the migration in the Sudan estimated at 300 km (Wickens, 1982). The relative abundance of vegeta­ tion stimulated human (Sudan Neolithic) settlement in the Sahara; the Sahel was at the time an area of savannah, with deep freshwater lakes and rivers all along its southern borders. From 9 kyr BP, dense forests began to win back the land, spreading out from their ‘refuges’ in less than a thousand years to take over an area well beyond that at present covered. In Ghana, around Lake Bosumtwi, the forest rapidly reappeared abruptly at around 9 kyr BP, and lake levels remained very stable between 9.2 kyr and 3.2 kyr BP, with a maximum between 8 kyr and 6 kyr BP (Talbot and Johannessen, 1992). Forests spread up the coasts of Liberia and Guinea and on to Senegal; in central Africa, near the Darfur; and towards Lake Victoria (Nyanza) across the western Rift, where dense mountain forests expanded greatly between 9 kyr and 6 kyr BP, with a marked optimum at 7-6 kyr (Roche, 1989). At the same time, on the coast west of the Congo basin, mangroves and coastal forests developed between 9 kyr and 6 kyr BP (Giresse and Lanfranchi, 1984).

Eastern Africa In eastern Africa, Ethiopia was warm, wet and had renewed vegetation, after the ‘wild Nile’ episode (cf. Figure 13.3) of torrential rains (Vermeersch et al., 1991). Glaciers were no more, and heavy rain kept the Rift lakes filled around 9.5-8.5 kyr, with Lakes Zwa, Langana, Abyata and Shala forming a single body of water. In the

Sec. 13.1]

Palaeoenvironments in Africa

307

Dankali depression, Lake Abaya, which was 160 m above its present level, experienced pronounced shrinkage, but reached a second high level between 7 kyr and 6.5 kyr BP (Gasse et al., 1980). Lake Rudolf (Turkana), fed by the Omo river, reached high levels (+80 m) between 10 kyr and 8 kyr, and was part of the Nile drainage area, as Lakes Albert and Victoria (Nyanza) had been since 12.5 kyr BP. The latter was at its maximum extent between 9.5 kyr and 6.5 kyr (Kendall, 1969), with dense vegetation, much of it evergreen forest, on its shores; after 6 kyr BP seasonal contrasts became more marked and the dense vegetation waned, giving place to semi-deciduous forests. Glaciers on Mount Kenya went back above an altitude of 4000 m between 8 kyr and 5kyr (Mahaney, 1989), and dense mountain forests reached their maximum extent near Lake Sacre at an altitude of 2438m (Coetzee, 1967). Evolution was broadly similar on other reliefs, like the Cherangani Hills, Mount Elgon, and Ruwenzori, where vegetation types moved higher (Livingstone, 1967); on Kilimanjaro the mountain rain forest reached its maximum development at about 6 kyr BP (Coetzee, 1967). The waters of the Rift lakes of Kenya and Tanzania stayed at very high levels, above those of the present day. On coasts bordering the Indian Ocean, as the forests flourished, the summer Somali Current was re-established, and associated upwelling was again a feature. Lake Kivu, where waters rose considerably between 10 kyr and 5 kyr BP, broke through the Bukavu sill around 9.5 kyr (Hamilton, 1982), and joined up with Lake Tanganyika by way of the Ruzizi river. Lake Tanganyika, into which the flow had begun as early as 12.5 kyr BP (Tiercelin et al., 1988), rose markedly between 9.5 kyr and 6 kyr. The thinner forests of the Zambezi were at their greatest extent between 12 kyr and 6 kyr BP (Vincens, 1989), and the temperature over Equatorial Africa has been estimated at 1.4°C greater than at present (Bonnefille et al., 1992). Southern Africa

In southern Africa, the period from 10 kyr to 8 kyr BP represented a climatic optimum (Van Zinderen Bakker, 1982). Almost the whole area enjoyed higher rainfall, which prevailed after 9 kyr BP, and around 8 kyr most of southern Africa was well-watered (Tyson, 1986), and remained so generally until 4 kyr BP. The Benguela Current off the coasts was more modest, with diminished upwelling. The northern Namib had rain, and between the dunes, lakes (pans) were fed by rivers running down through the Escarpment. The Kalahari basin, warmer and wetter than it is today, as ancient soils reveal (Heine, 1982), was again invaded by summer rains and vegetation, mostly savannahs, moving further south, with large lakes scattered in depressions, the largest one containing Lakes Ngami and Makgadikgadi. To the west, in Damaraland near Windhoek, there were wet conditions, particularly between 7 kyr and 6 kyr BP (Scott et al., 1991). Southern Africa was wet, with most of its rain in winter, but the interior saw little of it and was dry, with greater evaporation, and the interior plateau of Cape Province (lower Orange valley) was uninhabited between 9 kyr and 4.6 kyr BP (Beaumont, 1986). There was a major climatic change going on in the Mozambique Channel, at the latitude of the Comoro Islands, from 10 kyr BP onwards, with a re­

308

Palaeoclimatic variations and modes of general circulation

[Ch. 13

establishing of warm and humid conditions (Elmoutaki et al., 1992). Coral reefs appeared in the Channel, while forests flourished on Madagascar (Salomon, 1987). In summary, the environment of Africa (compared with today’s) at the time of the Holocene Climatic Optimum was characterised by: •

• •

•

general warming, the onset of which was earlier in the southern hemisphere where it manifested itself from 9 kyr BP onwards; in the northern hemi­ sphere this occurred at around 6 kyr, after a relatively brief but sometimes intense cool period centred around 7.5 kyr; a general increase in rainfall, and therefore in lake levels; however, with a few restricted areas having markedly less rain; considerably slower air and ocean current circulation, with weaker upwellings except in the northern part of the Indian Ocean, where the summer Somali upwelling was by contrast re-established; luxuriant growth of vegetation, with dense forests spreading rapidly out of their ‘refuges’, considerably larger areas of savannahs, fixation of dune structures, and far fewer regions remaining arid.

13.1.4 Palaeometeorological interpretation

The two situations pertaining in the LGM and the HCO, extreme periods during the last 18 000 years, have (almost) contrary features from the point of view of temperatures, rainfall, wind strengths, ocean conditions, and the extent of vegetation types and of climatic zones. Differences of such magnitude pose numerous problems of interpretation.

Questions

Various points merit reflection. Let us pick out the essential questions. 1

2

3

Temperature variations, especially at the time of the LGM, are large, but unequally distributed. Were they of local origin, or if not, what imported them? Zones of vegetation in southern Africa were situated, according to estimates, 400-500 km further south than their present position, and later, 300-500 km further north. The total migration was therefore of the order of 700-1000 km between the LGM and the HCO. This translation of climatic zones, roughly symmetrical and synchronous northwards and southwards, corresponds to an expansion or a contraction of the tropical part of Africa. What caused these migrations? The acceleration in air and ocean current circulations during the LGM is plain to see, as is their deceleration during the HCO. What was the cause of this, given the fact that their trajectories remained much the same? Was the spread of sand dune formations associated only with the force of the winds? Why did the Somali summer upwelling disappear during a period of strong winds, only to reappear when they were lighter?

Sec. 13.1]

4

5

6

7

Palaeoenvironments in Africa

309

When tropical fluxes were strong, the area they swept was smaller (LGM); when they moderated, the area was considerably increased (HCO). Why this funda­ mental paradox? Dense forests almost disappeared during the LGM, but held on in ‘refuges’ out of which they would spread strongly again: what were these refuges, and how were the trees sheltered? The LGM saw a lack of rainfall in the tropical zone, with values of about 25-30% less than today’s. However, the tropical area was much reduced, the temperature was not much lower, and the amplitude of the migration of pluviogenetic structures (notably the VME) was only slight. What was the cause of this paradoxical lack of rain? Rainfall was (in broad terms) extratropical in winter, or tropical in summer. During the LGM extratropical phenomena made deep incursions into Africa, especially in the north. This was, though, a time of severe and equally paradoxical lack of rainfall. What were the reasons for this?

