Biogeography: A Study of Plants in the Ecosphere [3 ed.] 9781315845227, 9780582080355, 9781138407190

Citation preview

Biogeography A Study of Plants in the Ecosphere Third Edition

Joy Tivy

Joy Tivy

Biogeography A StudyojPlants in the Ecosphere

Third Edition

Routledge Taylor & Francis Group LONDON AND NEW YORK

first published 1993 by Pearson Education Limited Published 2014 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN 52 Vanderbilt Avenue, New York, NY 10017, USA

Routledge is a11 imprint of th e Taylor & Francis Group, an i11jonna busi11ess Copyright © 1993, Taylor & Francis. AU rights reserved. No part of this book may be reprinted or reproduced or utilised in an y form orby any electronic. mechanical, or other means, now known or hereafter invented, includingphotocopying and rcc Xfj (2) modified modelwith E dependenton distance- Xn < Xf (from Pielou 1979,adaptedfrom Brown andKodric-Brown 1977)

of island specieshas long been a subject of ecological debate.MacArthur and Wilson (1967) developedan equilibrium theory designedto explain and predict the number of species as a function of a dynamic balance between rates of immigration and extinction (Fig. 13.1(A». In its simplest expressionthe theory assumesthat the numberof speciesis determinedby three factors: (1) the number of speciesin the mainlandsourceregion; (2) the size (area)of the island; and (3) the distanceof the island from the mainland.The equilibrium numberof species, however, would be expectedto increaseon a larger island at the same distance from the mainlandspeciespool; and to decreaseon an island of a given size with increasingdistancefrom the speciespool. As illustratedin Fig. 13.1(B), the larger island would have a lower extinction rate than the smaller one; while the more remote island would have a lower immigration rate. However, although species numbersattain an approximatelyconstantmeanlevel, the compositionof the flora

and fauna would be expectedto changecontinuously.This is becausethe loss of

some specieswill be made good by a gain of an equal number of others. The resulting replacement or equilibrium turn(f[)er rate may be large or small

248

Island ecosystems

dependenton the equilibrium level of the island in question. The average immigration rate (I) of new speciesper speciesin the mainland sourcepool (P) given the number(S) of island speciespresentcan be predictedfrom the equation

I

p-s

or 1= (P - S)

the averageextinction rate per speciespresent(E/S); and the rate at which the numberof specieson the island increaseswith time:

dS

- = 1- E or dt

dS = (P _ S) - S = 0 at equilibrium dt

The validity and predictivevalue of the equilibrium model havebeensupported by empirical studies (Diamond and May 1976) and by field experiments (Simberloff and Wilson 1969, 1970) involving certain speciesgroups,particularly birds and insectson tropical islands. However, the equilibrium theory has been criticised on several grounds, not least that of its simplicity. It treats all species togetherwith the implicit assumptionthat they are equalin numbersand constant for a given island (Williamson 1981). Also, it assumesthat speciesnumber and emigrationrate are dependenton island areaand are independentof immigration. However,as Pielou (1979) points out, large islandsare more likely than small ones to be in the path of migrants.Also, immigration and emigrationare not necessarily independent. Particularly on small islands the species may be 'rescued' temporarily from extinction as a result of population enlargementconsequenton an influx of immigrants. Also, the theory does not take environmental,biotic or historical factors into account. The relationship between speciesnumbers and island area is less close on temperateislands, where temperatureis a more important limiting factor than on tropical islands where the number of vascular speciesis relatedto the size of the island and the meantemperatureof the coldest month. Also, islandswith a large variety of physical habitatshave the potential to supporta greaternumberof speciesthan a smallerone. As a result of empirical studiesof animal specieson tropical mangroveislands, Simberloff(1969) concludedthat a distinction could be madebetweena temporary equilibrium (when species werenon-interactive)followed by a decline in species numbersto a permanentequilibrium as competitive exclusion startedto operate and to result in the extinctionof thosepoorly adaptedand the immigrationof those well adaptedto the biophysicalenvironment.This sortingprocess,it is maintained, would give rise to a 'set of co-adapted'specieswhich was not just a random sample of the original speciespool but which was characterisedby a constant trophic structure related to the location and ecology of the particular island. Indeed, there are many workers who would maintain that ecologicaldiversity was more important than distancein determiningthe size of an island's native biota (Wace 1978). COMPETITIONAND NICHE SHIFT

Evidence for the importanceof competition in the organisation(structuring) of vertebrate communities is the niche shift (expansion or ecological release) characteristicof many island species (Gorman 1979). Island immigrants can

249

Ecosystems

spreadinto habitats not inhabited by their mainland parentsand occupy niches which on the mainland would have been occupied by other species (Ricklefs 1973). Niche shift describesthe responseof an island species(or ecotype)in terms of its feeding, breedingand biotic adaptationsto a habitatwhich differs from that of the areaof origin. Lack (1943) demonstratedthis phenomenonfor island birds in Britain (Table 13.2) and Williamson (1981) quotes the example of the red Table13.2 Lack'sexamplesof nicheshifts in Orkneybirds (from Williamson 1981) Species

Nonnalmode

Orkneymodeof occurrence

Fulmarusglacialis (Fulmar)

Nestson cliffs

Columbapalumbus(Woodpigeon)

Nestsin trees

North Ronaldsay, Sanday Nestson flat ground* andsanddunes Orkneymainland, Rousay Nestsin heather

Turdusphilomelos(Songthrush)

Nestsin bushesand trees

T merula(Blackbird) Anthusspinoletta(Rockpipit)

Habitat:woodsplus bushyplaces Seacliffs

Acanthiscannabina(Linnet)

Bushesandscrub

(Calluna) Westray,PapaWestray Nestsin walls and ditches* Most islands,rocky and wet moorland PapaWestray Out of sightof sea* Sanday,Stronsay, Westray Cultivatedlandwithout bushes,reedymarshes

The normalmodeis alsofound in Orkneyin all cases. * Also on othernorthernislandsof the British Isles.

squirrel (Sciurus var. leucenrus) and the pine marten (Martes martes). The red squirrel is a subspeciesof the more widespreadpalaearcticS. vulgaris. With the decline of its natural habitat in coniferous forests it had spread to deciduous woods. Its presentdecline in the latter in face of 'apparentcompetition' (a hotly disputedpoint) with the grey squirrel (s. carolensis)has beenattributedto the fact that it is better adaptedto coniferouswoods. The pine marten,also characteristic of both coniferous and deciduouswoods, expandedits niche on to rocky areas following widespreaddestructionof its natural habitatin Wales. ISLAND EVOLUTION

The original equilibrium theory assumedthat all island speciesoriginated by immigration and did not allow for evolutionarychange.However, this is true only for continentalislandswhich were probablyoversaturatedwith speciesat the time

250

Island ecosystems

of their separationfrom the mainland. Fossil evidencesupportsthe view that, on separation,speciesemigration would exceedimmigration and numbersin those islands would fall until an equilibrium level dependenton the size of the island and its distancefrom the mainlandwas attained. The biota of oceanicislands,however,is characterisedby a high percentageof endemic species (Table 13.3) most of which are the autochthonouslyevolved Table13.3 Percentageendemicflowering plantsin selectedareas(from Carlquist1974)

Continentalareas California WestVirginia

Continentalislands British Isles

Old continentalisland New Caledonia

Oceanicislands Canary FernandoPo Galapagos Hawaii JuanFernandez StHelena

Numbero/species

Percentendemic

5529 2040

38.0

1666

o

2600

(90)?

826 826 386 1721 146 45

53.3 12.0

Island-likeareas Afro-alpine flora South-westernAustralia

279

3886

o

40.9

94.4 66.7 88.9

? (90)?

It shouldbe notedthat sizeof flora anddegreeof endemismvary with sizeof area,degree ofisolation,physicalconditionsof the areaandadjacentsourcearea:endemismdoesnot correlatewith speciesdiversity.

descendants oflong-distancemigrants.Separatedfrom the parentpopulationsthey have evolved independentlyof them in responseto the variety of new habitats available, i.e. by adaptative radiation. The number of endemicsin a particular taxonomicgroup and on an islandvaries.The numberof endemicspecieswill tend to be more commonin thosegroupswhich have,or haveacquired,a low dispersal capacity, e.g. wingless insects, flightless birds and seedswithout aids for wind dispersal(Carlquist 1974; Gorman 1979). It will also dependon the age, size and degreeof isolation of the island. On large islandswith a varied range of habitats, the probability that a resident population will be of a size and persistenceto undergoadaptiveradiationwill be higher than on a smaller island (Pielou 1979). The island must be large enoughand with sufficient habitatvariation to provide effective genetic isolation. The minimum area required will obviously vary between taxonomic groups. It must also be remote enough to result III a

251

Ecosystems

sufficiently slow immigration of colonists to ensure that specIatIOn can occur before all the available niches are filled and/or adaptiveradiation by one species blocks that of anotherwith similar ecologicalrequirements.Radiation,however,is limited or indeed absentin some taxonomic groups and on someremote islands. Gorman(1979) notesthat certainanimalsat the limit of their dispersalrangesuch as frogs (in New Zealandand Fiji) havenot radiated.Evenon large islandssuchas Hawaii, with a high degreeof endemism,there are successfulcolonists with a small numberof species. Adaptive radiation will tend to result in decreasingpopulation size and increasingniche specialisationas exemplified by Darwin's famous finches with over 40 specieson Galapagosand the Hawaiianhoney-creepers and lobeliad plant group, with 6 endemicgeneracontainingover 150 species(Gorman 1979). With decreasingpopulation size the probability of the species becoming extinct increasesas a result of environmentalchangesor predation. This speculative evolutionary sequenceor taxon cycle, shown in Fig. 13.2, was illustrated on the Entry from primary sources IExpans;on I

population of distribution and relict status

Extmctlon Mountam specialisation

Early subspeciation Early expansion of new taxa

Full Solomons distribution

I Subspeciation Secondary radiation

Fig. 13.2 Model of the taxoncycle (from Ricklefs andCox 1972)

Greater and Lesser Antilles archipelago by Ricklefs and Cox (1972). They classified the bird speciesin terms of geographicdistribution into groupswhich were interpretedas reflecting the main stagesin the taxon cycle (Table 13.4), i.e. (I) Undifferentiatedspeciesspreadingrapidly on someor all islands; (II) Speciesextinct on a few small islands; clear speciespopulation differentiation on larger islands; (III) Speciespopulationsfragmentedinto separatespeciesor subspeciesoccupying reducedareas; (IV) Descendantspeciespopulationsendemiconly on one island; have become relict endemics.

