Practical Geophysical Technology and Application for Lithological Reservoirs (SpringerBriefs in Petroleum Geoscience & Engineering) 9811641994, 9789811641992

Citation preview

SPRINGER BRIEFS IN PETROLEUM GEOSCIENCE & ENGINEERING

Huandi Wang Ming Li Yadong Wu Jianrong Gao

Practical Geophysical Technology and Application for Lithological Reservoirs 123

SpringerBriefs in Petroleum Geoscience & Engineering Series Editors Jebraeel Gholinezhad, School of Engineering, University of Portsmouth, Portsmouth, UK Mark Bentley, AGR TRACS International Ltd, Aberdeen, UK Lateef Akanji, Petroleum Engineering, University of Aberdeen, Aberdeen, UK Khalik Mohamad Sabil, School of Energy, Geoscience, Infrastructure and Society, Heriot-Watt University, Edinburgh, UK Susan Agar, Oil & Energy, Aramco Research Center, Houston, USA Kenichi Soga, Department of Civil and Environmental Engineering, University of California, Berkeley, USA A. A. Sulaimon, Department of Petroleum Engineering, Universiti Teknologi PETRONAS, Seri Iskandar, Malaysia

The SpringerBriefs series in Petroleum Geoscience & Engineering promotes and expedites the dissemination of substantive new research results, state-of-the-art subject reviews and tutorial overviews in the field of petroleum exploration, petroleum engineering and production technology. The subject focus is on upstream exploration and production, subsurface geoscience and engineering. These concise summaries (50-125 pages) will include cutting-edge research, analytical methods, advanced modelling techniques and practical applications. Coverage will extend to all theoretical and applied aspects of the field, including traditional drilling, shale-gas fracking, deepwater sedimentology, seismic exploration, pore-flow modelling and petroleum economics. Topics include but are not limited to: • • • • • • • • • • • • • • • • • • •

Petroleum Geology & Geophysics Exploration: Conventional and Unconventional Seismic Interpretation Formation Evaluation (well logging) Drilling and Completion Hydraulic Fracturing Geomechanics Reservoir Simulation and Modelling Flow in Porous Media: from nano- to field-scale Reservoir Engineering Production Engineering Well Engineering; Design, Decommissioning and Abandonment Petroleum Systems; Instrumentation and Control Flow Assurance, Mineral Scale & Hydrates Reservoir and Well Intervention Reservoir Stimulation Oilfield Chemistry Risk and Uncertainty Petroleum Economics and Energy Policy

Contributions to the series can be made by submitting a proposal to the responsible Springer contact, Anthony Doyle at [email protected].

More information about this series at http://www.springer.com/series/15391

Huandi Wang · Ming Li · Yadong Wu · Jianrong Gao

Practical Geophysical Technology and Application for Lithological Reservoirs

Huandi Wang Petroleum Industry Press Beijing, China Yadong Wu China National Oil and Gas Exploration and Development Company Ltd. Beijing, China

Ming Li The Research Institute of Petroleum Exploration and Development, PetroChina Beijing, China Jianrong Gao The Research Institute of Petroleum Exploration and Development, PetroChina Beijing, China

ISSN 2509-3126 ISSN 2509-3134 (electronic) SpringerBriefs in Petroleum Geoscience & Engineering ISBN 978-981-16-4199-2 ISBN 978-981-16-4197-8 (eBook) https://doi.org/10.1007/978-981-16-4197-8 © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, 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. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Preface

The discovery of lithological reservoirs accounts for an increasing proportion and a large proportion of proven reserves submitted each year are from lithological reservoirs. After many years of exploration, petroleum geologists have revealed the diversity and complexity of continental basin deposits. This kind of diversity comes from the multi-stage tectonic activity of basin-forming and the complexity of basin reconstruction, which leads to the formation of several types of basins, such as depression, fault depression, and foreland, has been formed. Since Mesozoic, the continental basin has experienced a variety of climatic conditions, including humid climate, arid climate, marshy environment, offshore environment, and saline environment. The superposition of these factors makes the lithological types of continental basins in China rich and varied, and the lithological traps are complicated and changeable. The terrestrial sedimentary reservoir model and theory of structural traps have been difficult to meet the needs of guiding lithological reservoir exploration. Compared with conventional structural reservoirs, lithological reservoirs have more concealment, more complex accumulation rules, more difficult to explore, and higher requirements for exploration technology. The purpose of compiling this book is to develop and innovate the theory of characteristic lithological reservoir, to form the corresponding prediction technology and exploration method, and to obtain the results of universal guiding significance for the exploration of the entire lithological reservoir. In the field of lithological reservoirs, there are still a series of problems, in theory, technology, and exploration. In theory, there is no systematic understanding of sandstone control factors, reservoir control factors, enrichment conditions, and reservoirforming laws. In terms of technology, the description of the sandstone body and reservoir prediction, oil and gas layer protection, low permeability fracturing, and high yield are not fully matched. On the exploration, lithological reservoir exploration procedure, especially the determination of the secondary exploration zones and industrialization mapping requirements is not clear on the field research and application of the east of the basin in China, especially in the exploration of the Songliao basin is more. A complete set of techniques has been established from true 3D structure and sequence interpretation to visual good placement design. Compared with the basins in western China, due to the complexity of basin structure and the diversity of reservoir and reservoir types, the supporting methods and technologies for v

vi

Preface

lithological reservoirs have not been established yet. Based on the successful exploration experience of lithological reservoirs in eastern China, this book summarizes the relevant techniques to promote the exploration of lithological reservoirs. After many years of exploration of lithological reservoirs, under the guidance of the analysis of accumulation controlling factors and distribution law of lithological reservoirs, this book has summarized a set of exploration methods for lithological reservoirs based on geological analysis by using geophysical techniques, which have been applied and popularized in practice and achieved good results. With the continuous improvement of the exploration degree of petroliferous basins in China, it has become an important research direction to strengthen the application of sequence stratigraphy techniques and methods and to study the formation conditions and distribution laws of lithological reservoirs. The establishment of a set of relatively perfect theory and matching technology has a very broad application prospect for promoting the exploration of lithological reservoirs. Studies at home and abroad have proved that sequence stratigraphy is an effective method to search for lithological reservoirs, and it is an important research direction to combine sequence stratigraphy theory with modern seismic interpretation techniques, especially with various seismic reservoirs prediction techniques. The preface, Chap. 1 and 2 were written by Huandi Wang; Chap. 3 and 4 by Huandi Wang, Ming Li, Yadong Wu; Some of the illustrations are provided by Jianrong Gao. The whole book is modified and revised by Huandi Wang. We take this opportunity to thank many people who have helped, in different ways, in the preparation of this book. In particular, We would also like to express our appreciation to Lisa Fan, senior editor of Springer International Publishing Company for her patience and understanding. Due to the proficiency limitation of the author, we expect the kind comments and opinions from readers regarding any mistakes or incorrectness in this book will be corrected in our next edition. Thank you. Beijing, China

Huandi Wang Ming Li Yadong Wu Jianrong Gao

Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Study on the Formation of Lithological Reservoirs . . . . . . . . . . . . . . 1.2 The Core Technology of Lithological Reservoir Exploration . . . . . . 1.3 Reservoir Prediction Technology Based on the Geological Concept and Quantitative Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 6 9 10

2 Geophysical Exploration Methods for Lithological Reservoirs . . . . . . 2.1 Method Thinking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 High-Resolution Sequence Stratigraphy . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Sedimentary Background Analysis . . . . . . . . . . . . . . . . . . . . . 2.2.2 Contrast of Sequence Division . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3 Sequence Interface Tracking Closure . . . . . . . . . . . . . . . . . . . 2.2.4 Identification of Seismic Sequence Interface . . . . . . . . . . . . . 2.2.5 Identification of Drilling/Logging Sequence Interface . . . . . 2.3 Sequence Constrained Reservoir Prediction Techniques . . . . . . . . . . 2.3.1 Reservoir Characteristic Curve Reconstruction . . . . . . . . . . . 2.3.2 Seismic Attributes Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3 Interpretation of Inversion Data . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Cognition and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11 11 13 14 15 16 16 17 17 18 21 24 26 26

3 Lithological Reservoir Exploration Technology . . . . . . . . . . . . . . . . . . . . 3.1 Research of Sequence Stratigraphy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1 Sequence Level and Its Geological Significance . . . . . . . . . . 3.1.2 Procedures for the Study of Sequence Stratigraphy . . . . . . . . 3.2 Seismic Attribute Analysis Technology . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1 Seismic Attribute Classification . . . . . . . . . . . . . . . . . . . . . . . . 3.2.2 Seismic Attribute Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Coherent Volume Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.1 Coherent Volume Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.2 Fundamental Principles of Coherence Computation . . . . . . . 3.3.3 Technical Processes and Procedures . . . . . . . . . . . . . . . . . . . .

29 29 30 30 36 37 38 49 49 51 52 vii

viii

Contents

3.4 3D Visualization and Virtual Reality Technology . . . . . . . . . . . . . . . . 3.4.1 Concepts and Principles of 3D Visualization . . . . . . . . . . . . . 3.4.2 Basic Visualization Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.3 Visualization of Full 3D Interpretation Technology . . . . . . . . 3.4.4 Virtual Reality Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Seismic Inversion Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5.1 Concepts and Classification of Seismic Inversion . . . . . . . . . 3.5.2 Basic Principles and Application Conditions of Seismic Inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5.3 Comparison of Different Seismic Inversion Methods . . . . . . 3.5.4 Reconstruction Method of Reservoir Characteristics . . . . . . 3.5.5 Seismic Inversion and Sequence Stratigraphy . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

56 57 58 59 62 65 66 70 77 80 87 89

4 Realization and Application of Geophysical Technologies for Lithological Reservoirs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.1 Realization of Geophysical Technologies for Lithological Reservoirs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.1.1 High-Resolution Sequence Stratigraphy and Prediction and Evaluation of Lithological Traps . . . . . . . 93 4.1.2 Seismic Imaging and Identification of Concealed Geological Bodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 4.1.3 Prediction Technology of Reservoir Pore Growing Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 4.1.4 Main Understanding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 4.2 Application of Geophysical Technologies for Lithological Reservoirs in Yingtai Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 4.2.1 Analysis of Geological Background in the Study Area . . . . . 96 4.2.2 Establishment of Stratigraphic Sequence Framework . . . . . . 97 4.2.3 Division of Stratigraphic Sequence . . . . . . . . . . . . . . . . . . . . . 98 4.2.4 Study on Sedimentary System . . . . . . . . . . . . . . . . . . . . . . . . . 98 4.2.5 Reservoir Inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 4.2.6 Oil and Gas Distribution and Prediction of Favorable Facies Zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 4.3 Cognition and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Chapter 1

Introduction

Concerning concepts of the lithological reservoir has put forward very early, and its connotation, characteristics, and classification are replenished and perfected constantly, but in the past have found lithological reservoir, with a certain chanciness, mostly tend to be following the find structural reservoirs exploration ideas and methods, and found nothing to do with structure or on the background of the tectonic lithological reservoir. In recent years, with the increasing number and scale of lithological reservoirs discovered at home and abroad, explorers have gradually differentiated them from structural reservoirs in practice, developed related theories, and explored methods and techniques suitable for lithological reservoir exploration.

1.1 Study on the Formation of Lithological Reservoirs Since the 1960s, some countries in North America, Western Europe, and other countries have been forced to look for oil in hidden traps due to the sharp decline in the oil reserve-production ratio. Therefore, to find and capture the hidden traps, which are mainly lithological and stratigraphic, becomes the main target to explore oil and gas potential in exploration mature basins. Lithological reservoirs were first classified as obscure reservoirs, a relatively fuzzy concept. Subtle reservoirs were first proposed by Karl in 1880. Wilson proposed in 1934 that non-structural traps are “reservoirs closed by changes in porosity in rock formations”. Lai put forward the concept of stratigraphic trap and published a paper entitled “stratigraphic oilfield” in 1936. In 1972 Halbert called the reservoirs formed by stratigraphic traps, unconformable traps, and paleotopographic traps as subtle reservoirs. For nearly 30 years, with the development of world petroleum exploration technology and the deepening of the scientific research work, the hydrocarbon reservoirs are defined to further expand to: under the condition of existing exploration method and technology level, more difficult to identify and describe the reservoir © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 H. Wang et al., Practical Geophysical Technology and Application for Lithological Reservoirs, SpringerBriefs in Petroleum Geoscience & Engineering, https://doi.org/10.1007/978-981-16-4197-8_1

1

2

1 Introduction

type, it covers the stratigraphic and lithological reservoirs and complex fault-block reservoirs and low amplitude gentle anticline oil–gas reservoir type, etc. (Li et al. 2014). Because of the geological background, trap mechanism, exploration theory and technique of the lithological reservoir, and the huge reserves and broad resource prospect, it is necessary to define and study the lithological reservoir conceptually. At present, lithological traps are generally defined as follows: lithological traps are traps that lack four-azimuth closure and cannot be found by the exploration strategy of looking for structural traps. If they are related to the structure, they develop in an unexpected place (such as the position of the lower flank of the structure). It is a trap that cannot be defined solely by structural closure, including a single lithological trap, stratigraphic trap, and lithological composite trap with a structural background. The upper of Fig. 1.1 shows lithological composite trap with the structural-lithological reservoir as background. This kind of trap is the majority, the structural background is favorable to the enrichment and preservation of oil and gas. The underpart of Fig. 1.1 shows a single lithological trap, the developed sandstone body is surrounded by mudstone strata, forming a relatively independent and single lithological reservoir. In the current lithological trap classification, the main mechanisms of trap formation can be divided into the following categories: lateral facies change trap, lateral sedimentary tip out the trap, overlying/concealed outcrop trap, channel/gully filling trap, diagenetic trap, fissure trap, and hydrodynamic trap. Specifically, it can be divided into 18 types: lateral pinch-out, lateral facies change, channel filling, regional sub-outcrop, valleys, structure filling flanking the overlap on the unconformity, cementation, regional unconformity of overlap, cracks on the deep basin gas, edge cutting, palate structure sub-outcrop, dolomitization/corrosion, coal bed methane adsorption, clastic rocks configuration, incised valley filling, hydrodynamic, asphalt sealing, etc. (Wang 2005). Among these trap types, lateral sedimentary pinch-out, lateral facies change, channel filling, and regional concealed outcrop are the most common, accounting for about 57% of the total. However, the number of traps with high frequency is not necessarily the most abundant. The types of traps with large reserves of individual traps include the overlapped traps on structural flank unconformity, the overlapped

Fig. 1.1 Lithological reservoir

1.1 Study on the Formation of Lithological Reservoirs

3

traps on regional unconformity, the bituminous plugging traps, and the deep basin gas traps. The existing data show that most of the formation traps have a tectonic dip of less than 2, which can account for 60% or even more of the total. At the same time, due to the large formation trap area but small effective formation thickness, coupled with the lack of natural oil–gas driving energy, more than 70% of reservoirs in the formation lithological trap have been subjected to secondary oil recovery or enhanced oil recovery technology. According to the reservoir characteristics of lithological traps, sandstone reservoirs account for 63.44% and carbonate reservoirs account for 26.25% of the 320 traps in the United States. According to the statistics of 1177 traps in the former Soviet Union, sandstone reservoirs account for 85% and carbonate reservoirs account for 15%. Besides, according to the statistics of 174 known lithological reservoirs in the production age, they are distributed from Ordovician to Tertiary, but most lithological reservoirs are distributed in Cretaceous, Tertiary, Carboniferous, and Permian, accounting for 80% of the total number of lithological reservoirs in these four eras. According to the basin background statistics, the lithological traps found in foreland basins, craton basins, passive continental margin basins, and rift basins account for nearly 85% of the total, among which foreland basins are the most developed, accounting for 55% (Wang 2005). In the late 1970s, China began to pay more attention to subtle reservoirs, and relevant papers and monographs have been published. In these works, based on the foreign counterparts’ understanding of concealed reservoirs, domestic scholars have systematically defined and described the concept, classification, characteristics, and distribution rules of concealed traps according to the characteristics of domestic reservoirs. Based on the distribution and exploration situation of oil and gas in China, most of the examples and research work of hidden traps are mainly concentrated in the eastern oil-bearing basins. Before the 1990s, the exploration of concealed oil and gas reservoirs in China has not received much attention and attention, and structural oil and gas reservoirs are still the focus of exploration. From the middle and late 1990s, the exploration of concealed oil and gas reservoirs in China started from the old oil areas in the east to the new areas in the west. In particular, CNPC and Sinopec, two major oil companies, organized a large number of competent scientific researchers to participate in the exploration. In petroleum exploration meetings in recent years, relevant experts have put forward several suggestions to develop the concealed trap exploration technology mainly based on lithological traps and strengthen the exploration of lithological reservoirs, which further clarified the exploration potential and future exploration technology requirements of lithological traps. Many scholars have put forward their views on lithological traps and some related concepts. See Jianyi (1984) in the “east China continental basin formationlithological trap hydrocarbon accumulation (with) distribution and the exploration of the research program” the article expounds on the formation of lithological trap and concealment of the formation of the geological background and distribution, and in the non-structural reservoirs (1986), the formation of lithological trap difference

4

1 Introduction

in the structural trap are collectively referred to as nonstructural traps. In practical exploration and production, terms such as subtle reservoirs, non-structural reservoirs, and stratigraphic lithological reservoirs are often used in a mixed manner, sometimes referring to all non-tectonic traps, sometimes including structural traps that are difficult to be identified by geophysical techniques at the time, with mutual confusion and ambiguity. In 2003, Chenzao Jia “China lithological reservoir resource potential and exploration technology”, to define the research object, to avoid confusion and in line with international standards, suggested that do not use the word “subtle trap”, clearly put forward using the concept of the lithological reservoir, the lithological trap the specification in the future is to use the term is necessary. Because each author’s starting point is different, the meaning of these terms is overlapping or synonymous. Lithological trap refers to the trap formed by impermeable layer or lateral shielding under sedimentation or diagenesis, which changes the lithology or physical property of reservoir rock mass. In the past decades, the formation conditions and distribution rules of lithological reservoirs have always been the focus of domestic research, and domestic experts and scholars have expounded them from different perspectives and levels. In China’s oil and gas exploration practice, the understanding of lithology reservoirs is deepening and improving constantly, the definition of lithological reservoir range (connotation) is extended, the traditional stratigraphic traps are placed within the lithological traps. This is because the meaning of the stratigraphic trap is relatively fuzzy. With the development of modern sedimentology and sequence stratigraphy theory, once known stratigraphic traps can be explained from the angle of sequence stratigraphy to the sequence of the interface. Therefore, in general, the practice of oil and gas exploration shows that there are the following types of lithological traps in Continental basins in China: (1) the lithological traps formed by interface control, ➀ the lithological traps below the unconformity, ➁ the stratigraphic overlying lithological traps, ➂ the lithological traps defined by sequence interface; (2) Lithological traps formed by sedimentary processes (Fig. 1.2), ➀ various types of underwater fans (submarine fans, pelvic floor fans, slope fans, etc.), ➁ deep valleys, ➂ lenticular lithological traps (channel sandstone bodies), ➃ updip pinch-out lithological traps, ➄ reef bodies; (3) Lithological traps formed by diagenesis, ➀ large sand-gravel rock diagenesis traps, ➁ lateral closure of thin sandstone bodies, ➂ fractured lithological traps formed by fault transformation in tight layers, ➃ karst lithological traps, ➄ lithological traps formed by uneven dolomitization; (4) Unconventional lithological traps, ➀ igneous bodies, ➁ buried mountains, etc. (Wang 2005). Because the change of interface and lithology control the formation and distribution of lithological reservoirs, the trend of finding lithological reservoirs at home and abroad is to use sequence stratigraphy, especially high-resolution sequence stratigraphy, to study the interface at all levels to predict the distribution of favorable sandstone bodies. Secondly, the change of rock physical properties caused by diagenesis is also an important factor in the formation of lithological traps. For example, “desserts” in low-permeability reservoir sandstone bodies are the result of diagenetic differentiation.

1.1 Study on the Formation of Lithological Reservoirs

5

Fig. 1.2 Lithological traps associated with sedimentation

The lithological reservoir distribution prediction of overall foreign technology idea is: under the framework of sequence stratigraphy, the data integration of outcrop, seism, logging, core, fluid properties, fluid pressure, and hydrocarbon shows; the establishment of the fine geological model based on the recovery of high-resolution sequence stratigraphy framework, the effective geological interpretation to the highresolution seismic data, and the effective prediction to the lithological reservoir distribution. At present, the establishment of the fine geological model under a highresolution sequence stratigraphic framework, the development of diagenetic trap prediction technology, and the prediction of the high precision seismic reservoir are the hotspots of lithological reservoir research abroad (Wang 2008). Foreign high-resolution sequence stratigraphy is based on three dimensions: outcrop, core, logging, and high-resolution seismic section. By the fine stratigraphic division and correlation, the one-dimensional drilling information has turned the basis of 3D stratum relationship prediction. The genetic and stratigraphic correlation framework of regional oil fields and reservoir level reservoirs is established to evaluate and predict the distribution of reservoirs, cap layers, and source layers. As for the accuracy of sequence division, foreign countries seek to establish sequence lattices of more than 4 levels of sequence and to take the reservoir unit with thickness below 10 m as the target of lithological trap prediction. In the research method, the outcrop is combined with the downhole. The three-dimensional distribution and reservoir heterogeneity of reservoir groups and physical compartments of different levels, such as river channels, deltas, or turbidimetric fans, are depicted utilizing advanced technologies such as shallow drilling, ultra-shallow seism, permeability measurement, resistivity measurement, gamma measurement, ground-penetrating radar, and global positioning system. The base-level cycle analysis is used to establish a fine prototype geological model of outcrop-type simulants in underground oil production

6

1 Introduction

reservoirs. Adequate good control is available in the exploration mature area. A highresolution sequence stratigraphic framework can also be restored by using drilling and 3D seismic data, to establish fine geological models such as the 3D distribution model of a skeleton sandstone body and reservoir geological model. The development of lithological traps is largely controlled by diagenesis, and some traps are mainly controlled by diagenesis, which is called the diagenetic trap. Diagenesis includes all Physico-chemical processes from burial to metamorphism. It is strictly controlled by basin types, sandstone body types in different sedimentary environments, rock types in parent rock areas, burial history, and diagenetic environments (temperature field, stress field, fluid field). According to the current research and development trend at home and abroad, it is required that the diagenesis research breakthrough the current petrological category, that putting it in the geological background of the whole basin, studying the diagenetic evolution characteristics of clastic reservoirs in different types of basins, and studying the controlling effect of geothermal field, tectonic stress field and fluid field in basins on reservoirs.

1.2 The Core Technology of Lithological Reservoir Exploration The lithological reservoir exploration technology in the world can be generally divided into three stages. In the first stage, before the 1920s, the surface geological survey was the main method, and oil seedlings were the main clue to discover oil and gas reservoirs. In the second stage, from the 1930s to the 1970s, the reservoir’s discovery mainly relied on the geological interpretation of wellbore data and the review of old oilfields and Wells, and quite a few lithological reservoirs were discovered by chance. In the third stage, seismic technology has played a major role in lithological reservoir exploration since the 1980s. In general, due to the limitations of exploration techniques, the exploration of lithological reservoirs has been mainly based on geological evaluation and analysis for a long time and based on the idea of structural traps. Due to the concealment of lithological traps, the success rate of exploration is generally low. Due to the rapid development and wide application of the high resolution 3D seismic and sequence stratigraphy technologies, It is possible to directly identify lithological trap targets according to geophysical data before the first well is drilled so that large-scale exploration and deployment for lithological reservoirs can be carried out and the success rate of lithological reservoir exploration can be greatly improved. With the rapid development of seismic exploration technology, 3D seism has been widely used, and 3D pre-stack depth migration technology and 3D visual interpretation technology have been developed. The development of submarine cable technology enables the realization of offshore multi-wave seismic exploration. With the great progress of high-resolution seismic technology, reservoir description technology has become mature day by day. The four-dimensional seismic technology

1.2 The Core Technology of Lithological Reservoir Exploration

7

extends the seismic exploration technology to the stage of oil and gas development and becomes an important method to monitor the development of oil and gas. At the same time, the accuracy of seismic technology is increasing, which makes it an indispensable means and technical method in lithological trap identification. According to the statistics, the lithological traps discovered by seismic means and methods account for 33%, while in the sea, the proportion is as high as 90%. Various seismic technology as horizontal seismic section, the time difference velocity analysis, seismic attribute technology, seismic trace inversion, complex seismic trace analysis, seismic simulation, seismic lithology modeling, AVO technology, Crosswell seismic technology, full 3D seismic interpretation technology, 4D seismic technology, directly or indirectly, the recognition of lithological trap play an important role. For example, Widuri oilfield in Indonesia takes lateral sedimentary pinch-out as the main reservoir trap, and this trap is discovered by revealing the distribution and lateral pinch-out characteristics of delta plain channel with seismic amplitude attribute. Another example is the Everest oilfield in the UK, which was discovered by using AVO bright spot analysis technology. It is also a lateral sedimentary pinchout trap, and condensate gas was pinch-out by a turbidite fan at the bottom of the basin. Gas-filled reservoirs showed AVO polarity reversal and amplitude migration increased. For lithological traps without obvious abnormality on the seismic profile, it is an important link to accurately describe and characterize the reservoir shape, distribution range, and lateral contact relationship. In recent years, through the combination of many seismic techniques, the series of reservoir prediction techniques developed specifically meet the needs of accurate positioning and quantitative evaluation of lithological traps. There are two common and more effective techniques in the series of reservoir prediction techniques, one is the seismic multi-attribute analysis technique, the other is the reservoir seismic inversion technique. Seismic multi-attribute analysis technology is developing rapidly, and there are hundreds of attributes available for analysis, with various methods. However, according to different geological backgrounds and oil and gas characteristics, the selection of attribute types and methods has a particular emphasis, among which some methods such as coherent volume technology, visualization, and frequency division technology have obvious application effects. Seismic inversion usually falls into two categories: pre-stack and post-stack. In the past 20 years, great progress has been made in post-stack seismic inversion, and many mature technologies have been formed. According to the function of well logging data, it can be divided into four categories: direct seismic inversion, seismic inversion under well-logging control, joint good logging inversion, and well-logging interpolation extrapolation under seismic control. It can be divided into recursive inversion, logging constrained inversion, and multi-parameter lithological seismic inversion. The high precision seismic inversion technology can accurately transform the interfacial seismic profile to the strata seismic profile, thus realizing the accurate prediction and evaluation of lithological traps (Fig. 1.3).

8

1 Introduction

Fig. 1.3 Optimization of the reservoir inversion method. a Model constraint velocity inversion, the vertical and horizontal resolution is high and the good point resemblance is good; b seismic feature velocity inversion, the vertical and horizontal resolution is not as good as the model constraint velocity inversion (Wang 2005)

In today’s exploration, the prediction and description of strata lithological traps are not solved by a single method and technique, which requires the integration of multiple disciplines and methods and techniques. On the premise of fully understanding the concept and characteristics of stratigraphic traps, the establishment of a sequence stratigraphic model with correct stratigraphic correlation and facies distribution, and the application of high-precision reservoir prediction technology or other geophysical methods will open up a new dimension for the exploration of stratigraphic lithological reservoirs. CNPC has established a whole set of exploration methods and technologies in the process of lithological reservoir exploration, mainly including (1) high-resolution 3D seismic acquisition technology; (2) High-resolution sequence stratigraphic analysis technology; (3) Seismic reservoir prediction technology; (4) Comprehensive evaluation technology of lithological traps; (5) Special drilling, logging, fracturing technology, etc. Among them, seismic multi-attribute analysis technology and reservoir seismic inversion technology are the two core technologies in lithological reservoir exploration research.