'Explanations’ We shall deal with only the most frequently-offered explanations. Much of the responsibility is laid at the door of ‘subtropical anticyclonic cells ’, often considered as ‘barriers’, allowing cold air into the tropics, or excluding it (cf. Chapter 3). Thus, “frequent meridional exchanges imply a weakening of anticyclones, while con­ versely, less frequent meridional exchanges imply a reinforcement of tropical anticyclonic cells”. This (erroneous) concept leads, for example, COHMAP (1988) to consider that around 18 kyrBP, “subtropical highs were weaker”, though they were “stronger at 9 kyr and 6 kyr BP” (exactly the opposite of what in fact happened). Similar confusions attend the link between these highs, and subpolar lows (one of which is the 'Icelandic')'. COHMAP (1988) thus proposes, for about 18 kyr BP, an association between “deeper lows and weaker anticyclones”. But the effect of intense meridional exchanges (if only on a directly observable seasonal scale) was, simul­ taneously, deeper lows and stronger anticyclones, and vice versa. No large-scale migration of anticyclonic cells is detected over Africa for the period of the LGM, but rather an anticyclonic reinforcement both northwards and southwards, with associated flux trajectories remaining very much identical (as evidenced by dune alignments). Thermal transfers and the considerable acceleration of circulation confirm the polar origin of subtropical anticyclonic cells: presumed subsident move­ ments (descending branches of Hadley cells) are completely incapable of delivering such an intense flow, nor could they bring on such a degree of cooling! The lack of rainfall in the tropics at the time of the LGM is associated with a supposed slackening of tropical fluxes. The (erroneous) hypothesis according to which the monsoon was “weaker 18 kyr BP” (COHMAP, 1988) obliges us to dis­ tinguish sharply between, on the one hand, the force of the fluxes, and on the other, their area of extension. More ‘local’ reasons have also been put forward, for example the southward shift and maintenance in a southerly position of the ITCZ (i.e. the meteorological equator) - the effect of which should be a great (but not synchronous)

310

Palaeoclimatic variations and modes of general circulation

[Ch. 13

increase in rainfall south of the equator, which is not what is observed. The displace­ ment of the tropical easterly jet (TEJ) is again proposed: the presumed subsidence responsible for aridity being therefore displaced southwards. On the understanding that this jet is replenished by energy liberated by the Indian monsoon over Asia, what becomes of it when the monsoon no longer reaches the continent (Figure 13.9b), as was the case during the LGM? So there are many hypotheses, often simply reflecting trends and/or sometimes founded upon simplistic systems, unconnected with observed reality. Given the magnitude of these changes, and knowing the real mechanisms of pluviogenesis, can we (without mixing up correlations and co-variations), seriously countenance the statement that “the factor at the root of aridity seems to be a slight variation in the salinity of North Atlantic surface water” (Durand, 1993)? As if salinity could be self-modifying, and also be able to control the activity of pluviogenic structures ... Can we also assume a “palaeo-Walker-circulation” to explain the Younger Dryas, without having proved its existence in the present (cf. Chapter 5)? To these explana­ tions, centred on Africa, are added the interpretations produced by models which are supposed to integrate Africa into general evolution. Questions and explanations abound, but the fundamental problem is how to determine the common cause, and the means of transmission, of climatic variations. 13.2 VARIATIONS IN INSOLATION AND IN MODES OF GENERAL CIRCULATION Is the common cause of climatic changes to be found in variations in insolation associated with orbital parameters? Looked at on a global scale, the answer is no, as it has varied by only 0.6% over the last million years (Genthon et al., 1987).

13.2.1 Variations in insolation In the tropics Can the cause lie within an increase in insolation in the tropics? The answer is still no, if we study variations in tropical insolation (Figure 13.6): at the Tropic of Cancer, insolation has varied during the last 20000 years between 4560 W/m12 (2 kyr BP) and 4601 W/m2 (11 kyr), i.e. by 41 W/m2 or 0.008%; at the Equator, between 4928 W/m2 (9 kyr) and 4944 W/m2 (20 kyr), i.e. by 16W/m2 or 0.003%; at the Tropic of Capricorn, between 4550 W/m2 (11 kyr) and 4608 W/m2 (20 kyr), i.e. by 58 W/m2 or 0.012% (after Davis, 1988). These are insignificant variations, out of proportion with known climatic modifications at the tropics. What they do show, however, is:

1

2

that tropical thermal evolution is not a phenomenon of local origin, but is the result of extratropical influences moving towards the Equator; that the tropics have always had more or less the same potential in perceptible and latent heat available for meridional transfers.

Sec. 13.2]

Variations in insolation and in modes of general circulation

311

4700

4650

4600 CM

4550 E S 4500

4450

1—I—

o o o o co

o o o cn CXI

-4~ ■ f

-4---f--

F- 4- ■-4---1---- H-F-

o o o o o O o o o o o o o o o o o o o o o o o o o o o o o o o o o co CXI T— o cn co co h- CD tn CXI CXI CXI CXI CXI CXI CXI CXI CXI T— •»-

• T- -F --4——r—

-4—--r

o o o O o o o o o o o o o o o o o o o o o o o o o o o co CXI ▼— o cn CD tn T— ■»— ■»— •»“

O O o co

4 H T'- 1 4400 1- - ^4 o o O o o o o o o o O o o o o o o o o o o o co CXI T— 1^ CD tn

years BP

Figure 13.6 Tropical insolation at 23°N and 23°S during the last 30000 years (represented by the sum of monthly averages of daily insolation in W/m2).

At high latitudes By contrast, variations in insolation at high latitudes have been much greater. Figure 13.7 shows respective variations in mean summer insolation near the north and south poles (after Davis, 1988). Though the general evolution is more or less broadly comparable, differences appear in the case of high latitudes (White and Steig, 1998; Steig, 2001). During the last 30 kyr, insolation at about 85 N varied between 1944W/m2 (24kyrBP) and 2195W/m2 (11 kyr), i.e. a rise of 251 W/m2, i.e. 13% of the minimum value. Two maxima appear at about 85 S, where insolation varied between 1944 W/m2 (30 kyr) and 2145 W/m2 (3 kyr), i.e. by 201 W/m2, a rise of 10% of the minimum value. The two curves representing insolation intersect three times: at around 28 kyr, the common value being 1970 W/m2; around 17 kyr, at 2082 W/m2 (i.e. 3% higher than today in the north, and 2% lower than in the south); and around 6 kyr, at 2123 W/m2. A slow decrease in insolation is occurring today at polar latitudes. The common cause of palaeoclimatic variations is a solar one, in association with orbital parameters; it is the result of modifications in insolation at high latitudes, the response differentiated by the dynamic factor. MPHs are the vehicles propagating cold polar air en masse in the direction of the tropics, and they are responsible for the transfer (along their leading edges) in the opposite direction of tropical energy to the poles, a phenomenon of the past as well as of the present. As a function of, on the one hand, seasonal variations in present-day general circulation, orchestrated by the polar thermal deficit, and, on the other hand, long­ term variations in the amount of this deficit (Figure 13.7), two modes of general

312

Palaeoclimatic variations and modes of general circulation

[Ch. 13

2200 2150

2100 E 2050 ■V. $ 2000

1950 1900

years BP Figure 13.7 Polar insolation at 85°N and 85°S during the last 30 000 years (represented by the sum of monthly averages of daily insolation in W/m2).

circulation can be characterised as a function of: the polar thermal deficit; initial characteristics of MPHs; and the speed of the meridional exchanges they impel: the rapid and slow modes.

13.2.2 Rapid general circulation (cold scenario)

The simple initial proposition is that, firstly, in a meteorological hemisphere in winter, phenomena are intensified because of the increased polar thermal deficit (Figure 5.6). So a rapid circulation mode combines, schematically, two winter meteorological hemispheres, i.e. a global situation that corresponds to a large thermal deficit, all year round and simultaneously (with more or less marked nuances) at both northern and southern high latitudes (Figure 13.8). The climatic pattern is cold, even in summer, and seasonal thermal contrasts are much reduced, such is the dominance of the cold. A rapid circulation mode (Figure 13.8) coincides with a large thermal deficit in high latitudes throughout the year.

•

•

MPHs are vigorous, deep, and of vast size, maintaining their low temperatures and considerable coherence over longer distances. They transport more polar cold air, are more rapid and have more distinctly meridional trajectories, penetrating further towards tropical margins. The greater dynamism of MPHs causes an intensification of cyclonic circulation at their leading edges, leading to a more intense return of tropical energy towards the poles. This transfer involves both perceptible and latent heat, the

Sec. 13.2]

Variations in insolation and in modes of general circulation

313

Key to figures 70 and 72 : W westerly jet E tropical easterly jet (TEJ) ______ MPH w\ AA--------------------- flux diverted back towards pole (warm air) __ L_ meteorological equator (MC) ~high-altitude tropical highs trade inversion (Tl) ____ - trade and/or monsoon (lower layers)*------------------

Figure 13.8 Rapid mode of circulation (with large polar thermal deficit) - diagrammatic (cf. Figure 13.10).