The taxon cycle has beenexplainedin different ways. Ricklefs and Cox (1972) maintain that competitionis the main factor 'driving' the cycle in that the initial competitive advantageof a particular species declines as predators (including parasites)and more competitive speciesevolve. Others (e.g. Diamond and May 1976) argue that the blanketing invasion of an archipelagoby mainland bird

252

Island ecosystems Table13.4 The numberof endangeredor extinctWestIndian speciesof birds asa

function of the taxoncycle (from RicklefsandCox 1972) Stagesoftaxoncycle I

II

III

IV

428

289

229

57

Numberofpopulationsin dangeror extinctsince1850

0

8

12

Percentage

0

3

5

13 23

Total numberof islandpopulations

speciesand subsequentimmigrants of the same speciescould be the result of involuntarylong-distancetransportby strongwinds (Pielou 1979). ISLAND ECOSYSTEMS

The ecosystemsof oceanic islands differ from those in comparablecontinental physical environmentsin (1) the impoverishmentof their flora and fauna and of their life forms; (2) the relative simplicity of their trophic structure; and (3) the influence of the seaon their energyflows and nutrient cycling. The poverty of the biota (particularly the lack of mammals)of oceanicislandshas alreadybeennoted. There is also a paucity of the more specialisedlife forms such as epiphytes, saprophytesand semi-parasitesof flowering plants (Carlquist 1974). As a result insular food chains tend to be shorter and simpler than in similar continental habitats.Many niches may be unoccupiedbecause immigrantspecieshave been selectedmore for dispersabilityor reproduction thanfor competitiveability, and in the higher trophic levels consumerstend to be generalistfeeders(Wace 1978). The interactionbetweenoceanicisland ecosystemsand the surroundingmarine ecosystemis of considerableimportance becauseof the high shoreline to area ratio. Thereis a constantand relatively large transferof nutrientsand energyfrom the seato the island and vice versa.Small islands,in particular, can be compared to 'high-energy' coastal environmentsin terms of exposureto salt spray, high wind and wave force and to tidal variations of sea level. Wind and wave energy effect the import to, and export from, the island of organicenergyand of nutrients. Also, the nutrient cycle is considerablyinfluenced by marine life - particularly birds and mammals(seals)and reptiles (turtles). The latter are all carnivoreswhich occupy the top marine trophic level. While they do not feed on the island, it providestheir breedingand restinghabitat.They effect, via faecesand corpses,an importantnutrient input. The amountof dried nitrogenplus excrement- guanoproducedby birds and reptiles and of rock phosphatedeposits,particularly on tropical islands, testifY to the importanceof the nutrient input. In addition, the coral reefs and atolls which developin associationwith oceanicislandsin tropical areasrepresentthe accumulationof calcium and other nutrient transfersfrom the seato the land. However, althoughnutrient input is high, lossesfrom the island to the seaare

253

Ecosystems

also large. This is becausethe size of the land area is small in relation to the energy-exportingagentsof wind, wave and surfacerunoff of precipitation. Also roostinglbreedingcolonies of the animals previously referred to are often of exceptionally high density and their movementfrom these to feeding grounds maintain a part or the whole of the island habitat in a disturbed condition. Although the nutrient budgetmust vary from one island to anotherdependenton its location and size, it is consideredthat in general,becausethe islandsare being continuouslyerodedaway by the sea,nutrient transferfrom the island to the sea will, in the long term, be greaterthan those from the sea to the land. And this negativebalancehas further beenexacerbatedby the impact of humanactivities.

ISLAND BIOGEOGRAPHYTHEORYAND CONSERVATION Since its publication in the mid-1960s the island biogeography theory has stimulatedtwo closely interrelatedlines of inquiry. One is its relevancefor islands other than marine ones,i.e. for the land-boundedisolates or habitat islands. The other is its potential applicationto the designand managementof naturereserves (Kent 1987). The latter, in particular, has elicited a flood of publicationswhich, thoughnow somewhatdiminished,still continues.Most havetendedto concentrate on the implications of the theory for nature reservedesignwithout always having due concernfor the validity of the theory, the predictionswhich could be basedon it and!or for the managementthe problemsinvolved. In his recent and most comprehensive review and critique of island biogeographytheory, Shafer (1990) notes that while the theory has provided a conceptualframework for the considerationof a number of related ideas, its practical applications for conservationare, in fact, fairly limited. As has been previously noted, the theory is a simple one, based as it is on the relationship betweenspeciesnumbersand the size and relative isolation of marineislands;and its corner-stone - the equilibrium theory - has not been convincingly substantiated.Of the predictionswhich have beenbasedon it, i.e. that: 1. 2. 3. 4.

The numberof speciesincreaseswith size of area; Speciesrichnessdecreaseswith distancefrom source; The numberof specieswill fluctuate; The turnover rate will depend on the balance between extinction and immigration; 5. Colonisationwill be non-random. Shafercontendsthat only the first is unarguable. SIZE OFAREA

All otherthings beingequal,the larger the areadelimited of a particularecosystem the greaterwill be the numberof speciescontainedwithin it, up to a point. Studies of minimum areasrequired to sample ecosystemsindicate that speciesnumber

254

Island ecosystems

increasesup to a point at which it levels off, becoming very slow or ceases. However,given the rateof speciesand habitatreductioncurrentlytaking place,the larger the reservethe better. Since it will be lessvulnerableto externalimpacts.It is more likely to containviable populationsof a greaternumberof taxa, as well as representatives of the highertrophic levels. Ultimately the optimum and maximum size neededwill dependon the particular size of the organismswithin a taxon or group of associatedtaxa, their relative abundance,evennessof distribution and their ecologicalrequirements.The capabilityof existing reservesto be increasedis negligible, while the size of future (as of existing) reserveswill inevitably be limited by other than ecological constraints affecting species numbers. Many workers have found a closer correlation with habitat diversity and latitude or a combinationof area,habitat diversity and latitude. Habitat diversity appearsto be a more important factor than size on the GalapagosIslands. Many relatively small high-altitude volcanic islands have speciesdiversity high in proportion to size becauseof their diverse range of climatic zones from tropical forest to alpine, as on Tenerife (Canaries). Size, however, is not the only variable affecting SLOSS - single large or severalsmall. A designproblem relatedto that of reserveareais whetherSLOSS reservesof equal areawill contain more species.This questionhas spawnedas manydiverseopinionsas thereare ecologistsand conservationistscombined.They spana continuumfrom thosein favour of one or anotherto thosewho considerit of little practical value for conservation(Murphy 1989). Many advocatesof the small-sizedreservespoint to the increasedspeciesrichnessas a resultof the longer combined edge than in the single large reserve. However, again, as Helliwell (1976a, b) notes, much dependson the size of the speciesand the nature of its habitat. SHAPEOF AREA

Many ecologistshave noted the superiority of a circular over a linear shape;the latter certainly gives the maximumareato boundaryratio. It has also beenargued that a long thin reserve is likely to suffer from the so-called 'peninsulaeffect' whereby the number of speciesdecline with distancealong an actual peninsula away from the mainlandsourcearea.Thereis, however,no evidencethat the latter will contain more speciesthan the former. Nevertheless,as in the case of the SLOSS debate,the advantagesof a longer edge-effeaassociatedwith the linearshapedreserve is consideredan advantagefor the managementof large game animals(e.g. deer) for whom the ecotonebetweenthe completeshelterof a closed wood and the openpotential feeding ground is particularly important. SPACINGOF AREAS

Much considerationhas also beengiven not only to the size but to the spacingof reservesand the degreeto which they shouldbe aggregatedor clustered,aswell as to the advantagesand disadvantagesof connectingcorridors betweenthem. As importantas the spatialpatternsof the reservesare their relationshipsto the matrix

255

Ecosystems

of other unmanagedand managedhabitats within which they are set and their local, regional and continentalsignificance.