1.3 Reservoir Prediction Technology Based on the Geological …

9

1.3 Reservoir Prediction Technology Based on the Geological Concept and Quantitative Model Because lithological traps involve many aspects such as lithological pinching and lithofacies change, reservoir prediction, reservoir morphology, and distribution description are particularly important. How to establish a close connection between geology and geophysics and how to integrate geological concepts and models into geophysical methods and techniques are fundamental to the accuracy of reservoir prediction and the accurate interpretation of results. In recent years, some domestic experts and scholars have made important breakthroughs in this field, the most important of which is the technology of reservoir characteristic reconstruction. This technique fully combines geology, logging, and seismic data, and makes the data body obtained after inversion have a higher resolution in the vertical direction, thus improving the accuracy of reservoir prediction. The steps of reservoir prediction based on the geological concept and quantitative model are as follows: (1) Establish geological concept model through sedimentary background analysis and sedimentary system research; (2) Reconstruct the reservoir characteristic curve to establish the connection between logging and seism; (3) Under the guidance of geological concept model, coherent, visual and frequency division interpretation are carried out through horizon calibration, and quantitative sequence grid model is established; (4) Sequence constraints, seismic attribute analysis, and seismic inversion; (5) Well logging oil and gas evaluation and hydrocarbon detection (Wang 2008). In recent years, with the deepening of exploration and development in the oil and gas field, the requirement of seismic reservoir prediction technology has become higher and higher, and the high-precision reservoir seismic inversion technology has emerged, especially the logging constraint inversion and reservoir characteristic reconstruction technology has gradually become mature. This is a kind of seismic inversion technology based on the model. Based on known logging data, drilling data, and geological law, the geological model of the underground reservoir is constructed, and then the seismic response of the model is obtained, and the seismic response of the reservoir model is compared with actual seismic data. According to the difference between the model and the actual seismic data, the model parameters are modified repeatedly until the model response best fits the observed seismic data. Finally, a reasonable geological model is derived from the given observation data, to predict and describe the horizontal and vertical distribution and physical property changes of underground reservoirs. From the perspective of three-dimensional space, the sequence constrained reservoir prediction technology, which organically combines the sequence stratigraphy theory with the seismic reservoir prediction technology, can greatly improve the quantitative degree of sequence stratigraphy research results, thus improving the prediction accuracy of lithological stratigraphic reservoirs, and developing and applying the quantitative sequence constrained reservoir prediction technology.

10

1 Introduction

References Li M et al (2014) Geophysical exploration technology: applications in lithological and stratigraphic reservoirs. (1st edn) Hardcover, Elsevier, 480 pages Wang H (2004) Present situation and the countermeasure analysis of subtle reservoir exploration. Oil Geophys Prospect 39(6):739–744. https://doi.org/10.13810/j.cnki.issn.1000-7210.2004.06.023 Wang H (2005) Geophysical exploration technology and application for lithological reservoirs: a case study of southern Songliao Basin. China University of Geosciences (Beijing) Wang H (2008) The sequence constraint prediction technology and application for lithological reservoir. China Pet Explor 13(4):36–42

Chapter 2

Geophysical Exploration Methods for Lithological Reservoirs

At present, the exploration target has become more and more complex, and the exploration is getting harder and harder. Therefore, it is necessary to study and develop different exploration techniques and countermeasures for different exploration targets according to their geological characteristics and exploration technical difficulties. For lithological reservoirs, the key technology of exploration lies in the imaging and identification of lithological traps and the prediction of heterogeneous reservoirs. By improving the resolution and accuracy of seismic imaging, developing and improving the high-resolution reservoir feature inversion prediction method, and carrying out key techniques such as seismic attribute analysis, a series of techniques for identifying lithological reservoirs and geophysical comprehensive description are established. Reservoir prediction technology is one of the core technologies of lithological reservoirs. It can be simply divided into seismic inversion technology and seismic attribute analysis technology, which can make a qualitative and quantitative prediction of lithology, thickness, physical property, and oil–gas content of reservoirs.

2.1 Method Thinking In the exploration of lithological reservoirs in some oil fields, the identification of sandstone bodies within sequence framework mainly adopts the seismic phase identification technology (i.e. the “phase plane” method) proposed by the seismic stratigraphy in the late 1970s, and the preparation of sequence unit sandstone body distribution maps is mainly qualitative or semi-quantitative. Although seismic reservoir prediction technology has been widely used, it does not use sequence stratigraphic framework constraint inversion and only reflects the distribution of multiple sets of composite sandstone bodies, and it is difficult to predict the distribution and contact relationship of a single sandstone body. Sequence analysis and reservoir inversion © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 H. Wang et al., Practical Geophysical Technology and Application for Lithological Reservoirs, SpringerBriefs in Petroleum Geoscience & Engineering, https://doi.org/10.1007/978-981-16-4197-8_2

11

12

2 Geophysical Exploration Methods for Lithological Reservoirs

are separated and not used organically together, which affects the accuracy of seismic reservoir prediction. Through research and practice, it is realized that under the constraints of highresolution sequence framework, the application of various reservoir prediction techniques and geological modeling to the same phase axis can improve the accuracy of reservoir prediction and help identify the pinch-out line of the sandstone layer and the distribution of isolated sandstone body. Based on sequence interface comprehensive tracking closure, the various reservoir inversion technique, seismic waveform classification, such as slicing technology, 3D visualization interpretation technology, and many other new methods of lithological reservoir exploration, the application of high-resolution seismic data, solve the problem of identification of thin reservoir, to improve the quantitative level of sequence analysis. In this section, an idea is put forward, that is, waveform classification seismic phase analysis and reservoir inversion, especially reconstruction inversion of the characteristic curve, are mainly used to solve the problems of plane phase identification and vertical resolution in the isochronous lattice with sequence constraints, which has a relatively obvious effect on the recognition of lithological reservoirs (Fig. 2.1).

Fig. 2.1 Sketch map showing the research route

2.2 High-Resolution Sequence Stratigraphy

13

2.2 High-Resolution Sequence Stratigraphy Sequence stratigraphic analysis technology plays an important role in oil and gas exploration and development. In the exploration of lithological stratigraphic traps, the isochronous framework of sequence stratigraphic and system tract of the whole basin should be established first, the spatiotemporal distribution and evolution characteristics of the sedimentary system should be clarified, and the general background of sequence stratigraphic of the whole basin should be clarified. According to the general rules of lithostratigraphic trap development, the favorable areas and strata are determined. Then it is necessary to establish the isochronous stratigraphic framework of the high-resolution sequence for the main target strata in the key areas, and to make clear the lithological distribution and stratigraphic overlay relationship of the high-frequency sequence, to establish an accurate stratigraphic deposition model in the whole region, lay a solid foundation for seismic inversion and reservoir prediction, and make the reservoir prediction results more accurate and reliable. The lithological traps are further identified by combining the spatial distribution law and configuration relation of the reservoir, cap, and lateral sealing layer. In both classical and high-resolution sequence stratigraphy, the foundation of sequence stratigraphic analysis is the identification mark of the key sequence interface. The main purpose of sequence stratigraphic analysis is to establish a sequence stratigraphic framework by identifying and tracking the key interface, namely sequence interface, primary flood surface, and maximum flood surface. The sedimentary system and facies map under the isochronous framework are compiled, and the sedimentary model is established to predict oil and gas reservoirs. A sequence is a set of genetically related strata bounded by unconformities or their corresponding conformities. The identification of sequence interface is based on seismic profile and well log as well as core and logging data (Wang 2008). Seismic prediction and sequence stratigraphy should be combined to give play to their respective advantages. The seismic prediction should be controlled by the highresolution sequence stratigraphic framework, to ensure that the prediction results can reflect the actual sedimentary structure. In recursion inversion and logging constrained inversion, the low-frequency model and the initial model are usually established under the control of a large set of seismic geological strata. These seismic geological strata are usually the result of seismic tectonic interpretation. Some seismic reflection interfaces with strong seismic reflection characteristics are usually selected in the stratigraphic selection. Some of these seismic structure interfaces are sedimentary sequence interfaces, while some are not consistent with sedimentary sequence interfaces. In this way, the model established under their control is not consistent with the actual sedimentary sequence, thus affecting the accuracy of reservoir seismic inversion and prediction (Fig. 2.2).

14

2 Geophysical Exploration Methods for Lithological Reservoirs

Fig. 2.2 Reservoir prediction under the constraints of high-resolution sequence framework (the good curve in the figure is a lithology indicator curve, with the high value to the right being sandstone and the low value to the left being mudstone

2.2.1 Sedimentary Background Analysis Whether it is regional sequence stratigraphy or high-resolution sequence stratigraphy of the target block, the sedimentary background of the target segment in the study area needs to be analyzed firstly. There are three main purposes of analysis: first, to enhance the pertinence of research ideas and methods; Second, to reveal the difference of sandstone body and sequence development under different geological backgrounds from the perspective of genesis; Thirdly, to predict the basic characteristics of sedimentary facies and environment. The main content of the analysis includes three aspects: structure, climate, and provenance. The analysis of tectonic background is mainly to understand the plate tectonic background, basin type, tectonic pattern, and tectonic evolution history of the study area. The analysis of climatic background is mainly to understand the paleoclimatic conditions during the sedimentary period of the target strata in the study area. Paleoclimatology is an important factor that controls the weathering pattern of parent rock, transport medium, sedimentary medium, sandstone body type, and source rock scale. The analysis of provenance conditions is mainly to understand the spatial distribution relationship between sedimentary areas and provenance areas, the area of provenance areas, the tectonic activity of provenance areas, and the parent rock properties of provenance areas. The sedimentary characteristics of the same period are different from those of the provenance.

2.2 High-Resolution Sequence Stratigraphy

15

2.2.2 Contrast of Sequence Division Sequence division and contrast is the basic research content of sequence stratigraphy (Fig. 2.3). At present, there are two main problems in sequence stratigraphic division and correlation in China: (1) sequence classification disorder, different authors’ sequence division results in the same research area are not consistent, and the same author’s sequence division schemes in different areas in the same basin are not consistent; Drilling and seismic sequence division are not uniform. The basic procedures for the establishment of sequence stratigraphic framework are as follows: first, the seismic profile backbone grid-connected wells should be established, and the density of the grid should be able to control the distribution of the sedimentary system in the study area; The second is the longitudinal sequence division of typical wells. Wells located in the transition facies belt are generally selected. Sequence, system tract, and subsequence boundary are identified according to the transition or abrupt surface of sequence evolution of longitudinal sedimentary facies, and the longitudinal sequence division scheme is determined; The third is the sequence division, and comparison is completed in the same process. The sedimentary sequence comparison method, the stratigraphic leveling comparison method, and the ground layer stacking comparison method can be used to make the sequence division results of the joint well profile consistent. Fourth, sequence boundary unconformity surface or the greatest lake invasion surface can be identified according to the seismic unconformity feature or the macroscopic stratigraphic structure transformation surface. The fifth is to unify drilling and seismic sequence division by well-seismic interaction comparison.

Fig. 2.3 Seismic sequence division and contrast

16

2 Geophysical Exploration Methods for Lithological Reservoirs

2.2.3 Sequence Interface Tracking Closure There is no fundamental difference between the tracing closure method of sequence interface and the common seismic tectonic interpretation, and the technique of seismic stratigraphic interpretation can be used. Unlike the conventional seismic horizon interpretation, which focuses on tracing and interpreting seismic markers, the study of regional sequence stratigraphy emphasizes the identification of unconformity and the greatest lake intrusion, because these two interfaces are most closely related to lithological stratigraphic traps. The unconformities control the stratigraphic overlapping traps and unconformable shielding traps while the maximum flood surface is associated with various types of isolated sandstone body lithological traps. The high-resolution sequence comparison emphasizes the recognition of the most isochronous lake invasion surface. It has been proved by years of practice that in lithological stratigraphic exploration, high-resolution sequence stratigraphy has the following functions: (1) To establish a high-resolution sequence stratigraphic framework to make clear reservoir distribution and oil layer connectivity; (2) To analyze the formation pinch-out and fault zone, and to find out the distribution of different types of oil and gas reservoirs; (3) To provide isochronic formation interface with well constraints to improve the accuracy of reservoir prediction; (4) To reduce the layer tracking time-lapse phenomenon, and to improve the structure of the map precision; (5) To clarify the reservoir-cap assemblage further, and to define the exploration target layer; (6) To analyze the morphology, boundary, and reservoir forming conditions of lithological stratigraphic traps.

2.2.4 Identification of Seismic Sequence Interface The seismic sequence boundary is the response of unconformities and comparable conformities on seismic profiles. The types of reflection terminations indicating the bottom and top interfaces of the sequence on the seismic profile include onlap, downlap, toplap, and truncation. In most cases, the onlap is a good sign for the increase of accommodating space and the rise of the water surface. Toplap is a sign that the water surface is relatively static; The truncation is a good indicator of the decrease of accommodating space and the relative decrease of water surface. At the same time, the migration to the center of the basin or the sudden downward migration of the shore onlap is the most reliable indicator of water surface decline.

2.2 High-Resolution Sequence Stratigraphy

17

2.2.5 Identification of Drilling/Logging Sequence Interface In terms of comprehensive logging data, sequence interface is mainly identified according to lithological change, lithofacies association change, mudstone color, and paleontological characteristics. Due to the difference in the location of the basin where the well is drilled and the sequence development, the sequence interface also shows different characteristics and should be recognized with different markers. In the well-developed parts of the system tract, such as the lacustrine wave base area and the slop-deep lake area, the sequence interface of them is progradation type (lacustrine base area or slope zone) or aggradation type (deep lake area) quasi sequence group, and the interface is progradation or aggradation type quasi sequence group. In some parts lacking the lower system tract, such as the coastal zone, their sequences are progradational quasi-sequence groups under the interface and regressive quasisequence groups on the interface. If the strata have been subjected to obvious denudation, partial or complete denudation of the system tract of the preceding sequence may lead to the complex longitudinal superposition of the quasi-sequences. In continental sequence stratigraphic analysis, the instability of stratigraphic development makes it more difficult to classify the stratigraphic sequence. To classify the stratigraphic sequence as reasonably and reliably as possible, it is necessary to combine seismic data with drilling data, well logging data, and all other available test and analysis data for comprehensive study.

2.3 Sequence Constrained Reservoir Prediction Techniques Reservoir prediction technology can be simply divided into seismic inversion technology and seismic attribute analysis technology. Seismic inversion is the core technology of quantitative reservoir prediction, which mainly solves the problem of vertical reservoir resolution and prediction. Seismic attribute analysis techniques include conventional seismic attribute extraction and analysis (such as amplitude, frequency, phase, etc.), waveform classification, 3D visualization, seismic coherence cube, pattern recognition, and other generalized seismic attribute analysis techniques. Seismic attribute analysis technology mainly solves the problem of reservoir distribution range and boundary recognition laterally, and its prediction is generally based on qualitative analysis (Wang 2008). With the rapid development of complex reservoir prediction technology based on high precision seism, it plays an increasingly important role in lithological reservoir seismic exploration. Due to the complexity of the geological conditions of lithological reservoirs, the diversity of trap conditions, the mutability of lithological space and the heterogeneity within the reservoir, the seismic data and geological, well drilling, well logging, and other multidisciplinary data must be used for seismic exploration of lithological reservoirs. Constrained by a high-resolution sequence grid, many new technologies, such as reservoir characteristic curve reconstruction technology,

18

2 Geophysical Exploration Methods for Lithological Reservoirs

Fig. 2.4 Seismic reservoir prediction constrained by sequence interface

inclination detection technology, seismic waveform classification technology, and 3D visual interpretation technology, are used to solve the identification problem of thin reservoirs by using high-resolution seismic data, to improve the quantitative level of sequence analysis. This provides a powerful research tool for the exploration of lithological reservoirs, and the sequence constrained reservoir prediction technology expands the horizon of the interpreters to a greater extent, greatly improves the understanding degree of underground geology of the interpreters, and thus effectively improves the exploration benefits of lithological reservoirs (Fig. 2.4).

2.3.1 Reservoir Characteristic Curve Reconstruction In recent years, the methods of seismic inversion mainly include recursive inversion, logging constrained inversion, and multi-parameter lithological seismic inversion. These inversion methods have different characteristics and applicability, so the most suitable inversion method should be selected according to the characteristics of the study area and the geological problems to be solved. With the increasing degree of oil and gas exploration and development, more and more attention has been paid to the study and evaluation of unconventional reservoirs (such as volcanic rocks, metamorphic rocks, weathered crust, mudstone, conglomerate, and other reservoirs) and thin interbeds of sandstone and mudstone, which are also the main reservoirs to be replaced by reserves in the future. These reservoirs are characterized by complex formation conditions, multiple controlling factors of reservoir performance, diversified reservoir space types, and strong reservoir heterogeneity, etc., so it is difficult for acoustic moveout to meet the high accuracy requirement of reservoir prediction. Therefore, the conventional acoustic impedance inversion technique which is constrained by the acoustic moveout logging curve is not very accurate to the unconventional reservoir prediction, and the traditional reservoir prediction technique faces serious challenges.

2.3 Sequence Constrained Reservoir Prediction Techniques

19

Fig. 2.5 Analysis and processing of reservoir characteristic curve

In the study of long-term reservoir prediction, a set of methods to improve the vertical resolution of reservoir prediction by reconstructing the reservoir characteristic curve is explored through analyzing the mutual correspondence between different logging responses and lithology and seismic information (Fig. 2.5). Reservoir characteristic curve reconstruction is based on a comprehensive study of geology, well logging, and seism aiming at specific geological problems and inversion targets, based on petrophysics, it is optimized from a variety of well logging curves and reconstructed a curve that can reflect reservoir characteristics. Theoretically, spontaneous potential, natural gamma ray, compensated neutron, density, resistivity, and other logging curves in the conventional logging series can be used to identify reservoirs, establish a good correlation with acoustic moveout, and convert them into quasi-acoustic moveout curve through mathematical statistics, to realize reservoir characteristic curve reconstruction (Wang 2005). Vertical resolution is the key to reservoir prediction in lithological reservoir exploration. From the current situation of lithological exploration and reservoir prediction technology, acoustic logging curves are often used to constrain reservoirs in conventional good constraint inversion. When the velocity difference between the reservoir and surrounding rock is not obvious, that is, the vertical resolution of the original logging curve is not high, which will inevitably reduce the accuracy of the inversion results. In practice, it is found that acoustic logging curves correspond to other logging responses containing lithological information. Therefore, it is proposed to improve the vertical resolution of reservoir prediction by reconstructing reservoir characteristic curves.

20

2 Geophysical Exploration Methods for Lithological Reservoirs

The essence of reservoir characteristic curve reconstruction is to get the characteristic curve which can better reflect the formation of lithological characteristic information through reconstruction. The key is to turn the fine change of reservoir lithological information revealed by logging into an operable seismic model. Reconstruction methods include theoretical model reconstruction and statistical regression model reconstruction. The following six aspects must be considered when reconstructing the reservoir characteristic curve: (1) Selecting logging response that can reflect the lithological difference, such as natural gamma ray, spontaneous potential, deep resistivity, etc.; (2) The selection of logging curves must be combined with local geological characteristics, such as high debris content or high potassium feldspar content in the area not using natural gamma; Spontaneous potential cannot be applied when the relative relationship between formation water salinity and drilling fluid filtrate salinity varies greatly. The available resistivity of the stratum with obvious difference between reservoir and non-reservoir; (3) Before reconstruction, logging curves must be corrected, such as the return of spontaneous potential baseline; (4) The low-frequency model of formation acoustic wave must consider the variation law of the low-frequency model of the acoustic wave with depth caused by under compaction, secondary pores, etc.; (5) The method of curve reconstruction must be optimized by combining with logging lithological section; (6) The reconstructed reservoir characteristic curve must make the composite record match the good bypass. For low-amplitude structural and lithological thin interbedded areas, the difficulties and key technologies in the study are as follows: (1)

(2)

(3)

Resolution issues. To interpret low-amplitude structure and lithological thin interbedded sandstone bodies accurately, high-resolution seismic data bodies with high fidelity are required as the basis for the prediction of high-precision sedimentary layers. Identification and detection of micro-fracture. For complex lithologicaltectonic and structural-lithological thin-bed reservoir prediction, some small faults, even micro faults, and fault extension are very important, especially for the deployment of development wells. Therefore, high interpretation precision is required, true 3D interpretation is applied (high-resolution data body meets the interpretation requirements), and fine fault identification and detection technology are also required (Liang and Wang 2003). Prediction of the inner thin reservoir. Seismic inversion is one of the most important methods for reservoir prediction. Generally speaking, acoustic impedance inversion is essential and the basis of reservoir parameter calculation. However, due to the limitation of the resolution of acoustic curves, acoustic impedance inversion alone cannot meet the requirements of thin reservoir prediction. Therefore, it is necessary to find a way out in the aspects of reservoir feature reconstruction and multi-parameter inversion.

2.3 Sequence Constrained Reservoir Prediction Techniques

(4)

21

Identification of lithological traps. For high fidelity and high-resolution seismic data, the existing seismic phase and seismic attribute identification and prediction techniques are the key techniques to solve lithological traps.

The basis of reservoir characteristic curve reconstruction is the different reflection of the attributes of different geological bodies in different physical fields, the display of the same geological body’s attributes in different physical fields is correlated, and the different attributes of the same geological body have different emphasis in different physical fields. Reservoir characteristic curve reconstruction is to obtain a characteristic curve that can better highlight the resolution of the reservoir by synthesizing relevant information that is beneficial to reservoir prediction according to the reservoir prediction target. The premise of reservoir characteristic curve reconstruction is that the original acoustic moveout cannot well reflect the change law of formation lithology (or inversion target), while other good logging responses which have a good correlation with moveout can better indicate the change law of lithology (or inversion target). The method of reservoir characteristic curve reconstruction has some limitations at the same time, so it is necessary to optimize the reconstruction method according to the regional geological background, logging response characteristics and their mutual relations, and the indication degree of lithology (or inversion target) change law.

2.3.2 Seismic Attributes Analysis Seismic attributes refer to the geometric, kinematic, dynamic, and statistical characteristics of seismic waves derived from pre-stack or post-stack seismic data through mathematical transformation. For a long time, the use of seismic data is only limited to the pick-up of seismic waves in the same phase axis, to realize the description of the geometric shape and structural characteristics of oil and gas reservoirs. More information about lithology, physical properties, and fluid composition is hidden in the seismic data. As is known to all, the characteristics of seismic signals are directly caused by the physical characteristics of rocks and their variations. Therefore, the relevant information (reservoir lithology, physical properties, fluid composition, etc.) is indeed hidden in the seismic data, although various distortions and even irrecoverable distortions may occur. It has always been people’s pursuit to carry out seismic attribute analysis, make a calibration, eliminate data distortion, and pick up information about lithology and physical properties hidden in these data, to give full play to the potential of seismic data. Especially when it is more and more urgent for people to understand the heterogeneity of lithological stratigraphic reservoirs, the abundant spatial variation information of seismic data becomes more and more precious. With the development of reservoir interpretation technology of seismic data, especially the need for three-dimensional seismic data volume analysis, people’s understanding of seismic attributes is more and more profound, and the number of seismic attributes

22

2 Geophysical Exploration Methods for Lithological Reservoirs

Fig. 2.6 RMS amplitude attribute plan

increases sharply. Based on the commonly used seismic attributes, new attributes are added continuously. At the same time, there are more and more methods for attribute calculation and analysis. Seismic attribute analysis has been successfully used to predict reservoir lithology, oil and gas properties, and reservoir physical properties. Different seismic attributes may represent different geological meanings. For example, some attributes may be sensitive to some reservoir environment, while others may reflect oil–gas characteristics, while more seismic attributes may have no specific geological meaning, but are merely mathematical deduction of some attributes. For more targeted applications, it is important to understand the physical and mathematical implications of these seismic attributes (Fig. 2.6). Seismic attribute analysis is on the carrier of seismic attributes from seismic data to extract the hidden information and to convert the information to another information related to lithology, physical property, or reservoir parameters, for the direct service about geological interpretation and reservoir engineering, to give full play to the potential of seismic data, and to improve the capabilities of reservoir prediction, characterization, and monitoring. It consists of two parts, namely seismic attribute optimization and prediction. The prediction can be petroleum, either lithology or lithofacies, or reservoir parameter estimation. The former emphasizes the clustering and classification function of seismic attributes, which is mainly realized by pattern recognition. The latter emphasizes the estimation function of seismic attributes, and the main method is the approximation of function and neural network (Wang 2001).

2.3.2.1

Coherent Cube Technology

As an important technical means of three-dimensional seismic interpretation and lithological analysis, coherent cube technology develops and applies very rapidly, and has achieved good results in fault analysis and some lithological reservoir studies. The application of the coherence technique in fault interpretation and the combination can control and avoid the randomness of fault interpretation and combination,

2.3 Sequence Constrained Reservoir Prediction Techniques

23

Fig. 2.7 Coherent cube technology—fault interpretation

and greatly improve the accuracy and efficiency of fault interpretation. At the same time, the multi-channel similarity of seismic attributes can be used to highlight the continuity of strata by emphasizing the irrelevance of seismic data in the display, which can well reflect the lithological changes of strata, such as pinching and intrusion. The combination of a coherent attribute with the detection technology of dip angle, azimuth, and seismic multi-attribute analysis technology also has a certain effect on some geological problems that are difficult to be solved in the conventional interpretation. It has become one of the most widely used and mature technical means in lithological reservoir analysis (Fig. 2.7).

2.3.2.2

Three-Dimensional Visualization Technology

3D visualization technology is the true 3D interpretation under sequence constraints, which has two characteristics: first, high accuracy, there is no intersection closure problem of vertical and horizontal survey lines during automatic tracking, and there is no accuracy problem of point and line interpretation. The second is to avoid the point of view of different interpretation personnel, basically, according to the performance characteristics of the data body to explain, it is the development direction of seismic interpretation. True 3D volume interpretation is the interpretation of data volume. It starts from a 3D visualization display, and takes a geological body or 3D research block as a unit, and adopts the spatial visualization interpretation combining point, line, plane, and volume. The visualization-based 3D data body interpretation is more scientific, more efficient, and more reliable in both full 3D structure interpretation and geological body interpretation. Although it is possible to interpret 3D seismic data in terms of encrypted 2D seismic interpretation, 3D seismic interpretation is not equivalent to encrypted 2D seismic interpretation. This is not a viable method of interpretation, and the result is poor quality and inefficiency. The two-dimensional interpretation of three-dimensional data bodies restricts vision and interpretation to vertical and horizontal lines. For the habit of interpreters, the two-dimensional vertical section is more intuitive, but the amount of data is too large to see, so the

24

2 Geophysical Exploration Methods for Lithological Reservoirs

Fig. 2.8 3D visualization

data is interpreted line by line. For full 3D data interpretation, however, the more data, the better, because the seismic interpretation to solve the problem is always multiple solutions, never enough data to uniquely determine the geological conditions of the underground reflection interface. True 3D interpretation technology plays an important role in the identification of low-amplitude traps, which can effectively identify traps with an amplitude of more than 7 m (Fig. 2.8).

2.3.3 Interpretation of Inversion Data 2.3.3.1

Lithological Reservoir Interpretation

High-resolution sequence stratigraphy and quantitative prediction of seismic reservoirs are performed using seismic, well logging, core, and logging data. The ultimate purpose of feature curve inversion is to understand the spatial distribution of reservoirs, so it is necessary to interpret the feature curve inversion data (Li et al. 2014). In the interpretation of reservoir thickness, the threshold values of different rock strata are firstly determined according to the statistical results of rock velocity in the reservoir section. Then, based on the precise calibration and the interpretation of the control layer of the target segment, the time window is opened along with the auxiliary layer of the target segment. The high-speed layer with similar acoustic impedance

2.3 Sequence Constrained Reservoir Prediction Techniques

25

Fig. 2.9 The multi-parameter lithological seismic inversion technique greatly improves the resolution of thin layers

and similar distribution to a known reservoir and in line with sedimentary law is interpreted as a reservoir. Reservoir thickness maps and effective sandstone thickness maps are completed according to the above interpretation methods (Fig. 2.9).

2.3.3.2

Favorable Reservoir Extrapolation Interpretation

Based on the interpretation of the thickness of the small layers and the accurate calibration of the favorable reservoirs, the extrapolation interpretation is carried out. First of all, starting from the well, the joint well section is taken as a transition to expand to the whole region. Based on the reservoir acoustic impedance values of different layers, the reservoir is explained by extrapolation based on the well-point reservoir. According to the velocity variation range of different layers on the acoustic impedance profile, the impedance value range is delineated by adjusting the color code, and then the reservoir distribution range is traced. To eliminate the influence of other lithology, lithological curves are added to the inversion data body, and seismic records are superimposed on the impedance section, to control the extension and distribution range of the reservoir in the wellness area. Inversion, seism, and logging are integrated and complement each other. The conventional seismic profile, instantaneous phase profile, and inversion profile are compared and interpreted in reservoir extrapolation. The acoustic curve, acoustic impedance curve, spontaneous potential curve, and the lithological profile reflecting oil test and well logging are compared and interpreted, and then the reservoir is demarcated to the seismic profile and inversion profile. Following the above reservoir interpretation methods, inversion, seism, logging, and other data are fully used to complete the interpretation of favorable reservoirs.