•

•

•

•

•

latter through the intermediary of water vapour vigorously diverted toward high latitudes. Note that the more meridional trajectories of MPHs means that they can reach and divert warmer subtropical or tropical air, with its greater volume of precipitable water over oceans. This transfer towards the poles may also involve continental air which, in drought-inducing conditions (because of enlarged AAs and the reinforcement of the unproductive character of the trade inversion), can export large amounts of dust. The intensity of this movement of dust, transported by more vigorous fluxes both in the lower layers and at altitude, is evidenced by the large amount of dust in polar ice cores in cold periods in Greenland and the Antarctic. Disturbances in mid-latitudes are more violent because of accentuated thermal contrasts and the increased strength of MPHs, causing more powerful updrafts and deeper and wider lows. Westerly jets, abundantly supplied by stronger updrafts, are in their turn accelerated, and shifted towards the tropics. Anticyclonic agglutinations (AAs) are, in their turn, more vigorous, as they are fed by more powerful MPHs, and are larger in area both northwards and south­ wards over both oceans and continents. Over the land, AAs are strongly reinforced in winter, enhancing anticyclonic stability for long periods. These continental anticyclonic situations are just as frequent in summer, and bring droughts (cf. Chapter 4). Generally, in both winter and summer, more stable AAs form at more tropical latitudes and their unproductive character (inversion) affects wider areas, especially above the tropical zone where the desert is enlarged in the direction of the Equator. Tropical circulation, vigorously fed by MPHs, is strongly accelerated, but trades cover relatively less space, their energy being transmitted onward to possible monsoons. The latter are also reduced in area. At the heart of the tropical zone the situation is paradoxical: fluxes are more rapid and stronger, but the area they sweep is smaller, as AAs approach each other from north and south, causing the tropical zone to shrink. There is also vigorous opposition from the flux blowing the other way in the adjacent

314 Palaeoclimatic variations and modes of general circulation

[Ch. 13

— —MPH r direction of motion of MPHs weak monsoon, > strong monsoon------------- weak trade-------- -► strong trade

||

Figure 13.11 Differential migration of the meteorological equator: rapid and slow modes (diagrammatic).

layers. The meteorological equator is generally shifted north of the geographical Equator (Figure 13.11b).

Ocean circulation at the surface is now decelerated, as MPHs are less dense and trades slower. The great gyres are further from the Equator, ocean currents are decelerated (especially at the eastern sides of oceans, where upwellings are weaker). Density currents at high latitudes are less intense, and less CO2 is absorbed, raising the CO2 concentration in the atmosphere. All oceanic circulation is decelerated, and thermal transfers are less intense. This mode of circulation, with greater seasonal contrasts, was active during the period of equilibrium in insolation centred on 6 kyr BP: It also applied in the period centred on 11 kyr BP, though with less balanced insolation, favouring the northern latitudes (Figure 13.7), though they still retained extensive ice sheets. The two modes, rapid and slow, of general circulation, resulting from variations in insolation in high latitudes, provide answers to questions about: • • • • • •

the intensity of meridional exchanges; the propagation of climatic modifications; the expansion and contraction of the tropical zone; modifications in wind speeds, ocean currents and the intensity of upwellings; the paradox involved in strong fluxes with only slight migration of the ME; and the increased or reduced ability of MPHs to appropriate tropical precipitable water and transfer it to higher latitudes.

They do not however explain the discrepancy in time between the minimum of insolation (Figure 13.7) and the maximum cooling (LGM), the latter (paradoxically)

318

Palaeoclimatic variations and modes of general circulation

[Ch. 13

coincident with insolation greater than that of today in the northern hemisphere. The discrepancy is simply connected with the fact that water freezes, and ice floats ...

13.3 GLACIATION AND DEGLACIATION

In 1990, Duplessy and Morel reminded us that “the onset of glaciation still wears its cloak of mystery”. Nothing has changed: the Laboratory of Glaciology in Grenoble added its judgment: “How does the Earth move into its periods of glaciation? This question has interested the scientific community for many years, but it remains an enigmatic one” (LGGE, 2004). Is this because conditions are being viewed only in situ, as they favour either winter accumulation or a diminution in summer melting, without reference to the absolutely indispensable meridional transfers? Is it because analysis is based upon unsuitable concepts that “no climatic model has ever managed to simulate snowfall ten times greater than today’s in Canada”? What is more, the question will continue to inhabit the realms of fantasy rather than science if we persist in imagining, a century after M. Milankovitch, that “when insolation is diminished in summer, snow which has fallen in the winter lies unmelted, and accumulates year after year this leads to the building of great ice caps” (Paillard and Parrenin, 2004)! How long would it take to build such an ice cap, even if the suggested mechanism were capable of doing it? How does such an ice cap come to be, especially if, as is supposed, “precipitation diminishes” during cold periods?

13.3.1 The onset of glaciation

The question - and it is a crucial one - of the origin of the water which forms the inlandsis is often debated, but without consensus. An example is that of the Barents Sea ice sheet, an extension of the Fennoscandian inlandsis stretching to the Svalbard archipelago. It was 3000 metres thick, and formed relatively recently, originating about 25 kyr BP and reaching its maximum size 20 kyr BP (Siegert, 1997). With others, Hebbeln et al. (1994) noted: “Such a rapid growth of a large ice sheet requires significant amounts of moisture, but the origin of this moisture has been unclear”. This origin has been sought in the vicinity of the inlandsis, but since the Norwegian Sea is almost always covered in ice, the authors suggest that “seasonally ice-free waters were an important regional moisture source for the Barents Sea ice sheet”. Is this proposed local summer-time source really sufficient, given the poor evapora­ tion rate in cold air, especially deep within a period of glaciation? An accumulation of this depth in such a short time cannot be countenanced unless there is a powerful advection of large amounts of precipitable water from far to the south, channelled in this case north-eastwards between the ice-sheets of Greenland and Fennoscandia (cf. Figure 13.12). Khodri et al. (2001) also sought the cause in the ocean (thermohaline circula­ tion), and, always like Milankovitch, in the “persistence of snow in summer”. Their

Sec. 13.3]

Glaciation and deglaciation

319

relief barring passage to MPHs -l 1 i i i inlandsis (CLIMAP, 1981) - altitude in metres MPHs ■■■direction of motion of MPHs - —► flux diverted at leading edge of MPHs conitinent outlines are as at present, no account being taken of the 120 m fall in sea level

Figure 13.12 Glacial topography and dynamics of MPHs in the northern hemisphere during the Last Glacial Maximum (LGM).

model also suggests that “this enhanced the Equator-to-pole thermal gradient in summer, particularly over the North Atlantic where subtropical latitudes are warmer and high latitudes colder”. Here we have an enhancement of the ‘Equator-to-pole thermal gradient’ by the authors, though it should be a ‘pole-to-Equator thermal gradient’. A pressure gradient is set up that decreases as one goes further south, thereby preventing the establishment of any northward flux from the south! And such a transfer would still have to be established directly over such a great distance ... elementary! It is particularly surprising to read, a little further on: “the summer increase of the Equator-to-pole surface temperature gradient acts to enhance the annual northward transport of moisture by the atmosphere” (szc!). Now we see a transfer - northwards - which is supposed to take place in opposition to the pres­ sure field! This transfer, reckoned to furnish “optimal conditions for delivering snow” is unfortunately impossible, since the reasoning behind it is completely wrong,