CRITIQUE OF ISLAND BIOGEOGRAPHYTHEORY During the 1970's island biogeographytheory was hailed as an intellectual breakthroughin a discipline in which theoriesare notoriously difficult not only to formulate but, more particularly, to validate. There was a growing conviction indeed belief - that island biogeographycould facilitate the prediction of the number of species,and the rate of extinction, in reservesof a given size (Shafer 1990). It was seenas the key to conservationmanagementand decision-making problems.Various reservedesignguide-linesreputedlyderivedfrom island theory havebeenproposed(Diamond 1975;Wilson andWillis 1975) and adoptedwithout reservationand qualificationby the InternationalUnion for Conservationof Nature (IUCN) 1980 World ConservationStrategy. Further, it is now embodiedas an establishedprinciple in many reputable ecological, conservationand planning textbooks! There is now a growing realisationthat the theory relatedto marine islands is not necessarilyapplicable to the terrestrial habitat islands (or isolates) of which nature reservesare a particular example. And even its early advocatessuch as Simberloff and Abele (1976) have more recently seriously questioned the application of a limited and insufficiently validated theory to conservation problems. FRAGMENTATION AND INSULARISATION

However, even if the theory asksmore questionsthan it can at presentanswer,it has servedto stimulatedebateaboutthe relationshipsbetweenecologicaldataand conservationneeds(which will be discussedmore fully in the final chapter)and to highlight the processesof fragmentation and insularisation of the unmanaged habitats and of the landscapeas a whole. Since 1950, unmanagedhabitats,e.g. forest, woodland, heathland,grasslandand wetland, throughout the world are being reducedat an ever-increasingrate either by direct exploitation or gradual encroachmentof other land uses on formerly more extensive habitats. Shafer quotestwo striking examples- that of the fragmentationof forest in a township in Wisconsininto habitat islands during the period of Europeansettlement.The 55 small forest islands remaining in 1950 had by 1978 been fragmentedinto 111 forest islandswith an averagesize of 0.009 km2 • A more recentstudy (Webb and Le Haskins 1980) showedthat the areaof heathlandin the Poole Basin (Dorset) was reduced by 86 per cent; and by 1978 the 10 original blocks containing c. 40 000 ha had beenreducedto 160 fragmentswith an averagearea of 4 ha or more, 608 less than 4 ha, 476 of which were less than 1 ha. As a result not only have the remainingnaturalareasbeenreducedto remnantislandsin an otherwise managedmatrix but, becauseof the erosion,existing reserves arebecomingmore insularisedand isolated from theseislands outwith its boundaries.In thesecases

256

Island ecosystems

conservationmust inevitably be more concernedwith the managementproblems raisedthan reservedesign.

HUMAN IMPACT ON ISLANDS Isolation has protectedislands from human impact for a longer time than many other areas. Once broken, however, the depletion of island biota and the breakdownof the island ecosystemshave been devastatinglyrapid. Indigenous Polynesianpopulationswere early establishedon the large Pacific islands from Hawaii and Fiji to Samoa.It was, however, not until the circumnavigationof the world and exploration by large sailing ships in the sixteenth and seventeenth centuriesthat the long isolation of the oceanicislandswas broken. They became points from which supplies of water and food were replenished and later, particularly in the nineteenth century, where permanentor semi-permanent whaling stationswere established. As a result of these early contactsexotic (alien) specieswere accidentally(by escape)or deliberatelyintroduced.Rats (which early attainedpeststatus),catsand dogs are examplesof the former. In the latter casepigs, goats, sheepand cattle, deer, rabbits and donkeys were brought in to provide food supplies for transoceanicships and whalers. Many of the native animals lacked the ability to survive in face of the mammal predators, to compete successfullywith large populationsof domesticherbivoresor to resist the depredationof disease.Some native populationsdeclined drastically in number or becameextinct. Among the latter the flightless land birds suchas land rails, moas,owls and eagleswhich filled previouslyunoccupiedmammalnicheson islandswere particularlyvulnerableand suffered a rapid early decimationand extinction. It is estimatedthat of the ninety original speciesofland birds in New Zealand43 per cent are now extinct (Towns and Atkinson 1991). Their 'natural'nichesare now occupiedby feral ancestors. The impact on native island floras was no less severethan that on the fauna. Reductionor extinction of plant as of animal specieswas most marked in those with small populations,restricted habitats,narrow and specialisedniches. Many failed to compete with more vigorous immigrants, or to survive the increased incidenceand intensity of fire and grazingby domesticatedherbivoresor the loss of specialisedanimal pollinators essentialfor reproduction.On the larger oceanic islandssuchas Hawaii, Fiji and New Zealandthe replacementof native vegetation by exotic forest plantationsand large-scaleintensive commercialagriculture has beenaccompaniedby an increasein both native and exotic weedsand pestsand in the use of herbicidesand pesticidesto control them. Today the sourceand natureof humanactivities on island ecosystemsmay have changedbut the effect of the initial impactcontinues.The twentieth-centuryuseof remote oceanicislands for strategicpurposes,for research(biological, environmental and defence) and tourism continues to maintain the stresseson, and endangerthe survival of, the remainingsmall populationsof rare endemicisland species.

257

Chapter

14 Mountain ecosystems

The island conceptcan be applied to terrestrialas well as to marine ecosystems. Terrestrialor 'continentalhabitat' islands(McArthur and Wilson 1967) are those relatively small areasof land or water surroundedby, and isolated from, larger ecologically similar areasby an extensiveareaof dissimilar habitat. The type and degreeof isolation of a terrestrialisland, however,is not always so clear-cutas in the caseof the marineisland, nor can the effectivenessof the terrestrialbarrier be as readily as that of the sea.Among the most clearly delimited terrestrial assessed islands are those where wet or aquatic habitats(e.g. bogs, swamps,lakes/ponds, oases)are surroundedby dry or arid conditions, or highlands (mountains) are surroundedby lowlandswhich are sufficiently different in physical characterand extensiveenoughto make migration acrossthem impossibleor extremelydifficult for the majority of species.At the global scalethe mountainisland holds pride of place.

MOUNTAIN CHARACTERISTICS While mountains owe their identity to their height above sea level there is no universally acceptedbasealtitude, i.e. that abovewhich environmentalconditions are inevitably different from those at lower levels. Mountains are more clearly distinguishedfrom their surroundinglowlandsby their interzonalcharacter.Rapid changein climatic conditionswith altitude is reflectedin an altitudinal zonationof the biota which, up to a point, appearsto replicate that from the equatorto the poles (see Fig. 14.1). The mountain world provides a reduced model or a microcosm of a large part of the total world environment. The ecological importance,if not dominance,of relief is greaterin mountainsthan in any other habitat. Variation in altitude, slope and aspectresults in a diversity of sharply contrastingmeso-and microhabitatsand a biotic variability within mountainsthat far exceedsthat in the surroundinglowlands. At the global scale, mountains combine the roles of a 'pioneer front' and a 'museum' (Rougierr 1962). The

258

5

4 3

Highest

Highest snowline Highest

Highest snowline

6

Highest

7

Highest

8

Highest

9

Highest

10

Highest Highest

OOOm

Highest Highest

Highest Highest

Mountain ecosystems

2 2 0

90"

60"

N

30°

Highest snowline

0°

30°

60Q S

Highest maximum timber line

Lowest snowline

Fig. 14.1 Global cross-sectionof alpineregionsshowinghighestsummits,snow-lineand timber-line (Ives andBarry 1974)

former is a function of the severity of environmentalconditions which at high altitude eventually become limiting for life. In the latter, mountains and particularly isolated mountain tops are characterisedby a high proportion of endemic species many of which are relicts of a formerly more widespread distribution. Finally, althoughmountainshave long providedrefugia for peopleas well as for plants and animals, and despite their recent rapid exploitation, they retain some of the most extensiveand still relatively 'natural' wildernesshabitats in the world today. BIOCLIMATIC ZONES

Increasein altitude is accompaniedby a generaldeteriorationof environmental conditions.The proportion of direct to diffused solar radiation and of ultraviolet radiation (UVR) increases.Ambient and surfacetemperaturesdecreaseand the length and warmth of the growing season decrease.In temperate latitudes cloudinessand precipitationincreaseand an increasinglyhigherproportionfalls as snow while the depth and persistenceof the snow-coverincreases.Wind-speed also increaseswith altitude. The resulting environmentalgradientis reflected in three distinct and usually clearly delimited bioclimatic zonesas illustrated in Fig. 14.2: (1) the montanezone; (2) the alpine zone; and (3) the zone of permanent snow and ice. The tree-lineor upperlimit of the montanezoneand the permanent

259

Ecosystems

snow-line or lower limit of permanentsnow, glaciers and ice-caps increasein altitude from near sea-leveltowardsthe equator.However, at anyoneplace, the actual location of theseboundarylines and the biota of the montaneand alpine zonesare dependenton the rate of climatic changewith altitude and relief (i.e. slope (or gradient)aspectand relative exposureof the groundsurface).Mountains are characterisedby such a tremendousdiversity of topoclimates(local climates) and microclimates(i.e. climate near the ground) as to invalidate the conceptof a distinct mountainclimate.

DAY

MORNING

W W

W

W MORNING

MORNING

W W

W W

W W

W

W

W

Fig. 14.2 Diurnal slopewind circulationin (A) a temperatemountainvalley and (B) on an isolatedtropicalmountain.S = snowline; I = inversionof temperature;eN = cummulonimbusclouds:cloud base(basedon Flohn 1969)

Topoclimate The slopeof the groundin relationto the directionand incidenceof solarradiation is undoubtedlythe major topoclimatic variable. In comparisonto level surfaces, the heat and associatedwater balanceon sloping ground at a similar altitude is modified becauseof the resulting differencesin the angle of incidence of solar radiation.The astronomicallypossibledurationand total of daily solauadiationis modified by the screeningof the horizon in mountains.Relatively gentle slopes can have a marked effect on the reduction of direct radiation. Also slopes at differing altitudes can initiate secondary(local) patternsof air circulation which further influence the heat and water balance. A measureof the comparative differencebetweenthe topoclimateof sloping comparedto horizontal surfacesis provided by the conceptof equivalentslopes (Geiger 1969). The equivalentslope 260

Mountain ecosystems

is expressedin termsof the numberof degreesof latitude, eithernorth or south,of a given slope where a level surface at the same altitude will have a similar topoclimate. Consequenton aspect, equatorward-facingslopes experience a sunnier, warmer and drier climate than those shady, cooler and more humid poleward-facingslopes(Table 14.1). Theseclimatic differencesare accompanied Table14.1 Soil moistureasa percentagedry weight of groundat 5-10 cm on the Grosser Stanfenberg(HarzMts), 17 September1953(from Geiger1969) Slopedirection Heightabovesea-level(m)