26

2 Geophysical Exploration Methods for Lithological Reservoirs

2.4 Cognition and Conclusion At present, oil and gas exploration objects have gradually shifted from structural reservoir to structural and lithological complex reservoir mainly composed of a thin mutual reservoir. The combination of continental sequence and seism has obvious advantages in the study and prediction of the distribution of thin interbed sandstone bodies in lithological reservoirs. Based on stratigraphic division and correlation, this section uses the research results of sequence stratigraphy and sedimentary microfacies to subdivide and contrast each small layer from the genetic perspective, and recheck and close each layer one by one depending on the marker layer. The main method of subdivision and comparison is to use logging data, and different electrical curves have different responses to reservoirs. Different regions due to the reservoir lithology and mineral composition, fluid properties, the influence of such factors as log responding to change is bigger, so should be fully researched before contrast litho-electric relations, optimizing the contrast curve, combined the basic characteristics of each other makeup for the shortage, for the fine structure research, the research of sedimentary microfacies, reservoir evaluation work has laid a solid foundation. By using the high-resolution inversion method of well-logging constraint, geological model of integrated sequence stratigraphic constraint, reference physical property curve, and natural gamma curve, reservoir characteristic curve can be reconstructed to distinguish thin sandstone layers. More traps can be found with the increase in the number of faults explained by using the techniques of fracture fine detection (coherence cube calculation) and dip angle detection. Different seismic time windows are analyzed by using waveform classified seismic phase analysis technology, which enables the interpreter to quickly scan the whole data, to quickly determine the target area, and to convert the seismic phase into sedimentary facies, thus effectively determining the distribution area of possible lithological reservoirs. Through the application, this section summarizes the results of combining sequence stratigraphy with the high-resolution seismic in time at present, organically combines geology and geophysics, and establishes a set of advanced, effective, economical, and practical sequence constrained reservoir prediction technology. The supporting application of this technology in the exploration of thin interbedded lithological reservoirs in the southern Songliao Basin of China has achieved remarkable results.

References Liang B, Wang H (2003) The predicting technology for micro-fracture reservoirs. Oil Geophys Prospect 38(4):400–404. https://doi.org/10.13810/j.cnki.issn.1000-7210,2003.04.012 Li M et al (2014) Geophysical exploration technology: applications in lithological and stratigraphic reservoirs. (1st edn) Hardcover, Elsevier, 480 pages

References

27

Wang H (2001) The prediction of oil and gas by the supervised artificial neural network. Uranium Geol 17(1):48–55 Wang H (2005) Geophysical exploration technology and application for lithological reservoirs: a case study of southern Songliao Basin. China University of Geosciences (Beijing) Wang H (2008) The sequence constraint prediction technology and application for lithological reservoir. China Pet Explor 13(4):36–42

Chapter 3

Lithological Reservoir Exploration Technology

Lithological reservoir exploration technology, is based on high-resolution 3D seismic data, controls multi-parameter lithology inversion by starting from sedimentary facies belt system and reservoir characteristics reconstruction inversion through the research of sequence stratigraphy, and utilizes coherent volume technology, seismic attribute analysis technology, and 3D visualization technology to describe the change and distribution of reservoirs within the scope of lithological traps, to predict the lithological reservoirs.

3.1 Research of Sequence Stratigraphy Sequence stratigraphy is the science of establishing a genetically interrelated isochronous stratigraphic framework based on seismic, drilling, and outcrop data, and of comprehensively interpreting stratigraphic distribution patterns, sedimentary environments, and lithofacies within this framework. Sequence stratigraphy deals with the relationships of rocks within the chronostratigraphic framework of strata bounded by erosional or non-sedimentation surfaces or with comparable conformities. Based on seismic, drilling, or outcrop data, combined with the associated sedimentary environment and lithofacies characteristics, sealevel change is taken as the main controlling factor to comprehensively explain the stratigraphic distribution model. From the perspective of oil and gas exploration, the central idea of sequence stratigraphy research is to make full use of seismic, drilling, logging, outcrop, and testing data to establish an isochronous stratigraphic framework as fine as possible. In this framework, the development, evolution, and distribution of sedimentary systems are analyzed based on the genetic relationship, to improve the predictability of lithological traps.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 H. Wang et al., Practical Geophysical Technology and Application for Lithological Reservoirs, SpringerBriefs in Petroleum Geoscience & Engineering, https://doi.org/10.1007/978-981-16-4197-8_3

29

30

3 Lithological Reservoir Exploration Technology

3.1.1 Sequence Level and Its Geological Significance Sequence stratigraphy has been extensively explored in basins of different ages, different tectonic types, and different paleogeographic backgrounds since the late 1980s. The practice results show that different levels of sequence stratigraphic units can be divided into different types of basins. In the study of sequence stratigraphy, sequences are generally divided into giant sequences, supersequence group (supersequence), sequence, and fourth-level sequences (parasequence). The first and second level sequences are generally recognized to be controlled by global and regional tectonic factors, and their interfaces are usually regional unconformities, representing important discontinuities. This situation is obvious in both Marine and terrestrial strata; The third level sequence is the basic sequence in sequence stratigraphic units (Li et al. 2002). There has been a basic consensus among geologists on the meaning and classification criteria of sequence stratigraphic units at all levels. However, there is still no reasonable explanation for the genesis of sequence stratigraphic units, especially the third-level sequence. Based on the statistics of a lot of practice, the duration of sequence stratigraphic units at all levels is given roughly, but the amplitude of oscillation is large. Nevertheless, the duration of sequence stratigraphic units is important in determining their classification (Wang 2005).

3.1.2 Procedures for the Study of Sequence Stratigraphy To meet the technical requirements of the exploration of terrestrial lithostratigraphic reservoirs, this section takes Songliao Basin as the dissection point and puts forward the research program of sequence stratigraphy. There are six research steps, including sedimentary background analysis, sequence division and correlation, sequence interface tracking and closure, sequence constrained seismic reservoir prediction, sedimentary facies comprehensive analysis, accumulation law, and target evaluation. In the research process, it is emphasized that all kinds of geological data and geophysical data should be fully combined. By fully combining sequence analysis with various modern seismic quantitative interpretation techniques, the precision and quantitative level of reservoir prediction and target evaluation of lithostratigraphic reservoirs are improved.

3.1.2.1

Sedimentary Background Analysis

Whether it is regional sequence stratigraphy or high-resolution sequence stratigraphy of the target block, first of all, it is necessary to analyze the sedimentary background of the target strata in the study area. The main purpose of the analysis is threefold: The first is to enhance the pertinence of research ideas and methods; The second

3.1 Research of Sequence Stratigraphy

31

is to reveal the differences in the development of sandstone bodies and sequences under different geological backgrounds from the perspective of genesis; The third is to predict the basic characteristics of sedimentary facies and environment. The main contents of the analysis include three aspects: the tectonic background analysis, the climate background analysis, and the provenance condition analysis (Wang 2005). The tectonic background analysis mainly focuses on the plate tectonic background, basin type, tectonic pattern, and tectonic evolution history of the basin. It has been proved that there are great differences in the types of main sandstone bodies and the distribution of lithostratigraphic reservoirs in basins formed under different plate tectonic settings. The tectonic framework of the basin plays an important role in controlling the plane distribution of sandstone bodies. Different types of sandstone bodies are often developed in different structural belts. Tectonic evolution is the dominant factor controlling the evolution of the longitudinal sedimentary environment, which leads to the difference of sandstone body types and sequence composition characteristics in different periods. The analysis of climate background is mainly to understand the paleoclimate conditions of the sedimentary period of the target strata in the study area. Paleoclimate is an important factor controlling the weathering mode of parent rock, transport medium, sedimentary medium, type of sandstone body, and scale of source rock. The results show that the parent rock is mainly weathered by physical weathering, and the intermittent flood period and wind force are the main media of deposition and transportation under arid climate conditions. The types of sandstone bodies are usually single, mainly composed of gravity flow deposits, mainly composed of alluvial fan and fan delta. Under humid climate conditions, the parent rock is dominated by chemical weathering, and mainly develops normal tractive flow sandstone bodies, and there are many types of sandstone bodies. In the case of the same tectonic subsidence amplitude and area, the scale of the lake basin water body and the scale of the source rock body is small due to the salinization of the sedimentary environment in the arid climate background. Under humid climate conditions, the scale of the water body and source rock mass is larger (Li et al. 2005). Provenance condition analysis is mainly to understand the spatial distribution relationship between the sedimentary area and the provenance area, the tectonic activity of the provenance area, and the nature of the parent rock in the provenance area. The control configuration of the sedimentary area and provenance area is different, and the sedimentary characteristics in the same period are different. The provenance and tectonic activity have a direct influence on sedimentation. For example, during the Himalayan movement in the western basins of China, the Tianshan and Kunlun orogenic belts were revived, and coarse clastic wedges with large thickness were often formed in the piedmont of the foreland basins. Although a lot of achievements have been made in the study of many basins, it is possible to ignore the analysis of the sedimentary background and get an understanding that is not consistent with reality. For example, Naruto Uplift in Jizhong Depression is a high uplift lacking the coverage of Paleogene, but it did not exist in the period of Kongdian deposition in Paleogene. It does not separate Langgu sag from Baxian sag, nor is it a provenance area. The Niutuo Uplift was gradually uplifted in

32

3 Lithological Reservoir Exploration Technology

the Oligocene, which caused the erosion of the Paleogene sedimentary strata. Some researchers do not understand the structural evolution history of the Jizhong Depression, and the sedimentary facies map of the Kongdian Formation compiled by them, a series of sandstone bodies are distributed around the Niutuo bulge, to run counter to the facts (Li et al. 2014).

3.1.2.2

Sequence Division and Correlation

Sequence division and correlation are the basic research content of sequence stratigraphy. At present, there are two main problems in sequence stratigraphic division and correlation: (1) The sequence classification is chaotic, and the results of sequence division of different authors in the same study block are not consistent, and the sequence division scheme of the same author in different blocks in the same basin is not consistent. (2) The division of drilling sequence and seismic sequence is not uniform. Because of the disorder of sequence classification, the following suggestions are proposed: (1) The sequence classification corresponds to the classification of “five-level cycle” which is widely popular in petroleum geology and is consistent with the tectonic evolution stage. (2) Regional sequence stratigraphy research, Taking a basin or depression (depression) as the research unit, the purpose of the research is to find new hydrocarbon-bearing strata and favorable exploration direction. The sequence is divided into 2–3 levels with an emphasis on the identification of sequence boundary unconformity and the maximum lacustrine transgression surface. (3) The high-resolution sequence stratigraphy research, With zones or targets as the object of study, the purpose of the study is to find favorable traps and position the well, it should be divided into four levels of sequence (or quasi-sequence) (Zou et al. 2004a, b). The sequence division scheme of different blocks in the same basin is not unified, which means that the sequence framework established is unequal in time-span. To solve this problem, we should pay attention to two aspects: (1) Regional sedimentary background analysis should be paid attention to in the high-resolution sequence division of local areas so that the sequence division of local areas is consistent with the sequence division of the whole basin. Therefore, even if we only study the sequence stratigraphy of a 3D seismic block, we need to pay attention to the analysis of the tectonic evolution history of the whole basin. The sequence division of the local block is consistent with that of the basin. In particular, we should pay attention to the division of 2–3 sequences, which is consistent with the regional tectonic evolution history of the basin. There is usually an unconformity between sequence boundary and basin margin. (2) Wells and seismic interaction correlation, the drilling sequence division, and seismic division are unified through the methods of synthetic record calibration, litho-electric combination and wave group characteristic correlation, stratigraphic structure correlation, and so on (Zou et al. 2002). The basic procedure for establishing a sequence stratigraphic framework is as follows: Firstly, the seismic profile grid of the base shaft should be established. The

3.1 Research of Sequence Stratigraphy

33

density of the grid should be able to control the distribution of the sedimentary system in the study area. The second is the vertical sequence division of typical Wells, typically, Wells in the transitional zone are selected, sequence system tract and parasequence boundary were identified according to the transition or abrupt surface of the evolution of the longitudinal sedimentary facies sequence, and the division scheme of the longitudinal sequence was determined. The third is the sequence correlation of the combined well profile, sequence division, and correlation is completed in the unification process. The sedimentary sequence correlation method, layer leveling correlation method, and stratigraphic superposition correlation method can be used to make the sequence division results of the linked well profile consistently. The fourth is to identify the sequence boundary unconformity surface or the maximum lacustrine transgression surface according to the characteristics of seismic unconformity or the transformation surface of the macroscopic stratigraphic structure. The fifth is to unify the division of drilling and seismic sequence through well-seismic interaction correlation (Fu et al. 2005).

3.1.2.3

Sequence Interface Tracking Closure

The closure method of sequence interface tracing is not fundamentally different from the conventional seismic horizon interpretation, and all seismic horizon interpretation techniques can be used. Different from the conventional seismic stratigraphic interpretation, which focuses on tracing and interpreting seismic markers, regional sequence stratigraphy emphasizes the identification of unconformities and the maximum lacustrine transgression surface, because these two interfaces are most closely related to lithostratigraphic traps. The unconformities control the stratigraphic overlap traps and unconformable shielding traps, and the maximum flooding surface is associated with the lithological traps of various types of isolated sandstone bodies (Fig. 3.1). The high-resolution sequence division and correlation emphasizes the identification of the most isochronal transgression surface. Many years of practice have proved that in the exploration of lithostratigraphic reservoirs, high-resolution sequence stratigraphy plays the following roles: (1) Establishing a high-resolution sequence stratigraphic framework and clarifying the reservoir distribution and reservoir connectivity; (2) Analyzing the pinch-out and fault-fault intervals and finding out the distribution of different types of oil and gas reservoirs; (3) Providing wellconstrained isochronous stratigraphic interfaces and improving the accuracy of reservoir prediction; (4) Reducing the phenomenon of horizon tracking penetration and improving the accuracy of structural mapping; (5) To further clarify the reservoir-cap assemblage and define the exploration target strata; (6) To analyze the morphology, boundary, and reservoir forming conditions of lithostratigraphic traps (Wang 2008).

34

3 Lithological Reservoir Exploration Technology

Fig. 3.1 Formation interface is on the maximum flooding surface or the sequence interface

3.1.2.4

Seismic Reservoir Prediction Constrained by Sequence Interface

At present, the identification of sandstone bodies in the sequence framework is mainly based on the seismic facies identification technology proposed by seismic stratigraphy in the late 1970s. The sandstone body distribution map of the sequence unit prepared is the mainly qualitative map or semi-quantitative map. Seismic reservoir prediction technology has been widely used, but it does not use sequence stratigraphic constrained inversion. It only reflects the distribution of multiple sets of composite sandstone bodies, and it is difficult to predict the distribution and contact relationship of a single sandstone body. Sequence analysis and reservoir inversion are separated and cannot be used together, which affects the accuracy of seismic reservoir prediction (Wang 2005). Through many years of research and practice, it is realized that under the constraints of high-resolution sequence framework, using various reservoir prediction techniques, geological modeling to the in-phase axis can improve the accuracy of reservoir prediction, which is conducive to identifying the pinch-out line of sandstone layers and the distribution of isolated sandstone bodies (Fig. 3.2). Based on sequence interface comprehensive tracking closure, the various reservoir inversion technology, seismic waveform classification, such as slicing technology, 3D visualization interpretation technology are applied to sequence analysis, and a series of sandstone thickness and sandstone percentage contour maps can be compiled, improving the quantitative level of sequence analysis.

3.1 Research of Sequence Stratigraphy

35

Fig. 3.2 Inversion section under the constraints of sequence stratigraphic isochron framework

3.1.2.5

Comprehensive Analysis of Sedimentary Facies

Any data analysis of sedimentary facies has the problem of multi-solution, which should be combined with core, well logging, seism, and other data analysis. The data of core, outcrop, well logging, and logging is the basis for determining the types of sedimentary facies, and the results of seismic facies, seismic waveform classification, and seismic lithology prediction are the basis for determining the distribution range of various sedimentary facies. The maps provided by sequence stratigraphy and traditional sedimentology are the same in form, but the prediction function of the research results is different. In the study of underground geology, the traditional sedimentology mainly uses the drilling data to map according to the sedimentary model, which only provides the concept to describe the sedimentary characteristics of the study area, and lacks the strong prediction function. Sequence stratigraphy is the preparation of sections and plans of various sedimentary facies under the constraints of sequence interface. Due to the use of quantitative reservoir prediction results such as seismic lithology inversion, the sedimentary facies maps prepared are of predictive significance and can be used as a direct basis for well location deployment.

3.1.2.6

Accumulation Law and Target Evaluation

At present, the study of sequence stratigraphy usually takes the establishment of sequence model as the final result, and no systematic industrial map of objective evaluation has been prepared completely, and no systematic study has been done on the relationship between continental sequence and the distribution of lithostratigraphic oil and gas reservoirs. After years of research and practice, it is concluded that sequence stratigraphy should complete the following aspects when evaluating the target and studying the

36

3 Lithological Reservoir Exploration Technology

accumulation law: (1) Through the study of the evolution law of regional sequence, we can clarify the law of regional souring-reservoir-cap assemblage in the longitudinal direction, define the sequence of main exploration targets, and explore new areas of oil and gas exploration; (2) Through high-resolution sequence stratigraphy, the reservoir and cap assemblage can be further identified and the exploration target strata can be confirmed; (3) Comprehensive evaluation and analysis of favorable sandstone bodies under the constraints of sequence framework include reservoir connectivity, physical property prediction analysis, oil and gas bearing analysis, etc.; (4) Analyzing the conditions of lithostratigraphic traps, and finding out the overlap line of the inner layer of the sequence framework, the distribution of trap boundary and the oil-accumulation tectonic background; (5) Analysis of oil and gas in traps, including oil and gas sources, migration channels, the amount of trap resources and the risk of oil and gas.

3.2 Seismic Attribute Analysis Technology Seismic attributes refer to the geometric form, kinematics, dynamics, and statistical characteristics of seismic waves derived from pre-stack or post-stack seismic data by mathematical transformation. For a long time, the use of seismic data is only limited to picking up the seismic wave in the phase axis to realize the description of the geometric form and structural characteristics of oil and gas reservoirs. Seismic data hide more abundant information about lithology, physical properties, and fluid composition. It is well known that the characteristics of seismic signals are directly caused by the physical properties of rocks and their variations, therefore, the reservoir lithology, physical properties, fluid composition, and other relevant information may be distorted, even irreversible distortion, but it is indeed hidden in the seismic data. It has always been people’s pursuit to analyze seismic attributes, make a calibration, eliminate data distortion, and pick up the information about lithology and physical property hidden in these data, to give full play to the potential of seismic data. Especially when people are more and more urgent to understand the heterogeneity of stratigraphic lithological reservoirs, the rich spatial variation information of seismic data is more precious (Wang 2004a, b). In recent years, with the progress of seismic data reservoir interpretation technology, especially the need for 3D seismic data volume analysis, people’s understanding of seismic attributes is more and more profound, and the number of seismic attributes has increased sharply. Based on the commonly used seismic attributes, new attributes are added constantly. At the same time, there are more and more methods for attribute calculation and analysis. Seismic attribute analysis has been successfully used to predict reservoir lithology, oil, and gas-bearing properties, and estimate reservoir physical properties.

3.2 Seismic Attribute Analysis Technology

37

3.2.1 Seismic Attribute Classification At present, nearly ten types of seismic characteristic parameters can be extracted from seismic data volumes, such as amplitude, frequency, phase, polarity, impedance (or velocity), and other categories, and each category contains many kinds of parameters (Fig. 3.3). There are many available seismic attributes. According to statistics, there are more than 300 kinds of clearly defined seismic attributes, and there are dozens of commonly used attributes. In the face of so many seismic attributes, people often have nothing to do in practical application. It is helpful to understand and use seismic attributes correctly to classify a large number of seismic attributes reasonably. Geophysicists have done a lot of work in this field. There are many classification schemes for seismic attributes from different starting points. To facilitate the practical application of seismic attributes, the research results of seismic attributes are summarized, and some commonly used seismic attributes are summarized according to their functions (Table 3.1).

Fig. 3.3 Amplitude attribute and average instantaneous phase attribute

Table 3.1 Function table of common seismic attributes Geological Bright spots, characteristics dark spot, sand bodies

Petroliferous property

Sequence

Faults and cracks

Unconformity

Commonly used attributes

Instantaneous phase, the slope of the instantaneous frequency, the slope of the reflection intensity, the slope of the power spectrum

Instantaneous frequency, decibel-based reflection intensity

The first, second, third spectral park frequency, limited frequency bandwidth energy, KLPC1 average correlation, correlated kurtosis

Correlated KLPC2, correlated KLPC3

Instantaneous amplitude, reflection intensity, the maximum value of the peak amplitude

38

3 Lithological Reservoir Exploration Technology

3.2.2 Seismic Attribute Analysis The purpose of seismic attribute analysis is to extract the hidden information from seismic data with seismic attributes as the carrier and to convert this information into information related to lithology, physical properties, or reservoir parameters, and that can be directly used for geological interpretation or reservoir engineering. Thus, the potential of seismic data can be fully exploited, and the ability of seismic data in reservoir prediction, characterization, and monitoring can be improved. It consists of two parts: seismic attribute optimization and prediction. Prediction can be either oil–gas bearing, lithology or lithofacies prediction, or reservoir parameter prediction (estimation), the former emphasizes the clustering and classification function of seismic attributes, mainly through pattern recognition, while the latter emphasizes the estimation function of seismic attributes, mainly through function and neural network approximation. With the increasing ability of seismic attribute extraction and the increasing importance of lithological reservoirs, the application of seismic attribute analysis technology to reservoir property prediction has been widely used. The general process of seismic attribute analysis is summarized as follows: (1) Determine the time-depth relationship between drilling and seismic data, that is, horizon calibration; (2) stratigraphic tracing and interpretation to determine the time window, and seismic attribute extraction; (3) seismic attribute optimization, to select the least number of attribute combinations for prediction; (4) Seismic attribute analysis to establish the statistical relationship between seismic attributes and geological characteristics, to predict the reservoir characteristics between Wells under the guidance of intensive seismic data. As mentioned above, seismic attribute analysis techniques include horizon calibration and interpretation, attribute extraction, attribute optimization, and attribute analysis. In addition to horizon calibration interpretation, which can be understood as the division and tracking of sequence interfaces for the target layer, the remaining three parts are the three core steps of seismic attribute analysis technology. The following will discuss in detail the three aspects of attribute extraction, attribute optimization and attribute analysis.

3.2.2.1

Seismic Attribute Extraction Method

Seismic attributes include profile attributes, layer attributes, and data volume attributes, and their extraction methods have their similarities and differences. The extraction of profile attributes is mainly accomplished through special processing processes, such as complex seismic trace analysis, trace integration, seismic inversion, and so on. The extraction methods of layer attributes are as follows: Instantaneous horizon attributes extraction, single-channel time window attributes extraction and multichannel time window attributes extraction. The instantaneous horizon attributes are derived from the special processing methods such as complex seismic trace analysis

3.2 Seismic Attribute Analysis Technology

39

and picked up at the horizon; The single-channel time window attributes are picked up along a variable time window; The multi-channel time-window attributes picking method is similar to the single-channel time-window attributes picking method. In addition to defining the upper and lower boundaries of the extraction window, it also defines the number of traces and modes extracted horizontally. As a special case, a time slice can be thought of as an isochronous horizon with no physical meaning (Wen et al. 2003). The extraction method of the data volume attribute is the same as that of the horizon seismic attribute except that time slice is used instead of horizon.

3.2.2.2

Seismic Attribute Optimization

With the deepening of the application of seismic attributes, more and more seismic attributes have been proposed. Seismic attribute analysis usually goes through a process from less to more and more to less. From less to more means to extract as many possible attributes as possible, that may be relevant to the reservoir prediction during the early stages of the prediction design. In this way, useful information can be utilized to improve the performance of reservoir prediction by drawing on the experience of various experts. However, the infinite increase of attributes also hurts reservoir prediction. The reasons are as follows: (1)

(2) (3) (4)

Some seismic attributes may have nothing to do with the target layer itself, but reflect the changes of other strata. These attributes can only interfere with the prediction of the target layer; The increase of attributes will bring difficulties to calculation because too much data will occupy a large amount of storage space and calculation time; A large number of attributes will certainly contain a lot of interrelated components, resulting in information duplication and waste; The number of attributes is related to the number of training samples. In terms of pattern recognition, when the sample size is fixed, too many attributes will lead to the deterioration of the classification effect.

Therefore, it is necessary to select the best subset of seismic attributes from the seismic attribute set for specific problems, that is, seismic attribute optimization problem. Seismic attribute optimization methods can be divided into dimensionality reduction mapping of seismic attributes and selection of seismic attributes. The former usually uses K-L transform and principal component decomposition, which construct a few effective new seismic attributes from a large number of original seismic attributes. The drawback is that the physical meaning of the original seismic attribute no longer exists. The selection of seismic attributes is to optimize the attribute by experience or mathematical method. Before the selection of seismic attributes, the objective function must be designed. The objective function is determined flexibly according to the reservoir prediction method and seismic attribute selection method, and different

40

3 Lithological Reservoir Exploration Technology

methods have different objective functions. It can be divided into expert optimization, automatic optimization, and mixed optimization, and it is the key to quantitative seismic facies analysis. In general, it should be performed on 3D migration data recovered by true amplitude. When the stratum is near horizontal and the structure is not complicated, the seismic attributes can also be extracted by stacking profile. It should be noted that the purpose of extracting seismic attributes is to explain the pale sedimentary environment and sedimentary facies, therefore, any major geological event after sedimentation, especially tectonic movement, poses a threat to the seismic properties that reflect the original stratigraphic characteristics. In addition, as many techniques as possible should be used to extract more seismic attributes from different aspects. Although these seismic attributes do not have direct geological significance, the comprehensive analysis of these seismic attributes can reveal the characteristics of seismic waves corresponding to various geological phenomena. (1)

(2)

(3)

Expert optimization: Generally speaking, oilfield experts have a good understanding of the relationship between reservoir information and seismic attributes in a certain area, and can select seismic attributes based on experience. Sometimes experts can put forward several groups of better seismic attributes or combinations of seismic attributes, but it is difficult to conclude which group is the best. This can be calculated by the error rate (pattern recognition) or prediction error (function approximation or neural network) and compare, selecting the smallest error rate or prediction error as the best seismic attribute or combination of seismic attributes. Automatic optimization: Because the relationship between the solved problems and seismic attributes is complex, it is difficult to select them empirically. To obtain the optimal effect of reservoir prediction, it is necessary to optimize the combination of seismic attributes. The measure of “best” is to minimize the rate of error or prediction error. To obtain the optimal attribute combination, only by using the enumeration method, the error rate or prediction error of various attribute combinations (a total of 2D-1) is compared to select the optimal attribute combination, which is the combinatorial optimization problem. When D is large, 2D-1 is too large to be solved by the enumeration method. Therefore, we have to look for a suboptimal solution with less computation. The commonly used methods are attributed comparison method, sequential forward method, sequential backward method. Genetic algorithm and RS theory decision analysis method are new methods for optimizing seismic attributes. Mixed optimization: To overcome the limitation of expert knowledge and experience and reduce the calculation amount of automatic optimization, the combination of expert optimization and automatic optimization can be used for seismic attribute optimization. The most common method is to combine expert optimization with the optimal search algorithm to obtain the optimal solution to the combinatorial optimization problem.

From the above discussion, it can be seen that It is generally impossible to determine the best subset of seismic attributes. However, practical experience shows that

3.2 Seismic Attribute Analysis Technology

41

a simple attribute selection (optimization) method is likely to get better results than random attribute selection. It should be pointed out that in the seismic attribute dimensionality reduction mapping, the new seismic attributes are constructed from a large number of original seismic attributes, while in the seismic attribute selection, the selected seismic attributes are accepted from the original set of seismic attributes.