320

Palaeoclimatic variations and modes of general circulation

[Ch. 13

turning on its head the very simple principle of the force of the gradient! And, as if this were not enough, the authors also point out that “precipitation is therefore enhanced over land”, but they associate the increased precipitation with “a decreased Icelandic low ... (and with)... weaker cyclonic activity” (sic), which obviously runs counter to meteorological reality! This article, which (oddly) appeared in Nature, is symptomatic of efforts to simulate glaciation via models, which efforts serve to reveal the faults (sui generis) of the models themselves. To this we can add, as in the present case, a surprising lack of knowledge of the rudiments of meteorology, a fault often encountered in palaeoclimatology. Possible changes in general circulation are always suggested: for example, the IPCC remarks that “very rapid and important changes in temperature, generally associated with changes in oceanic and atmospheric circulation, occur during the last glacial period and during deglaciation” (Sc. Basis, Chapter 2, 2001). However, these changes in atmospheric circulation are not described. How could they be, since the models take no account of general circulation? Consequently, phenomena are usually considered in situ, as shown by the claim of Paillard and Parrenin (2004), quoted above, that, in glacial periods, “precipitation diminishes”. This ‘relationship’ is nothing more than the application, on the scale of the basic cell of models, of the simplistic principle ‘T/R’: icold=less evaporation — less rain, and it takes us no further than the hypotheses already formulated by Adhemar and Milankovitch. Also unrecognised is the importance of polar latitudes in driving the aerological dynamic. In evidence, look at the curve in Figure 12.11 of June insolation at latitude 65°N (Petit et al., 1999). Why 65°N, when the subject is the Antarctic? Is it simply because “according to Milankovitch it is the high northern hemisphere latitudes which ... are the most sensitive to changes in insolation” (Paillard and Parrenin, 2004)? Or because some “traditionally invoke insolation at 65°N as the preponderant factor in forcing towards glacial conditions” (LGGE, 2004)? So - no individualised meteorological hemispheres, no meteorological equator (a fundamental disconti­ nuity), and circulation over the Antarctic controlled by ... the north pole (!), and not by the aerological dynamic in the three southern units of circulation which meet at the south pole? Conversely, it is also imagined that a “stable anticyclone ... with easterly surface winds” (COHMAP, 1988) might have been able to persist on glacial highlands. Such stability is beyond imagination, since such an anticyclone could never maintain itself, given the extreme mobility that we see today of the air over Greenland and the Antarctic, where violent catabatic winds accompany the departure of every MPH. An article by Bartiein et al. (1998) on the simulation of the palaeoclimate from 21 kyr BP in North America envisaged “the generation of a glacial anticyclone above the inlandsis", an ‘anticyclone’ that - if it existed - would (as a result of the direction of rotation of the associated winds) hinder advection onto the inlandsis of precipitable water from the south! It therefore does not seem that decisive progress has been made by numerical models in the explanation of past climatic changes. Bard (2004) seemed to echo this sentiment when writing about climatic transitions: “In spite of the use of ever more sophisticated models in recent years to study the amplitude, duration and initial

Sec. 13.3]

Glaciation and deglaciation

321

conditions of these sudden transitions, these questions remain open, and debate among the modellers is lively”. But is it possible to restore reality, if no account is taken of that reality itself and we fall into the same errors caused by the concepts of models: • • • •

• •

envisaging phenomena in situ; assuming simplistic relationships in basic cells; seeing statistical correlations as physical relationships; failing to individualise the agents of the general dynamic, of circulation, and of transfers, especially that of atmospheric water; misapprehending the real processes of the weather; and lacking an overall perception of phenomena in a well defined aerological context?

13.3.2 Dynamical processes of glaciation

The building up of mountains of ice more than 3000 metres high requires the inter­ vention of powerful meteorological phenomena, continually acting over thousands of years. The importation of extraordinary volumes of water - as water vapour, and then clouds, and finally as abundant falls of snow, which piles up and is stored as ice - results from a very strong acceleration of meridional exchanges. By reason of their dimensions, energy and regenerative power, the only climatic agents capable of organising this intense, repeated transfer are the Mobile Polar Highs and, more specifically, the cyclonic circulation engendered within the low-pressure corridors formed at their leading edges, bringing precipitable water potential gathered up in tropical latitudes and transporting it towards the polar regions. Climate models do not take MPHs into account, and therefore cannot simulate the dynamic conditions responsible for the formation of the inlandsis. Conditions pertaining to this transportation of air across the North Atlantic may be investigated with reference to the meteorological situations causing snow­ storms along the eastern coasts of the United States and Canada (Kocin and Uccellini, 1990), and blizzards such as those of 1888 (Kocin, 1988) and 1993 (Forbes et al., 1993), both labelled ‘blizzard of the century’. Similarly, conditions have been very severe during many recent winters. For instance, in Photo 36 (Chapter 8), taken on 5 February 1996, we see the immense MPH that brought a record-breaking cold spell to North America. Temperatures fell to —51 °C in Chicago, and to — 15°C in Louisiana and Florida. On its leading edge, this MPH caused a particularly intense, direct and rapid transfer (restricted in the main to the lower layers), of perceptible and latent heat from the tropics across the Gulf of Mexico and the Caribbean in the direction of the north-eastern Atlantic. Less insolation in polar latitudes, a summer phenomenon (obviously not a winter one, strictly speaking, if the Sun does not rise), leads to the constant renewal of this type of situation during all seasons. The thermal deficit is accentuated, meaning more powerful MPHs: so the onset of glaciation takes place in the context of a (more and more) rapid mode of circulation (Figure 13.8). MPHs venture deep into

322

Palaeoclimatic variations and modes of general circulation

[Ch. 13

the tropical zone to pick up their precipitable water, which is well conserved beneath trade inversions with their reinforced ‘unproductivity’. The tropical zone is generally left short of precipitable water, and rainfall levels progressively decrease. However, the extratropical zones benefit from this water, though not all of them, as certain areas are favoured because of the trajectories found within the units of circulation; these areas preferentially harbour low-pressure areas at the north-eastern edges of the units (Figure 5.2, Chapter 5). Less insolation in polar latitudes also means that the altitude at which precipitation advected by MPHs freezes becomes lower and lower, and this is quite low down in these latitudes anyway, even in summer. As the ice massif gains height, dynamic updraft will be reinforced and the rain-snow boundary is reached sooner, increasing the proportion of snowy precipitation. However, above the ‘optimum snowfall’ level, the amount of precipitation is diminished. The accumulated ice, with its higher albedo and its radiative capability, encourages the cooling caused by the lower insolation, and increases the energy of MPHs, which are vigorously ejected by the increasing falling gradient. This in its turn promotes the return of moist and warmer air associated with Iqw pressure, and the intensity of the transfers increases. All the while, the ice stored within the inlandsis is depleting the meridional ex­ changes; as ever greater quantities of their immediately available water are extracted, sea levels progressively fall.

13.3.3 Antarctic glaciation Ice floats on water, and an inlandsis is soon dislocated when the ice is no longer on the land. Antarctica is an island, and its inlandsis has not been able to develop significantly since the LGM (growing only slightly); and here is the key to the growing radiative imbalance of the hemispheres due to the accumulation of ice in the northern hemisphere, where the inlandsis was able to form on land masses. This difference explains why the post-glacial warming was able to proceed more rapidly in the southern hemisphere (White and Steig, 1998). The question of when the ice of the Antarctic began to form is not an easy one to answer, since the main mass of the Antarctic ice cap has remained unmelted ever since it was formed, 60 million years ago, and it may have happened even earlier, in aerological conditions very different from today’s. Also, the extent of the ice sheet has seen little variation, not venturing far from the land mass beneath it during cold periods, which were characterised essentially by an increase in volume. The dynamic of the exchanges to and from the most southerly latitudes is fairly easy to schematise, since the Antarctic continent is centred upon the pole and is isolated far from other continents, except in the case of the southern tip of South America. This geo­ graphical configuration, and the easy passage of MPHs with very great initial energy in both their low-pressure corridors and their associated closed Southern Ocean lows, form part of a regular evolution in tandem with cosmic parameters, as shown by the Vostok curves in Figure 12.11. However, the zonal band of heavy southern rains (on average greater than 1000 mm annually) lies currently between about 40°S

Sec. 13.3]

Glaciation and deglaciation

323

and 60°S, along the corridor into which the lows associated with MPHs move. Consequently, most of the rain falls into the sea, while the Antarctic continent, situated mostly within the 70°S circle, receives less than 300 mm annually, except in the north of the Antarctic Peninsula, where precipitation may attain values of 500 mm. Values decrease towards the interior of the continent, less so in western Antarctica, and more rapidly with altitude to the east of the Antarctic Cordillera; over most of the Antarctic Dome, values stay below 100 mm (rain-equivalent). The dynamic of the weather over the Antarctic ice cap does not exhibit uniform behaviour:

•

•

The eastern Antarctic (to the east of the Antarctic Cordillera), which contains 85% of the volume of the ice cap, reaches a height of more than 4800 metres, and “there has not been any important modification in the thickness of the ice since the end of the last glaciation” (Lorius, 1983). This part of the inlandsis has seen little in the way of variations on the palaeoclimatic scale (Bindschadler, 1998). A kind of ‘glacial immunity’ is conferred upon it by its geographical position (and especially its great distance from South Africa and Australia), its latitude and its altitude. Its altitude, with very low temperatures in the central part of the Antarctic Dome, assures a snow coefficient of 100% but, as at Vostok, only moderate snowfalls of the order of 50 mm per year are possible, which are unrepresentative of the general pluviometric pattern. The western Antarctic, sometimes referred to as the WAIS (West Antarctic Ice Sheet), has held its own during the last three interglacial periods: 125 kyr BP, when conditions were warmer than during the Holocene Climatic Optimum (HCO); 220 kyr BP; and 320 kyr BP, with a possible fourth period 420 kyr BP (Postel-Vinay, 2002). This last is thought to have been the warmest interglacial period of the last 500000 years. Having attained its maximum volume about 20 kyr BP, it is thought to have lost two-thirds of its ice mass during the last deglaciation (Bindschadler, 1998). As Figure 13.7 shows, after the minimum of insolation around 30 kyr BP, at 85°S the volume of ice was at its greatest about 20 kyr BP, and a maximum of insolation from 20 kyr to 3 kyr brought about rapid deglaciation (Figure 12.1 Id), especially in the WAIS. A slow decrease in insolation has continued for the last 2000 years at southern polar latitudes. Most of the mass of this ice sheet, with its much more modest relief, sits on a rocky base, which is below sea level. The ice projecting beyond the land forms shelves, such as those found in the Weddell Sea and the Ross Sea. This part of the ice cap, which is the most vulnerable (because the ice is floating on water), is sensitive to variations in insolation, and grows during cold periods with their more abundant precipitation, and shrinks during warm periods when melting takes place and precipitation is less abundant.

The unusual behaviour of the WAIS is worth analysing. With good reason: for example, before stating somewhat rashly that an iceberg originating from the Larsen ice-shelf was “the most visible sign of the warming of our planet” (Le Monde, March 2002), the so-called ‘scientific’ journalists, and also some ‘scientists’, and the IPCC,

324

Palaeoclimatic variations and modes of general circulation

[Ch. 13

might have benefited from a close study of the dynamic of the past, and of today. We shall return to these matters in Chapter 14.

13.3.4 Glaciation in the north

Although the glaciations in the south and in the north were roughly synchronous, they were certainly not identical in character. Neither was their geographical spread the same, since one was limited to the Antarctic continent, while the other extended widely into continents near the Arctic, but the inlandsis did not cover the pole itself (cf. Figure 13.12). Their respective dynamical characters were also different: the southern dynamic was particularly simple, while encounters between MPHs and mountain masses (of both rock and ice) created a more complex situation within the northern dynamic. The Greenland inlandsis

There have certainly been variations in the volume of the inlandsis of Greenland, but in spite of its latitude, it has been preserved during the Quaternary period. So Greenland has always featured in the picture of air circulation in the North Atlantic aerological space, and is still pivotal. Since this island, lying between 82°N and 61 °N, is quite ‘off centre’ vis-a-vis the North Pole, relatively speaking, its ice should by now have melted, not least in its southern areas which are at the same latitude as Scandinavia. The reason for the longevity of the ice is not a directly climatic, but rather orographical. The ice is spread across a basin surrounded by mountains. These mountains: •

•

on the one hand prevent direct contact between the ice and the sea, thereby preventing large-scale ‘calving’; on the other hand, the altitude of the ice surface (on average, above 2000 metres), ensures temperatures are low enough to preserve most of the ice, with a snow coefficient of almost 100%.

Greenland’s great extension in latitude means that, in the south, precipitation exceeds 1000 mm per year, while in the north the mean annual figure is only about 100 mm, a roughly south-north decline matching in direction that of the height of the snow/rain boundary. Consequently, climatic conditions are not homogeneous, and ice core samples from different parts of Greenland are not (in terms of latitude) immediately comparable in significance. Neither are they comparable a fortiori to the Vostok samples, since there is a general synchronicity but no precise simultaneity (White and Steig, 1998).

Dynamics of the northern glaciation

In the Northern Hemisphere, the presence of land masses and the disposition of high relief create individualised entities of circulation in the lower layers (Figure 5.1). The

Sec. 13.3]

Glaciation and deglaciation

325

trajectories of the MPHs within these entities determine the directions of diverted cyclonic fluxes, and thereby the locations where precipitation and possible accumula­ tion of ice will be heaviest. In the North Pacific, the Rockies channel the precipitable water towards Alaska. In the North Atlantic, air moving up from the south is directed into Canada, Greenland, the Norwegian Sea and Scandinavia. In northern Eurasia, the limit of the ice of western Siberia (Thiede and Mangerud, 1999) is determined by the impoverishment of the precipitable water reaching it from the Atlantic by way of the Norwegian Sea and Western Europe, and from the Mediterranean across central Europe. The impoverishment in the latter case is due to the presence of the CaucasusZagros-Himalaya mountain barrier and the ‘continentalisation’ of the air that occurs after the relief has been crossed. Variations in polar insolation determine the vigour of MPHs and the resulting intensity of this circulation. Compared with the present era, the Eemian interglacial period, which lasted from 130 kyr BP until 120 kyr BP, was a time of greater summer insolation (13% more) in northern high latitudes, and higher temperatures (at least 2°C higher). The Eemian was also warmer than the Holocene Climatic Optimum (HCO, around 6 kyr BP). The ensuing deterioration in the climate unleashed the most recent (Wurm) glacial period, which lasted for about 1 000 000 years and was at its height about 20 kyr BP. About 115 kyr BP, insolation in northern high latitudes (with northern summer at Earth’s aphelion) was 9% less than it is today, and the ensuing lower temperature was the first act: MPHs were thereby strengthened, resulting in an intensification of the transportation of tropical energy towards the more favoured areas of each unit of circulation. A consequence in Scandinavia, for example, was a rise in air temperature (Mangerud, 1991), and sea surface temperatures were 1° or 2°C higher than today’s (Ruddiman and McIntyre, 1979), an evolution worth bear­ ing in mind when contemplating the current situation. Then the weather became colder and more and more violent, and MPHs more energetic, with stronger cyclonic circulation from the south and abundant precipita­ tion at high latitudes, increasing in volume as the lowered snow line encouraged more and more snowy precipitation. The ice cover already present in Greenland and on Ellesmere Island spread across Baffin Land and the Labrador plateaux. Because of the trajectory (similar to today’s) preferentially followed by MPHs across North America between the Rockies and Greenland, the earliest invasion of the ice took place across eastern Canada and Greenland (where advections of tropical air came to a halt). At the same time, intense streams of warm, moist Pacific air along the Rockies towards Alaska led to the amassing of ice on those mountains. Northern Eurasia was invaded later, the ice spreading from the Scandinavian highlands. Cooling in the polar regions was progressively amplified by the increased albedo of the ice, enhancing the radiative effects of reduced insolation. Slowly, the ice of the inlandsis became thicker, the Laurentide ice sheet met the ice of the Rockies and, gradually, the channel between the Canadian Arctic and the Atlantic was closed (Figure 13.12). MPHs slid both northwards into the Arctic and southwards into the Atlantic from these new, impressive ice mountains, which reached heights of 3800 metres in Canada, and 2500 metres in Scandinavia, northern Russia and western Siberia.

326

Palaeoclimatic variations and modes of general circulation

[Ch. 13

Glacial relief and MPHs The slow build-up of these imposing mountains of ice gradually altered patterns of circulation, and especially the directions in which MPHs moved out (Figure 13.12). The narrowing of the corridor between the Laurentide ice cap and the Rocky Mountains caused a progressive acceleration of the cold air of MPHs moving from the Arctic towards the Gulf of Mexico, until the corridor was sealed. Other passages also became blocked, the first being that between Greenland and Baffin Land (via the arc of Ellesmere) and, later, the gap between Scandinavia and Scotland: an almost continuous barrier was erected between the Arctic and the Atlantic. The ice attained its greatest volume around 20 kyr BP, or approximately 4000 to 5000 years after the minimum of insolation which occurred at latitude 85°N about 24 kyr BP (Figure 13.7). At the time of maximum glaciation, the Arctic Basin was divided from the Atlantic by an impassable ‘chain’ of ice mountains, with only one narrow corridor left opening into the Norwegian Sea (Figure 13.12): powerful MPHs, truly ‘polar’ in nature and therefore very cold, moved out across eastern Siberia, China and the Pacific. In the case of the Pacific MPHs, southerly advections on their leading edges resulted in a much greater accumulation of ice along the Rockies and on the western inlandsis. In America itself, the ice spread beyond the Great Lakes, as far south as latitude 40°N (the latitude of Valencia in Spain, or Naples in Italy). Just as happens nowadays in Greenland, ‘American’ MPHs, in this case not really ‘polar’, moved off the icy highlands, which formed a ridge at about 50°N (Figure 13.12). Cooling increased the pressure of MPHs upon the surface of the sea, accelerated the winds, enlarged the ice fields and increased the density of cold waters: this could only lead to an intensification in ocean circulation. As evidence of this, we may cite the acceleration of the cold, driving Canaries and California Currents, and the increased vigour of their associated upwellings. In their turn, the Alaskan Current, the Gulf Stream and the North Atlantic Drift/Norwegian Current (all warm currents) were necessarily accelerated. As happened with warm air during the onset of glacia­ tion, warm waters were also vigorously fed into the Gulf of Alaska and towards the Norwegian Sea, at least early on, before a general cooling took hold with a considerable extension of the ice pack. This intensification of circulation, the cold in high latitudes and the proximity of the inlandsis and the ice fields, all enhanced the absorption of carbon dioxide by dense currents. The lower CO2 concentration (simple consequence) was accompanied (in time) by lower temperatures.