N

ENE

500-540 450-499 400-449 350-399

22 25

21 20 25 27,30

26

S

w

14

14,15 14

21

18

by those in the soil climate and associatedbiota. In deep, steep-sidedalpine valleyswith a latitudinal orientation,the tree-lineand the transientsnow-lineattain higher altitudes than on the shadedslopes.Also, slopeswith a westerly (in the northern hemisphere, easterly in the southern) aspect have higher ground temperaturesthan thosefacing east(or west in southernhemisphere).In the latter a higher proportion of the morning solar energyis expendedin evaporationfrom humid soil, while the latter drier slopesreceiveintenseradiationin the afternoon. Maximum ground temperaturesare displacedtowards the west and south-west (northernhemisphere). Periodicvariation in the radiation and heatbudgetsof slopesinfluencesthe air circulation and results in characteristicvalley climates. As shown in Fig. 14.3, during the day and particularlyduring the summer,the warmersurfacelayer of air in the valley bottomsmovesupslope.At night, cold air tendsto move, albeit very slowly, from the colder summitsdownslope.The steeperthe slope the more rapid the flow which, in mountains,often occursin periodic 'surges'or 'avalanches'of cold air. The coldestair collects in depressionsand valley bottomswhere, in the absenceof other air movements,temperaturesdrop below freezing and a sharp temperatureinversion forms between the cold valley bottom air and a warmer thennal belt above (Table 14.2). In the latter the risk of late spring and early autumn frosts is greatly reduced,and the growing seasonis longer and warmer than on the valley slopesabove or below. In tropical and subtropicalregionslocal mountain and valley winds can reach gale-forceintensity, particularly when they are funnelled in deepgorges,suchas thosesome5000 m deepwhich extendfrom the Altiplano of Brazil to the Amazonlowlands (Flohn 1969).Above the mountain summits local upslope winds are horizontally integrated into reverse anti-wind systemsof varying scales. Over large mountain massesthey give rise to large relief-inducedand thermally driven mountainsystemswhich at particulartimes of the year can dominatethe weatherpattern. Mountains also exert a passiveinfluence on air movementparticularly during

261

262

749 591 646 651 640 625 619 608

1307 (peak 1312) 1157 1008 925 850 796 658 622 (valley)

1955

(mm)

MayOct.

Height of station abovesealevel (m)

(117) 131 127 132 120 119 92 96

Total (em)

(9) 10 43 49 46 41 65 55

% of melting snow

Freshly]allen snow

Precipitation

3.4 4.5 4.7 5.5 6.0 6.3 9.1 11.7

-25.2 -20.1 -19.4 -16.6 -18.5 -19.1 -23.6 -28.1

Daily temperature range in may Absolute min. (C) (C) 5.5 6.9 7.6 8.0 8.1 7.9 6.4 4.3

Mean min. (C) 8.2 9.1 10.0 10.5 11.0 11.1 16.8 10.6

Mean (C)

Mean temperature May-Oct. 1955

Air temperature

12.1 12.3 13.1 13.9 14.6 15.1 16.8 16.8

Mean max. (C)

Table 14.2 Observationsin the thermalbelt on the GrosserFalkensteinHarz Mts (from Geiger 1969)

(118) 144 156 157 159 158 109 97

15 12 16 18 22 11 6

Length of frost-free period (days) Spruce(em)

9 11 13 15 12 8 6

Beech(em)

Phenology, increaseoflength ofshootsofplants 5-6 years old (mean, 1956-57)

Ecosystems

Mountain ecosystems

Tropical Andes

Brazi l S Africa Australia

S. E. Asia I

Deciduous Deciduous forest

Europe

Patagonia New Zealand

6 Deciduous

500

ec

i

0

(Nohh) 70 50 Humid tropical forest Sub-tropical evergreen Sub-tropical deciduous

s

1

o

D

uou

2

st re foous du

cid

500

De

500

0

30 0

00

500

Cloud forest Mixed deciduous and evergreen Deciduous forest

500

700

P

(South)

Paramo Boreal forest Arctic (Antarctic) Alpin e

Fig. 14.3 Diagrammaticnorth-southglobal cross-sectionillustratingvariationin altitudeof mainvegetationzones(redrawnfrom Rougierr1962)

stormy weather.Wind-speedcan be acceleratedto storm or gale force particularly where it is funnelled along narrow valleys or over cols. Dependenton wind direction, slopes can be exposedor shelteredfrom high wind-force. Exposure increasesevapotranspirationand exacerbatesthe depressiveeffect of altitude and aspecton temperatures.It also exertsa considerableinfluence on the distribution of precipitationin mountains.The correlationbetweenaltitude and precipitationis much less close than that between altitude and temperature.While there is a tendencyfor precipitationto increasewith altitude, in temperateand subtropical climatic regions,this is dependenton the location, altitude, massand orientationof the mountainin relation to the global air-masscirculation. Mountain rangescan initiate uplift of unstable air massesand result in an increasing amount of orographicprecipitationon windward slopesand a rapid decreasein precipitation as a result of the rapid adiabaticwarming of descendingair on the lee-sideor rain shadow.Dependenton orientation,sharp contrastsbetweenhumidity and aridity can occurover shortdistances.In the tropical zone(c. 10°Nand S of the equator), however, convectional uplift determinesthe upper limit of the cloud base and resultsin a precipitationmaximum at c. 3000 m (Geiger 1969). Snow The actualamountof precipitationwhetherin the form of rain or snowwithin the mountainareais dependenton the relative degreeof exposureto, or shelterfrom, the prevailing wind. Soil temperaturesare determinedmore by snow depth than altitude. Snowprovidesa protectionagainstfrost. On exposedridges and summits from which snow is blown the ground can freeze to considerabledepth during

263

Ecosystems

winter. A high proportion of winter precipitation in mountainsfalls as snow. At high latitudes winter temperatureconditions are such that snowfall is the main form of winter precipitation and the duration of a snow-covera regular seasonal occurrence.Suitableatmosphericconditions(i.e. high humidity plus temperatures near or below 0 °C at ground level) only occur at a critical· altitude which is dependenton the regional climate and the temperaturelapse-ratewith altitude. Snow-covercan be seasonaland of generallycyclic occurrenceor permanent.The boundarybetweenthe snow-coveredand snow-freegroundis the snow-linewhose altitudinal limit varies both temporally and spatially. A seasonalsnow-coveris characterisedby a transientsnow-line the altitude of which can vary annually,and during the courseof one winter seasonit is dependenton fluctuating fall, aspect and shelter.The permanentsnow-line,abovewhich a coverpersistsfrom one year to the next, may vary annually but neverthelessattains a maximum position in relation to landform and altitude and maintains a long-term mean boundary position betweenthe seasonaland permanentsnow-cover.At the global scalethe snow-line increasesin altitude from the poles where it is, theoretically, at its lowest level towardsthe equator(Fig. 14.2).

VEGETATION ZONATION

In spite of the local environmentaldiversity the mountain biota is universally characterisedby an altitudinal sequenceof relatively distinct zones (or stages), each with a homogeneity particularly of vegetation form and composition. Although the numberand nomenclatureof zonesvary, dependenton altitude and zonal climate on the one hand and on the focus and scaleof particularstudieson the other, five main zones are commonly identified in the EuropeanAlps and North American Rockies.

1. Hill (or basal) zone: flora and fauna related to that of lowlands; dominanceof cultural elements;interregionalvariation of vegetationcommunities; 2. Mountain (or montane)zone: distinguishedby presenceof speciesendemicto a greater or lesser extent to mountains; nearly always forest-dominated, particularlyby deciduousbroad-leavedtrees; 3. Sub-alpinezone: comprisingone or more 'horizons'dominatedby conifers and an upperfringe of shrubs: (a) basalhorizon - spruce(Picea) dominated; (b) middle horizon - larch (Larix spp.) and pines (Pinus cembrot, P. arole). The undestoreyis, characteristically,composedof heath shrub species, numerousmossesand lichens; (c) transitional summit horizon - shrubs(Ericaceae,Rhododendron,etc.) an abundanceof mossesand lichens, rosette alpine herbs and some airelles and dwarf trees (Pinus mugho), birch (Betula), willow (Salix) and alder (;lInus); 4. Alpine horizon: without treesor upright woody plants, the vegetationis sward-

264

Mountain ecosystems

like characterisedby a continuous cover dominated by grassesand sedges, small forbs with rosette or cushion growth, short stems, reducedleaves and large brilliant flowers. A further distinction can be made betweenthe welldrainedand the marshywaterloggedhabitats; 5. Snow horizon: betweenthe alpine horizon and the zone of permanentsnow; discontinuousvegetation cover with an increasedabundanceof mossesand with increasing severity of environmental conditions lichens adapted to particularmicrosites. However, not all mountains conform to this alpine zonal model. The biogeography of each mountain depends on not only the local and regional environmentalconditionsbut also on the effectof pastenvironmentalconditionson the biota. The absolutealtitude, numberandwidth of the vegetationzonesvary. At the global scale the altitudinal limits increasefrom the pole to the equator; at equivalentlatitudes,from the exterior to the interior oflarge mountainmasses;and from coastalto landwardslopes.Thereis a markedcontrastbetweenthe zonalflora of the mountainsof northerntemperateand Boreal climatic regionsand that of the intertropicalandsoutherntemperateregions.Also, sincethe mountainrangesof the former were floristically linked during the colderclimatic phasesof the Quaternary period, the speciescompositionparticularlyof the montaneand alpine vegetationis lessvariecl than in the long-isolatedmountainsof the southernhemisphere.Among the biotic characteristicscommonto all altitudinal gradientsare: 1. A decline in the number of vascularspecieswith height; but becauseof high habitat diversity total speciesrichnessof mountainsis greaterthan that of any other terrestrialenvironment; 2. A decreasein the structuralcomplexity of the vegetationcover; 3. A high degreeof endemism; 4. The dominanceof conifers in the upper montanehorizon; 5. The developmentof alpine meadows(or grassland)which have no latitudinal equivalent.