3.2.2.3

Seismic Attribute Analysis Method

Since the emergence of seismic methods in the 1920s, as a necessary means of exploration, seismic methods have made great contributions to the search and discovery of oil and gas resources. At present, the application of seismic technology has been extended from structural morphology analysis to reservoir evaluation, from structural reservoir exploration to lithological reservoir exploration. As one of the core technologies of lithological reservoir exploration, seismic attribute analysis plays an important role in reservoir description and evaluation. Its functions are mainly manifested in three aspects: oil and gas prediction, lithology or lithofacies prediction, and reservoir parameter estimation.

Oil and Gas Prediction Method With the continuous introduction of various new mathematical tools and signal processing techniques to seismic exploration, the subsurface geological information extracted from seismic data has been greatly increased. The idea of looking directly for subsurface oil and gas reservoirs based on reflected seismic data naturally arose. The first is the “bright spot” technique, which emerged in the 1970s, in which the amplitude and polarity of reflected waves are used as attributes to identify oil and gas reservoirs. Later, a variety of techniques for comprehensive detection of oil and gas reservoirs using a variety of seismic attributes have appeared. Since the 1980s, pattern recognition technology has been widely used, including statistical pattern recognition, neural network pattern recognition, and other oil and gas prediction technologies (Wang 2001). (1)

Statistical pattern recognition method. Oil and gas statistical pattern recognition is a kind of technology that, according to the difference of seismic wave kinematics and dynamic characteristics (such as waveform, amplitude, frequency, phase, etc.) between oil and gas reservoirs, extracting various seismic attributes from seismic data, and using a multivariate statistical method to predict the location and range of oil and gas reservoirs. Because conventional seismic interpretation methods often encounter many difficulties and are not effective in reservoir studies, the common difficulty is that the reservoir is thin and indistinguishable, another difficulty is that some changes in reservoir characteristics are too weak to be detected by the naked eye on seismic records. The pattern recognition technology uses a variety of seismic

42

(2)

3 Lithological Reservoir Exploration Technology

attributes to judge the changes of the reservoir, so it has a high comprehensive resolution ability. Its realization process can be divided into the learning process and prediction process. For the unknown category of seismic trace, the optimized seismic attributes are calculated in the learning process, and the classifier is used to classify and predict the category of the seismic trace. There are many kinds of classifiers, and two linear discriminant methods, namely Fisher and Bayes, are common. BP neural network pattern. BP neural network pattern recognition for oil and gas prediction is a new method that appeared in the early 1990s. It has the ability of self-learning, self-adaptation, and strong fault-tolerant ability, so it is a good method for oil and gas prediction.

The pattern recognition workflow of artificial neural network (Fig. 3.4) is summarized as follows:

Fig. 3.4 Pattern recognition workflow of artificial neural network for seismic data

3.2 Seismic Attribute Analysis Technology

➀ ➁ ➂ ➃ ➄

43

The target horizon is obtained by processing and interpreting the seismic data in the working area; Open a time window along the interpretation horizon of seismic profile to extract seismic characteristic parameters; Select the wells with appropriate characteristic parameters and known oil and gas conditions as the learning sample training network; After the convergence of network training, the input–output relationship mapping is obtained; The trained network is used for the prediction of unknown regions to complete the rapid evaluation.

The processing of working area data generally includes the study of post-stack data and the in-depth understanding of working area geological conditions. It is more important to evaluate the seismic data in the working area. The quality of seismic data is an important factor affecting the result of evaluation and prediction, so it is reasonable and necessary to evaluate the quality of seismic data in the working area before the artificial neural network pattern recognition. The quality of seismic data can be represented by three physical parameters: signal-to-noise ratio, resolution, and energy. If a high signal-to-noise ratio, high resolution, and moderate target energy can be achieved, the quality of this kind of seismic data is good. The interpretation of seismic data is a very important process because the pattern recognition results are meaningful only when more accurate horizon positions are obtained. The interpretation of the data requires the completion of work area loading, logging correction, contrast tracing, and other works, and the rough structure map of the target interval is obtained. After the seismic data interpretation is completed, the seismic attributes of the target interval must be extracted and analyzed. For 2D seismic lines, a time window of appropriate size is opened along the interpreted horizon, and various seismic attribute parameters are calculated in the time window. The size of the time window depends on the thickness of the reservoir. In general, the size of the time window should not be less than 1.5 times the thickness of the reservoir, nor more than 3 times the thickness of the reservoir. The calculation of parameters is usually large, so the optimization of parameters is very important before the calculation. If it is threedimensional data, in the process of parameter selection, the whole three-dimensional data body is calculated for each calculation, which requires a considerable amount of calculation. Since the mapping relationship between seismic attributes and oil and gas content should be relatively stable in a working area, to reduce the amount of calculation, it is usually adopted to select several typical survey lines in the working area to calculate their attributes and extract and analyze their attributes. After finding the appropriate parameters, the calculation of parameters and attribute analysis is carried out for the whole three-dimensional working area. The establishment of parameter combination includes two processes: one is to select the appropriate parameters to participate in the classification and prediction calculation of pattern recognition by calculating the seismic attribute parameters in the time window; the other is to select the parameters by analyzing the prediction

44

3 Lithological Reservoir Exploration Technology

results. By combining the parameters obtained after these two choices, a parameter combination for pattern recognition is obtained. After the parameter combination is obtained, the work to be done is to design the artificial neural network model. The design of the network model includes the selection of parameters such as the number of network layers, the number of input nodes, the number of output nodes, and the number of nodes in the middle layer. The principle of model design is to best classify different classes in the sample space, and the number of output nodes is usually two or three, that is, whether oil and gas exist or not. After the appropriate network model is obtained, samples should be selected to train the network. The selection of samples is also very important. Proper samples can not only accelerate the convergence of network training but also, more importantly, affect the results of classification prediction. The selection of samples should be comprehensive, and typical seismic trace or well information that can represent the characteristics of oil and gas in the study area should be selected as far as possible. This process must be repeated several times until a satisfactory trial result can be obtained. After the network training converges, the weights and thresholds of the network can be obtained. Finally, the weights and thresholds of the network are used to calculate the unknown region, and the predicted results are obtained. Testing the prediction with as many known Wells as possible, and repeating the process until you get an optimal prediction. For the three-dimensional work area, after completing the above process for the typical survey line, the network parameters and weight coefficient obtained should be used to predict and calculate the work area, and finally get the predicted results.

Lithology and Lithofacies Prediction Methods It is generally accepted that seismic parameters, such as reflection structure, geometry, amplitude, frequency, continuity, and interval velocity, represent certain lithological assemblage, bedding, and sedimentary characteristics of the sediments producing reflection. Therefore, the prediction method of lithology and lithofacies of seismic attributes is to establish the statistical relationship between seismic attributes, lithology, and lithofacies, and the plane distribution of lithology and sedimentary facies is predicted by analyzing the reflection characteristics and wave mechanics characteristics of seismic data. The traditional method of seismic facies analysis is described by naked eye observation. However, with the continuous improvement of seismic data acquisition technology, seismic information contained in seismic profiles becomes more abundant. Much of this information is not detectable by looking at seismic profiles with the naked eye, the geological characteristics of this seismic information can only be explained by using the seismic data processing technology and computer technology to extract and analyze them, and by certain mathematical methods. Therefore,

3.2 Seismic Attribute Analysis Technology

45

quantitative lithology and seismic lithofacies analysis are produced and studied by multivariate statistical methods. (1)

(2)

Multivariate statistical method for lithology and lithofacies prediction. Determining several types of sedimentary facies in a seismic sequence is also the key to quantitative seismic facies analysis. The traditional method is determined only by drilling and logging facies analysis, however, for a basin or a large block or an area with few Wells drilled, this approach is insufficient. Therefore, sedimentary facies type should be carried out according to the following aspects. ➀ Analyze single well sedimentary facies according to drilling and logging data to determine the sedimentary facies type of the study layer in the good area. ➁ Determine sedimentary facies type according to the seismic facies model. For example, in the continental faulted basins in eastern China since Mesozoic and Cenozoic, the alluvial fan facies or fan-delta facies are generally found on one side of the fault, if the fault side is disintegrated or weakly reflected seismic facies. ➂ According to the sedimentary tectonic evolution history of the basin, the sedimentary environment and sedimentary system of the basin can be inferred, and the possible sedimentary facies types and their spatial positions can be analyzed. ➃ Then, the seismic attributes are extracted according to the learning channels, and the discriminant function of the learning channels is established by the multivariate statistical method. Since these learning channels correspond to the sedimentary facies of the well, the established discriminant function is the discriminant function of the sedimentary facies. If there are many Wells in the study area, it is possible to establish several discriminant functions, each of which corresponds to a different sedimentary facies. ➄ According to the discriminant function of each sedimentary facies, the discriminant analysis of the seismic attributes of each seismic trace in the target strata in the study area is carried out, and the plane distribution of the sedimentary facies in the target strata is finally determined. Seismic facies analysis method based on Kohonen neural network. In quantitative seismic facies analysis using attributes, the traditional methods emphasize the role of Wells, that is, well calibration and well sample monitoring. This implicitly assumes that the categories of finite good points represent all the categories of the entire work area and that there is no seismic phase outside of the good points, which is not true. Besides, such an assumption may lead to missing information in the following two aspects: ➀ the change degree of seismic information in the whole working area (that is, the overall change degree of seismic data); ➁ The overall variation degree of seismic data is spatially distributed. For the use of seismic information classification, is to do not know the entire area of seismic information can be divided into several categories, how the distribution of each category. This makes it impossible to assess the magnitude of the seismic signal variation at the good point, moreover, meaningful seismic signal changes cannot be extrapolated if they cannot be related to the overall seismic signal changes.

46

3 Lithological Reservoir Exploration Technology

So, how to evaluate the overall variation of seismic signals? Seismic facies maps can be obtained by the unsupervised neural network. With the seismic facies diagram, the seismic information variation degree at the good point can be related to the seismic facies and the overall variation degree of seismic information. Thus the relative value of the good point can be confirmed or disproved. The method can be divided into two levels, ➀ Seismic facies analysis is carried out by using the similarity of seismic waveforms. ➁ Waveform similarity and seismic attributes are used to analyze seismic facies. The seismic phase neural network analysis method based on waveform similarity is based on the various characteristics of seismic waves in the interior of the time window. Seismic trace waveform is the basic property of seismic data. It contains all relevant information, such as reflection mode, phase, frequency, amplitude, and so on. It can be considered that any changes in physical parameters related to seismic wave propagation can be reflected in seismic trace waveform changes, The variation of seismic trace waveform can be characterized and measured by the variation of sample values over time. It emphasizes the general morphology and relative changes of seismic trace, and the map showing the similarity distribution of seismic trace waveform is very similar to the facies map showing similar geological characteristics, so it is called seismic facies map. The steps are as follows. ➀ Kohonen neural network technology is used to extract typical seismic tracks from some seismic data sets. They can represent the overall change of the entire area, The typical seismic trace is arranged in the order of gradual waveform change. Each track then specifies a value or a color to form a set of model tracks. The number of the model trace can be determined according to the complexity of actual seismic data. ➁ Each track in the seismic data set is correlated with all typical seismic tracks. And we assign the value or color to the typical seismic track that has the maximum correlation value with the actual seismic track. Thus, similar seismic tracks have the same or similar values or colors, forming a seismic facies map. It is the similarity diagram of the actual seismic trace and the typical seismic trace. This method has the characteristics of no good data and fast analysis, and it can analyze different time windows. The interpreter can quickly scan the entire data, quickly identify the target area, and can carry on more detailed research work to the target area. At the same time, compared with previous seismic facies analysis, this method has enhanced the characteristics of quantitative and objective.

Reservoir Parameter Estimation Method Reservoir parameters include reservoir thickness, porosity, permeability, saturation, sand and mudstone content, etc. Reservoir parameter estimation methods can be roughly divided into four categories: (1) Kriging method using logging data only; (2) Linear regression method combining well-logging data with seismic attributes; (3) Geostatistical method combining well-logging data with seismic attributes; (4) Neural network approximation method combining logging data with seismic attributes. The first method is only suitable for the situation with more well data, but

3.2 Seismic Attribute Analysis Technology

47

Fig. 3.5 Multi-attribute neural network prediction

it is difficult to describe the change details of the reservoir. With the improvement and refinement of the accuracy of reservoir description, this kind of method has been used less and less. The latter three methods all emphasize the combination with seismic attributes, which represent the development trend of reservoir parameter estimation, and have evolved from a single attribute to a multi-attribute. It can be said that the reservoir parameter estimation method based on multiple seismic attributes is the future development direction (Figs. 3.5, 3.6 and 3.7). Traditional reservoir parameter estimation methods attempt to find the physical relationship between reservoir parameters and seismic attributes, and then derive the mathematical relationship based on the physical relationship under the hypothesis (which is difficult to satisfy in practical application), and then use the single attribute to predict the parameters on this basis. However, in the numerous seismic attributes, some of the seismic attributes of the physical and geological significance are very clear, the use is quite direct. As seismic reflection amplitude, it is normally related to the lithology, porosity, and saturation of the reservoir. However, the definition of many seismic attributes is very clear, the physical interpretation of their numerical changes is very vague, their relationship with underground lithology and physical properties is not obvious, there is no quantitative relationship between them directly derived from the theory or the theoretical approximate formula. Practical experience shows that there is usually a good relationship between seismic attributes and logging lithology and physical property parameters, so we can change the traditional research method based on theoretical or theoretical approximate relationship and adopt the

48

3 Lithological Reservoir Exploration Technology

Fig. 3.6 Seismic attribute inversion porosity profile

Fig. 3.7 Seismic attribute inversion porosity time slice

data-driven method instead. The so-called data-driven method uses a large number of seismic attributes and well log lithology and physical property interpretation results at the good point to establish relationships among them using geo-statistics, neural networks, pattern recognition, artificial intelligence, simulated annealing, and genetic

3.2 Seismic Attribute Analysis Technology

49

algorithms. Then the lithology or physical property parameters beyond the good point can be derived from the seismic attributes based on the relationship, which can be interpreted by logging calibration and residual correction. It emphasizes the “digital” relationship between reservoir parameters and seismic properties, rather than the physical relationship, thus avoiding the time delays associated with finding the physical relationship. At the same time, it avoids the unrealistic assumptions introduced to derive the corresponding mathematical relations of physical relations (Huang et al. 2004).

3.3 Coherent Volume Technique Coherence volume processing technology can generate new coherent data volume by quantifying the coherent attributes of seismic data volume to highlight and emphasize the irrelevance of seismic data. In general, the special feature of this technique is the analysis of discontinuous strata such as faults and lithology. Conventional seismic profiles and seismic attributes along the vertical section of the dip and the horizontal section of the fault and the in-phase axis are easy to explain, but the vertical section along the strike and the horizontal section of the parallel fault and the in-phase axis is difficult to explain. The coherent discontinuous display solves these problems well. It can accurately reflect the discontinuous characteristics of underground strata and explain the geological and hydrocarbon reservoirs such as quantitative faults, lithological anomalies, and carbonate salt fractures. As an important technical means of 3D seismic interpretation and lithology research, coherent volume technology has been applied and developed very fast. It has become a conventional means of interpretation and has been applied in fault interpretation and even lithology interpretation, and is one of the basic techniques widely used in seismic data interpretation.

3.3.1 Coherent Volume Concept Coherence volume analysis is a new 3D seismic interpretation technique developed in the middle of the 1990s. Coherence is a measure and attribute of the similarity between multi-channel seismic data. Coherence volume technology is to use coherence between 3D data volume data to display the continuity and interruption of data. The computation of coherence can provide a quantitative value of data similarity when the coherence is weak or disturbed by noise. By quantifying the coherent attribute of seismic data volume, the coherent operation is carried out for waveform, and a new coherent attribute volume that is different from the conventional seismic amplitude data volume is generated. This kind of data volume can be used to interpret complex faults and concealed strata lithology, but these complex geological features can’t be recognized and interpreted in conventional seismic data. The special feature

50

3 Lithological Reservoir Exploration Technology

of coherence technology is that it highlights those transverse discontinuous and irrelevant seismic and geological features, such as faults, deltas, river channels, etc. It can reflect a variety of underground geological conditions more objectively and truly, help researchers to analyze and understand problems from the overall concept, and improve the efficiency and accuracy of interpretation work. The concept of coherence is a measure of the similarity between multiple channels of data. In the visual display of multichannel data, coherence is detected by the interpreter. One of its most effective and important applications is the detection of the in-phase axis, especially when the amplitude of the in-phase axis is small and hidden in the noise. In addition to detecting this metric, a quantitative value is also assigned to the intensity of the in-phase axis. If its valuation is an energy normalized metric, it can be easily converted into the signal-to-noise ratio. Therefore, coherence can be used to evaluate the data quality. Its calculation can provide the quantitative value of data similarity in the case of weak coherence or interference by noise, and cross-correlation is the basis of its calculation. For the analysis of discontinuous waveform data, it is easy to interpret the vertical section along with the dip and the horizontal section cut along the fault and the phase axis, but it is difficult to interpret the vertical section along the strike and the horizontal section cut along the phase axis of the parallel fault, and the quantitative discontinuity of the coherent volume solves these problems better. At present, the calculation of coherent volume is the calculation of three-dimensional seismic amplitude data volume, mainly using the multi-channel similarity of seismic data to extract the coherent coefficient data volume. The discontinuity of seismic data is highlighted by the display of three-dimensional, plane and section, including the display of plane, section, and arbitrary line, etc., and the display of unrelated anomalies is emphasized. Its premise is that the strata are continuous and the seismic wave changes gradually, so adjacent channels and lines are similar to each other. When local layer continuity is destroyed and changes, such as fault, pinch out, invasion, deformation, etc., the seismic wave changes, showing the mutation of edge similarity. Through the multidimensional display of the coherent volume, the geological targets related to fault structure, sedimentary strata, stratigraphic physical properties, and even fluid changes can be identified. In recent years, as an important means of 3D seismic interpretation and lithology analysis, coherent volume technology has developed and applied very rapidly, and has achieved good results in fault analysis, stratigraphic lithology body, and special reservoir in some areas. By applying coherence technology to fault interpretation and combination, the randomness of fault interpretation and the combination can be controlled and avoided, and the accuracy and efficiency of fault interpretation can be greatly improved. At the same time, the use of multi-channel similar seismic attributes can emphasize the irrelevance of seismic data in the display, highlighting the continuity of the strata, which can well reflect the lithological changes of the strata, such as pinch-out, invasion, etc. Combined with the detection technology of stratigraphic dip angle, stratigraphic azimuth angle, and seismic multi-attribute analysis technology, the coherent attribute has a certain effect on some geological problems that are difficult to be solved in conventional interpretation and has become one of

3.3 Coherent Volume Technique

51

the most widely used and mature technical means in 3D seismic fault interpretation and lithology analysis.

3.3.2 Fundamental Principles of Coherence Computation At present, there are many algorithms to generate 3D seismic coherent attribute data, which are mainly based on the horizontal direction. According to the signalto-noise ratio of the data and the stability of the algorithm, it mainly includes three coherent algorithms: C1, C2, and C3. These three coherent algorithms have their advantages and disadvantages: The C1 coherent algorithm calculation speed, low demand on computer memory, but affected by noise interference, poor stability, low resolution; The C2 coherent algorithm has the strong anti-jamming ability and high resolution, but the computation is large and the transverse resolution is low; The C3 coherent algorithm has better stability, stronger anti-interference ability, and higher resolution, but it is not suitable for the calculation of large dip formation data because of the large amount of computation. The C3 coherence algorithm is based on the covariance matrix of the C2 algorithm to calculate the coherence value of 3D seismic data volume.

3.3.2.1

C1 Coherent Algorithm

In most coherent computation software, the coherent algorithm is based on the traditional normalized energy cross-correlation principle to calculate the coherent volume, also known as the first generation algorithm, whose calculation principle is relatively simple and easy to understand. In general, complex seismic attributes, including measurements of seismic wave amplitude, frequency, and phase, have been successfully used to map seismic lithology. The multichannel relation (including the cross-correlation technique) is also applied to the automatic picking up of 3D seismic reflectors. However, the seismic coherence measure is on the inside of the numerical algorithm, as a kind of application properties, local seismic lateral changes, 3D seismic fault skeleton is quite effective for delineating coherent data volume, and this algorithm can highlight the tiny changes of formation, including the point of distributary channel sand dam, canyons, slump structure, and tidal discharge three-dimensional images.

3.3.2.2

C2 Coherent Algorithm

The second-generation algorithm, the C2 algorithm, can perform similarity analysis on any tracing number to estimate its coherence. In addition to more robust coherence and dip/azimuth calculations in noisy environments, vertical analysis time windows

52

3 Lithological Reservoir Exploration Technology

can be limited to only a few time samples and can accurately calculate thin and small changes in information characteristics. The C2 coherence algorithm is widely used in recent years. It performs similarity analysis on arbitrary multi-channel seismic data and calculates its coherence. In addition to the more stable calculation of coherence, dip, and azimuth under noise environment, it can limit the range of time data sampling points in the analysis of vertical time window, to better calculate the subtle changes of formation characteristics. For the second-generation algorithm C2, considering the reflector tendency and azimuth angle of the whole data volume, the fault in a certain direction can be displayed more clearly. Especially in the vertical direction, the non-zero mean crosscorrelation algorithm can reduce the confusion between the characteristics of the overlying layer and the underlying layer, and improve the vertical resolution greatly. The C2 algorithm adopts multi-channel processing technology, which has good stability and is suitable for low signal-to-noise ratio data. It can also improve signalto-noise ratio and resolution by adjusting window size. At the same time, the secondgeneration algorithm (C2) is based on three-dimensional similarity, which provides a good method for calculating seismic coherence, and the coherent values obtained are clearer. By using an analysis window of any size, a balance is struck between the conflicting requirements of maximizing lateral resolution and improving SNR.

3.3.3 Technical Processes and Procedures Coherent data volume is generated by analyzing the local waveform changes in the direction of 3D seismic horizontal and horizontal survey lines and calculating the coherent values of a 3D seismic data point by point. After the generation of three-dimensional coherence, statistics are used to draw coherent spatial coherence changes, such as the reflection of section and lithology, from the incoherent and random in-phase axis. Due to the existence of faults and the change of lithology, the data of the seismic trace truncated by faults are suddenly interrupted, resulting in the weak coherent contour along the section. Stratigraphic structures produce similar interruptions that allow channel boundaries and delta deposits to be distinguished and interpreted. The basic workflow of coherent volume analysis can be carried out as follows: (1)

(2)

(3)

The 3D seismic data are added to the 3D visualization system, and the animation browsing is carried out to determine the research horizons and targets, and the geological tasks are defined. According to the geological task, coherence algorithm and processing parameter tests are carried out to determine the window and parameters of coherent volume processing, and coherent volume data is processed to form three-dimensional coherent data volume. The coherent data is added into the 3D visualization system to browse the animation. It can be browsed according to the main survey line, contact survey

3.3 Coherent Volume Technique

(4)

(5)

(6)

(7)

53

line, and time to find out the distribution location and distribution direction of faults and lithological strata, especially the development location and spatial distribution relationship of small faults and lithological bodies. Focus on observing and analyzing the time slice of seismic coherent data. The fault and lithological body are explained one by one on the time slice, and the three-dimensional space framework of the fault and lithological strata is established by the coherent volume. This process requires rich experience of interpreters and corresponding understanding of petroleum geology, but there is no concept of horizon, and it depends on the coherence of the data volume. It is therefore efficient and objective (Fig. 3.8). Combined with the interpretation of the 3D seismic profile, the rationality of the interpretation results is checked and verified. After the spatial fault interpretation is completed, the main seismic lines, the contact seismic lines, and the arbitrary seismic lines are cut to determine the positions of the sections, faults, and lithologies, and the rationality of the coherent volume interpretation and the conventional seismic amplitude volume interpretation is mutually verified. At the same time, the integrated display of coherent data volume and its interpretation results is combined with an arbitrary cutting display to grasp the spatial distribution of fault and lithological body and its consistency with geological understanding, so that the interpretation results are more reasonable and reliable. Automatic tracking of horizon seed points was carried out, and the interpretation results of horizon interpretation were input to interpret coherent data body fault plane combination and lithological body along with the reflection layer of wave peak or trough. The results of fault and lithology interpretation are exported to the conventional seismic data, and the corresponding description maps of structure, stratigraphic deposition, and reservoir evaluation are compiled.

Fig. 3.8 Horizontal slice of the coherent volume

54

3 Lithological Reservoir Exploration Technology

Two methods for improving SNR are suggested: One is suitable for structural analysis and the other is suitable for stratigraphic analysis. In the case of steep dip and nearly vertical faults, the time window can be enlarged and the number of coherent channels can be adjusted. The second method to improve the SNR is to extract the coherence generated by the whole data body along the interpretation horizon. The in-phase axis of a single reflector changes less than the horizontal time slice. Extracting coherent data corresponding to zero-crossing will greatly aggravate the influence of seismic noise in the coherent display. Most of the automatic picking software tracks the maximum peak or very small valley value of the seismic in-phase axis to maximize the signal-to-noise ratio. The leveling data are first used and then the coherence along the horizon is analyzed. To some extent, this method is more sensitive to automatic picking errors than the method of first calculating a coherent data body and then extracting the results. Because of the characteristics of lateral variation of thin clastic sandstone body, when applying coherent calculation, we must pay attention to which changes are caused by faults and which changes are caused by lithofacies changes. It is necessary to explain the rich experience of personnel and the understanding of oil and gas geological law. This technique, combined with dip and azimuth detection techniques, can be used to accelerate the display of the characteristics of overpass, retrograde, and unconformity. Variations in dip/azimuth and coherent features indicate folding, erosion, block slope shift, block rotation, or adjacent block descent. When facing sequence boundary extraction, dip/azimuth and coherent data can be used to analyze the product, fan body, and natural complex. In some cases, these dip/azimuth and coherent data are related to the structural characteristics of conventionally defined stratigraphic boundaries. In other cases, it can be used to accelerate the analysis of overpass, regressive and unconformable surfaces. The calculation of coherence and dip/azimuth data can be done independently before the interpretation of seismic data, or they can be gathered along the interpretation horizon to provide a general understanding and evaluation of the geological background. In this geological exploration model, dip/azimuth and coherent data volumes can pick up key dip and strike lines across major structures or interpretation horizons at an early stage of a project; In the interpretation model, these dip/azimuth and regular discontinuities can be associated with the strata and sequence boundaries, thus enabling us to obtain the progradation and overlap models of the internal structures of the three-dimensional bodies. The conventional seismic trace attributes such as the wavelet envelope, phase, frequency, and bandwidth of the reflection layer can be applied to the analysis of the local plane reflection layer by combining the extraction results of other seismic attributes along with the layer. Based on the well-established methods of geological statistics and cluster analysis, the basis of quantitative three-dimensional sequence stratigraphic analysis is formed. Because coherence calculations highlight discontinuities, coherence techniques can produce an unbiased image of the fault, Conventional time slicing is useful when interpreting faults along with a dip, but it is very difficult to accurately pick up faults parallel to strike on a time slicing. The coherence volume technique can distinguish faults at any orientation, which is helpful for more accurate and efficient interpretation.