Deglaciation in the north

Insolation in high northern latitudes gradually improved from 22 kyr BP, increasing by 13% to reach its culmination at about 11 ky ago at 85°N (Figure 13.7). Around 10 kyr BP, a remarkable situation existed: simultaneously, northern hemisphere summer occurred at the Earth’s perihelion (11 kyr) and the Earth’s axis was strongly tilted (9 kyr). There was abundant precipitation during the first phase of deglaciation, associated with still vigorous MPHs. Warming elevated the rain-snow boundary so that precipitation was increasingly in liquid form and, as every mountain-dweller and

Sec. 13.3]

Glaciation and deglaciation

327

winter sports enthusiast knows, snow and ice lose the protection of their albedo when rained upon and melt very rapidly as the comparatively warmer water flows across them. As the ‘American’ MPHs departing from the inlandsis gradually lost their strength, the intensity of meridional exchanges was correspondingly slowed as was the advection of water, bringing about a progressive decrease in rainfall and cloudiness as insolation increased. The thickness of the inlandsis rapidly decreased, and it was melting ten times faster than it had formed. However, deglaciation was not a steady process. It was interrupted by intervals of (sometimes severe) cold, as insolation increased. We can mention the two main intervals, the first between 12.7 and 11.5 kyr, and the other between 8.4 and 8.0 kyr BP, and there were other more short-lived events. As a consequence of glacial inertia, the most favourable period of the Holocene Climatic Optimum (HCO) did not occur until 6 kyr BP (i.e. 5000 years after the maximum of insolation) (Figure 13.7). Northern polar insolation has been decreasing since 11 000 years ago and, at 85°N, its value is lower today than it was at the beginning of deglaciation. The retreat of the ice caps created new opportunities for the formation and movement of MPHs. These conditions throw light upon the main fluctuations of deglaciation associated with increased insolation. As the height of the ice mass decreased, catabatic winds were slowed and ‘American’ MPHs became less energetic (and relatively warmer). Conversely, however, corridors appeared one by one through the glacial relief, renewing communication between the Arctic and the Atlantic, and allowing colder Arctic MPHs to move across America (advected cooling). Access to the Norwegian Sea (Figure 13.12) increased between 17 and 15 kyr BP, as the ice retreated between Svalbard and Scandinavia, and a further corridor was open between the Scottish and the Scandinavian ice caps (Siegert, 1997). The so-called ‘Scandinavian’ trajectory followed by MPHs now gave direct access to Europe, and probably contributed to the first cold episode of the Older Dryas. (Some) severe cold returns From about 15 ky ago, a passage began to form between the glaciers of the Rockies and the Laurentide icefield (see ngdc.noaa.gov/paleo/pollen, 2004). Through this corridor, which widened only very slowly (it was still a narrow defile about 14 kyr BP), Arctic MPHs were soon vigorously streaming southwards past the Rockies. These MPHs, colder than their ‘American’ ice field counterparts, brought on the cold episode of the Younger Dryas (12.7 kyr to 11.5 kyr BP), abundant evidence of which is to be found from regions in and around the North Atlantic (Rodbell, 2000). This rapid return to cold conditions even as insolation was reaching its maximum (Figure 13.7) can be explained only in dynamical terms. The coming of these powerful Arctic MPHs, channelled along the eastern side of the Rockies and towards the Gulf of Mexico, saw a short-lived return of the glacial dynamic, i.e. an upsurge in the importation of water from the tropics and a return to very rainy conditions. This was a warm, localised episode, followed by a further cold period when the Laurentide icefield briefly advanced. Rainfall during this warmer episode increased by 50%, and the weather was violent in character; the temperature in southern Greenland rose by

328

Palaeoclimatic variations and modes of general circulation

[Ch. 13

7°C (Dansgaard et al., 1989) as warm air was advected from the south. A similar situation occurred in Canada’s Maritime Provinces (Mott et al., 1986). In the interior of Greenland, the accumulation of snow had doubled by the end of this episode (Alley et al., 1993). This warming was succeeded by marked cooling across North America, mainly to the south of the Laurentide icefield, and in Europe, where glaciers began to advance again (Anderson, 1997). In northern Africa, partly influenced by the North Atlantic aerological unit, the Younger Dryas spelled the end of the wet ‘wild Nile’ episode, and dry conditions briefly reasserted themselves (Leroux, 1994c). The opening up of the corridor between the icy Rockies and the Laurentide ice sheet, allowing some Arctic air to flow across America, also gave some relief from the cold to Asia, and explains the often mentioned time-lag between Atlantic and Pacific phenomena. In central China, the onset of the Younger Dryas saw an “abrupt reversal” of conditions, and put an end to the accelerated deposition of loess brought from the moraines along the leading edge of Eurasian inlandsis, which was in rapid retreat, melting back to Scandinavia by 11 kyr BP. The loess was channelled preferentially towards the Gobi (as the ‘yellow wind’) through the sill of Dzungaria (Figure 13.12). This wind dynamic gave way to an episode when the deposition of dust was reduced and soils were formed in rainy conditions (Zisheng et al., 1993, Ding et al., 1998). Less powerful MPHs meant that moist air made its way to the interior, a prelude to the re-establishment of the Chinese summer monsoon. Conditions of severe cold returned again between 8.4 kyr and 8.0 kyr BP. About 10 kyr BP, although the corridor along the foot of the Rockies was now well open, a large cap of ice was still centred on Hudson Bay, though much larger than the Bay itself. By 9 kyr BP (Barber et al., 1999), this ice still covered the entire Bay, Baffin Land and the Labrador plateau, but by 8.2 kyr BP the cap was breaking up: Hudson Bay was ice-free and substantial deposits persisted only in the Labrador area and Baffin Land. Now Arctic MPHs were able to move off eastwards, an important consequence as it now became possible for cold air to flow into the Davis Strait and Baffin Bay. Greenland, the north-eastern Atlantic and Europe then became rapidly colder (Alley et al., 1997; von Grafenstein et al., 1998; Barber et al., 1999). In Saharan Africa, the Holocene Climatic Optimum (HCO, from 9 kyr BP until 6 kyr BP) was also interrupted by a short, cooler period, separating the warm, wet Tchadian and Nouakchottian episodes (Leroux, 1994c); conversely, the summer monsoon was re-established in China. The climatic oscillations which characterised the last glaciation, such as the so-called ‘Heinrich Events’, the post-glacial cold spells mentioned above, and the ‘Dansgaard-Oeschger Events’, have been interpreted in many different ways. Proposed causes include: the contribution of fresh water flowing from lakes and rivers, meltwater from the ice or from icebergs, ice surges and earthquakes, etc.; but alongside these ‘freshwater and salt-water solutions’ (or rather, instead of them), we find no aerial explanation! This is really too much! When discussion centres on the weather, it is obviously first and foremost to the aerological dynamic that we should turn, to try to unravel meteorological-climatic phenomena. What is needed is a truly climatological approach to the analysis of the dynamic of deglaciation, with reference to variations in insolation at high latitudes, the

Sec. 13.4]

Palaeocirculations over Africa

329

changing intensity of circulation (MPHs, returning fluxes), the evolution of glacial relief and/or the thermal and aerological consequences of volcanic activity. Such is the approach of D. Soto (Thesis, 2008), currently working at the Climatology, Risks and Environment Laboratory (LCRE). Orbital parameters, and their glacial consequences in the North, shifting and modifying the rhythm of insolation, are responsible for the palaeoclimatic evolution of Africa.

13.4 PALAEOCIRCULATIONS OVER AFRICA

Africa’s palaeocirculations are a result of interference between general conditions, and of specifically African conditions modifying this general impetus.