ALPINE (OR MOUNTAIN) TIMBER-LINE

A distinction is frequently madebetweenthe mountain(or altitudinal) timber-line and the latitudinal tree-line (seep. 177). The former is the upperlimit of tree and forest growth; the latter the extreme limit of growth of trees or shrubs of 2 m height or more. The tree-line, however, is an ecotoneof varying distinctiveness and width betweenthe sub-alpineforest and the alpine zones. As such it is an importantfocus of biogeographicalstudy by ecologists,forestersand geographers. Wardle (1974: 306) considers it to be 'the sharpesttemperaturedependent boundaryin nature' and hence may provide a sensitive guide to environmental change. The sub-alpineforest varies in specific compositionfrom one part of the world

265

Ecosystems

to another. In temperateand Boreal regions it is dominated by a few hardy conifers.With increasingaltitude the growth height of the treesdecreasesand the forest is eventuallysupersededby a krummholz- a densethicket oflow, contorted, horizontally spreadingtrees and shrubs - or by an area of sparselydistributed small but upright tress comparable to the tundra elfin forest. Coniferous krummholz is particularly widespread in north temperate mountains and is extensiveon the Rocky rangesin Colorado and Wyoming. Developmentalforms vary from tree islands(wherethe lower branchesof adjacenttreesbecomelayered) and low flaggedandprostratekrummholzto the isolatedcushionkrummholzunder the most severe conditions. In some species krummholz is environmentally produced,in others it has beenproved to be genetically determined.In tropical highlandsthe sub-alpineforest is composedof arborescentforms of generawhich in extra-tropical regions are representedby small herbaceousspecies. These arborescentmegaphytesinclude speciesof Euphorbiacae,Senecioand bamboo. On a global scalethe altitude of the timber line varieswith latitude (Table 14.3) and with degreeof continentality.It increasesfrom the polesto the equator,inland from the coast,and from the exterior to the interior of large mountainmasses.In the interior of the Rockies and the Himalayas the tree-line attains altitudes comparableto those in the tropics. While the latter has been ascribed to the massenbergeffort (Le. decreasein the temperaturelapse-ratein larger and higher mountains)the higher summertemperaturesof continentalcomparedto oceanic

Table14.3 Altitude andmain speciesof selectedtimber-linesat varyinglatitudes(from Wardle 1974) Location

Altitude(m)

63°NSweden

1000

60°N Alaska

900

Main species Birch (Betulapubescens) Sitkaspruce(Piceasitchensis)

50°N British Columbia

1850-1900

Abieslasciocarpa(krummholzto 2500)

50°N Rockies(Alberta)

2150-2300

Englemannspruce(P. engelmannit)

47°N Switzerland

1900-2000

PinuscembraandP. abies

41-44°N Caucasus

Up to 2500

Betulaverrucos

38°N S. Nevada

3300

Pinusalbicaulis (krummholzto 3750)

28°N E Himalayan

3800

Larix griffithii

19°NMexico

Mean3950

3900-4100

Pinushartweggi Polylepi Podocarpuscompactus

4900

Polylepitomentella

36°S SnowyMts (Aust.)

1850-2000

Eucalyptusniphophila

42°S SouthIsland(NZ)

1200-1300

Nothafogusmenziesii

Max. 4100 9°SAndes 6° SNewGuinea 19°S N. Chile

266

4100

Mountain ecosystems

climatic regimesareconsideredas,if not more,important.The alpine tree-linehas, as in the tundra, beencorrelated(p. 95), with the 10 °C isothermfor the warmest month of the year by many authors.At the local toposcalethe tree-linetendsto be depressedon exposedconvexslopesand to be higher on shelteredconcaveslopes. Inverted tree-lines are found in depressionsand valley floors where treeless vegetation occurs below the sub-alpine forest zone becauseof unfavourable conditionscreatedby nocturnaltemperatureinversionsduring the growing season. Well-markedtemperatureinversionsin temperateregionsmay havelittle effect on vegetationduring the growing seasonbecauseof short nights and temperatures above O°c. However, in tropical regions, the less cold-toleranttrees are more liable to damageby long night-frostswhich can take place at any time of the year. The factors and processeslimiting tree growth at high altitude have been as contentiousan issue as in the caseof the tundra tree-line. Table 14.4 providesa Table14.4 Comparisonof Arctic andalpineenvironmentsandvegetation(from Webber 1974)

Component Latitude Altitude Solarradiation (averageJuly intensity) (0.4-0.7 ~m for growingperiod) Photoperiod(maximum) Air temperature(July mean) Soil temperature(maximum) Precipitation(annualmean) Wind-speed(annualmean) Waterstress Permafrost Averagelengthof growingperiod Numberof commonvascularplants Microhabitatdiversity Averagearealvascularproduction Averageratio of above-to below-ground biomass Averageareanetphotosynthetic efficiency

Arctic tundra (Point Barrow, Alaska)

Alpinetundra (Niwot Ridge, Colorado}

71°N 7m

400N 3549m

0.30 cal cm-2 min- 1 10 x 107 cal m- 2 84 days 3.9°C 2.5°C 107mm 19 !an hour-1 -4to -5 Universal 55 days 40 Small 100 g m- 2 year-1

0.56 cal cm-2 min- 1 9 X 107 cal m- 2 15 hours 8.5°C 13.3 °C 1021 mm 37 km hour-1 -6to -8 bars Sporadic 90 days 100 Large 200 g m- 2 year-1

1:8

1 : 12

0.5%

0.5%

concisereview of the main theories.On the one hand, it was initially considered that tree growth was limited by an insufficiently long period of warmth for the completionof shoot growth so that the requisite degreeof winter hardiness(Le. ability to withstandcold and desiccation)could be initiated by exposureto the first slight winter frosts without risk of damage.Indeed,only four to five speciescan

267

Ecosystems

survive the extremewinter at the high continentaltree-linein the Rockies.On the other hand, recentwork has put more emphasison the tree regenerativecapacity. Seed production is low and variable and seedlingscan only survive if they can withstand or avoid the particularly stressful microclimate at the tree-line. Temperaturesnearthe groundare very muchhigher during the day and very much lower at night than the ambient temperatureabove. As a result, plants are subjectedto a large diurnal temperaturerangeand to soil frost-heavein the winter. Wardle (1974) notesthat microclimatemeasurements in New Zealandshow that seedlingsof some speciesof Nothofaguswill only becomeestablishedin shaded sites provided by the more environmentallytolerant maturetrees.In this casethe tree-line will be composedof erect trees without a krummholz zone. The latter only developsbelow the regional tree-line where, despitea constantexposureto high wind-force, speciesseedlingscan becomeestablished,and survive, but with a slow growth rate and a wind-shearedform. Among other factors that might limit tree growth, carbondioxide assimilation balance,soil temperaturesand increasedUV -B radiationhavebeensuggestedbut not supported by conclusive evidence. Wardle (1974: 398) summariseshis preferredexplanationof the alpine timber-line as occurring 'at an altitude where the environmentaltolerancesof vascularplants and, in particular, their ability to ripen their shootsso as to be able to withstand seasonallyadverseconditions are quite abruptly reached'. The tree-line varies temporally as well as spatially in responseto climatic change,to natural disturbanceand increasingly to the effect of human impact. Evidencefor tree-line fluctuation as a result of climatic changeis relatively sparse and scattered.Vegetationchangededucedfrom pollen analysissuggeststhat the tree-linecould havebeendepressed,dependenton latitude,by 900-1400m during the coldestphasein the Pleistoceneperiod. Dating fossil tree stumpsin the northeasternHighlandsof Scotlandhave revealedthat in the post-glacialBorealperiod the tree-line reached700-800m, 100 m higher than the presentnaturaltree-line (Pears1985). On a shortertime-scaleand at the local site-levelnaturaldisturbanceby regular and severe snow avalanches,intense gales, plagues of phytophagus insects, outbreaksof diseaseand intenselightning-setfires can all depresstree-linesfor longer or shorterperiods of times. In addition, the effects of natural catastrophes have beenexacerbatedby depressionand obliteration of tree-linesconsequenton deforestationparticularlyin tpe long-settled,semi-aridareasof the world from the

northernAndes to the westernMediterraneanand centralAsia, and as aresult of the intensificationof burning and grazing. ALPINE ZONE

The Alpine zone is that area above the mountain tree-line which is not permanentlysnow-covered.It shareswith the arid desertsand the polar tundrasa severityof environmentalconditionsthat becomemarginal for life. Similarities in the physical environment,the flora and the vegetationof the tundra and alpine zones(Table 14.4)havereceivedconsiderableemphasisto the extentthat the latter

