Relative Fidelity Processing of Seismic Data

Methods and Applications

Xiwen Wang, Xiwen Wang

• Publisher: Wiley-Blackwell
• Year: 2017
• ISBN:
  9781119052920
• Edition: 1
• Topic: Physical Sciences
• Subtopic: Geology & Earth Sciences

This book presents a comprehensive overview of relative fidelity preservation processing methods and their applications within the oil and gas sector. Four key principles for wide-frequency relative fidelity preservation processing are illustrated throughout the text. Seismic broadband acquisition is the basis for relative fidelity preservation processing and the influence of seismic acquisition on data processing is also analyzed. The methods and principles of Kirchhoff integral migration, one-way wave equation migration and reverse time migration are also introduced and illustrated clearly. Current research of relative amplitude preservation migration algorithms is introduced, and the corresponding numerical results are also shown.

RTM (reverse time migration) imaging methods based on GPU/CPU systems for complicated structures are represented. This includes GPU/CPU high performance calculations and its application to seismic exploration, two-way wave extrapolation operator and boundary conditions, imaging conditions and low frequency noise attenuation, and GPU/CPU system based RTM imaging algorithms. Migration velocity model building methods in depth domain for complicated structures are illustrated in this book. The research status and development of velocity model building are introduced. And the impacting factors are also discussed. Several different velocity model building methods are also represented and analyzed. The book also provides the reader with several case studies of field seismic data imaging in different kinds of basins to show the methods used in practice.

1
Study on Method for Relative Fidelity Preservation of Seismic Data

1.1 Introduction

The reservoir of lithology formation, under the control of multiple factors such as regional structures and depositional facies belt, shows some distribution regularity and also has significant exploration potential. However, the exploration challenge is relatively great due to their complex concealment. In recent years, China has made great breakthroughs and discoveries in this field, proving favorable and huge remaining resource potential; this has gradually become the major field of reservoir gain in China [1–4].

The lithology analysis of these lithology reservoirs poses high requirements for the seismic data processing [5–7]. The first challenge is to increase the resolution of seismic data, so as to identify the geologic features like thin sand layers and sand body pinch‐out points, on the processed seismic data. Second, to preserve the amplitude, and not damage the module of amplitude relation between adjacent seismic traces on processing flow. However, previous study mainly focused on identifying thin layers, increasingly widening the frequency band of seismic data and improving the dominant frequency. Therefore, a lot of high‐resolution seismic processes have been developed for the purpose of increasing seismic dominant frequency.

Increasing the resolution of seismic data is mostly realized by deconvolution method [8–19], which is an important means in seismic data processing. In order to increase the resolution of seismic data, an operator for high frequency deconvolution is devised to identify thin layers. However, which principles should be used to determine the frequency bandwidth and dominant frequency of operator for high frequency deconvolution? Is it possible to increase the dominant frequency of the operator for high frequency deconvolution without restriction? These are the questions that remain to be discussed.

Figure 1.1 presents the comparison of lithology reservoir seismic profiles through SN31 wellblock in the Luxi area of Junggar Basin. In Figure 1.1a, a high‐resolution and high‐S/N seismic processing technology was adopted. The lithology reservoir of J2t0 of SN31 wellblock was interpreted from the seismic profile shown in the figure (it also proved to be sand reservoir through drilling), but from the profile, the continuity of J2t0 reservoir was favorable (in the location of Well S302, the J2t0 extended for more than 1 km to the updip direction of the reservoir until the lithology reservoir pinched out). For this purpose, a number of wells were planned for this complete lithology reservoir. Wells S302 and S302 shown on the figure were among them, but their drilling effectiveness varied greatly. The reservoir in Well S303 is very good and the well is a high oil producer, while reservoir in Well S302 is poor (lithology varied, and J2t0 reservoir is mainly mudstone).

Figure 1.1 Comparison between (a) conventional seismic profile and (b) relative fidelity seismic profile through wells S302 and S303.

From the analysis of the data on Figure 1.1, it was believed that the over emphasizing on the high resolution and high SNR during processing deteriorated the fidelity of processed results, directly resulting in the distortion of lithology reservoir seismic response that was reflected on seismic profile.

On seismic profile, the display reliability of seismic response on lithology reservoirs is subject to the fidelity of seismic processing. Currently, it is very difficult to realize absolute seismic fidelity preservation processing, but relative fidelity preservation processing is possible. Figure 1.1b shows the relative fidelity preservation profile that was reprocessed as per the flowchart in Figure 1.13, in 2006. As clearly shown in the figure, the reflection of J2t0 reservoir in Well S302 is very weak, indicating that the reservoir is no longer present. The pinch‐out point of J2t0 reservoir identified on seismic profile is nearly 300 m to downdip direction of lithology reservoir in Well S302, which matches with the data of Well S302 that has been completed.

This has raised a question: the processing of seismic data is the key in exploration of lithology reservoirs. Only when the seismic response of reservoir properties is truly reflected on the profile after seismic processing to the greatest extent can the seismic data be effectively used to identify the lithology reservoirs.

In view of the above problems, we made analysis by selecting a 2D high‐resolution pilot line of 8 km from the Shidong area in the Junggar Basin across Wells Shidong‐2 and Shidong‐4, on the results of high resolution and high SNR processing and relative fidelity preservation processing. Based on this, we came up with a processing method for relative fidelity of seismic data.

Relative fidelity preservation processing[5] of seismic data should pay attention to the following items: (1) Protection of effective frequency band. The widening of effective band should rely on SNR of seismic data. In order to widen the low S/N seismic data to high frequency band, it is necessary to control the bandwidth and dominant frequency of the deconvolution operator. The main function of deconvolution is to increase the energy of high frequency components within the effective band so as to prevent notch frequency at effective high frequency band. (2) Protection of low frequency, especially that of 3–8 Hz. The scenarios that suppress low frequency data of 10 Hz and below in high resolution and high SNR processing, or give significant loss of effective low frequency information with adoption of strong f‐k noise elimination should be avoided. (3) Amplitude preservation. Modification modules like RNA should not be applied to avoid damaging the lateral relationship of seismic channel amplitude. (4) Phase preservation. The module should not be used as it...

Table of contents

• Cover
• Title Page
• Table of Contents
• Preface
• 1 Study on Method for Relative Fidelity Preservation of Seismic Data
• 2 Method and Principle for Seismic Migration and Imaging
• 3 Study of Reverse Time Migration Method for Areas With Complicated Structures Based on the GPU/CPU System
• 4 Study and Application of Velocity Model Building Method for the Areas with Complicated Structures
• 5 Case Study
• Index
• End User License Agreement