13.4.1 Circulation at the time of the Last Glacial Maximum

Sea ice covered the North Atlantic north of latitude 42°N, and the American and Scandinavian ice sheets extended across 19.6 million km2 (respectively, 13 million and 6.6 million km2), more than twice the size of the Sahara (8 million km2). The glaciated land mass to the north of Africa maintained the energy of meridional-trajectory MPHs and American MPHs, most of which had crossed the ice sheets; hence the vigour of AAs over the Atlantic, northern Africa and the Mediterranean, and the Arabian peninsula (Figure 13.13). Cold, dense air was permanently advected, and anticyclonic stability pushed pluviogenesis onto the southern margins of agglutinations. Precipitable water was captured and moved northwards towards the polar regions along the leading edges of MPHs, and then above those same MPHs and lower-level AAs, whilst aridity was the rule below. Relief projecting above the agglutinations escaped this lack of rain, extracting part of this potential in the form or rain and snow, especially if, like the Atlas Mountains, it presented a continuous barrier in the path of tropical moisture diverted towards the pole (Figure 13.13b). The accelerated maritime trade, which increased the strength of upwelling, and the reinforced continental trade caused sand to be transported at ground level over great distances, building the longitudinal dunes of the Ogolian-Kanemian. The north Indian Ocean trade (kaskasi) was distinctly more energetic than it is today (Fontugne and Duplessy, 1986; Sarkar et al., 1990), particularly in winter but also in summer, and prevented the establishment of the summer Indian monsoon (Van Campo, 1984; Prell and Van Campo, 1986), a consequence of which was the weakening or dis­ appearance of the Somali upwelling. The powerful nature of the northern trades kept the ME in a very southerly position. During the winter (Figure 13.13a) the surface line of the ME was displaced, at sea to the latitude of the Gulf of Guinea and to the southern edge of the Congo basin; at this time, the thermal lows which today fringe the dense forests had become (before disappearing) incapable of either countering the dynamic nature of northerly (as well as southerly) influences or main­ taining an oceanic flux across the forested area. As a result, the dense forests, which

330

Palaeoclimatic variations and modes of general circulation

................ MPHs direction of motion of MPHs — — — surface line of inclined met. equ : IME ■ k I land over 1000 metres — — —

Figure 13.13

[Ch. 13

—► tropical circulation (trade and/or monsoon) mean position of vertical met. equ : VME Interoceanic Confluence (IOC)

Palaeocirculation in Africa during the Last Glacial Maximum (LGM).

Sec. 13.4]

Palaeocirculations over Africa

331

cannot thrive for longer than 2-3 months in a dry atmosphere, were swept by the cooled harmattan and gave way to heterogeneous wooded areas. Dense forests remained only in sheltered orographical locations, i.e. where there was protection from drying fluxes (especially the harmattan), on western slopes of massifs. In sum­ mer, the northward movement of the Atlantic monsoon, normally attracted in by continental thermal lows, was hindered by the north’s anticyclonic character, limit­ ing the development of the IME. The decrease in rainfall across Africa had its origin essentially in the establishment of direct exchanges between the tropics and the poles, drawing tropical precipitable water from the leading edges of MPHs, and in the pronounced diminution of westerly components in the circulation. In western and central Africa the reduced rainfall was the result of a reduction in the transfer of water inland by the Atlantic monsoon, unable easily to penetrate the continent, and succeeding only in the summer (Figure 13.13b). Trade circulations, predominantly easterly, prevailed over East Africa; Hamilton (1982) particularly noted a reversal in direction of rain­ bearing winds near mountain summits (at altitude, the vector flux is at present the westerly Atlantic monsoon). The northerly and southerly trades brought precipitable water evaporated from the Indian Ocean towards the VME, which remained close to the Equator throughout the year. As a result of the energy and positive contribution of structures involved, there was abundant rainfall, and the levels of Lakes Nakuru and Manyara rose; Lake Bogoria was from three to four times deeper than at present; Lakes Albert and Victoria (Nyanza) had outlets into the Upper Nile. In this cooler context, glaciers were well supplied (although the moderate displacement of the VME reached only as far as southern Ethiopia). However, precipitable water was soon exhausted in the westward flow, whilst the trade fluxes became continentalised on the plateau. By 15 kyr BP, at the end of the period under consideration, lake levels fell, and Lakes Albert and Victoria (Nyanza) were without outlets. This downturn in events can be attributed to a loosening of the ‘vice’ confining the motions of the ME into a narrow band, and to a wider distribution of precipitation of a more seasonal character. The widening of the band within which the VME migrated led to the period of the ‘wild Nile’ (Figure 13.3), during which torrential rains from the VME, lying across Ethiopia, which was then devoid of vegetation, caused massive flooding along the Nile, with large amounts of fresh water and silt pouring into the Mediterranean (corresponding to the biblical ‘Deluge’). In the Southern Ocean, the ice-pack began at 55°S. MPHs strengthened the maritime trade channelled along the Namib, and upwelling increased in vigour as a result; dunes were formed as far north as the mouth of the Congo. The greater depth of MPHs meant that their upper parts could locally cross the Escarpment, which was not difficult around the Orange valley according to the evidence of dune orientation, and flow in a layer eastwards onto the southern African plateau (Figure 13.13b). MPHs proceeding up the Mozambique Channel provoked violent incur­ sions through the valleys and onto the plateau, greatly invigorating the southern trade in its work of moulding the sands of the Kalahari. In the south of south­ ern Africa, more abundant rain was associated with the more profound influence of MPHs. The surface line of the meteorological equator (IME) was kept north of the

332

Palaeoclimatic variations and modes of general circulation

[Ch. 13

Kalahari basin by the dynamic nature of air from the south. The penetration of the Atlantic flux (monsoon) into the Congo basin was hindered, or even halted, by the continentalised southern trade, which pushed the Interoceanic Confluence back to the Great Escarpment by the Namib, and along the western edge of the Congo basin (Figure 13.13b). Over eastern Africa and the Indian Ocean this southerly influence discouraged the formation of the Madagascar monsoon, and the Mozambique Channel and Madagascar were invaded by cold air.

13.4.2 Circulation at the time of the Holocene Climatic Optimum Circulation at the time of the HCO is relatively easy to describe, because its characteristics are almost the opposite of those discussed in the preceding section. No particular schema needs to be presented, as there are great similarities with the present-day situation (Figures 4.5 and 4.6). HCO circulation is characterised by the weaker nature of its centres of action and fluxes, and by an amplification (more marked than today’s) in the migration of discontinuities (ME, IOC) and areas swept by tropical fluxes. The warmer nature of this period led to a weakening of the power of MPHs and AAs; an effect of this weakening, associated with higher levels of evaporation and increased precipitable water and the capture of less tropical energy by MPHs, was more rain. Slower trades meant less upwelling of deep waters near the Atlantic coasts, and wind mobilisation over the continent showed a marked decrease. The thermal factor played an important part, in collaboration with the zenithal motion of the Sun, in creating deep thermal lows drawing air in. Forests spread again as the circulation calmed, to become a determining factor in the definition of the surface pressure field. Around their edges, the dense forests maintained thermal lows, bringing in humid fluxes across the continent and, as they expanded, the lows were shifted both northwards and southwards to re-establish the Interoceanic Confluence across the south of the Congo basin and determine the surface position of the IME, which migrated across a progressively wider area. The lower-layer thermal factor led to great amplification of transequatorial circulations such as the Atlantic, Madagascar and Indian monsoons, the appearance of the summer Somali upwelling providing evidence of the re-establishment of the Indian monsoon over the Indian Ocean. The meteorological equator migrated widely along its whole structure, allowing the deployment of the IME structure; its increased movements in the middle layers (VME) gave a more distinctly seasonal character to the abundant rainfall. At the beginning of the period, around 9 kyr BP, insolation was greater in the northern hemisphere in summer, but the continuing presence of residual ice sheets in America and Scandinavia together with wide seasonal thermal contrasts still acted to make northern MPHs relatively powerful, especially in winter. Precipitation over northern Africa involved two interfering areas of pluviogenesis, tropical and extratropical, and the likelihood of rain in the Sahara was high throughout the year. Lake Chad received a significant amount of water flowing off the Tibesti until about 8 kyr BP (the period of the retreat of cold water diatoms). Around 6 kyr BP, as an

Sec. 13.5]

Conclusion

333

effect of radiative equilibrium between the two hemispheres and the disappearance of the northern continental ice, MPHs and AAs became weaker, allowing the tropical zone to achieve its maximum enlargement, whilst the diminution in the intensity of meridional exchanges meant that Africa could retain the advantages of tropical precipitable water. On the other hand, its seasonal character became more obvious, with tropical precipitation across most of Africa, a (stormy) summer wet season on its margins, and a dry winter season with intense evaporation. In northern Africa, rain from the Mediterranean, linked with MPHs, worked its way southwards after crossing the barrier of the Atlas Mountains. In the south, the area of the Cape was well watered, but the more shallow MPHs could not easily reach the plateau (and more especially the lower Orange valley), which received only a continentalised eastern trade, with strong evaporation. This was one of the few regions which did not enjoy the advantages brought by the HCO; another was undoubtedly Somalia, where the re-establishment of divergence and the lower-layer jet must also have meant less rainfall than in the LGM.