268

Mountain ecosystems

was (and indeed still is in many texts) commonly referredto as the Arctic-alpine zone.Thereare, however,importantand ecologicallysignificantcontrastsbetween the two environments.Theseinclude the greaterintensity of solar radiationwith a particularlyhigh proportionof direct radiationand of ultraviolet light than at highlatitude lowlands. Snowfall and wind-speeds are much higher than those experienced in the tundra while the occurrence of permafrost is sporadic. Consequentlargely on varying altitude, aspect and slope the heterogeneityof microhabitatsand the ecologicalimportanceof the microclimateis evengreaterin the alpine zone than in the tundra. Both areas,however, are treelesswith a lowgrowing vegetation cover which becomesmore open and discontinuouswith increasing environmental severity. The flora of both is composed of a high proportion of cryophytes,the most abundantof which are herbaceousperennials (particularly grassesand sedges),small or prostrate shrubs, and cryptogams. However, the alpine flora is much the richer of the two biomes. In addition, the alpine vegetation of tropical mountains is both structurally and floristically different from that of temperatemountains.It is composedmainly of grassland with tall, columnar,arborescentor sub-arborescent life forms belongingto genera such as Lobelia and Senecioin Africa and Espeletiain South America. The alpine zone in the tropics, however, is limited in extent and has beenstudied less than that in temperatelatitudes. The vegetationof the alpine zone in the northernhemisphereis composedof a complexmosaicof plant associationsdependenton gradualor steepmicroclimate gradientssuchas degreeof exposureto, or shelterfrom, direct solar radiationand wind, the magnitudeof the daily rangeof groundandplant temperature,the annual frequency of the soil freeze-thawcycles, the length of the growing season,the depth and duration of snow-coverand the availability of snow meltwater on the one hand and on microvariations in the physical condition and the chemical compositionof the mineral substratumon the other. The most favourablehabitats with the longestgrowing seasonare thoseon gentlebut well-drained south-facing slopes above the tree-line. These carry closed, species-rich and relatively productive alpine-meadow(or alps). In northern Europe and Eurasia they have long been used for summergrazing by transhumantlivestock and as a sourceof winter fodder (hay) for the valley farms. The most severeenvironmentalconditionsare associatedwith persistentsnowbankswhere the thaw commenceslate and the growing seasonis very short and with the highestwindsweptridges and plateaux.At theselimits only cryptogams can survive the long winter cold, the short cool growing seasonand, on snow-free ground,often intensesummerdrought.As in the tundra,cryptophytic mossesand lichens can take advantageof the highestdaytime temperatureswhich occurin the interfacebetweenthe soil and the air and when thereis sufficient ice meltwaterto permit very slow growth. Capable of withstanding either winter or summer desiccation alpine lichens are even longer lived (up to 1300 years has been estimated for the speciesRhizocarpongeographicum) than those in the tundra (Billings 1974). The absoluteupper limit for plant life is reachedbetween5000 and 7000 m dependenton latitude and aspect. However, most exposedalpine summitsare 'bald'.

269

Ecosystems ORIGIN OF THE ALPINE BlOTA

Although some alpine plant specieshave a wide range of distribution such that they can be found in both hemispheres,most are spatially restricted to a few mountainsystems.The closegeneticrelationshipbetweenthe Arctic flora and that of the alpine zoneof the north temperatemountainsof Eurasiaand North America has long been recognised.The origins of this markedly disjunct distribution of apparently similar species have been a subject of considerabledebate among biologists, geologists and geographers.It is now generally accepted that the presentalpine floras of the temperatezone are isolated remnantsof a formerly more widespreadcircumpolar crytophytic flora. The latter is thought to have evolved as the climate becamecolder in the late Tertiary period and to have reachedits greatestextentand replacedthe pre-existingnemoralflora over much of the land in high latitudes by the Plioceneand early Pleistoceneperiods (Love and Love 1974). The high frequency of occurrenceof polyploidy in the small Arctic-alpine flora is thought to be indicative of the relatively recentevolution of the cryophytic flora from pre-adaptedspecies.Inter- and post-glacialfossil flora reveal that as the ice-sheetsretreatedand the climate amelioratedthe formerly continuouscold-adaptedflora was disrupted.Remnantswere restrictedon the one hand to high latitudes, on the other to high altitude habitats.Some specieswith Ar..:tic-alpine affinities were able to persist in open unstablelowland coastal or riverine habitats,free of competitionfrom the larger more productivepost-glacial vegetation. The spatial continuity of the Arctic-alpine flora was, however, maintainedand carriedfar south,particularly on the highestnorth-southtrending mountainrangessuchas thoseof the westernCordillera and easternmountainsof North America, and the Urals of eastern Europe.Isolation of Arctic from alpine and of alpine islandsfavouredthe subsequentevolution of mountainendemics. The relationshipsbetweenArctic and alpine animalsare less marked than for plants.Cold adaptationis commonto both. While trophic and structuralnichesare similar in both biomes, alpine vertebrate mammals do not exhibit the cyclic populations common in the Arctic. In addition, they need, at the maximum altitudes, to be adaptedto low-oxygen levels and to be able to move on steep irregular and frequently unstable slopes. Mammals, in particular, tend to be exclusive to one or the other zone except where they meet in the high latitude mountains. A few species, such as the Arctic or mountain hare (Lepus timidus) are regardedas true relicts. It is, however, generallyheld that the alpine

fauna has by and large not evolved from the formerly continuousArctic-alpine biome. The main centre of evolution of alpine vertebratesis thought to have occurred in the mountains of central Asia. The Tibetan Plateau and adjacent mountains constitute the largest spatially continuous alpine area in the world. Habitat diversity is high. A large part of the area is dry 'alpine desert',and the alpine zone has a wider altitudinal extent than elsewhere.Although quantitative dataare lacking the numberof bird and mammalspeciesattainsa maximumin this area. In addition, it is the only alpine area which escapedcomplete biotic disruptionduring the Pleistocene.Local adaptationof montane(sub-alpine)forest and steppespeciesto alpine conditions also occurredon restricted and isolated

270

Mountain ecosystems

mountain tops throughout the world. In general the restricted area and patchy distribution of existing alpine habitatsmay accountfor the relative poverty of the vertebratefauna comparedwith that of the Arctic tundra. HUMAN IMPACT The diversity and isolation of mountain habitats are reflected in that of a wide range of distinctive human societieswhoseways of life reflect an early and close adaptationto the particular problemscreatedby a spatially variable and difficult physicalenvironment.Traditional methodsof movementand resourceuse tended to optimise the expenditureof human energy and the conservationof resources (particularly soil and water) on high steep slopesand where level terrain is at a premium. The intricate terracing of slopes for cultivation particularly in many subtropicaland tropical mountain areastestifies to the skill with which this was achieved. In many of the long-settled densely populated regions of the MediterraneanBasin and of the Middle and Far East, the mountainswere early deforestedto supply timber for fuel, ships and construction.Today, little remains of the original forest which hasbeenreplacedby shrub or grasslandmaintainedby fire and grazing or by bare slopesravagedby soil erosion as in North Africa, the Middle Eastand northernChina. The humanimpact on the mountainareasof the world has, withinthe last 200 years,increasedin intensity and extentwith the symbiotic developmentof modern methodsof communicationand resourceexploitationparticularlyfor tourism. The latter developedin the nineteenthcentury with the rise in popularity of summer residences,sanitoria, mountaineeringand alpine skiing. Increasedmobility and affluencein the post-SecondWorld War yearshaveseenthe exponentialgrowth of the ski industry particularly in the alpine zones of mountainswith a seasonalor year-long snow-cover of adequate duration, extent and depth. The indirect ecological impact of the constructionof roads, buildings and uplift facilities has been greater than the direct effect of 'piste skiing' on the mountain biota. Increasedeaseof accessto the alpine zone in spring and summer,however, has exposedthe biota to the direct physicalimpact of trampling. Alpine developments have resultedin the drastic disturbanceof the vegetationin an environmentwhich is particularly vulnerable becauseof steep slopes, a substratumsubjected to frequentfreeze-thawcycles and a biota, as in the tundra,living at the margin and where recoveryfrom habitat disturbanceis all the more difficult.

271

Chapter

15 Aquatic ecosystems

The most fundamental contrast within the biosphere is that between the environmental conditions and associatedorganisms of terrestrial and aquatic ecosystems.Although the latter cover c. 75 per cent of the earth'ssurface,plant biomassis dominanton land, animal biomassin water (seeTable 15.1). However, Table15.1 Comparisonof sizeandproductivityof terrestrialandaquaticecosystems

Habitats

Area (J(I km2)

*Approx. volume (J(I km3)

NPP (]O9 t year-I)

(iff tyear-I)

P:C+D

Terrestrial Aquatic Ratio

145 365 1 : 2.5

14.5 1445.0 1: 99

1l0.5 59.5 1: 0.54

867 3067 3: 5401

1 : 0.001 1: 20

SP

* Basedon an averagedepth4000 ill andassuminga terrestrialinhabitedzone100 ill deep. NPP = netprimaryproduction;SP= secondary(animal)production;C = consumers;D = detrivores;P = primaryproducers. while the NPP of aquaticecosystemsis only abouthalf that from the land, aquatic animal productionis over 3.5 times that from the land. Nevertheless,despitethese contrasts,ecologicalprocessesare similar in both habitatsand are influenced by the samecomplexof variables.