3.3 Coherent Volume Technique

55

In the description of faults and hidden structures, the coherence time slice is much clearer than the conventional 3D seismic data slice, and the combination of the two is very effective for solving such problems as the spatial characteristics of horizons. When studying stratigraphic sequences, a better conception of transgression and regression can be obtained by clearly identifying sedimentary bodies such as beaches and deltas. Coherent properties can be used to describe reservoir boundaries and close with logging data for lithological feature projection. The analysis of coherent data volumes can accurately map the spatial location of faults and stratigraphic structures, which is of great help to well placement design. It can indicate the drilling trajectory and completion location of the well based on the identification of faults and fault zones. In the process of fault interpretation, the subjective judgment intervention and experience factors of interpreters are reduced, and the fault space automatic interpretation based on the distribution of irrelevant data zones is replaced, which not only improves the fault resolution greatly, but also improves the working efficiency greatly, and the interpretation results are more objective and reasonable. The following problems should be paid attention to when using coherent volume technology: First of all, seismic data should be evaluated. Only the fault interpreted by high signal-to-noise ratio data can be trusted, otherwise, it is difficult to interpret coherence. Irrelevant data anomalies may not always be faults, but may also be caused by lithological changes or other geological phenomena. Therefore, the specific analysis should be carried out in combination with a vertical profile when interpretation, instead of generalizing. Secondly, not all faults can be recognized in the coherent data volume. When the vertical drop is within an apparent period or an integer multiple of it, the faults in the coherent volume do not reflect. This is because when the vertical fault distance is the apparent period of the seismic wave or an integer multiple of the apparent period of the seismic wave, the wave peak on the reflection profile is continuous with the peak, and the coherence coefficient is very large, resulting in no reflection at the fault. These two points must be paid attention to when using coherent volume technology. For fractured reservoirs, which are highly heterogeneous, when coherence is applied, because the basic principle is to indirectly predict the fracture zone by identifying the fault, rather than directly predicting the fracture, inevitably, some misinformation is mixed in, so we must be careful to conclude from other sources and results, to find reliable information related to fracture development, to make the coherent explanation more in line with objective reality. Coherence processing and interpretation technology have been widely used in oil and gas exploration and development, which plays an important role in solving the geological situation of complex areas and the increasing amount of seismic data. It not only improves the interpretation efficiency and precision of seismic data, so that 3D seismic data can be fully applied, but also can highlight the discontinuity of data, quickly and accurately identify the fault and stratigraphic sedimentary characteristics, and directly describe the reservoir intuitively and accurately. Coherence processing and interpretation technology have become an indispensable and conventional technique in 3D seismic data interpretation (Fig. 3.9).

56

3 Lithological Reservoir Exploration Technology

Fig. 3.9 Coherent attribute plan. The linear very weak coherent zones (blue) show mainly faults; Strongly coherent zones (red) are dense limestone zones; The weak coherence zones (yellow-green) may be the reservoir development zones

3.4 3D Visualization and Virtual Reality Technology 3D visualization technology was born in the mid-1980s, a set of computer data processing, image display, and other leading edges of comprehensive technology. It is an image display tool to display, describe and understand many underground geological phenomena and characteristics by using 3D seismic data volume. It enables us to describe objective data, real-time processing and display, and timely and effective analysis results with real static or dynamic images in three-dimensional dynamic information, and to find out the geological and physical information contained in them and their interrelationships from a large number of scattered and complex data. It enables geologists, geophysicists, and reservoir engineers to “enter” into the data body that describes and characterizes the stratigraphic characteristics, to have a deeper understanding of the occurrence, development, and impact of various phenomena. Virtual reality technology is the technology that the scientific and engineering circles pay attention to in the 1990s. Its rise creates a new research field for the development of human–computer interaction interface, provides a new interface tool for the application of intelligent engineering, and provides a new description method for large-scale data visualization in petroleum exploration and development and other industries. The characteristic of this technology is that the computer creates an artificial virtual environment. This kind of virtual environment is a three-dimensional space formed by computer graphics, or another real environment is programmed into the computer to produce a realistic “virtual environment”, so that the user is in the visual sense of immersion in the virtual environment. The application of this technology has improved the way people use computers to process multi-project data, especially in the need to deal with a large amount of abstract data, which can be seen and felt in the virtual environment to complete.

3.4 3D Visualization and Virtual Reality Technology

57

3.4.1 Concepts and Principles of 3D Visualization At present, 3D visualization technology has been widely used in the field of petroleum exploration, such as geology and geophysics. Utilizing the advanced display technology of the infinite real-time graphic workstation, it can view the seismic data in all directions, not only as a quality control tool but also as a way to display and describe many geological features in the subsurface. The changes of complex structures or stratigraphic sedimentary reservoirs can be displayed on the graphic workstation in real-time. It provides a new effective means of data analysis and results in an expression for petroleum exploration researchers. With good practicability and effectiveness, it has become a basic technical method and tool in seismic data interpretation and reservoir description and research. It is widely used in all fields of geology and geophysics and its engineering, both large structures and fluid in pores can be displayed on visual tools. It is a means to describe and understand a model, a representation of the data body, not a simulation technique. The application of visualization technology enables us to use a large amount of data to check the continuity of the data, identify the authenticity of the data, and put forward useful abnormal information, which provides a useful tool for rapid analysis and understanding. At the same time, it is also a means of quality control and plays a bridge role for the better effect of data and multidisciplinary communication and collaboration. Because of its application, it saves a lot of time for seismic interpreters and improves the benefits of exploration and development. With the rapid development of computer and image display technology, 3D visualization technology has gradually become mature and has become the key technology in reservoir prediction and description methods. It can directly and quickly display and describe sedimentary facies zones such as channel, delta, alluvial fan, and thick sandstone reservoir distribution, and it is very suitable for rapid and accurate identification and description of the thick lithological body (generally greater than 15 meters). 3D visualization must be done on 3D seismic data volumes. This kind of 3D data can be either the conventional amplitude data volume, or the velocity or acoustic impedance data volume after inversion processing, or the data volume of various attributes after extraction processing, such as coherent volume, instantaneous amplitude, etc. Then, in a 3D visual environment, we comprehensively analyze various data and research results to determine exploration targets and design drilling well locations. The 3D visualization technology makes the prediction and description of seismic reservoirs reach a new realm, which realizes the substantial leap from 2 to 3D. 3D visualization is based on the concept of volume to display seismic data volume, various attribute bodies, three-dimensional spatial contact relations of faults, traps and reservoirs, spatial drilling tracks, etc. The basic principle is not complicated. Generally speaking, there are two types of visualization, namely, surface visualization and 3D volume visualization. Surface visualization is usually used to show seismic profiles, layers, faults, well curves, etc. 3D volume visualization, as the name implies, is based on the concept of volume to display such as seismic data volumes, various

58

3 Lithological Reservoir Exploration Technology

attributes, 3D reservoirs and traps, faults, spatial contact relationships, spatial good trajectories, etc. In general, visualization refers to 3D volume visualization in a general sense.

3.4.2 Basic Visualization Methods Seismic is the most important technical means in oil and gas exploration. Seismic data reflect the geophysical characteristics of various underground geologic bodies. Due to the discreteness, locality of data, and the limitation of technical means, the research on the underground geological structure, sedimentary characteristics, and reservoir description has been limited to two-dimensional space for a long time, even when we are faced with three-dimensional data. The 3D visualization technology is based on the basic structure of 3D seismic data. By changing the visualization parameters, the 3D data body presents the best display image, so that researchers can quickly and accurately detect and display the geological feature information in the 3D data body. It can greatly improve the ability to analyze and study the three-dimensional data volume and the spatial imagination and technical analysis mean of studying the pale structure, pale landform, and the restoration of the original basin. Data volume visualization methods: there are two different methods of visualization of data volume, namely image fixation or ray calculation and target fixation or combination. The first approach is to use ray processing calculations, where the visible parts are those rays passing through the volume from front to back, and the cumulative visual properties are those volume elements encountered. Compared with the second method, this method requires much more computation and has poor compatibility. Moreover, this method only makes the low precision approximate calculation for the center point of the volume element (the sampling point) that is not encountered. The second processing method is to calculate the level volume element in the data body. In the plane, the level volume elements are arranged from the back to the front. In the two-dimensional image, the volume elements synthesize the important layer interface, but also being able to compute layer by layer as images, and looking at it layer by layer on the screen. This approach allows you to see the internal organization of the data body, fixing or slowing down the viewing speed to better examine the intermediate results. There are five methods commonly used in the interpretation of visualization technology, including the automatic tracking of the inner layer of the volume element, the visualization of the target layer, the visualization of the isochronous volume, the visualization of the layer and the monitoring section, and the visualization of the multi-attribute volume.

3.4 3D Visualization and Virtual Reality Technology

59

3.4.3 Visualization of Full 3D Interpretation Technology Visualization can be simply summarized as two kinds of visualization techniques for different data display methods, that is, three-dimensional volume display and geometric volume or surface display. Geometry display is still not separated from the traditional 3D data 2D display method. It allows geological interpretation of the 3D data volume and rapid display of the 3D interpretation results and adjusts the interpretation accuracy according to the display results. The 3D volume display is a real 3D volume display method, which can visualize the seismic data volume. 3D volume display includes two concepts: using computer 3D graphic display technology to complete infinite diversity of image processing and analysis and research of 3D visual display results. It has the characteristics of simple and easy operation, intuitive display results, and clear geological characteristics. By processing the display parameters of the 3D data volume, including color, light, transparency, etc., the 3D visualization technology enables the data of various 3D seismic attributes to be quickly and rapidly displayed on the workstation, which can directly display the structure of the strata and the characteristics of the sedimentary reservoir. The researchers directly study and analyze the problems according to certain geological concepts in the three-dimensional space, and then describe and evaluate the structural, sedimentary, and reservoir characteristics of the exploration targets. It provides a new technical means of geological and geophysical integrated research for explorers. New technical means bring new research ideas. The basic workflow of 3D visualization technology (Fig. 3.10) can be decomposed into five key steps. Fig. 3.10 Basic workflow of 3D visualization

60

3.4.3.1

3 Lithological Reservoir Exploration Technology

Basic Work

The basic work includes the understanding of regional oil and gas geological law and the establishment of the 3D seismic data volume of the study block. It is necessary to analyze and evaluate the petroleum geological characteristics and data quality of the study area, to have a certain understanding of the characteristics of stratigraphic sedimentation distribution, the main controlling factors and rules of reservoirs, and the formation and evolution characteristics of structures of the target strata in the study area, and to analyze and collate zonal data, well logs, fault plugging properties, etc., to understand the early horizon calibration and preliminary interpretation results of seismic data. At the same time, the 3D visualization work needs to prepare the corresponding 3D volume data, including the target processing to maintain the amplitude of the data volume, acoustic impedance inversion data volume, processed a variety of attributes, etc. The visualization data volume is a 3D overlay seismic data volume file suitable for 3D visualization display.

3.4.3.2

Data Volume Browsing

In the beginning, it’s necessary to set the appropriate proportion of color bar, line, track, and time direction, and to adjust the size of the data volume in the window. Then, to start to scan and browse the data from three directions: inline, crossline, and time, to check the quality of the data, and to determine the research layer and target. Next, it’s for us to determine the target layer segment, and to conduct the overall browsing of the whole work area in different directions line by line, to have an overall understanding of the quality and characteristics of the data. At the same time, special targets (such as bright spots, flat points, etc.) are found and identified in the whole work area to determine the distribution zones of stratigraphic deposition and structural development (such as salt dome, large fan body, etc.). Once the research target area and layer segment are determined, the research target layer segment can be directly redefined and loaded into memory. If the amount of data in the working area is too large, data re-sampling can be carried out by an appropriate data extraction method. You can also nip and tuck away the shallow and deep layers to ensure the accuracy and speed of the 3D visualization.

3.4.3.3

Layer Interpretation

Based on browsing, first of all, the interpretation results of the previous layer are called out, and the interpretation of the target control layer can also be carried out by using the waveform automatic tracking technology. Automatic layer tracking technology has two good criteria for layer picking: The first is the establishment of “seed points”, through the layer calibration to determine the layer with better reflection characteristics, the establishment of “seed points".

3.4 3D Visualization and Virtual Reality Technology

61

Second, the selected layer can be tracked and compared horizontally in the working area, and the waveform characteristics and amplitude are required to have a good correlation. It can be said that the automatic tracking method requires a higher signalto-noise ratio of data (Yuan et al. 2005a, b). The interpretation of control layers is the basis of 3D visualization. It is very important for fault analysis, attribute analysis along with layers, and a detailed description of isochronous reservoir targets.

3.4.3.4

3D Visualization

The 3D target visualization includes three parts: stereoscopic display along with the layer, isochronous data volume perspective, and detailed description of target tracking. This is the core of 3D visualization techniques, which can reveal changes in structural development and sedimentary characteristics in detail that cannot be recognized by traditional interpretation methods. It is an important sign of the application level of 3D visualization technology to master and use this characteristic technique well (Fig. 3.11). The stereoscopic display along the layer is the visualization of the target layer. By adjusting the parameters such as color, illumination condition, and observation angle, the target layer of the research can be displayed on the graphic workstation with the best stereoscopic visual effect. This allows researchers to better understand the structural development characteristics and fault distribution. At the same time, horizon leveling can be carried out, which can not only further study the structural development and sedimentary characteristics of the strata but also use the locking window visualization method to more convenient and faster visualization analysis of the large dip angle strata. In the case of large dip angle and large tectonic fluctuation,

Fig. 3.11 3D Visualized display of acoustic impedance and interpretation layer rationality

62

3 Lithological Reservoir Exploration Technology

the stereoscopic display along the layer has obvious advantages in the superposition analysis of tectonic and seismic attributes, the rationality of fault and structure interpretation, and the analysis of the matching relationship between structure and reservoir. Isochronous data volume perspective is the visualization of isochronous objects. Transparency and color are important parameters that affect the visual effect of a data volume. According to the specific geological conditions and the quality of seismic data, the range of isochronous data volume and the visualization parameters are determined, and the 3D seismic attribute body is reasonably selected. Locking the time window up along the layer is a key step to establish the isochronous geological body using volume visualization technology. On this basis, the two key parameters of transparency and color can be adjusted accurately and flexibly, so that researchers can quickly browse the isochronous geological bodies in the three-dimensional visualization environment, understand the spatial distribution of faults, sedimentary bodies, and reservoirs, and then directly evaluate the geological parameters of oil and gas exploration, identify geological targets for further detailed study. Target tracking and picking is a fine visual interpretation and characterization of geological targets. It includes interpretation of plane and section, seed point tracking of seismic reservoir target, thickness estimation, etc.

3.4.3.5

Results Display

The main purpose of the results display is to display the geometric features and relative position relations of the three-dimensional visualization results, adjust the color and light parameters well, and display the visualization results perfectly and accurately. On the other hand, early results can be checked to help find problems that are not properly explained, and the evaluation scheme of the geological target body and well trajectory design are reconceived. With the rapid development of interpretation methods and techniques and computer science, higher requirements have been put forward for seismic interpreters. Seismic interpretation is the process of transforming seismic data into geological language. The interpreters are required to master and apply new geological and reservoir theories and concepts, new interpretation methods, and technologies, be familiar with 3D data, workstations, and flexible color display, and have rich geological imagination and logical thinking ability, to solve geological problems closer to the objective reality.

3.4.4 Virtual Reality Technology Virtual reality systems are also known as immersive visualization. It’s a leap forward in the latest qualitative development of 3D visualization. It immerses the observer, operator, and decision-maker into the multi-dimensional image of digitized

3.4 3D Visualization and Virtual Reality Technology

63

information, and improves the analysis and understanding of information utilizing perspective, exact spatial coordinates, and all-around human–computer interaction. The rapid development of computer hardware technology matched with the continuous improvement of computer software systems makes it possible to produce real-time animation of sound and images based on large data sets. The design of human–computer interaction systems continues to innovate, novel and practical input–output devices continue to enter the market, all these have laid a good foundation for the development of virtual reality systems. Just because virtual reality systems have an extremely wide range of application fields, people are full of expectation and interest in the rapid development of the broad application prospect of virtual reality systems.

3.4.4.1

Application of Virtual Reality Visualization System in Oil and Gas Exploration

Essentially, virtual reality is an advanced computer user interface. The core content of virtual reality technology is that it provides users with various intuitive and natural real-time sensing interaction means such as sight, hearing, and touch to facilitate the operation of users to the maximum extent, thereby reducing the burden of users and improving the working efficiency of the whole system. The purpose of dynamic environment modeling technology is to obtain the 3D data of the actual environment and establish the corresponding virtual environment model by using the 3D data obtained according to the application’s needs. The acquisition of 3D data can use CAD technology (regular environment), and more environments need to use non-contact visual modeling technology, the organic combination of the two can effectively improve the efficiency of data acquisition. From two-dimensional graphics, three-dimensional visualization to the modeling perspective of virtual reality, the natural sciences and engineering sciences have produced unprecedented tremendous promotion. Users can interact, model, and experience in an immersive virtual environment, including scene flight, multi-dimensional immersion, and target tracking. In a word, that is, personally on the scene. Three-dimensional visualization is a great step forward compared with twodimensional graphics, but it still uses two-dimensional screens to express threedimensional displayed information, which provides far less information and human–computer interaction means. International experts have analyzed that onedimensional or two-dimensional displays can only enable people to obtain 5% to 10% of the actual information at a certain moment, while large-scale virtual reality systems can enable the observer’s brain to obtain 80% of the actual information. Just because of this huge difference in information, virtual reality visualization technology itself has been realized in many industries, but with the rapid improvement of computer graphics and image processing functions in recent years and the sharp decline in the price of computers, people’s actual feeling to this technology is more and more from virtual to reality.

64

3 Lithological Reservoir Exploration Technology

Visualization techniques have been widely used in 3D seismic interpretation over the past decades. It has developed from the visualization of 3D seismic data to the visualization of multi-disciplinary and multi-domain 3D data bodies. With the development from a traditional visual environment to a real virtual reality environment, many geophysical software companies have developed their virtual reality system one after another. The rapid development of the computer and information industry has continuously promoted the technological progress of the petroleum industry and changed the working process and way of geoscience research. The introduction of the high-performance parallel computer enables the extensive application of timeconsuming seismic data processing methods (such as pre-stack depth migration, reverse-time migration, etc.), which significantly improves the seismic signal-tonoise ratio, resolution, and imaging quality of complex geological structures. This change in working mode greatly improves working efficiency, shortens the research cycle, and improves the level and reliability of the results. For oilfields from the perspective of international development of innovative technology and information, one of the key points is the rapid transformation from large computing centers to Internet distributed immersive visualization systems centers. In such an integrated computing, network, and virtual reality system, traditional data processing and interpretation analysis, traditional single-person workstation interpretation analysis will be transformed into multidisciplinary teams immersed in a variety of subsurface data volumes and geological reservoir models of visual, interactive three-dimensional images. Experts can use an acoustic immersive immersion or various inductive interaction directly to mobilize and analyze data, looking for trap reservoir, looking through the spatial distribution of its reservoir, calculating the reserves and error location, comparing various risk model of development, designing the well location and drilling trajectory, integrating seamlessly oil well production data and reservoir mathematical model results, finding the rest of the reservoir and hidden reservoir, and then reducing greatly the development cost. It is also important that in such an immersive visualization system center, technical experts and management decision-makers are no longer limited to the traditional way of reviewing report atlas and listening to multimedia presentations to synthesize decisions. Together with professionals, they immerse themselves in the work area traps and around the reservoir, using voice control or other interactions, and even going along the tracks to be laid, touching those reservoirs, going in there to check the results, looking at the modeling and simulation results of different approaches. This reduces risk and optimizes decision-making.

3.4.4.2

The Difference Between Oil and Gas Exploration Virtual Reality Visualization System and Conventional 3D Visualization

Visualization and modeling analysis with virtual reality systems are fundamentally different from conventional 3D visualization.

3.4 3D Visualization and Virtual Reality Technology

(1)

(2)

(3)

65

With more information, the data distribution enhanced by the threedimensional sense enables people to obtain more understanding and analysis and reasoning under intuitive induction, and the conclusions drawn from this are often non-existent under the condition of quasi-three-dimensional visualization. In the virtual reality environment, the realization of interactive operation brings users a sense of participation when they are in their place, which is a leap forward compared with the conventional three-dimensional visualization in which the mouse clicks on the menu or the interactive operation of graphics. Conventional 3D visualization is usually limited to the two-dimensional computer monitor screen, which is operated by a single person, while virtual reality system is usually realized by the large-size screen and projection space equal to or beyond the height of the human body, which enables many behaviors requiring team cooperation, such as oil field exploration and development decision-making, to be realized.

In a word, virtual reality visualization technology has four advantages: ➀ Virtual reality makes use of various immersion systems, such as a fully immersed cave, the new user interface also includes head and hand tracking, users manipulate and interact with data through the body’s natural movements, such as marching, pointing, and grasping, at run time, the user can be connected to the virtual world without console input, creating a very vivid working environment in which the user is completely focused on the task being performed; ➁ Powerful 3D visualization and highly immersive environment enable us to increase the understanding of complex 3D data and models; ➂ The size of the display area allows the whole team to collaborate in the data space; ➃ The new tracking user interface can improve the work efficiency and shorten the work process and cycle. Like other emerging science and technology, virtual reality technology is also a cross and integration product of many related disciplines. Its research content involves artificial intelligence, computer science, electronics, sensors, computer graphics, intelligent control, psychology, and so on. We must be soberly aware that although the technical potential in this field is huge and the application prospect is also very broad, there are still many unsolved theoretical problems and insuperable technical obstacles. In a complete virtual reality environment, the virtual reality system will act as a powerful system for multi-dimensional information processing and become a powerful tool for geologists and geophysicists to think and create as well as to deepen their existing concepts and acquire new concepts.

3.5 Seismic Inversion Technique Seismic inversion technology which is a new subject rising in the 1980s is developed with the continuous application of seismic technology in oilfield exploration and development. Compared with the traditional seismic exploration methods, the basic

66

3 Lithological Reservoir Exploration Technology

theory and principle of the method are the same, but the seismic inversion technology is different from the early stage of exploration in geological tasks, basic data involved and research methods adopted. In the early stage of exploration, the geological task of seismic work is to search for structures and find out traps, and its focus is on the fluctuation changes of the reflection interface. When the seismic information is mainly used for the travel of the reflection layer, the main work content is structural interpretation. With the deepening of exploration, the core of reservoir geophysics is to study reservoir heterogeneity. Its geological task is to predict the reservoir and its parameters, focusing on the lateral changes of reservoir characteristics, using more seismic information. The work is based on reservoir seismic inversion and information extraction to carry out seismic comprehensive interpretation. New seismic inversion techniques are emerging and developing rapidly. The main methods used include recursive inversion, logging constrained inversion, multiparameter lithology seismic inversion, etc. All of these inversion methods have different characteristics and applicability. Therefore, the most suitable inversion method should be selected according to the characteristics of the study area and the geological problems to be solved.

3.5.1 Concepts and Classification of Seismic Inversion 3.5.1.1

Seismic Inversion Concepts

Most of the exploration and development work of oil and gas fields is focused on the reservoir, but seismic exploration has only used the acoustic characteristics of rock strata to determine the lithology interface for a long time, which makes it difficult to combine the seismic and oilfield geology. To make the seismic data to be directly compared with drilling data, it is necessary to convert the conventional interfacial reflection profiles into rock logging profiles and convert the seismic data into a form that can be directly compared with drilling data. The technology to realize this conversion is seismic inversion. Different companies have different business software such as Seislog, Velog, Glog, Strata, Emerge, Jason, ISIS、SLIM, and PARM. Figure 3.12 is an example of a wedge-shaped sandstone body that is difficult to visualize on a conventional seismic profile and is mistaken for angular unconformity formation erosion. While in the seismic inversion section, it is clear that this is a high-velocity wedge-shaped sandstone body. The basic principles of seismic inversion techniques can be illustrated in Fig. 3.13. Different layers have different values of velocity and density, and the product of velocity and density is called acoustic impedance. Reflected waves can be generated as long as there is a difference in acoustic impedance between different layers. Assuming that the seismic trace on the seismic profile is the normal incident trace, that is, the seismic incident ray is perpendicular to the rock layer interface, then the normal incident reflection coefficient is calculated by the following equation:

3.5 Seismic Inversion Technique

67

Fig. 3.12 Seismic inversion techniques transform the interface seismic profile into a rock profile. a Wedge-shaped sandstone body; b conventional seismic profile; c seismic inversion profile

Ri =

ρi+1 νi+1 − ρi νi ρi+1 νi+1 + ρi νi

(3.1)

where Ri is the reflection coefficient of the bottom interface of the first layer, ρi is the density (g/cm3 ) of the i layer, νi is the velocity (m/s) of the i layer, ρi+1 is the density (g/cm3 ) of the i + 1 layer, and νi+1 is the velocity (m/s) of the i + 1 layer. Seismic waves go through a filtering system from excitation, propagation to reception. After a very sharp pulse passes through this filtering system, it becomes a continuous pulse waveform with a certain length, which is usually called a wavelet. This process is equivalent to replacing the reflectance “bar” at each reflectance position with a wavelet whose polarity and amplitude depend on the positive and negative values of the reflectance. Since rock layers are usually thin, the spacing between top and bottom reflection coefficients is much smaller than the wavelet length, reflected wavelets from different interfaces overlap each other and stack together to form

68

3 Lithological Reservoir Exploration Technology

Fig. 3.13 Formation and inversion principle of seismic records

a seismic record, this is a convolution process, which is used to make synthetic seismograms. Finally, due to the spherical divergence, absorption attenuation, and transmission loss of seismic waves, the actual field recorded on magnetic tape has strong amplitude in the shallow layer and weak amplitude in the deep layer, and the difference is too big to be explained. Only after amplitude attenuation compensation, seismic trace with a similar amplitude difference between deep and shallow layers can be obtained, that is, the result seen in the seismic profile. So that’s how the seismogram is formed. The production of synthetic seismograms or synthetic seismic profiles (seismic models) by this process is called forward modeling. Inversion is to estimate the inverse wavelet of a wavelet and use the inverse wavelet to carry on the convolution operation with the seismic trace, which is usually called deconvolution, to get the reflection coefficient. Then, put the reflection coefficient into the recursive formula derived from Eq. (3.1): ρi+1 νi+1 = ρi νi

1 + Ri 1 − Ri

(3.2)

Thus, the acoustic impedance of each layer can be calculated recursively layer by layer, which realizes the transformation from interfacial reflection profile to rock layer profile. The above is the basic principle of inversion, but also the traditional meaning of seismic inversion.

3.5 Seismic Inversion Technique

69

With the deepening of exploration and development of oil and gas fields, the requirements for seismic reservoir description technology are increasingly high, and the high-precision reservoir seismic inversion technology arises at the historical moment. In particular, the logging constrained seismic inversion, multi-parameter lithological seismic inversion, and reservoir characteristics reconstruction techniques are becoming mature. This is a model-based seismic inversion technique, which constructs a geological model of the underground reservoir under the constraints of known logging, drilling data, and geological laws, and then seeks the seismic response of the model. The seismic response of the reservoir model is compared with the actual seismic data, and the model parameters are modified repeatedly according to the difference between the model and the actual seismic data until the model response fits the seismic observation data in the best way. Finally, a suitable geological model reconstructed from the given observation data is obtained to better predict and describe the vertical and horizontal distribution and physical property changes of underground reservoirs. Seismic data contains abundant information on the lithology and physical property. Through seismic inversion, the seismic data of the interface type can be converted into the logging data of the rock type. It can be directly compared with the results of drilling and logging, and then the geological interpretation can be carried out by taking the rock layer as a unit. In this way, the advantages of intensive seismic data in the transverse direction can be fully utilized to study the spatial variation of the reservoir characteristics.

3.5.1.2

Types of Seismic Inversion Methods

Seismic inversion can be divided into pre-stack inversion and post-stack inversion from the perspective of the seismic data used in seismic inversion, it also can be divided into travel-time inversion and amplitude inversion according to the seismic information used in inversion, and it can be divided into structural inversion, acoustic impedance inversion, and reservoir parameter inversion, etc. according to the geological results of inversion (Wang et al. 2020). Pre-stack inversion mainly includes tomography based on travel time, AVO analysis based on amplitude, elastic impedance inversion (Fig. 3.14), etc. Post-stack inversion mainly includes structural analysis based on travel time and acoustic impedance inversion based on amplitude information. Seismic inversion has made great progress and many mature methods and techniques have been formed. According to the role played by logging data, it can be divided into four categories: Direct seismic inversion without good constraints, seismic inversion under logging control, combined logging and seismic inversion, and logging interpolation and extrapolation under seismic control are used in different stages of oil and gas exploration and development respectively. The logging constrained inversion is a joint logging seismic inversion method. Based on the seismic frequency band, the frequency range of the inversion acoustic impedance is extended to the low frequency and high frequency, respectively, by combining logging and geological model information.

70

3 Lithological Reservoir Exploration Technology

Fig. 3.14 Elastic impedance inversion

According to the different calculation methods and inversion ideas used in the inversion calculation, seismic inversion has many names, such as seismic lithology simulation, generalized linear inversion, wide-band constrained inversion, sparse pulse inversion, nonlinear inversion (neural network inversion, genetic algorithm inversion, simulated annealing inversion, etc.), single-channel inversion, multichannel inversion, etc. With the development of seismic inversion technology, new calculation methods and inversion ideas continue to emerge, and the names of seismic inversion will also emerge endlessly. However, in the final analysis, many overlapping seismic inversion methods can be roughly divided into three categories: recursive inversion, logging constrained inversion, and multi-parameter lithology seismic inversion.