13.5 CONCLUSION Analysis of past climatic evolution facilitates our understanding of present-day mechanisms, not just in the case of Africa but for the whole tropical zone: forests have also shrunk in Amazonia (Servant et al., 1993; Lips and Duivenvoorden, 1994), and India no longer experienced the monsoon at the time of the LGM (Hashimi and Nair, 1986; Sutra, 1997; Figure 113.9). By concentrating on changes, analysis has shown that, as in the case of the seasons today, general circulation is driven by variations in insolation in high latitudes, as it was in the past, whether or not there was synchronicity in both hemispheres. The magnitude of the polar thermal deficit determines the mode, be it rapid or slow, of meridional exchanges carried out through the intermediary of MPHs. Palaeoclimatology shows us that the concept of the MPH and its application to general circulation, which explains particularly the ‘mysterious’ mechanism of glaciation, can help us to understand the key mechanisms of the climate within a defined context of circulation. It can be applied on all spatial and temporal scales. Even if some aspects of past climates are still unclear, and ambiguities remain, our study of them shows that the dynamic of the climate is rigorously organised. We read about the “collapse of the climate”, the “breakdown in the climatic machine” or the climate “turning”, but such terms are a mere perversion of language. The dynamic of the weather and of the climate always follows the same principles, both in the past and the present, and any modifications are not of their nature, but of their intensity. So in order to discuss their ‘breakdown’, we first of all have to know how they really work: what are the real mechanisms? Before summoning up catastrophist scenarios (such as Bard’s “Can the Climate be Turning on Us?’’ (2004)), more suitable for the tabloid press than for scientific debate, we should be pondering the lessons of the past!

334

Palaeoclimatic variations and modes of general circulation

[Ch. 13

Palaeoclimatology also tells us that natural climatic changes have precedents, of magnitudes far removed from those of current phenomena, and throws light on the general trend of today’s climate. In fact, since about 4800 years ago (Figure 13.3), the climatic system, still almost a warm scenario with a relatively slow circulation, has gradually, with many remissions, slid towards a cold scenario with rapid circu­ lation. And more specifically in Africa, is the Sahel drought one of the manifestations of the progress of this climatic evolution? Does the period from 1930 to 1960, a contemporary climatic optimum, or again, the period 1980-2000, constitute the most recent remission in the long history of climate since the end of the Holocene optimum?

14 Recent climatic evolution

If we are to believe the IPCC, recent climatic evolution is characterised by ‘global warming’, caused by the greenhouse effect and leading to an upheaval in the climate. However, weather, and climate, composed of successive weathers, obey exact mechanisms that are unchanging in generally comparable conditions. The climate does not ‘break down’: there are variations in intensity - not in its nature. We must seek to understand those mechanisms (cf. Parts I and II). Moreover, no parameter evolves in isolation, and the weather is an entity to be studied as a whole, unless it is convenient to dissociate. So no analysis of the climate can be based on just one parameter (today, for example, temperature is singled out), without its validation by bringing in other elements of weather. Any parameter taken and observed in isolation gives an imperfect picture of phenomena. All measurements are imperfect, and their representativeness is limited. As a parameter, rain is the most discontinuous in space and time; it records the occurrence and the efficacy of the meteorological phenomena responsible, but only at the level of the observing station. The climatic message is different according to whether the rain falls all the year round (the annual total may be unchanged, even though there are wide seasonal differences), or whether it is tightly concentrated into one season, as in marginal climatic domains like the Sahel. Information about the rainy season may help to build up a picture of the possible displacement of pluviogenic structures. Temperature, which is much influenced by local factors, is estimated using two daily values: maximum temperature (Lx) and minimum temperature (tn), whether these values are for a period of some hours or even minutes or seconds. Note that mean temperature is conventionally equal to tx + tn/2. If, in summer, there are above-average temperatures for the season (e.g. in France in 2003, or around the Norwegian Sea where the temperature curve is rising, or when in Alaska ice is melting), warming due to the greenhouse effect is immediately invoked. If a period is colder than normal (as was the case in March 2008 in France which had record cold temperatures; in mid-November from the Atlantic to Russia when a

336

Recent climatic evolution

[Ch. 14

wave of cold caused the deaths of more than 200 people; or in Canada’s Maritime Provinces where the temperature curve is falling), the cooling is immediately seized upon as evidence of a return to the Ice Age. Recourse to such shorthand and to simplified or even downright simplistic relationships, short-circuiting the real chain of processes, does not ask the question: do we know which of the facts involved have global significance and which local? Within this shifting reality with constantly alternating heat and cold through space and time, how truly representative is the recent rise in mean temperature (IPCC) since the 1980s (Figure 12.13)? Is this a real rise, or just an artefact of the method of calculation which may involuntarily, through the number of observing stations, show a bias towards regions which are getting warmer? There is bound to be imprecision in climatic analysis; it obliges us to seek further levels of information in observed parameters and to consider general climatic evolution in a given aerological space, i.e. one in which physical connections between parameters are known. Only in this way will we take account of weather, through analysis of the dynamics of phenomena, and of climate (expressed through means) which is the resultant of weather. Climate analysis, which is more or less imperfect for given parameters, depends also on the geographical context involved. As we have already stressed, the notion of a ‘global climate’ is a fiction. The climate cannot be defined on such a planetary scale. We shall also analyse the facts considered by the IPCC as emblematic of the presumed ‘climatic warming’. This approach will involve successive examples (it is impossible to include everything), through analysis of climatic facts currently considered to be of major importance: • • • •

the the the the

14.1

great Sahel drought; dynamic of the Antarctic; dynamic of the Arctic/North Atlantic/Europe/Mediterranean space; dynamic of the space Pacific.

DYNAMICS OF THE GREAT SAHEL DROUGHT

The IPCC (2007) notes that “drying has been observed in the Sahel”, and ascribes it to climatic warming. This needs to be carefully analysed.

14.1.1

Sahelian pluviogenesis

Sahel means ‘shore’, in this case the southern land ‘shore’ of the Sahara (Figure 13.2), and this area is an example of directly observable climatic change. Its geo­ graphical location, at the margins of tropical pluviogenesis, gives it a single, brief, wet season with only one chance of rain in the year, and variations between years are wide. If the hoped-for rain, normally between 100 and 500 mm, does not materialise, then disastrous consequences are immediately seen in the landscape. Even the possibility of some winter rain (cf. Chapter 8) cannot compensate for the loss of

Sec. 14.1]

337

Dynamics of the great Sahel drought

the rainy season. Vegetation and crops die, ground water levels fall, lakes dry up, flocks wander in search of grazing, and people go hungry. These images are familiar to us,] and, although they have sometimes been rather exaggerated, they are often, unfortunately, true. It was soon realised that the normal levels of 1931-1960 were an illusion. They represent a pluviometric optimum that drew nomadic populations northwards, and they later moved back southwards as the drought set in, facing dramatic competition for land. The ‘danger signal’ of 1967-1968 had gone unnoticed, and was soon forgotten in rainy 1969. Then, to the surprise of all, the great drought came. Similar (though less severe) events had occurred in the 1910s, and later in the 1940s. So the rain would return ... but we now know that this hope was vain. Figure 14.1 shows that rainfall (at a maximum from 1930 to 1960), already very irregular, has decreased in successive stages since the 1970s; distinct shortages occur in 1972-1973, 1982— 1984, and from 1989 onwards. Figure 14.2 (Morel, 1995) reminds us that the lack of i

;

i

i

i

i

Ill 4 L l , l id j... J 1 I fll IT 1 ' 1 nr 1 1 V* 1. . Ill'll ..•1

.1^

1 .

III II

1.1

ill i

4rsOLTJOtr>c3LOC>tr>C3Lr>OLr>iz>ir>(z>LOcz>ir>0Lr*