THE PHYSICAL HABITAT Wateris in many respectsa more favourableand certainlya more equablemedium for life than is the land. Desiccation is unknown except in transitional areas 272

Aquatic ecosystems

betweenland and water where, becauseof varying water-levels,organismsmust be adaptedto exist for varying periods of time in and out of water. Further, organismsliving in water are immersedin a solution that contains many of the elements essential for their existence. Oxygen and carbon dioxide are readily soluble and available in water. Relatively easily absorbedfrom the atmosphere, their replenishmentand distribution from the surfacedownwardsis facilitated by water mixing as well as by photosynthesis,respirationand decomposition.Of the two, carbon dioxide is more readily soluble than oxygen. It reactswith water to form carbonic acid (H zC03) and can be easily fixed in carbonate(-C03) or bicarbonate(- HC03) form. The concentrationof carbondioxide in wateris c. fifty times higher than in the atmosphere.That of oxygenis, however,on averagemuch less, 0.001 per cent in fresh water comparedto 20 per cent in the atmosphere. Expressedanotherway, the averagedissolved oxygen concentrationof water is 10 ppm, i.e. forty times less than the weight of oxygen in an equivalentvolume of air. Also, while the oxygen concentrationin the atmosphereis uniform, it is extremely variable in water, and is also less evenly distributed in water than carbon dioxide. Its concentrationis higher in surface than in deeperwater, as a consequenceof the concentration of photosynthetic activity in the upper illuminated zone, and in cooler than warmer water. However, exceptin localised areasin very deeplakes and seas(e.g. fiords, the Black and Mediterraneanseas) where bottom water tends to be stagnantand oxygen deficient, water circulation ensuresa replenishmentof oxygen sufficient to maintain some life even in the deepestoceans.In contrastto lakes and oceans,rivers are relatively shallow and more turbulent. They exposea greatersurfaceareaper volume to air and hence, under unpolluted conditions, the amount of oxygen varies little throughout the length and depth of the water body. SALINITY Water contains varying amounts of all the minerals found in the earth's crust. Someforty-five mineral elementsare known to be presentin water; someoccur in larger quantitiesthan others. In the seathe most abundantare sodium (30.9 per cent) and chlorine (55.3 per cent) - the main determinantsof the salinity of seawater. Together, with smaller quantities of magnesium, sulphur, calcium, potassiumand bromine, they account for over 99 per cent of the total mineral content of sea-water.The remainder, occurring in minute and often variable quantities, include the nutritive elements, such as nitrogen, phosphorusand potassium, essential for plant growth and whose availability determines the nutrient statusand hencethe potential fertility of the aquatichabitat. The salinity of sea-wateris determined by the amount of salt or sodium chloride (NaCI) measuredin gramsof salt per 1000 g of water (per cent); thereis a continuum in the biospherefrom marine through brackish to fresh water with a salinity less than 0.05 per cent. However, given the volume of sea-water,the salt concentrationis not excessiveand the range of salinity in the open seas is comparativelylow. The most saline (athalassic) conditions occur in inland lakes (e.g. Great Salt Lake, Dead Sea)and semi-enclosedseasin arid areassubjectto 273

Ecosystems

constantly high temperaturesand evaporation. Among the latter the highest salinities recordedare those for the Red Sea(average40-41 per cent). Minimum salinity is found in brackish coastalwaterswhere the inflow of large volumes of fresh water from the land or from melting ice causesdilution. In the open seas salinity rangesfrom only 37.5 per cent in tropical to 33 per cent in polar seas. With similar temperatures,saline water is denserthan fresh water and in the absenceof movementthe former tendsto sink below the latter and mixing of the fresh and salt water may be relatively slow. TEMPERATURE

Water is physically, as well as chemically, a more uniform and less stressful environmentfor life than the land. The specific heatof water is higher than that of solids; it absorbsheat and losesit more slowly than the land surface.Spatial and seasonalvariationsof temperatureare, however,dependenton the size, depthand mobility of the particular water body. The difference between the surface temperatureof the warmestseas(c. 32°C in the PersianGulf) and the coldest (c. 2 °C in Polar areas) is about 30°C, comparedwith a temperaturerange of 87-90 °C on the land. The annual range of oceanictemperaturerarely exceeds 10°Cand is normally much less. However, most of the heat derived derivedsolar radiation is absorbedin the surface layers; hence, the seasonaland latitudinal variations of temperatureexperiencedat the surfacedecreaserapidly with depth. Below about 100 m there is relatively little annual fluctuation and low temperaturespersist over extensiveareas.On the floors of the deepestoceans, temperaturesvary by only a few degreesaboveor below 0 0c. The relative uniformity of temperaturein the oceans is facilitated by the continuity of the water area, and by the large-scalecirculatory movementsin the oceanbasins.Vertical movement,i.e. overturning, is engenderedby variations in water density resulting from differencesin salinity and temperature.Cold and salinewaterbeingdenserthanwarm/andfresh watertendsto sink below the latter. Also, under the influence of the rotation of the earth and the friction of winds, surface oceanic waters are kept in constantcirculation (see Fig. 15.1). Water heatedin tropical latitudestendsto flow away in greatanticlockwiseor clockwise eddiestowards the colder polar north and south latitudes.At the equator,colder bottom water (originating in the polar regions) wells up to take its place. Upwelling also occurs along the west coasts of the southern continental land masseswhere oceancurrentsand prevailing offshorewinds 'push'warmersurface water away from the continentaledge. Heat radiation is completely absorbedin the top 1-2 m of water. In slowmoving water bodies,the surfacewater heatsup more rapidly than that below. If wind action is insufficient to mix the water, a distinct upper isothermallayer (the epilimnion) of varying depth, with a temperatureover c. 4 °C and hence a lesser density, forms above cooler water below (the hypolimnion). The two become separatedby a rapid temperature gradient or thennocline (+ 1 °C m- I ). The resulting direct stratification may be a relatively short-lived or more persistent phenomenonduring the summerin temperatewater bodies (oceansand lakes) or

274

Aquatic ecosystems

Conlin ental shell

Cold currents

Warm currents

Warm currents currents Warm currents currents

Fig. 15.1 Oceancurrents

persistfor most of the year in tropical waters.The temperaturerangebeweenthe epilimnion and the hypolimnion may vary from 8 °C in temperateto 16°C in continentallakes subjectto severewinters and hot summers.Although the range may only be c. 4 °C in the tropics, the stratification may be more stable than in waterwith muchwider temperaturevariations.This is becausethe changein water density per degreeof temperaturechangeis so much greaterat high than at low temperatures.The stability of the thermocline is dependenton the depth and mobility as well as on the temperatureof the water mass. In shallow (less than 15 m deep) fresh or saline water the ratio of the epilimnion to the hypolimnion will be very much higher than in deepwater and stratification may not occur. In tropical climatesthe thermoclinemay be permanentor, at least, semi-permanent. Cooling of surfacewater and the consequentconvectionaloverturning together with wave action disrupt the thermoclinein temperatelatitudes.In polar areasthe developmentof a markedand seasonallypersistentthermoclineis inhibited by low temperature.In contrast, during winter freshwaterlakes and, under very severe conditions, brackish coastalwater may freeze over, trapping slightly warmer (i.e. abovefreezing) water below. LIGHT

Although a greater percentageof the total solar radiation reaching the earth's surfacefalls on water rather than land surfaces,illumination is much feebler than on land. Loss by reflection can be high, particularly at high altitudes where the angle of incidenceof the sun's rays is low (90 per cent loss, comparedto 2 per cent in calm conditions with an overheadsun accordingto Boney 1989). Light which passesacrossthe surfaceis absorbedvery rapidly by water, yellow humic substances(gilvin), plants and inorganicparticulatematter. In addition, scattering by plants and suspendedinorganicparticlesimpededownwardpenetration.Light 275

Ecosystems

also changesin spectralcompositionand irradiancewith depth. Red and ultraviolet light are absorbedin the upper layers of clear ocean water; blue-greenlight penetratesto greatestdepths.However,blue light is rapidly absorbedin waterwith a high proportion of gilvin. Maximum penetrationis by greenlight (Boney 1989). Irradianceor the rate of supply of radiantenergyper unit surfacearea(expressed as J m-2 S-I (min-lor mol m-2 S-I) decreasesexponentially with depth. In offshore water light penetrationcan be at leasttwice that in turbid inshorewater. In the clear tropical water of the open oceanlight of high intensity canpenetrate four or five times deeperthan in shore. Gross photosynthesis

photosynthes Gross

S

1 Gross photosynthes

Gross photosynthes

Z

Z

Fig.15.2 Characteristicrelationship betweengrossphotosynthesi s andwater depth:Zs = depthatwhich a white disc disappearsfrom view of observerat the surface; Ze= limit of euphoticzone,i.e. light compensationpoint; S = surfaceinhibition particularlyon sunny daysdueto ultravioletlight; I = irradiance(from Moss 1980)

Two ecologically significant light zonesillustrated in Fig. 15.2 can be present in a water body:

1. The euphotic (or photic) zone whose lower limit is that at which light intensity is insufficient for photosynthesisto proceedat a rate which compensatesfor respiration, i.e. the light compensationpoint. The averagelight compensation point has beengenerallydefined as that at which intensity is 1 per cent of the subsurfacelight. The depth at which this occurs varies from 10 to 15 m in turbid marginal water, to a maximum in clear, open water of about 100 m, which roughly correspondsto the average depth of the outer edge of the continentalshelf. In shallow coastalwaters and lakes and in many rivers, the euphoticzone can extendfrom the surfaceto the water bottom; 2. The disphotic (or aphotic) zonewith insufficient or no light for photosynthesis . While the distribution of species of aquaticplantsand animalsis determinedby their adaptationto salinity, that of primary productivity is influenced more by the nutrient status of the water. Most open oceanic areasare oligotrophic (nutrientpoor, particularly in phosphorusand nitrogen). Marine water becomesincreasing eutrophic (nutrient-rich) from off to in shore becauseof the input of mineral

276

Aquatic ecosystems

elementsfrom the adjacentland and because,in certain areas,cold nutrient-rich water wells up at the surfacefrom oceandeeps.In contrast,the nutrient statusof fresh water is more directly related to the chemicalcompositionof the substrate, the natureof the adjacentvegetationand, increasingly,the input of phosphatesand nitrates as a result of human activities. Oligotrophic lakes and rivers are characteristicof poor, acid soil and/or rock and of particularly deep,rock-bound lakes with permanentlyor seasonallycold but relatively stagnantbottom water. Eutrophic fresh water is more usually associatedwith shallow lakes or slowmoving rivers over mineral-rich sedimentsand in which there is little or no temperaturestratificationof the water. The most nutrient-richwatersare thosein brackish coastal lagoons and estuariesat the interface between the land and the sea.