3.5.2 Basic Principles and Application Conditions of Seismic Inversion 3.5.2.1

Recursive Inversion

The seismic inversion method based on reflection coefficient recursive calculation of formation acoustic impedance (velocity) is called recursive inversion. The key of recursive inversion is to estimate the formation reflection coefficient from seismic records and obtain the acoustic impedance information which is best consistent with the known drilling well.

3.5 Seismic Inversion Technique

71

In the recursive inversion method, the logging data mainly play the role of calibration and quality control and do not directly participate in the inversion operation. Therefore, recursive inversion is also called direct inversion, or seismic inversion under logging control.

Basic Principle Recursive inversion is a process of seismic data conversion, and the resolution, signalto-noise ratio, and reliability of the results are completely dependent on the quality of seismic data. Therefore, the seismic data used for inversion should have a wider frequency band, lower noise, relative amplitude retention, and accurate imaging. Logging data, especially acoustic and density logs, are the standard for comparison and interpretation of seismic lateral prediction. Before inversion processing, it should be carefully edited and corrected so that it can accurately reflect the physical characteristics of rock strata. The technical core of recursive inversion is to correctly estimate the formation reflection coefficient (or eliminate the influence of seismic wavelet) from seismic data, and its acoustic impedance is an important petrophysical parameter, which can be directly compared with drilling for reservoir lithology interpretation and physical property analysis. The typical methods are stratigraphic deconvolution, Sparse-spike Inversion and deconvolution, and phase correction in the frequency domain. The stratigraphic deconvolution method uses the least-square method to estimate the mathematically optimal wavelet or reflection coefficient based on existing logging data (acoustic and density) and borehole seismic records. The advantage of this method is that the underdetermined problem of the wavelet solution is changed into a deterministic problem. The inversion results which are most consistent with the good logging can be obtained in the range of the good logging section. There are two main limitations: Firstly, this method completely ignores logging errors and seismic noise. These factors, especially the objective existence of the former, make it more difficult to determine the wavelet. Secondly, the estimation of the formation deconvolution factor is the best approximation in the mathematical sense in the calculation window. The difference between the actual processing range and the time window has gone beyond the applicable scope of the method. Even at this wellpoint, it is no longer possible to obtain error minimization inversion results. It is not difficult to see that the main factors affecting the recursive inversion based on formation deconvolution are the quality of well logging data, the signal-to-noise ratio of seismic data, and the consistency of seismic noise. Sparse Spike Inversion is a recursive Inversion method based on Sparse pulse deconvolution. It mainly includes maximum likelihood deconvolution, L1 mode deconvolution, and minimum entropy deconvolution. Such methods aim at the problem of underdetermination of seismic records, the basic assumption that the formation reflection coefficient is composed of a series of strong axes superimposed on the Gaussian background is put forward, under these conditions, the subsurface

72

3 Lithological Reservoir Exploration Technology

strong reflection coefficient and seismic wavelet are estimated by different methods. The advantage of this method is that the reflection coefficient is calculated directly from seismic records without drilling data, and the recursive inversion is realized. The drawback is that it is difficult to get the final result consistent with the logging curve. The recursive inversion method based on frequency-domain deconvolution and phase correction avoids the underdetermined problem of calculating the wavelet or reflection coefficient, taking the coincidence degree between the borehole inversion results and the actual logging curve as the basic basis for parameter optimization, thus, the reliability and interpretation of the inversion data are guaranteed and it is the leading technique of recursive inversion.

Application Conditions The recursive inversion method based on the direct transformation of seismic data retains the basic characteristics of seismic reflection quite completely and does not have the problem of multiple solutions of the model-based method. It can reflect the spatial variation of lithofacies and lithology, and better reflect the physical property change of the reservoir under the condition of relatively stable lithology. The recursive inversion method has a wide range of applications. In the early stage of exploration, under the condition of less drilling, the sedimentary system of the strata was determined by petrographic analysis of inversion data. According to the characteristics of the reservoir revealed by drilling, the lateral prediction is carried out to determine the evaluation well location. At the early stage of development, under the condition of the thick reservoir, the recursive inversion data can provide more reliable structure, thickness, and physical property information for geological modeling, and optimize the design of the scheme. In the reservoir monitoring stage, the time-delay seismic inversion velocity difference analysis can help to determine the spatial variation of reservoir pressure and physical property, and then infer the oil and gas front. Due to the limitation of seismic frequency bandwidth, the resolution of recursive inversion data is relatively low, which cannot meet the needs of thin reservoir research.

3.5.2.2

Logging-Constrained Inversion

Under the geological condition of the thin reservoir, due to the limitation of seismic frequency bandwidth, the accuracy and resolution of the direct inversion method based on ordinary seismic resolution can’t meet the requirements of oilfield exploration and development. The logging constrained seismic inversion technology can supplement the limited seismic bandwidth with rich high-frequency information and complete low-frequency components of logging data, using known geological information and well logging data as constraints, high-resolution stratigraphic lithology

3.5 Seismic Inversion Technique

73

data can be calculated (Fig. 3.15). It provides a reliable basis for the fine description of reservoir depth, thickness, and physical property.

Basic Principle Logging constrained inversion is a model-based seismic inversion technology, and its approach is shown in Fig. 3.16. This method starts from the geological model and uses the model optimization iterative deflection algorithm, through constant modification and updating of the geological model, the forward synthetic seismic data of the model is in good agreement with the actual seismic data, the final model data is the inversion result.

Fig. 3.15 Logging constrained seismic inversion

Fig. 3.16 Flow diagram of logging constrained seismic inversion method

74

3 Lithological Reservoir Exploration Technology

The noise in the actual seismic records will strongly affect the inversion results. An alternative method is the conjugate gradient method. The main advantages of using the conjugate gradient method to solve the formation acoustic impedance are as follows: (1) The algorithm is accurate and stable; (2) The ill-posed problem in large matrix processing is avoided by not doing matrix inversion; (3) with a strong anti-noise ability; (4) It is easy to implement the constraint conditions in the solution process. Therefore, instead of a direct solution, the conjugate gradient method is used for logging constrained inversion. The formation acoustic impedance information is obtained by iteratively modifying the formation model and successive approximation (Yuan et al. 2005a, b).

Application Conditions The logging constrained seismic inversion technique combines the seismic and logging organically, breaks through the limitation of seismic resolution in the traditional sense, and theoretically can obtain the same resolution as the logging data. It is the key technology for the fine description of the reservoir in the development stage of the oil field. Multiple solutions are an inherent characteristic of logging constrained seismic inversion methods, which mainly depends on the degree of agreement between the initial model and the actual geological conditions. Under the same geological conditions, the more Wells are drilled, the more reliable the results are, the fewer Wells are drilled, and the greater the prediction error is. For the above reasons, logging constrained inversion is usually applied to areas with a large number of Wells. Generally, at least 6 Wells are required and the distribution of good points is relatively uniform, to ensure the quality of inversion results. If the lateral change of the geological body is more drastic, the number of Wells should be increased. In principle, the initial model established according to the known drilling and logging data should reflect the basic framework and geological law of the underground geological body correctly, and the lateral variation details of the geological body are mainly depicted by seismic inversion. Seismic data play two main roles in logging constrained seismic inversion: one is to provide horizon and fault information to guide the logging data interpolation and extrapolation to establish the initial model; the other is to constrain the geological model of the effective seismic frequency band to converge in the correct direction. The higher the resolution of seismic data is, the finer the horizon interpretation is likely to be, and the closer the initial model will be to the actual situation. Meanwhile, the range of effective control frequency bands will be larger, and the area of multiple solutions will be correspondingly reduced. Therefore, improving the resolution of seismic data is an important way to reduce multiplicity for itself. In the logging constrained seismic inversion method, improper emphasis on the two concepts is easy to cause misunderstanding. One is to emphasize that the resolution is up to a few meters, it sounds brilliant to the layman, but it has no practical meaning. Since the method itself starts and ends with the model and theoretically has

3.5 Seismic Inversion Technique

75

the same resolution as the good log, the question is how best to reduce the multiplicity. The second is to emphasize how similar the actual log and the good side inversion results are to show the reliability of the inversion, which can be misleading. The first step in the initial modeling process is logging data correction and then wavelet extraction. The logging model is made only after the synthetic record is most similar to the good bypass. In practice, the model near the well cannot be significantly modified, so this comparison is not practical.

3.5.2.3

Multi-parameter Lithology Seismic Inversion

Conventional seismic inversion techniques, whether recursive inversion based on seismic trace or logging constrained inversion based on the geological model, are guided by the theory of convolution model. In principle, they can only retrieve the information of acoustic impedance (including velocity and density). To a certain extent, these conventional seismic inversion techniques have solved many difficulties and bottlenecks in oil and gas exploration and development, such as reservoir prediction and description, reservoir heterogeneity research, oil, and gas-bearing prediction, etc. The key to the success of these conventional seismic inversion techniques is the difference in acoustic impedance between the reservoir and surrounding rock, but the geology of oil and gas exploration is very complex. In many cases there is little or no difference in the acoustic impedance between the reservoir and surrounding rock, it is difficult to distinguish the reservoir from the surrounding rock and to identify the favorable reservoir based on acoustic impedance alone. It has been found that some electrical curves, such as natural gamma, spontaneous potential, and resistivity, are sometimes more sensitive to lithology differentiation. Can we find a way to directly invert lithology seism by linking seismic information to other lithology curves? Scientists have put forward a variety of lithology seismic inversion methods after continuous exploration. Multi-parameter lithology seismic inversion is to select several reservoir parameters which are sensitive to lithology differentiation and establish a relationship with seismic information based on analyzing lithological and electrical characteristics of the reservoir. The whole lithology parameter data body is predicted by nonlinear function mapping and neural network technology to achieve the purpose of lithology identification and reservoir prediction.

Basic Principle The commonly used software systems of multi-parameter lithology seismic inversion are primarily InverMod by Jason and Emerge by Hampson Russell. Jason’s InverMod method used principal component analysis to create a nonlinear functional mapping between seismic characteristics and reservoir lithology parameters to perform overall lithology inversion of seismic data. Hampson-Russel’s Emerge is to establish the

76

3 Lithological Reservoir Exploration Technology

relationship between various seismic attributes and reservoir lithological parameters through neural network technology, to achieve the purpose of overall prediction of lithological parameter data. Jason’s InverMod is used as an example to introduce the basic principle and application conditions. InverMod multi-parameter lithological seismic inversion is a model-based multiparameter lithology seismic inversion technique. It obtains reservoir parameters from geological, seismic, well logging, drilling, drill core, cuttings, logging, outcrop, and other types of data, and establishes 3D reservoir parameter models. Based on the acoustic, density, resistivity, SP, natural gamma, porosity, permeability, and oil saturation data, this software overcomes the shortcomings of conventional seismic inversion techniques, which can only obtain acoustic, density, and acoustic impedance. Therefore, InverMod can effectively perform reservoir prediction and description. The processing flow of multi-parameter lithology seismic inversion of the InverMod system is shown in Fig. 3.17. The main technical steps include (1) geological modeling, (2) principal component analysis, and (3) model estimation.

Application Conditions The InverMod multi-parameter lithology seismic inversion technique is used to establish the relationship between seismic information and reservoir parameters under the control of the geological model by using principal component analysis and model estimation techniques. This characteristic makes it break through the concept of the convolution model in the traditional sense, so it can not only reflect the conventional reservoir parameter curves such as acoustic wave, density, and acoustic impedance

Fig. 3.17 The flow chart of multi-parameter lithology seismic inversion in InverMod

3.5 Seismic Inversion Technique

77

but also reflect a variety of lithology geological information such as gamma, spontaneous potential, resistivity, porosity, permeability, and oil saturation. This technique broadens the field of combining well-logging information with seismic data and is a further development of the inversion technique. The InverMod multi-parameter lithology seismic inversion technique is suitable for seismic prediction of all kinds of complex reservoirs, especially for thin reservoirs with obvious differences in one or several lithology parameters, and the accuracy of inversion results is high. However, the physical significance of the inversion algorithm involved in this technique is not very clear, and the exact relationship between multi-parameter lithological geological information and seismic reflection information is not studied enough, which affects the popularization and application of this technique. In practical application, it should also be noted that since the multi-parameter lithological seismic inversion is based on the statistical analysis of well lithological parameters and seismic information, the reliability of the inversion results is naturally related to the good pattern density. Usually, there should be more than 6 Wells in the research area, and there are too few well points, so the results of the principal component analysis are not universal, and it is easy to produce the false impression of generalization.

3.5.3 Comparison of Different Seismic Inversion Methods 3.5.3.1

Recursive Inversion

Among the recursive inversion methods based on frequency domain deconvolution and phase correction, Seislog is a typical software, and its processing flow chart is shown in Fig. 3.18.

Fig. 3.18 Recursive inversion processing flow chart

78

3 Lithological Reservoir Exploration Technology

The key technologies include deconvolution in the frequency domain to recover the amplitude spectrum of formation reflection coefficient, phase correction to make the good side inversion track fit the good log best, and low-frequency modeling to reflect the variation trend of formation acoustic impedance.

Frequency Domain Deconvolution and Phase Correction According to the convolution model of seismic records, it is necessary to eliminate the influence of seismic wavelets to extract formation reflection coefficient from seismic records. The process in Fig. 3.18 is realized through two steps of frequency domain deconvolution and phase correction. The function of frequency domain deconvolution is to eliminate the influence of wavelet amplitude spectrum so that seismic records can best approximate the amplitude spectrum of formation reflection coefficient within the effective frequency band. In Fig. 3.18, the approximate method of constant phase correction is adopted for the elimination of wavelet phase, and the optimal phase is usually determined by constant phase scanning, that is, the optimal coincidence between the relative velocity of seismic inversion and the actual logging relative velocity is taken as the criterion to determine. In Fig. 3.19, the left side shows the application effect of phase scanning and optimal phase correction. It is not difficult to see that the optimal phase correction significantly improves the similarity between the inversion trace and the actual logging, which indicates that the phase correction can effectively eliminate the influence of the wavelet phase spectrum.

Fig. 3.19 Recursive inversion quality control example

3.5 Seismic Inversion Technique

79

Establishing a Low-Frequency Model Due to the limitation of the seismic acquisition system, the seismic direct inversion results do not contain low-frequency components below 10 Hz, which need to be compensated by extracting from other data. There are usually three ways to introduce low-frequency information: actual acoustic logging filtering, seismic velocity analysis, and geological modeling. Considering the sparsity of actual drilling, the subjectivity of the geological model, and the rough nature of seismic velocity, the low-frequency information model is usually established by combining the three methods. On the right side of Fig. 3.19 is a comparison example of a depth domain, the correct relationship in the depth domain verifies the accuracy of the low-frequency model, and the reservoir details correspond to each other, which indicate the effect of phase calibration.

3.5.3.2

Logging Constrained Inversion

In essence, logging constrained seismic inversion is a joint inversion of seismicity and logging. The low and high-frequency information of the result comes from logging data, while the structural characteristics and middle-frequency range depend on seismic data. Multi-solution is an inherent characteristic of logging constrained seismic inversion, that is, the information outside the effective frequency band does not affect the final result of the synthetic seismic data. The key to reducing the multisolution problem of logging constrained seismic inversion is to establish a correct initial model. The accuracy of logging constrained seismic inversion results not only depends on the geological characteristics of the research target, the number of wells, the well location distribution, the resolution, and the signal-to-noise ratio of seismic data but also depends on the fine degree of processing.

3.5.3.3

Multi-parameter Lithology Seismic Inversion

Since multi-parameter lithology seismic inversion is directly aimed at lithology prediction and interpretation, the selection of the lithology characteristic curve is very important. The usual method is to collect the logging data of all wells in the study area as far as possible, analyze and study the corresponding relationship between each kind of electric survey curve and lithology, find out one or several curves that are sensitive to lithology differentiation, and take them as the target parameters of lithology seismic inversion for subsequent feature analysis and inversion processing.

80

3 Lithological Reservoir Exploration Technology

3.5.4 Reconstruction Method of Reservoir Characteristics Acoustic impedance inversion, which combines seismic, well logging, and geological information is not only easy to connect seismic data with well-logging data, but also can effectively study the changes of reservoir physical property parameters. It has become an indispensable key technology in reservoir prediction. Among various acoustic impedance inversion methods, the purpose of reservoir prediction is how to correctly apply these methods and get the most reasonable and reliable results in reservoir prediction. The key to improving the accuracy of reservoir prediction is to process the logging data effectively and get the characteristic curve that can reflect the variation law of the inversion target.

3.5.4.1

Method Proposal and Concept

The density values of adjacent layers in a certain range of continuous sedimentary strata have a small difference, and the difference of reflection coefficient is mainly reflected in the difference of velocity between adjacent layers. The larger the difference of velocity between adjacent layers is, the greater the reflection coefficient is and the higher the longitudinal resolution is. The accuracy of lithology identification is affected by the characteristic curve used in constrained inversion, and the vertical resolution is the key to reservoir prediction in lithological reservoir exploration. According to the status of lithological exploration and reservoir prediction technology, the acoustic moveout curve is the essential basic data for conventional impedance inversion based on logging constraints. When the resolution of the acoustic moveout curve for the reservoir is not high, the vertical resolution accuracy of conventional constrained inversion is low. This is mainly because the response of acoustic moveout logging used for constrained inversion can not reflect the lithological characteristics and variation law of formation well, and can also not match well with seismic information. With the continuous improvement of oil and gas exploration and development, more and more attention has been paid to the research and evaluation of unconventional reservoirs (such as volcanic rocks, metamorphic rocks, weathering crust, mudstone, conglomerate, etc.) and thin interbeds of sandstone and mudstone, which are also the main reservoirs to replace reserves in the future. These reservoirs are characterized by complex formation conditions, many controlling factors of reservoir performance, diverse reservoir space types, and strong reservoir heterogeneity. Therefore, it is difficult for acoustic moveout to meet the requirements of high precision reservoir prediction, the conventional acoustic impedance inversion technology constrained by acoustic moveout logging curve is not accurate enough to predict unconventional reservoirs, and the traditional reservoir prediction technology is faced with serious challenges.

3.5 Seismic Inversion Technique

81

In the research of long-term reservoir prediction, a set of methods to improve the vertical resolution of reservoir prediction by reconstructing reservoir characteristic curves is explored by analyzing the correspondence between different logging responses and lithology and seismic information. Reservoir characteristic curve reconstruction is based on the comprehensive study of geology, well logging, and seismic, aiming at specific geological problems and inversion targets, based on petrophysics, it selects and rebuilds the curve that can reflect the reservoir characteristics from various good logging curves. In theory, the logging curves of spontaneous potential, natural gamma, compensated neutron, density, and resistivity in the conventional logging series can be used to identify the reservoir and establish a good correlation with the acoustic moveout, which can be converted into the pseudo-acoustic moveout curve by mathematical statistics method to realize the reconstruction of the reservoir characteristic curve. The basis of reservoir characteristic curve reconstruction is that the attributes of different geological bodies are reflected differently in different physical fields, the same attribute of the same geological body is correlated in different physical fields, and the different attributes of the same geological body have some emphasis in different physical fields. Reservoir characteristic curve reconstruction is to obtain a characteristic curve that can better highlight the reservoir resolution by integrating relevant information that is beneficial to reservoir prediction according to the reservoir prediction target. The prerequisite of reservoir characteristic curve reconstruction is that the original acoustic moveout can not well reflect the change law of formation lithology (or inversion target), while other logging responses with better correlation with acoustic moveout can better indicate the change law of formation lithology (or inversion target). At the same time, the reconstruction method of reservoir characteristic curve has some limitations, so the reconstruction method must be optimized according to the regional geological background, logging response characteristics, and their mutual relations, as well as the indication degree of lithology (or inversion target) change law.

3.5.4.2

Fundamental Principle of Reservoir Characteristic Curve Reconstruction

The key to reservoir prediction is to establish a geological model with high precision and in line with production requirements, and the correct application of the good logging curve is very important. According to the basic principle of reservoir characteristic curve reconstruction, the steps and flow of reservoir characteristic curve reconstruction are put forward.

82

3 Lithological Reservoir Exploration Technology

Basic Principle The key to the reconstruction principle of reservoir characteristic curve is that reference curve and target curve must have similar (related) logging response characteristics to the same physical phenomenon, the only difference is that the reference curve reflects the reservoir better than the target curve, at the same time, there must be a good correlation between the two curves, and the greater the correlation, the more reliable the reconstructed curves are. In general, for a single sandstone body of 15 m, the results obtained by conventional impedance logging constrained inversion are reliable, while for a sandstone body of 5 m, the results obtained by conventional impedance logging constrained inversion are usually unreliable (Fig. 3.20). Reliable inversion results can be obtained by reservoir characteristic curve reconstruction. There are certain differences in the frequency of formation information contained in well logging data and seismic data. The good logging data contains low, medium, and high-frequency information, and the seismic data only has the medium frequency information is better, while the reconstructed reservoir characteristic curve should contain low, medium, and high-frequency information. Seismic data are the main data of inversion. With the seismic intermediate frequency information as the main information and the low and high-frequency information of logging as the compensation, the reservoir characteristic curve with good low, medium, and high-frequency information is reconstructed (Fig. 3.21), to establish the relationship between the reconstructed curve which can best reflect the reservoir characteristics and the seismic. Different good logging responses reflect different physical properties of the same formation from different angles, and all have their inherent differences. The foundation of reservoir characteristic reconstruction is the correlation between the logging curve and the inversion target. The difference reflects the different physical properties, but different logging responses are unified within a certain range. The acoustic curve with the same dimension is reconstructed by using the logging response related to lithology, and the reservoir characteristic curve is reconstructed by combining with

Fig. 3.20 Relationship between sandstone thickness and resolution reliability of conventional impedance constrained inversion

3.5 Seismic Inversion Technique

83

Fig. 3.21 Relationship between frequency and resolution of logging, seismic and reconstructed curves

Fig. 3.22 Schematic diagram of reservoir characteristic curve reconstruction

the formation acoustic low-frequency model so that it can reflect not only the characteristics of formation velocity and acoustic impedance but also the lithological difference, thus establishing a better relationship between reservoir characteristics and seism (Fig. 3.22). Two principles must be followed in the reconstruction of reservoir characteristic curves: (1) The multidisciplinary integrated, because of the reservoir geological characteristics in the study area, based on deep geological analysis, guided by the rock physics, make full use of lithology, electrical and radioactive logging information and acoustic properties, the relationship between reservoir characteristic curve reconstruction, and makes this curve has obvious reservoir characteristics, easy to identify; (2) Ensure that the reconstructed reservoir characteristic curve matches the synthetic record and well bypass, to ensure that the data body obtained after inversion has a higher vertical resolution and improve the accuracy of reservoir prediction. What is described above is only the basic principle of reservoir characteristic curve reconstruction in reservoir prediction. The reconstruction method of reservoir

84

3 Lithological Reservoir Exploration Technology

characteristic curve is different for different areas and reservoir types and the reconstruction of reservoir characteristic curve emphasizes the multidisciplinary synthesis and the pertinence of geological problems.

Work Procedure and Process Reservoir characteristic curve reconstruction includes 6 steps: (1) Analysis of regional geological background and logging characteristics; (2) Preprocessing and standardization of logging curves; (3) Establishing the low-frequency model of an acoustic wave changing with depth; (4) Correlation analysis of logging curves; (5) Reservoir characteristic curve reconstruction method optimization; (6) Reservoir inversion effect inspection and error analysis, model adjustment and curve reconstruction results output (Huang et al. 2005). According to the principle, steps, and operational feasibility of reservoir characteristic curve reconstruction in practical application, the process that should be followed in reservoir characteristic curve reconstruction is summarized (Fig. 3.23).

Fig. 3.23 Flow diagram of reservoir feature reconstruction

3.5 Seismic Inversion Technique

3.5.4.3

85

Reservoir Characteristic Curve Reconstruction Model

Based on correlation analysis of well logging curves, there are several basic methods to reconstruct reservoir characteristic curves by selecting reasonable models.

The Model Based on Original Good Logs Reconstructing Reservoir Characteristic Curve Different models are used to reconstruct the reservoir characteristic curves with different original logging curves. This section mainly introduces the reservoir reconstruction model suitable for sandstone-mudstone profile and fractured reservoir. (1)

(2)

The model suitable for sandstone-mudstone profile reconstruction: ➀ spontaneous potential, Spontaneous potential reflects reservoir permeability, after preprocessing and standardization, SP amplitude can reflect the quality of the reservoir, and the acoustic moveout of the reservoir is inversely proportional to SP (negative anomaly of the reservoir). The reconstruction of the reservoir characteristic curve based on spontaneous potential can be used in reservoir prediction to evaluate reservoir thickness and quality. ➁ Natural gamma, Natural gamma reflects lithology or reservoir permeability through formation radioactivity, after preprocessing and standardization, the natural gamma amplitude can reflect the quality of the reservoir under certain conditions, and the acoustic moveout is inversely proportional to the natural gamma. The reconstruction of reservoir characteristic curve by natural gamma is suitable for the areas with good reservoir resolution by natural gamma and is mainly used in reservoir prediction to evaluate reservoir thickness. ➂ Resistivity, Resistivity mainly reflects the electrical conductivity of rocks and fluids, after preprocessing and standardization, the resistivity amplitude can reflect the quality of the reservoir. In a certain range, the acoustic moveout of the reservoir is directly proportional to the resistivity. The reconstruction of the reservoir characteristic curve based on resistivity is suitable for the area with good reservoir resolution and is mainly used in reservoir prediction to evaluate reservoir oil and gas. The model suitable for fracture reservoir reconstruction: in fractured formations, there is a close relationship between fracture porosity and deep and shallow resistivity in fractured reservoirs, and there is also an obvious anomaly

86

3 Lithological Reservoir Exploration Technology

Fig. 3.24 Logging response and reconstruction curve characteristics

in the acoustic moveout of fractures (Fig. 3.24). Shallow resistivity and deep resistivity are well correlated with acoustic moveout, and reservoir characteristic curves can be reconstructed by deep and shallow resistivity logging. The Model Based on Porosity Logs Reconstructing Reservoir Characteristic Curve This model is suitable for the situation that the porosity can be obtained well by using the compensated neutron and density logs, but the accuracy of the porosity obtained by using the acoustic moveout is low. First of all, it should be emphasized that porosity here refers to effective porosity obtained by using acoustic moveout. There is a close correlation between acoustic moveout and effective porosity. When there are compensated neutron and density logs in the logging series, the effective porosity of the reservoir can be accurately determined by using compensated neutron and density logs through correction of oil and gas, lithology, etc. This porosity excludes the influence of lithology and oil and gas and can be used to classify reservoir thickness and evaluate reservoir quality.

3.5 Seismic Inversion Technique

87

The pseudo-acoustic moveout curve was reconstructed using effective porosity. This method is mainly used in reservoir prediction to evaluate gas-bearing reservoirs or lithology.