AQUATIC SUBSYSTEMS In most aquatic ecosystemsthree subsystems(Fig. 15.3) dependenton spatial variationsin the physicalhabitatand characterisedby distinctive communitiescan be recognised(Barnesand Mann 1980):

1. The openwater habitat (the marine pelagiczone); 2. The substratumor benthic habitat; 3. The fringe or edgehabitatof coats,water basinsor channels. The relative importanceof these subsystemsvaries from one type of aquatic ecosystemto another. PELAGIC HABITAT

The pelagic habitat is dominatedby two types of organismsphysically capableof living in water. One is the plankton ('drifting') organismswith no or limited meansof locomotion and whosemovementsare dependenton thoseof the water they inhabit. The other is the neckton,free-swimmingorganisms,fish and marine and freshwatermammals.There are, in addition, birds for whom the pelagic fish are an important if not the only food source. The plankton vary in size and function as indicatedin Table 15.2. Someare primary producers(algae);someare consumers(bacteriaand animal plankton); someare decomposers(bacteria).

Plant plankton About 90 per cent of primary production in fresh or marine pelagic habitats is undertakenby plant plankton (phytoplankton).The most abundantrepresentatives in the seaare minute algae,one-celleddiatoms,dinoflagellatesand coccoliths; in fresh waters diatoms, dinoflagellates,desmidsand blue-greenprokaryotes.The latter, either unicellular or filamentous,are nitrogen-fixing. Minute size providing a large area:volume ratio and a slow sinking rate facilitates suspensionand absorptionof mineral nutrients from the water. The efficiency of this form may

277

278 50 - 200 g/m 2

less than 50 g/m2

Fig 15.3. Net primary productivity patternof the oceans(from Bunt 1975)

over 200 g/m2

Ecosystems

Aquatic ecosystems Table15.2 Sizegradingof planktonicorganisms(from Boney1989) Maximumdimension(jJ,m)

Plankton'category'

Lessthan2 2-20 20-200 200-2000 More than2000

Picoplankton(algae,bacteria) Manoplankton(animalsandalgae) Microplankton(animalsandalgae) Macroplankton(animalsandalgae) Megaplankton(animals)

well explain the overwhelmingdominanceof plant life by the phytoplanktonin aquatic ecosystems.They can occur in such densities as to form surface concentrationsor 'blooms' which colour the water. The 'red-tides' in coastal waters, the appearanceof which occasionallyhit the headlines,are causedby population explosions of dinoflagellates, usually of toxin-producing species (Boney 1989). Plant plankton production varies both temporally and spatially - the most important factors limiting productivity are light intensity, the nutrient statusand the temperatureof the water. Moss (1980) distinguishesbetweenlight as the factor limiting the rate (of gross photosynthesis)and nutrients limiting the yield (net photosynthesis). Rate of gross photosynthesisdeclines with depth from a maximum near the water surface to the compensationpoint dependenton the initial light intensity and the turbidity of the water mass. In very bright sunlight surface inhibition may occur possibly as a result of the rapid absorption of ultraviolet light (Moss 1980). Rate of growth is also affected by temperature. Somephytoplanktonhave minimum requirementsbelow 0 °C; some,such as live on mud-flatsin tropical regions,can toleratetemperaturesof 30°C.In temperate The and cool latitudes seasonalrates of growth are partly temperature-related. correlation, however, is not perfect. On the one hand, seasonalphytoplankton 'blooms' in cool and cold water exceedthe maximum growth in warm tropical water. On the other hand, marked seasonalvariation in growth occurs in both Arctic and tropical water where the annualrange of temperaturein the euphotic zone is relatively low. Explosive growth has beenobservedin Lake Baikal (former USSR) before the ice breaks(Boney 1989). The most important factor limiting phytoplankton productivity is the availability of nutrients - particularly of nitrogen, phosphorusand iron (in that order) in seawater and of theseplus manganesein freshwaterlakes.The nutrient statusof water in the euphotic zone is dependenton the rate and efficiency with which the nutrients lost from this zone are replenished.In shore, fresh and salt water tends to be nutrient richer than offshore water becauseof the proximity to inputs by water flow from the adjacentland surface.Nutrient statusis also higher in thosepelagicareaswhereminerals(eitherfrom terrestrialinput or from organic decomposition)which have becomeincorporatedin bottom sedimentsare most rapidly recycled,i.e. returnedfor use by phytoplanktonin the euphoticzone.This is most effective where water mixing, as a result of turbulenceextendsfrom the surfaceto the baseof the water mass.Such is the casein most rivers, in shallow

279

Ecosystems

lakes and in epicontinentalseasoverlying extensiveareasof continentalshelf. It is also characteristicof areaswhere cold nutrient-rich water wells up from ocean deeps.In the oceansthis occursnearthe equator,wherewarmer,lessdensewater diverges (floats) north and southwards, around the margin of the Antarctic continentwhere surfacewater drifts north-eastwardsunder the influence of the strongwest-wind drift (Le. trade winds) and particularly along the westerncoasts of North and South America, Africa and to a lesser extent Australia (see Fig. 15.1). In theseareasthe continentalshelfis comparativelynarrow and deepwater lies relatively close in shore and surface water carried off shore by prevailing winds is replacedby upwelling of cold nutrient-richbottomwater. As indicatedin Fig. 15.3 and Table 15.3 thesenutrient-richareashavethe highest annualprimary productivity. Table15.3 Net primaryproductionin variousmarinehabitats(from McCluskey1981)

% area Net primaryproduction kcal m- 2 year-I 1000 g C year-I

Open sea

Coastal zone

Upwelling regions

Estuan·es(and coral reefi)

90 50

9.4 100

01 300

0.5 1000

16.3

3.6

0.1

2.0

In addition to thesespatialvariationsin marine productivity, seasonalvariations as a result of thermalstratificationare characteristicof both marineand freshwater bodies. This inhibits vertical water mixing and nutrient cycling for longer or shorterperiodsof the year. In warm tropical waters stratification may be virtually constantthroughout the year, resulting in low nutrient status and low primary productivity despitehigh temperaturesand light intensities.Annual productivity is higher, but more seasonal,in cool temperateand cold latitudes (and in highaltitude lakes) (see Fig. 15.4). In the former, increasingly favourable light and temperatureconditions for growth in late spring!early summer give rise to an explosive growth (the spring bloom) of phytoplankton. This rapidly depletes nutrients in the euphotic zone. Recycling must await surfacecooling and windinduced turbulence in the winter. Disruption of the thermocline may be accompanied (when light and temperature conditions are favourable) by a secondary,smallerautumnbloom. In cold Arctic and Antarctic marinewater, high turbulence inhibits the development of a marked seasonalstratification and primary productivity is more clearly a function of favourable light/temperature conditions.Constantor seasonalstratificationis characteristicof most lakes apart from the very large water bodiessubjectedto high wave turbulence. Compared to terrestrial plants, phytoplankton productivity is comparatively low. That in the oceanscontributes35 per cent of the total primary biological productivity of the biosphere.The averageproductivity (1.0 g cm-2 day-I) of the oceansis only slightly greaterthan that of the world's deserts;that of shallow seas and deep lakes is higher (0.5-3.0g cm-2 day-I). While the productivity of

280

Aquatic ecosystems

Solar radiation reaching sea surface Concentration of nutrients in surface layers

Convectional

Standing stock of diatoms

Convectional mixing Convectional mixing Winter

Spring

Convectional mixing

Summer Seasonal Thermocline

Spring

Spring

Convectional mixing

Fig. 15.4 Generaliseddiagram illustratingseasonal variationsof selectedparametersin

surfacewaterof temperateseas(adaptedfrom Tait 1972)

eutrophiclakes (3.0-10.0g cm-2 day-I) is more comparableto that of terrestrial vegetation (Boney 1989) plants other than phytoplankton make a greater contributionthan in the more extensivemarine pelagiczone. The phytoplankton provides all but a small proportion of the primary productionin the oceans.In lakes the proportionatecontribution dependson the size, depth and the nature of the substratumand tends to be less in shallow sedimentarythan in deep rocky lake basins. The nature of the phytoplankton results in plant-food relationshipsdifferent from those in terrestrial ecosystems. First, becauseof their microscopicsize, direct consumption(i.e. grazingby large herbivores) is limited. The majority of the aquatic herbivores are among the smallest of aquatic animals, of a size capable of capturing and ingesting phytoplanktonefficiently. For this reason,food chainstend to be longer in aquatic habitatsthan on land and the standingcrop (or phytobiomass)representsa smaller proportion of the primary plant productivity than in the case of land animals. Second,a higher proportion of the phytoplanktonproduction than of terrestrial plant production is consumedby herbivores; the grazing food chain is more important than the detrital chain and, at anyone moment, the animal biomass exceedsthat of the plant biomass.This is becauseconsumptiontendsto keeppace with production.The life spanof the phytoplanktonwhoseindividuals cells may, given favourable conditions, divide every 10-36 hours is very much shorterthan most of the organismswhich depend directly or indirectly on it. However, the percentageof the phytoplankton production 'grazed' varies with a number of factors of which the nutrient status of the water is particularly important. In eutrophic water where productivity is relatively high a large amount may be left

281

Ecosystems

ungrazed,eventuallyto be decomposedby the bacterioplanktonwhich inhabit the euphoticand disphotic zone. Zooplankton The conversion of the phytoplankton into food particles of a size that can be efficiently usedby larger aquatic animals is effectedby the zooplankton- a group that cannotbe comparedwith any similar group of terrestrial animals. They are definedas animalsincapableof maintainingtheir position againstwater movement (Barnesand Mann 1980). As such, they comprisean extremelydiverse group of organismsvarying in size, feedingcharacteristics,and life cycles (seeTable 15.4). Table15.4 Salinitymagnitudeof aquaticproductivityin differenthabitats(from datain Whittaker 1975)

Habitats

Salinity 8%0NaCI

Aquatic Saltwater 33-37(35) Openocean Upwelling areas Continentalshelf Algaebeds/reefs 5-35 Estuaries* Freshwater