3.5.5 Seismic Inversion and Sequence Stratigraphy Seismic inversion technology and sequence stratigraphy technology should be combined to give full play to their respective advantages. The seismic inversion should be carried out under the control of the high-resolution sequence stratigraphic framework so that the inversion results can truly reflect the actual stratigraphic sedimentary structure. In recursive inversion and logging constrained inversion, lowfrequency models and initial models are usually built under the control of a large set of seismic geological layers, which are derived from seismic tectonic interpretation. Some seismic reflection interfaces with strong seismic reflection characteristics are usually selected in the selection of layers. Some of these seismic tectonic interfaces are sedimentary sequence interfaces, while some are not consistent with the sedimentary sequence interfaces. As a result, the model built under their control is not consistent with the actual stratigraphic sequence, which will affect the accuracy of reservoir seismic inversion and prediction. Sequence stratigraphy analysis technology plays an important role in oil and gas exploration and development. When lithological trap exploration is carried out, the isochronous framework of sequence stratigraphy and system tract of the whole basin should be established first, the temporal and spatial distribution and evolution characteristics of the sedimentary system should be clear, and the sequence stratigraphic background of the whole basin should be clear, according to the general law of lithological trap development, the favorable areas and layers for its development are determined. Subsequently, the isochronous stratigraphic framework of high-resolution sequences should be established for the main target strata in the key areas, and the lithological distribution and stratigraphic superposition relations of high-frequency sequences should be clarified. Then, the accurate stratigraphic sedimentary model is established in the whole area, which lays a solid foundation for seismic inversion and reservoir prediction and makes the reservoir prediction results more accurate and reliable. Furthermore, the lithological traps can be identified by combining the spatial distribution and configuration of the reservoir, cap, and lateral sealing layers. Figure 3.25 is a well-tie seismic profile with one well crossing a 3D seismic work area in Bohai Bay Basin. Only two large sequence interfaces are explained and seismic constrained inversion is carried out. The inversion section prediction shows a large set of sandstone and mudstone, but the resolution is not high. The reddish-yellow are the predicted sandstones, the turquoise represents mudstone, the curve used in the well is the GR curve, the high GR value is mudstone, the good pillars are blue, the

88

3 Lithological Reservoir Exploration Technology

Fig. 3.25 Inversion results of well-tie seismic profile (constrained by model data under the control of large suite of layers)

Fig. 3.26 Inversion results of well-tie seismic profile (constrained by model data under the control of high-resolution sequence stratigraphic framework)

low GR value is sandstone, the good pillar is reddish yellow. Figure 3.26 is a seismic profile with detailed sequence stratigraphic framework interpretation, it is observed that due to the control of high-resolution sequence stratigraphic framework, the vertical resolution of seismic inversion profile has been greatly improved. Especially, when compared with the sandstone and mudstone curve of the well, it has a high coincidence degree, and multiple single sandstone bodies are also predicted, which is beneficial to the later characterization of a single sandstone body. Figure 3.27

Fig. 3.27 The superposition shows of well-tie seismic profile inversion results and seismic data

3.5 Seismic Inversion Technique

89

is the superposition display of reservoir seismic inversion results and seismic data under the control of a high-resolution sequence stratigraphic framework. It reflects the literal fidelity of the prediction to the seismic data, this kind of prediction result is reliable and accurate.

References Fu Z, Wang H et al (2005) Sequence stratigraphy research status and progress. Prog Explor Geophys 28(5):320–324 Huang X, Wang H et al (2004) The desert static correction method and application in Tarim basin. Geoscience 18(1):133–137 Huang X, Wang H et al (2005) Based on the data consistency prediction and suppression of freesurface multiples—theory research and preprocessing. Chin J Geophys 48(1):173–180 Li M et al (2000) The study of fractured carbonate reservoir prediction in an area in Tarim Basin. Geophys Prospect Pet 39(2):24–35 Li M et al (2002) Identifying and predicting the technology of deep volcanic gas reservoirs in the north of Songliao basin. Oil Geophys Prospect 37(5):477–484 Li M et al (2005) Methods and applications of geophysical exploration for lithological and stratigraphic reservoirs. Beijing, Petroleum Industry Press, pp 35–40 Li M et al (2014) Geophysical exploration technology: applications in lithological and stratigraphic reservoirs. (1st edn) Hardcover, Elsevier, 480 pages Wang H (2004a) Present situation and the prospect of key technology for oil geophysical exploration. China Pet Explor 9(1):41–51 Wang H (2004b) Present situation and the countermeasure analysis of subtle reservoir exploration. Oil Geophys Prospect 39(6):739–744. https://doi.org/10.13810/j.cnki.issn.1000-7210. 2004.06.023 Wang H (2005) Geophysical exploration technology and application for lithological reservoirs: a case study of southern Songliao Basin. China University of Geosciences (Beijing) Wang H (2008) The sequence constraint prediction technology and application for lithological reservoir. China Pet Explor 13(4):36–42 Wang Y, Wang H et al (2020) Pre-stack elastic inversion for “geomechanical sweet-spot” prediction in Longhupao region of Songliao Basin in China. Geophys Prospect Pet 59(1):114–121. https:// doi.org/10.3969/j.cnki.issn.1000-1441.2020.01.013 Wen S, Wang H et al (2003) The zero-offset seismic trace fitting technique at the time-frequency domain and the application of high-resolution data processing. Oil Geophys Prospect 38(5):512– 516. https://doi.org/10.13810/j.cnki.issn.1000-7210.2003.05.009 Yuan Y, Wang H et al (2005a) Introduction to improving the signal-to-noise ratio of seismic data processing technology. Oil Geophys Prospect 40(2):168–171. https://doi.org/10.13810/j.cnki. issn.1000-7210.2005.02.016 Yuan Y, Wang H et al (2005b) Denoising technology in the application of seismic data processing. J East China Inst Technol (Natural Science Edition) 28(1):12–16 Zou C et al (2002) New seismic techniques for exploration and development of oil and gas. Petroleum Industry Press, Beijing Zou C, Li M et al (2004a) Recognition technique and application of structure-lithology pool in the south of Songliao Basin. Acta Petr Sinica 25(3):32–36 Zou C, Li M et al (2004b) Depositional sequences and forming conditions of the Cretaceous stratigraphic-lithologic reservoirs in the Quantou-Nengjiang Formations, South Songliao Basin. Pet Explor Dev 4(2):14–17

Chapter 4

Realization and Application of Geophysical Technologies for Lithological Reservoirs

The prediction technology of complex reservoirs based on 3D seismic has been greatly developed, which plays an important role in the seismic exploration of lithological reservoirs. Because of the complexity of the lithological reservoir geological conditions, the diversity of trap conditions, Mutability of lithological space, and Heterogeneity within the reservoir, it is necessary to adopt the “high resolution, high signal-to-noise ratio and high fidelity” 3D acquisition, processing, and interpretation method for seismic exploration of lithological reservoirs. 3D seismic acquisition and processing, especially the integrated interpretation technology with visualization as the core, are developing rapidly, It provides a powerful research tool for the comprehensive analysis of seismic data, geological data, drilling data, and well logging data. Especially the organic combination of visualization technology and seismic attribute analysis, coherence, inversion, and other technologies, greatly expand the scope of the interpreters and improve the interpreter’s knowledge of subsurface geology, effectively improving the exploration efficiency of lithological reservoirs.

4.1 Realization of Geophysical Technologies for Lithological Reservoirs At present, the advanced and effective geophysical technologies for lithological reservoirs proved by practice are being vigorously promoted in all oil fields in combination with the exploration practice. The main technologies are as follows. Full 3D visualization construction interpretation: automatic horizon tracking technology, coherent volume technology. Target detection and recognition: frequency decomposition technology, formation recursive inversion technology, neural network recognition technology, seismic attribute decomposition technology.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 H. Wang et al., Practical Geophysical Technology and Application for Lithological Reservoirs, SpringerBriefs in Petroleum Geoscience & Engineering, https://doi.org/10.1007/978-981-16-4197-8_4

91

92

4 Realization and Application of Geophysical Technologies …

Sandstone body description and identification: well logging constrained seismic inversion technology, sequence framework geological model building technology, high-resolution 3D reservoir characteristic curve reconstruction technology, seismic multi-attribute extraction, and description discrimination technology. Effective trap identification: formation dip-angle detection technology, multiparameter superposition discrimination technology, visualization technology (Wang 2005). Other new technologies under development include: (1)

(2)

Pre-stack inversion will be the main technique in lithological reservoir exploration. Post-stack impedance inversion, as a leading technique for reservoir prediction, has played an important role in lithological reservoir exploration and reservoir description since its mass industrial application in the 1980s. However, post-stack inversion is limited because only acoustic impedance information can be retrieved. To obtain more elastic information of rock, the inversion content has been extended from post-stack to pre-stack, and from acoustic impedance inversion to elastic parameter inversion. Now the information getting from seismic prestack inversion is very rich, such as P-wave velocity, shear wave velocity, and the wave velocity ratio, density, lame constant, bulk modulus, and shear modulus. This information enhances the accuracy of prediction of lithology, thickness, and fluid, and the difference of different lithology and stratigraphy, which is more conducive to the reliable identification of reservoirs. Multi-wave and multi-component seismic technology will play an important role in lithological reservoir exploration. Fluid prediction is one of the most important and difficult problems in the exploration of lithostratigraphic reservoirs. Due to the addition of new useful information about shear wave and converted wave, the ability to identify lithology and fluid in complex reservoirs is greatly improved by multi-wave and multi-component seismic. Compared with simple P-wave exploration, it has great advantages in improving trap imaging, lithology prediction, fluid identification, fracture detection, anisotropy, and so on (He et al. 2021).

The key points of geophysical exploration of lithological reservoirs are high precision 3D seismic acquisition, fine sequence analysis, seismic interpretation and inversion, accurate identification, and objective description of sandstone bodies. Based on studying the distribution of macroscopic lithological traps in the basin, Using the existing abundant well logging and logging data as well as high resolution seismic and 3D seismic data, the high-resolution sequence stratigraphic analysis of lithological reservoir block is carried out. Taking refining sequence as the starting point of the research work, on this basis, the sedimentary microfacies and the prediction of the distribution of sandstone body are carried out to determine the pinch-out line of sandstone and the distribution of lithological reservoirs. Based on the exploration practice of lithological reservoirs, the relatively high abundance reservoir blocks are optimized utilizing reservoir micro evaluation and well-constrained seismic inversion. In areas and horizons with low exploration levels, comprehensive stratigraphic

4.1 Realization of Geophysical Technologies …

93

and sedimentary studies should be strengthened to search for lithological traps (Wang 2008). Specific research contents are as follows: (1) Identification and division of highresolution sequence interfaces; (2) High-resolution seismic data sequence interpretation and well-constrained seismic reservoir inversion; (3) Drilling and logging highresolution sequence stratigraphic analysis and correlation; (4) Distribution characteristics of sandstone bodies and sedimentary microfacies in finely stratified sequences; (5) Study on high-resolution sequence stratigraphy interpretation and lithological reservoir prediction techniques.

4.1.1 High-Resolution Sequence Stratigraphy and Prediction and Evaluation of Lithological Traps High-resolution sequence stratigraphy should first identify unconformities and maximum flooding surfaces. The identification of the two interfaces captures the interface most closely related to the distribution of lithostratigraphic reservoirs. The unconformity controls the stratigraphic overlap traps and unconformity shielding traps, and the maximum flood surface is associated with the lithological traps of various types of isolated sandstone bodies. Then through the study of the evolution law of regional sequence, we can make clear the law of regional souring-reservoircap assemblage in the vertical direction, define the sequence of main exploration targets, and explore new fields of oil and gas exploration. In the key research area, the core facies, logging sequence and logging facies, seismic sequence, and seismic facies are comprehensively analyzed based on 3D seismic, high-resolution seismic, logging, and core data. Based on the identification and correlation of sequence boundaries and flood surfaces, a fine sequence framework is established to analyze the spatial distribution characteristics of source rock, reservoir rock, and caprock, and to determine the basic factors controlling the formation of lithological traps and their distribution rules, such as lithological pinch-out line, reservoir physical property change zone, stratigraphic overlap line, and stratigraphic unconformity surface (Wang 2004a, b).

4.1.2 Seismic Imaging and Identification of Concealed Geological Bodies At present, the exploration target has become more and more complex from the past simple types and conditions, and the exploration is getting harder. Therefore, it is necessary to study and develop different exploration techniques and countermeasures according to the geological characteristics and technical difficulties of different exploration targets. For subtle reservoirs, the key exploration techniques are imaging

94

4 Realization and Application of Geophysical Technologies …

and recognition of subtle traps and transverse prediction of heterogeneous reservoirs. By improving the resolution and precision of seismic imaging, developing and improving the inversion and prediction methods of high-resolution reservoir characteristics, and carrying out multi-wave seismic processing and interpretation experiments, a series of geophysical comprehensive description technologies for subtle reservoir identification have been established.

4.1.3 Prediction Technology of Reservoir Pore Growing Zone The oil storage space of a reservoir is mainly controlled by the characteristics of sedimentary microfacies and the growth of secondary pores, which is the basis of the growth of primary pores and the formation of secondary pores. Therefore, based on the study of microfacies and the mechanism of diagenesis, the spatial distribution of the secondary pore growing zone or the relatively high porosity and high permeability zone and the lithological blocking area are predicted. Besides, structural fractures are one of the types of oil storage space that can’t be ignored, especially for the low permeability of tight reservoirs.

4.1.4 Main Understanding (1)

(2)

(3)

(4)

The application of the combination of sequence stratigraphy and 3D seismicity, the seismic inversion technology of the third-order sequence group and the high-resolution inversion technology of the fifth-order sequence, organically combine geology and geophysics and establish a relatively perfect set of advanced, effective, economical, and practical techniques for the identification and prediction of lithological reservoirs. Continental basins are characterized by frequent tectonic movement, unconformity, and sedimentary discontinuity, diverse types of sandstone bodies, small scale, and rapid phase transformation, which are favorable conditions for the formation of various types of lithostratigraphic reservoirs. The terrestrial lithostratigraphic reservoirs are mainly distributed around “three sides”. Lithological reservoirs are mainly concentrated near the largest lake flooding surface, and stratigraphic reservoirs are mainly distributed above and below the “unconformity surface”. The “fault plane” not only controls the growth of sandstone body but also the main channel of oil and gas longitudinal migration, and also plays an important role in controlling the formation of lithological and stratigraphic reservoirs. High resolution sequence stratigraphic isochronous correlation framework, in the process of the establishment of the book, is proposed for the first time wellseismic sequence analysis, comprehensive well constrained seismic inversion profile lithological cycle analysis and lake invasion as a control system of 3D

4.1 Realization of Geophysical Technologies …

95

Fig. 4.1 The multi-parameter lithologic seismic inversion under the control of sequence stratigraphic isochron framework can accurately reflect the lateral lithologic change because of the effective extraction of reservoir change information from actual seismic data

(5)

(6)

seismic inversion and so on animation slice technique, a comprehensive highresolution sequence division and correlation, tracking, closed system of highresolution sequence stratigraphy analysis method. It is of great significance to make full use of 3D seismic data to carry out high-resolution sequence stratigraphic quantitative analysis (Fig. 4.1). In the research process of sedimentary facies, Stratimagic seismic facies quantitative analysis technology is used to analyze the plane distribution and longitudinal evolution of sedimentary facies. With past single well, section or flat compared to qualitative, semi-quantitative analysis of sedimentary facies, the set of technical methods to make full use of a large number of 3D seismic data, to carry out the fine analysis of sedimentary microfacies types and distribution characteristics within the high-resolution sequence stratigraphic framework, to map rapidly, and to determine the lithology reservoir distribution of favorable reservoir facies belt have more important significance. Through high-resolution sequence stratigraphic analysis and reservoir forming condition analysis, the sedimentary microfacies, reservoir distribution, and oil and gas distribution law are predicted, and a new viewpoint is proposed for the distribution of lithological reservoirs in the study area, which has important guiding significance for the exploration of lithological reservoirs in the future.

4.2 Application of Geophysical Technologies for Lithological Reservoirs in Yingtai Area According to the characteristics of lithological reservoirs, the application of geophysical technology for lithological reservoirs generally adopts the following methods: (1)

(2)

Analysis of geological background in the study area. It mainly studies the structures and sedimentary environments that control sedimentation, analyzes sedimentary systems and rock types, evaluates sedimentary reservoirs, and predicts the distribution of favorable reservoirs. Establishment of stratigraphic sequence framework and classification of sequence stratigraphy.

96

(3)

(4)

(5)

4 Realization and Application of Geophysical Technologies …

The sequence of the study area is divided into four to five order sequences that control the distribution of rock mass in the study area, and the sequence of the study area is tracked in the plane. The main techniques are applied: singlegood evaluation, sequence framework analysis of connecting wells, interactive correlation and interpretation of wells and seism, sandstone body interpretation, and sedimentary facies interpretation. Study on the sedimentary system. The sedimentary system is a three-dimensional sedimentary body composed of sedimentary facies related to provenance and auction processes under certain geographical environments. The difference between sedimentary facies analysis and sedimentary system analysis lies in the different scales of sedimentological analysis. The scale of facies analysis is smaller, while the scale of sedimentary system analysis is larger. The application of the sedimentary system analysis method is helpful to better grasp the macroscopic sedimentary law of the basin, to restore the paleoenvironment of the basin as a whole, to reshape the palaeogeographic appearance of the basin, to find out the spatial sequence change and distribution law of sedimentary facies, and to provide a basis for guiding oil and gas exploration in the basin. Reservoir inversion. According to the regional geological characteristics and data conditions, the appropriate method is selected to carry out reservoir inversion and predict the spatial distribution of sandstone bodies under the restriction of sequence. All Wells should be used for reservoir inversion as far as possible. When the acoustic moveout does not reflect well the reservoir characteristics, the reservoir characteristic curve should be reconstructed. The main techniques used are logging constrained seismic inversion, multi-parameter seismic inversion, and high-resolution reservoir characteristic reconstruction seismic inversion. Analysis of favorable zones and prediction of targets. Through the analysis of reservoir-forming conditions, the regularity of oil and gas enrichment is studied and the regional distribution of oil and gas is predicted. The main techniques used include comprehensive analysis and evaluation, seismic attribute analysis and visualization, well location design, etc.

4.2.1 Analysis of Geological Background in the Study Area The Yingtai area in the south of Songliao Basin is located in Zhenlai County, Jilin Province. The regional structure is located at the north end of the Honggang terrace in the central depression of Songliao Basin, adjacent to the western slope area in the west and the Da ‘an-Gulong depression in the east. Since the Triassic, this area has experienced four stages: uplift and denudation, fault depression, depression deposition, and uplift. At the end of the Permian, the Songliao Basin was in a state of uplift and denudation, which caused the inversion

4.2 Application of Geophysical Technologies …

97

of the Paleozoic strata, and formed the NE-trending tectonic morphology under the influence of the NE-trending and NNE trending faults. From Late Jurassic to Late Cretaceous, the main shoreline shallow lake to semi-deep lake deposits was formed in the Yingtai area due to the large-scale shallow fault activity in the strata. Then the basement of the basin subsided as a whole, a wide range of sedimentary environments began to form. At the same time, under the influence of the tensile stress, based on the basement fracture in the early stage, a series of normal faults arranged in a bird line on the plane were formed. The stratigraphic thickness gradually thinned from west to east and north to south and formed a series of low-amplitude fault-nose structures, fault-block structures, and anticlines along the fault direction. In the Late Cretaceous, the compressive stress gradually strengthened, the basin was uplifted, and the overlying strata were reversed and folded, which made the Cretaceous and the overlying Tertiary form unconformity contact. At the end of Tertiary, the base was lifted again, and the Yingtai landform was formed, the exploration degree in this area is relatively low, and there are many small and concealed lithological reservoirs with great resource potential. The reservoir in the Yingtai area of Jilin oilfield is located in the most developed area of the Zhenlai-Tailai provenance depositional system in the north, which shows braided river delta depositional facies of Baokang river system delta front. The small thickness of a single sandstone body, rapid lateral variation, relatively developed faults, and small trap range makes it difficult to predict the reservoir. There are six main contents of horizontal reservoir prediction research: (1) Carrying out fine horizon calibration for the sandstone group and oil formation of the main target layer, and establish the connecting well calibration backbone profile; (2) Further refining and implementing the fault combination mode, identifying the distribution of fault system, and strengthening the interpretation and implementation of small faults; (3) Checking out the distribution and characteristics of the sandstone body, and further studying the distribution law of the reservoir, and describing the reservoir thickness and physical properties; (4) To apply seismic facies analysis techniques, combined with drilling, logging and seismic data, focusing on the identification of lithological traps and their morphology; (5) Selecting favorable structural traps and concealed traps for detailed geological and geophysical description and providing drilling locations; (6) Using all kinds of seismic information, combined with drilling and logging data, the reservoir prediction and reservoir description of the main sandstone layers and oil-bearing intervals are carried out (Wang 2005).

4.2.2 Establishment of Stratigraphic Sequence Framework According to the core, outcrops,  well logs, and seismic data, two large regional  angular unconformities T5 , T02 , two intra-basin angular unconformities T4 , T03 and 16 local unconformities have been identified in this area. The Upper Jurassic to Upper Cretaceous strata in Songliao Basin can be divided into one giant sequence, three supersequence groups, six super sequences, and 19 sequences, and on this basis,

98

4 Realization and Application of Geophysical Technologies …

58 system tracts and several parasequences have been identified. They constitute the sequence stratigraphic assemblage characteristics of Songliao Basin. The building characteristics of the super sequences have a regular evolution. Both the initial fault subsidence supersequence and the fault subsidence supersequence were formed in the fault subsidence stage in the early development of the Songliao rift basin, and they constitute the syn-rift subsidence, supersequence group. The faultdepression transformation supersequence, the lower depression supersequence, and the upper depression supersequence were all formed in the depression subsidence stage of the middle development of the Songliao rift basin. The overall development scale of the three supersequence subsidence lake basins gradually increased from the bottom to the top, and the lake basin reached the maximum size when the sequence XVI of the upper depression developed. Together they constitute the post-rift thermal subsidence supersequence group. The sequence XVII and the sequence branch above it belong to the rift atrophy supersequence group. The syn-rift subsidence supersequence group, post-rift thermal subsidence supersequence group, and post-rift atrophied supersequence group constitute the filling sequence of the Songliao rift basin and form the continental rift basin giant sequence.

4.2.3 Division of Stratigraphic Sequence As a typical continental rift basin, Songliao basin due to the nature of the basin evolution and phase difference caused the different environment, the diversity of the sequence stratigraphic characteristics of the Songliao Basin, both fault depression phase fault basin, and there was a depression period of depression basin, the former again there is a single fault type and complex fault depression, which in turn exist depression lake basin and depression rivers. There are two sequences in the initial fault depression period, four sequences in the fault depression period, two sequences in the fault-depression transformation period. The lower depression and the upper depression were divided into 2 super sequences and 9 sequences in the depression period. The atrophy stage can be divided into two sequences.

4.2.4 Study on Sedimentary System The key to the study of sedimentary system analysis is genetic stratigraphy, that is, stratigraphic division and correlation should fully consider the isochron of stratigraphic units, and explain the relationship between large sedimentary bodies and their surrounding sedimentary bodies according to the sedimentary environment and the synsedimentary structures that control the formation of sedimentary environment.

4.2 Application of Geophysical Technologies …

99

Based on the core observation of a large number of exploration Wells in the south of Songliao Basin and the analysis of single good facies and section facies, the stratigraphic framework of the basin is determined and the sedimentary facies of some sedimentary systems are studied by using the comprehensive information of geology, well logging, seismic analysis, and various laboratory tests. The sedimentary background of these sedimentary facies and the tectonic activity at that time were taken into full consideration, and the sedimentary facies assemblages with genetic connection were analyzed as a whole to determine the type and space–time distribution of the sedimentary system. This is the principle of depositional system division. According to the above principles, we made full use of geological, well logging, and seismic information to study the sedimentary system of different periods in the southern part of the basin, and concluded that the fault depression period in the southern part of the Songliao Basin was a near-source, multi-source, and multisedimentary system. In the short axis direction of the basin, there is Zhenlai Baicheng alluvial fan-braided river delta-lacustrine depositional system. In the long axis direction, the Tongyu-Baokang alluvial fan-fluvial-deltaic lacustrine sedimentary system was developed. The Changchun-Huaide alluvial fan-fluvial-delta-lacustrine sedimentary system and the alluvial fan-fluvial-shallow water delta-lacustrine sedimentary system are nearly oblique to the long axis of the basin. There are five facies types: alluvial fan, river (meandering river, braided river), delta (fan delta, braided river delta, shallow water delta, delta), lake (gravity flow), volcanic rock, and pyroclastic rock.

4.2.5 Reservoir Inversion In the south of Songliao Basin, there is a continuous depression with low water level and shallow water delta front sandstone body. The sandstone body varies greatly in the direction of parallel provenance and vertical provenance. The thickness of a single sandstone body is small, and the longitudinal resolution of conventional impedance constrained inversion is not high (Li et al. 2014). Due to lithology, diagenesis, and other factors, mudstone and sandstone can’t be separated effectively by using only velocity (acoustic moveout). But the study found that the spontaneous potential in this area can effectively divide the reservoir, combining spontaneous potential with velocity can further improve the ability to identify sandstone bodies, using the spontaneous potential to reconstruct the reservoir characteristics, the characteristic curves that can effectively identify the reservoir are obtained. It can be seen from the relationship between spontaneous potential, velocity, reservoir characteristic reconstruction curves, and well sideway seismic characteristics that reservoir characteristic reconstruction curves can not only effectively identify reservoirs, but also coincide with wave group characteristics of a well sideway seismic profile (Fig. 4.2). Conventional velocity curve and reservoir reconstruction characteristic curve were used for inversion respectively, and the inversion results of conventional acoustic

100

4 Realization and Application of Geophysical Technologies …

Fig. 4.2 Relationship between well logging curve, reconstructed curve, and characteristics of the well sideway wave group

constraints and reservoir reconstruction curve constraints were compared. It can be seen that the boundary of sandstone and mudstone of 10-20m sandstone body is ambiguous by using conventional acoustic impedance constraint inversion, while the boundary of sandstone and mudstone of 3–5 m sandstone body is visible by using reservoir characteristic reconstruction curve constraint inversion (Fig. 4.3). In other words, reservoir feature reconstruction can greatly improve the resolution ability of thin sandstone layers and lithology identification. This method has been applied to the prediction of the thin sandstone body or lithological reservoir in many areas.

4.2.6 Oil and Gas Distribution and Prediction of Favorable Facies Zones The development characteristics of sequence stratigraphy in the middle of a petroliferous basin control the filling sequence and sedimentary characteristics of the basin.

4.2 Application of Geophysical Technologies …

Acoustic impedance constrained inversion profile

101

Reservoir reconstruction constrained inversion profile

Fig. 4.3 Comparison of conventional impedance constrained inversion and reservoir feature reconstruction constrained inversion results

Therefore, the geological elements such as source, reservoir, cap, and so on that form the oil and gas pools have a certain distribution rule in the sequence framework. Through the study of sequence stratigraphy, it provides a good conceptual model for us to determine the type and distribution of the sedimentary system, and to predict the distribution and combination of favorable source rock, reservoir, and caprock. Then we can find out the distribution law of oil and gas in the sequence framework and the type of oil and gas reservoir, which can provide an important basis for the exploration of oil and gas reservoirs. Comprehensive analysis and evaluation, seismic attribute analysis and visualization, well location design, and other major techniques are used to predict the reservoir.

4.2.6.1

Oil–Gas Distribution Law

There is no essential difference between the geological conditions for the formation of lithological reservoirs and those for structural reservoirs. It also must have the production, storage, and cover ring transportation and protection conditions. However, due to the special source-reservoir-cap configuration of lithological reservoirs, various trap forms, specific sedimentary and tectonic environment conditions, and other organic combinations, the types of oil and gas reservoirs are diversified and concealed, which is far more complex than structural oil and gas reservoirs. The large lithological reservoirs in the south of Songliao Basin are mainly distributed in Changling Sag and its surrounding slope zone. The five factors controlling lithological accumulation are effective oil supply in the main source area, delta front facies zone, effective direct cap rock, lateral upper structure, and fault sealing.

102

4 Realization and Application of Geophysical Technologies …

The Delta Front Facies Zone Controlling the Distribution of Lithological Reservoirs There are several fluvial delta (or fan delta) depositional systems around Songliao Basin, which provide abundant reservoir space. In the south of Songliao Basin, there are mainly three confirmed deltas: the Baicheng-Zhenlai delta from the west; From the Baokang-Tongyu Delta in the southwest, and the Changchun-Huaide Delta in the southeast. In the delta plain facies zone near the provenance area, there are mainly distributary channels, fissure fan, distributary channel edge, and interdistributary bay microfacies. Sandstones are well developed with good transverse connectivity, and the sand-to-ground ratio is generally more than 50%. The delta front facies zone is the underwater part of the delta entering the lake area. The main sedimentary microfacies include underwater distributary channels, mouth bar, sheet sand bar, and far sand bar. Sandstone and mudstone are interlaced frequently. Compared with the delta plain facies belt, the sand-to-ground ratio is significantly reduced, generally 40–20%. The pre-delta facies zone entered the lake area, and mainly developed the pre-delta mud, argillaceous silt, and sandy gravity flow, which had no reservoir capacity. During the sedimentary period of Qingshankou Formation-Nenjiang Formations above The fourth section of Quantou Formation, along with the advance and retreat of the lake water, the delta front sandstone bodies also showed frequent regressive, accretionary, and progressive deposition, which created conditions for the formation of lithology and various structural and lithological complex traps related to lithological changes. The exploration practice shows that in the area where the sand-to-ground ratio is more than 40%, it is difficult to form effective traps because the reservoir is too developed and the structural amplitude is low, so the oil test results are mainly water production. In an area with a sand-to-ground ratio of less than 20%, the sandstone lenticular reservoir can be developed locally, but it is difficult to form a large-scale oil-bearing area with commercial exploitation value because of the significant deterioration of reservoir physical property. Only in various tectonic settings such as palaeo-uplift and slope zone, sandstones developed properly, and sandstones and mudstones interbedded frequently, which is the fundamental reason for the formation of the large area of structure-lithological oil and gas in the front facies belt. There are many oil-bearing layers because the sandstone bodies of different horizons are overlapped in space. The drilling process also confirmed that the discontinuous oil and gas show could reach several hundred meters, allowing for multi-zone exploration.

The Existence of Direct Cap Rock is a Necessary Condition for the Formation of the Large-Area Lithological Reservoir The lacustrine mudstones of the first section of Qingshankou Formation and Nenjiang Formation in the depression period of Songliao Basin are not only good hydrocarbongenerating rocks but also two main sets of regional cap rocks in the south of the Songjiang Basin. Most of the discovered oil and gas are distributed under two sets of

4.2 Application of Geophysical Technologies …

103

regional cap rocks (middle and lower oil assemblages). However, with some exceptions, the specific spatial distribution of oil reservoirs depends on the development degree of direct cap rock. The main oil layers in the Yingtai-Sifangtuozi area are mostly distributed in the first section of the Qingshankou Formation, the reason is that the top of the first section of the Qingshankou Formation develops a good set of direct mudstone cap. This set of cap rocks was formed during the maximum lacustrine transgression period and gradually thickened from the delta root to the lacustrine basin area. The thickness of the main oil area in the front facies zone was greater than 25 m. This direct cap makes the oil and gas accumulate on a large scale in the first section of the Qingshankou Formation, forming a large area of the lithological reservoir. The exploration results show that with the approach to the provenance direction, the sand-to-ground ratio of formation increases, the thickness of the direct mudstone cap at the top of the first section of Qingshankou Formation gradually becomes thinner (about 10 m), and the position of the oil layer also moves upward from the first section of Qingshankou Formation to the second and third section of Qingshankou Formation. For example, in Well Hei 43 and Well Hei 52 in the south of the Daqing well, the oil and gas in the first section of the Qingshankou Formation are active, but only a small amount of oil is obtained from the test results, mainly water. However, high yield oil flow was obtained in the second and third sections of the Qingshankou Formation. The reason is that the mudstone cap at the top of the first section of the Qingshankou Formation becomes thinner and the sealing condition becomes worse, which leads to the upward migration and accumulation of oil and gas. Therefore, the exploration of the delta front belt in the south of Songliao should not only pay attention to the prediction of sandstone, but also the prediction of the development of the direct cap rock at the top of the sandstone layer.

Tectonics and Faults Have a Good Sealing Effect in the Lateral Direction The tectonic evolution shows that in the middle and shallow layers of the southern Songliao Basin, the tectonic movement is not strong, and the Changling sag and its slope zone lack large structural zones, and the depression zone and gentle slope zone are mainly formed. The local structural types are mostly low-amplitude structures, which are not conducive to the gravity differentiation and accumulation of oil and water. However, this situation is greatly improved due to the regular distribution of fault development zones in the middle and shallow strata. The large area of oil-bearing lithology has formed several effective composite traps, namely, structure-lithological and fault-lithological reservoirs, which are sealed by faults and structures laterally. Structural studies show that under the combined control of regional weak extension and strike-slip, the normal faults developed during the sedimentary period of Qingshankou Formation-Nenjiang Formation in Songliao Basin basically spread NNW-SSE, mainly developed in the first and second section of Qingshankou Formation, faulted downward to the third and fourth section of Quantou Formation, and generally disappeared in the strata of Yaojia Formation upward. On the one hand,

104

4 Realization and Application of Geophysical Technologies …

these faults provide important channels for hydrocarbon migration, on the other hand, they provide conditions for the formation of effective lateral shielding and play the role of hydrocarbon enrichment. The exploration results show that the reverse normal fault which is opposite to the formation tendency is more effective in oil control.

Reasonable Configuration of Source, Reservoir, and Cap is Beneficial to Hydrocarbon Accumulation During the development of the Songliao Basin, under the action of vertical structure, several sedimentary cycles were formed. In terms of the formation, migration, accumulation, and preservation conditions of oil and gas, it constitutes several sets of corresponding source-reservoir-cap assemblages. Especially in the middle stage of basin development, namely, Quantou Formation and Nenjiang Formation, the oilbearing strata of Qingshankou Formation and Nenjiang Formation in two secondary sedimentary cycles constitute two source-reservoir-cap assemblages. The secondorder sourcing-reservoir-cap assemblage of Quantou Formation-Nenjiang Formations is caused by the second-order sedimentary cycle caused by the tectonic movement period and the generation, drainage, migration, and accumulation of oil and gas reservoirs. There are three types of assemblage: the mixed type of lower generation and upper generation, upper generation and lower generation, and mudstone fractured reservoir.

4.2.6.2

Prediction of Favorable Facies Zones

The distribution of lithological reservoirs in the south of Songliao Basin is closely related to the tectonic and sedimentary evolution of the basin. The characteristics of multi-cycle tectonic and sedimentary evolution formed multiple sets of hydrocarbon generation layers, the changes of the three provenances in different periods controlled the reservoir development. The exploration results show that the distribution of oil and gas has a strong regularity. In other words, oil and gas mainly revolve around the center of hydrocarbon generation depression and form oil and gas reservoirs in favorable tectonic lithofacies change zones with good reservoir physical properties. The distribution of oil and gas reservoirs is not single and isolated, but is usually distributed in certain parts of the structure and sedimentary system, and has the characteristics of the zonal distribution.

Sandstone Bodies in Different Structural Positions Form Different Types of Lithological Reservoirs Different types of lithological reservoirs are formed in different structural parts because of different types of sandstone bodies, the different distribution range of

4.2 Application of Geophysical Technologies …

105

Fig. 4.4 Lithological hydrocarbon accumulation model in the middle and shallow parts of the oil field

sandstone bodies, and different changes of lithological facies. For example, structurelithological reservoirs are mostly formed in anticlines, sandstone updip pinch-out reservoirs are mostly formed in the slope tectonic background of monoclines or depressions (synclines), fault-lithological reservoirs are mostly developed in the upper and lower walls of faults, and sandstone lens reservoirs are mostly distributed in hydrocarbon-generating depressions.

The Types of Lithological Reservoirs Formed by Rocks in Different Parts of Sedimentary Facies Belt Change Regularly According to the analysis of reservoir profiles that cut through the dry-retaining sandstone body laterally and longitudinally, the Gaotaizi oil reservoir transitioned from the low-amplitude structural reservoir in the sandstone core to the structurelithological reservoir and fault-lithological reservoir in the delta front to the updip pinch-out reservoir in sandstone and the pre-delta facies sandstone lens and fractured mudstone reservoir (Fig. 4.4). When the Putaohua reservoir was deposited, the palaeo-topography was gentle, and the main distributary channel was formed in a shallow-water delta. In the later period, the center of the depression moved southward, the river retreated, and the delta sandstone body formed in the early stage was damaged by water invasion, resulting in the formation of residual mouth bar and coastal bar. The Reservoir sandstone body is stacked vertically and has poor horizontal connectivity, which forms the pattern of longitudinal interaction between sandstone and mudstone and lateral contact between

106

4 Realization and Application of Geophysical Technologies …

sandstone and mudstone layer. Therefore, the planar distribution of the Putaohua reservoir is manifested as a sandstone lensed reservoir or fault-lithological reservoir under the background of local uplift or nose-like structure.

The Regressive Distribution of the Central Assemblage Depositional System Determines that the Lithological Reservoirs Are Distributed in the Basin During the sedimentary period of Sects. 4.1, 4.2 and 4.3 of Qingshankou Formation, with the lake basin shrinking, the shoreline gradually advanced into the basin, and the delta sandstone body also advanced into the basin. Along the direction of provenance, the delta front facies sandstone and mudstone deposits were successively transformed into pre-delta mudstone deposits. The spread point of this facies zone provides a shielding condition for the formation of reservoirs dominated by lithology control. Therefore, from the first section to the third section of the Qingshankou Formation reservoir, the plane distribution of the reservoir gradually advanced into the basin, and the buried depth of the reservoir changed from deep to shallow.

The Types of Lithological Reservoirs Are Different in Different Oil and Gas Assemblages The sandstone bodies of the Heidimiao oil layer in the Nenjiang Formation are mainly from the north and southwest of the basin, which is mainly composed of some sandstone lens reservoirs, structure-lithological reservoirs, and fault-lithological reservoirs. Sandstone bodies of the Ser Tu oil layer are derived from the Baokang depositional system in the west Yingtai and the southwest and formed structure-lithology, faultlithology, updip pinch-out sandstone, and sandstone lens reservoirs in coordination with the structure. The provenance of the sandstone bodies of the Fuyu reservoir is from the southeast and southwest of the basin, and they cooperate with the structure or fault to form structural lithology or fault-lithological reservoir. The sandstone bodies of Pupuhua and Gaotaizi oil reservoirs are from the Yingtai depositional system in the west and the Booking depositional system in the southwest, which are mainly composed of updip pinch out of sandstone, sandstone lens, faultlithology, structure-lithology, and fractured mudstone reservoirs (Fig. 4.5).

4.3 Cognition and Conclusion (1)

The application of a combination of sequence stratigraphy and 3D seismic, the seismic inversion technology of the third-order sequence group, and the highresolution high-precision inversion technology of the fifth-order sequence,

4.3 Cognition and Conclusion

107

Fig. 4.5 The inversion software is used to predict the oil-bearing range of the sandstone body, which shows that the sandstone body has good oil-bearing property

(2)

(3)

(4)

organically combine geology and geophysics and form an effective technical combination of lithology. Through the analysis of the geophysical characteristics of the reservoir, the effective identification and comprehensive evaluation of physical properties and oil and gas properties of the reservoir are summarized; Through the comprehensive analysis of different types of the lithological reservoir forming conditions, the evaluation and description of the lithological reservoir are finally completed. Technically, based on the high-resolution 3D seismic data, the conventional 3D seismic data analysis and interpretation software are flexibly applied, and the multi-factor coherence technology and sequence reservoir comprehensive inversion combined with wavelet transform prediction technology are developed to effectively solve the identification problem of subtle reservoirs. In the process of establishing the high-resolution sequence stratigraphic isochronous correlation framework, for the first time, an isochronous animation slicing technique for 3D seismic inversion is proposed, which adopts wellseismic sequence analysis, well-constrained seismic inversion profile lithological cycle analysis, and lacustrine transgression surface as top and bottom control, carrying out a high-resolution sequence stratigraphic analysis technology and method system that integrates high-resolution sequence division, correlation, tracking, and closure. It is of great significance to make full use of 3D seismic data to carry out high-resolution sequence stratigraphic quantitative analysis. In the process of sedimentary facies research, with past single well, section or flat compared to qualitative, semi-quantitative analysis of sedimentary facies, the technical methods to make full use of a large number of 3D seismic data, carried out within the high-resolution sequence stratigraphic framework of sedimentary microfacies types and distribution characteristics of fine analysis, rapid mapping, determine the lithology reservoir distribution of favorable reservoir facies belt has more important significance.

108

4 Realization and Application of Geophysical Technologies …

(5)

Through high-resolution sequence stratigraphic analysis, prediction of sedimentary microfacies and reservoir distribution, analysis of oil and gas distribution law, and accumulation conditions, the combination of continental sequence and seism has important guiding significance for future lithological reservoir exploration. The application of effective geophysical exploration techniques for lithological reservoirs has achieved remarkable results in thin-interbedded lithological reservoirs in Songliao Basin, China, which has laid a technical foundation for the exploration and development of such reservoirs.

(6)

References He F, Wang H et al (2021) Comparison of azimuthal anisotropy of P-wave and S-wave velocity based on wide azimuth seismic data. Oil Geophys Prospect 56(2):289–294, 301. https://doi.org/ 10.13810/j.cnki.issn.1000-7210.2021.02.010 Li et al (2014) Geophysical exploration technology: applications in lithological and stratigraphic reservoirs. (1st edn) Hardcover, Publisher, Elsevier, 480 Pages Wang H (2004) Present situation and the prospect of key technology for oil geophysical exploration. China Pet Explor 9(1):41–51 Wang H (2004) Present situation and the countermeasure analysis of subtle reservoir exploration. Oil Geophys Prospect 39(6):739–744. https://doi.org/10.13810/j.cnki.issn.1000-7210.2004.06.023 Wang H (2005) Geophysical exploration technology and application for lithological reservoirs: a case study of southern Songliao Basin. China University of Geosciences (Beijing) Wang H (2008) The sequence constraint prediction technology and application for lithological reservoir. China Pet Explor 13(4):36–42

Index

A Abnormal information, 57 Absorption attenuation, 68 Accommodating space, 16 Acoustic impedance, 18, 20, 24, 25, 57, 60, 61, 66, 68–71, 74–76, 78, 80, 83, 92, 100 Acoustic moveout, 18, 19, 21, 80, 81, 85, 86, 96, 99 Acquisition, 44, 63, 91 Alluvial fan, 31, 45, 57, 99 Amplitude, 2, 7, 17, 20, 22, 24, 30, 31, 37, 40, 41, 44, 46, 47, 49–51, 53, 57, 60, 61, 67–69, 71, 85, 97, 102, 103, 105 Amplitude attenuation, 68 Amplitude spectrum, 78 Anomaly, 49, 50, 55, 85 Automatic horizon tracking, 91 AVO, 7, 69 Azimuth detection, 54

B Bandwidth, 37, 54, 72 Basin margin, 32 Bright spot, 7, 37, 41, 60

C Canyons, 51 Characteristic curve, 9, 12, 19–21, 26, 79– 86, 96, 99 Characteristic curve reconstruction, 17–21, 81–85, 92 Characteristic parameter, 37, 43 Coherent, 9, 22, 23, 49–56

Coherent algorithm, 51 Coherent cube, 22, 23 Coherent volume, 7, 29, 49–53, 55, 57, 91 Compensated neutron, 19, 81, 86 Constrained inversion, 34, 70, 72, 80, 82, 87, 99, 101 Convolution, 68, 75, 76, 78 Cross-correlation, 50–52

D Data body, 9, 20, 23, 25, 43, 53, 54, 56–58, 64, 75, 83 Deconvolution, 68, 71, 72 Deep resistivity, 20, 86 Deformation, 50 Deltas, 5, 7, 50, 52, 55, 57, 97, 99, 101–103, 105, 106 Diagenesis, 4, 6, 94, 99 Discontinuous characteristics, 49 Downlap, 16

E Effective frequency band, 78, 79 Elastic impedance inversion, 69, 70 Erosion, 32, 54, 66 Exploration strategy, 2 Exploration target layer, 16 Exploration technology, 1, 3, 4, 6, 7, 29

F Fan delta, 31, 45, 99, 102 Fault-block reservoir, 2 Fault interpretation, 22, 23, 49–51, 53, 55

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2022 H. Wang et al., Practical Geophysical Technology and Application for Lithological Reservoirs, SpringerBriefs in Petroleum Geoscience & Engineering, https://doi.org/10.1007/978-981-16-4197-8

109

110

Index

Fault-lithological reservoir, 103, 105, 106 Faults, 4, 16, 20, 22, 26, 33, 37, 45, 49–58, 60–62, 74, 94, 96–99, 101, 103–106 Fault structure, 50 Fidelity, 20, 21, 89, 91 Flood surface, 13, 16, 93 Fluid change, 50 Foreland basin, 3, 31 Formation trap, 3 Forward modeling, 68 Frequency, 2, 13, 17, 20, 37, 41, 44, 46, 51, 54, 69, 71, 72, 74, 78, 79, 82–84, 87, 91 Frequency division, 7, 9 Frequency domain deconvolution, 77, 78

Invasion, 15, 16, 50, 94, 105 Inversion, 7–9, 11–13, 18–21, 24–26, 29, 34, 35, 48, 57, 60, 65, 66, 68–84, 87, 88, 91–94, 96, 99, 100, 106, 107 Inversion algorithm, 77 Inversion profile, 25, 67, 88, 94, 107 Irrelevance, 23, 49, 50 Isochronic formation interface, 16 Isochronous framework, 13, 87 Isochronous stratigraphic framework, 13, 29, 87

G Geological background, 2, 3, 6, 7, 14, 21, 31, 54, 81, 84, 95, 96 Geological interpretation, 5, 6, 22, 38, 59, 69 Geological model, 5, 6, 9, 26, 69, 73–76, 79, 81, 92 Geological modeling, 12, 34, 72, 76, 79 Geophysical method, 8, 9 Geophysical technique, 4

K Kinematics and dynamic characteristics, 41

H Heterogeneity, 5, 17, 18, 21, 36, 66, 75, 80, 91 Heterogeneous reservoir, 11, 94 High-frequency sequence, 13, 87 High-resolution, 4–6, 8, 11–14, 16–18, 20, 21, 24, 26, 29, 30, 32–34, 36, 43, 51, 72, 87–89, 91–96, 106–108 High-resolution seismic technology, 6 Himalayan movement, 31 Horizon calibration, 9, 38, 60, 97 Horizontal resolution, 8 Horizontal section, 49, 50 Hydrocarbon, 1, 3, 5, 32, 49, 102, 104, 105 Hydrocarbon detection, 9

I Immersion, 56, 63–65 Initial model, 13, 74, 75, 79, 87 In-phase axis, 34, 49, 50, 52, 54 Interpretation, 9, 12, 13, 16, 20, 21, 23–25, 33, 34, 36, 38, 43, 47–50, 53–55, 57– 62, 64, 66, 71, 72, 74, 79, 87, 88, 91, 93, 94, 96, 97, 107 Interval velocity, 44

J Joint well section, 25

L Lacustrine transgression surface, 32, 33, 107 Litho-electric relations, 26 Lithofacies characteristics, 29 Lithological distribution, 13, 87 Lithological information, 19, 20 Lithological profile, 25 Lithological reservoir, 1–6, 8, 11, 12, 17–19, 22–24, 26, 29, 36, 38, 41, 80, 91–95, 97, 100–108 Lithological trap identification, 7 Lithology, 4, 11, 14, 19, 21, 22, 25, 26, 29, 35, 36, 38, 41, 44, 45, 47–53, 66, 69– 72, 75–77, 79–83, 85–87, 92, 95, 99, 100, 102, 103, 106–108 Logging, 5, 8, 9, 13, 17–21, 25, 26, 29, 35, 43, 45–47, 49, 55, 66, 69, 71–76, 78–86, 92, 93, 96, 97 Logging constrained inversion, 7, 13, 18, 66, 69, 70, 73–75, 79, 82, 87

M Marker layer, 26 Migration, 7, 16, 36, 40, 94, 103, 104 Model estimation, 76 Mudstone strata, 2 Multi-component, 92 Multi-parameter inversion, 20 Multi-wave, 92, 94 Multi-wave seismic exploration, 6 Mutation, 50

Index N Natural gamma ray, 19, 20 Neural network, 22, 38, 40–48, 70, 75, 76 Neural network recognition, 91 Nonlinear inversion, 70

O Objective evaluation, 35 Objective function, 39, 40 Oil-bearing basin, 3 Onlap, 16 Optimization, 8, 38, 40, 41, 43, 72, 73, 84 Outcrop, 2, 5, 29, 35, 76, 97 Overlap, 2, 33, 36, 54, 67, 93

P Paleoclimatology, 14 Paleontological characteristics, 17 Paleotopographic trap, 1 Pattern recognition, 17, 22, 38–44, 48 Permeability, 4, 5, 46, 76, 77, 85, 94 Petrophysical parameter, 71 Petrophysics, 19, 81 Phase, 11, 12, 17, 21, 25, 26, 36, 37, 41, 45, 46, 50, 51, 54, 71, 72, 77–79, 94, 98 Phase scanning, 78 Physical property, 4, 9, 11, 21, 22, 26, 36, 38, 47–50, 69, 71–73, 80, 82, 93, 97, 102, 104, 107 Pinch-out, 2, 4, 7, 12, 16, 33, 34, 50, 92, 93, 105, 106 Porosity, 1, 46–48, 76, 77, 85–87, 94 Post-stack, 7, 21, 36, 43, 92 Post-stack inversion, 69, 92 Pre-stack, 7, 21, 36, 92 Pre-stack depth migration, 6, 64 Pre-stack inversion, 69, 92 Principal component analysis, 75–77 Processing, 19, 38, 41, 43, 44, 49, 52, 55–60, 63–65, 71, 74, 76, 77, 79, 91, 94 Progradation, 17, 54

Q Quantitative evaluation, 7 Quantitative model, 9 Quasi-sequence groups, 17

R Recursive inversion, 7, 18, 66, 70–72, 75, 77, 78, 87, 91

111 Reflection coefficient, 66–68, 70–72, 78, 80 Reservoir characteristic reconstruction, 9, 82, 96, 99, 100 Reservoir description, 6, 41, 47, 57, 58, 69, 92, 97 Reservoir morphology, 9 Reservoir parameter estimation, 22, 41, 46, 47 Reservoir prediction, 7–9, 11–14, 16–22, 26, 30, 33–35, 38–40, 57, 75, 76, 80, 81, 83, 85, 87, 92, 97 Reverse-time migration, 64 River channels, 5, 50 Rock profile, 67 S Salinization, 31 Sand dam, 51 Sandstone body, 2, 4, 6, 11, 12, 14, 16, 20, 26, 31–34, 36, 54, 66, 82, 88, 92–94, 96, 97, 99, 100, 102, 104–107 Sandstone reservoir, 3, 57 Sand-to-ground ratio, 102, 103 Saturation, 46, 47, 76, 77 Sea-level, 29 Sedimentary, 2, 4, 6, 7, 13–15, 20, 25, 29, 31, 33, 35, 40, 44, 45, 55, 57–59, 61, 62, 87, 93–104, 106 Sedimentary background, 14, 30, 31, 99 Sedimentary background analysis, 9, 14, 30, 32 Sedimentary facies, 14, 15, 26, 29–33, 35, 40, 44, 45, 57, 95, 96, 99, 105, 108 Sedimentary microfacies, 26, 92–95, 102, 108 Sedimentary strata, 32, 50, 80 Sedimentary system, 9, 13, 15, 29, 33, 45, 72, 87, 95, 96, 98, 99, 101, 104 Sedimentation, 4, 5, 29, 31, 40, 60, 95 Sedimentology, 4, 35 Seism, 5, 6, 9, 17, 19, 25, 26, 35, 75, 83, 96, 108 Seismic acquisition, 8, 79, 91, 92 Seismic amplitude attribute, 7 Seismic attribute, 7, 21–23, 36–51, 54, 59, 62, 76 Seismic attribute analysis, 9, 11, 17, 21, 22, 29, 36, 38, 39, 41, 91, 96, 101 Seismic attribute decomposition, 91 Seismic attribute extraction, 17, 38 Seismic attribute optimization, 22, 38–40 Seismic data, 5, 6, 9, 12, 17, 18, 20–23, 29, 36–38, 41–46, 49–60, 62, 64, 66, 69,

112 71–75, 77, 79, 80, 82, 88, 89, 91–93, 95, 97, 107, 108 Seismic facies, 34, 35, 45, 46, 93, 95 Seismic facies analysis, 40, 44–46, 97 Seismic geological strata, 13 Seismic imaging, 11, 93, 94 Seismic interpretation, 7, 22–24, 41, 49, 50, 62, 64, 92 Seismic inversion, 7–9, 11, 13, 17, 18, 20, 25, 38, 65, 66, 68–79, 87–89, 92, 94–96, 106, 107 Seismic lithology modeling, 7 Seismic lithology prediction, 35 Seismic multi-attribute analysis, 7, 8, 23, 50 Seismic multi-attribute extraction, 92 Seismic phase analysis, 12, 26 Seismic prediction, 13, 77 Seismic profile, 7, 13, 15, 16, 25, 32, 43, 44, 49, 53, 57, 66–68, 87, 88, 99 Seismic quantitative interpretation, 30 Seismic record, 25, 41, 68, 70–72, 74, 78 Seismic reflection interface, 13, 87 Seismic reservoir prediction, 8, 9, 11, 12, 18, 30, 34 Seismic section, 5, 7 Seismic sequence boundary, 16 Seismic signals, 21, 36, 45, 46, 64 Seismic simulation, 7 Seismic trace, 7, 38, 42, 44–46, 52, 54, 66, 68, 75 Seismic trace inversion, 7 Seismic waveform, 12, 18, 34, 35, 46 Seismic waveform classification, 12, 18, 34, 35 Seismogram, 68 Sequence analysis, 11, 12, 18, 30, 34, 92 Sequence boundary, 15, 32, 33, 54, 93 Sequence constraint, 9, 12, 23 Sequence division, 5, 15, 30, 32, 33, 95, 107 Sequence interface, 4, 12, 13, 16–18, 30, 33–35, 38, 87, 93 Sequence interface tracing, 33 Sequence stratigraphic analysis, 8, 13, 17, 54, 92, 93, 95, 107, 108 Sequence stratigraphy, 4–6, 9, 13–16, 26, 29, 30, 32, 33, 35, 36, 87, 93–95, 100, 101, 106 Signal-to-noise ratio, 43, 50–52, 54, 55, 61, 71, 79, 91 Simulation, 57, 64, 70 Slicing technology, 12, 34 Slump structure, 51 Songliao Basin, 26, 30, 96–99, 101–104, 108

Index Spatial distribution, 14, 24, 31, 53, 62, 64, 87, 93, 94, 96, 103 Spatial distribution law, 13 Spherical divergence, 68 Spontaneous potential, 19, 20, 25, 75, 77, 81, 85, 99 Stacking profile, 40 Statistical regression model, 20 Strata seismic profile, 7 Stratigraphic azimuth angle, 50 Stratigraphic deposition model, 13 Stratigraphic dip angle, 50 Stratigraphic overlay relationship, 13 Stratigraphic trap, 1, 2, 4, 8, 13, 16, 33, 36 Structural reservoir, 1, 3, 4, 26, 41, 101, 105 Structure-lithological reservoir, 105, 106 Subtle reservoir, 1, 3, 4, 93, 94, 107 Synthetic record calibration, 32 Synthetic seismogram, 68

T Target layer, 38, 39, 58, 60, 61, 97 3D visualization, 12, 17, 23, 24, 29, 34, 52, 56–62, 64, 65, 91 Threshold value, 24 Time-delay, 49, 72 Time-depth relationship, 38 Time-lapse phenomenon, 16 Tomography, 69 Toplap, 16 Tracing closure, 16 Truncation, 16 Turbidite fan, 7

U Unconformable surface, 54 Unconformable trap, 1 Unconformity, 2–4, 13, 15, 16, 30, 32, 33, 37, 54, 66, 93, 94, 97 Updip, 4, 105, 106

V Velocity analysis, 7, 79 Velocity inversion, 8 Vertical resolution, 12, 19, 52, 80, 81, 83, 88 Virtual environment, 56, 63 Virtual reality, 56, 62–65 Virtual reality environment, 64, 65 Visual interpretation, 6, 18, 62 Visualization, 7, 23, 57–65, 91, 92, 96, 101 Visualization interpretation, 23

Index W Waveform automatic tracking, 60 Waveform classification, 12, 17 Wave impedance inversion, 18, 20, 69, 70, 80, 92 Wavelet, 54, 67, 68, 71, 72, 75, 78, 107 Wavelet envelope, 54

113 Wavelet phase spectrum, 78 Weathering, 14, 31, 80 Well-logging, 7, 9, 17, 19, 24–26, 35, 46, 71, 72, 76, 77, 80–82, 85, 91, 92, 99, 100 Well-seismic interaction, 15, 33 Well-seismic sequence analysis, 94, 107