3D structural-orientation vector guided autotracking for weak seismic reflections: A new tool for shale reservoir visualization and interpretation Haibin Di 1 , Dengliang Gao 2 , and Ghassan AlRegib 3 Abstract Recognizing and tracking weak reflections, which are characterized by low amplitude, low signal-to-noise ratio, and low degree of lateral continuity, is a long-time issue in 3D seismic interpretation and reservoir char- acterization. The problem is particularly acute with unconventional, fractured shale reservoirs, in which the impedance contrast is low and/or reservoir beds are below the tuning thickness. To improve the performance of interpreting weak reflections associated with shale reservoirs, we have developed a new workflow for weak- reflection tracking guided by a robust structural-orientation vector (SOV) estimation algorithm. The new SOV- guided auto-tracking workflow first uses the reflection orientation at the seed location as a constraint to project the most-likely locations in the neighboring traces, and then locally adjust them to maximally match the target reflection. We verify our workflow through application to a test seismic data set that is typical of routine 3D seismic surveys over shale oil and gas fields. The results demonstrate the improved quality of the resulting horizons compared with the traditional autotracking algorithms. We conclude that this new SOV-guided auto- tracking workflow can be used to enhance the performance and effectiveness of weak reflection mapping, which should have important implications for improved shale reservoirs visualization and characterization. Introduction Three-dimensional horizon interpretation is one of the routine tasks in subsurface structural analysis and reservoir characterization from 3D seismic data (e.g., Sheriff, 2002; Herron, 2011). In the past few dec- ades, geophysicists have developed various versions of horizon-tracking algorithms (e.g., Harrigan et al., 1992; Leggett et al., 1996; Huang, 1997; Hoecker and Fehmers, 2002; Faraklioti and Petrou, 2004; Yu et al., 2008; Patel et al. 2010; Herron, 2011, 2015; Yu et al., 2011; Wu and Hale, 2015a). The algorithms have been successfully and widely used in routine seismic visualization and in- terpretation. The traditional workflow for horizon tracking works effectively for reflections with strong amplitude, high signal-to-noise ratio (S/N), and high de- gree of lateral continuity, which is typical of shallowly buried young reservoirs with strong contrast in density and velocity. However, in many cases of seismic char- acterization of reservoirs with weak contrast in density and velocity, particularly of unconventional reservoirs in deeply buried old sedimentary sections, the task of tracking reservoir horizons becomes difficult and chal- lenging. For example, in the Middle Devonian Marcellus Shale reservoir in the Appalachian Basin (USA) and the Lower Silurian Longmaxi Shale reservoir in the Sichuan Basin (China), the weak acoustic contrast and/or the thinness of reservoir beds make the reflection events too weak to identify. Incapability to track weak reflec- tions associated with shale reservoirs hinders our capability to visualize and characterize reservoirs (e.g., Halbouty, 1982; Li et al., 2005; Wang and Zhao, 2010; Wang et al., 2012). Tracking weak reflections has been a long-time issue in the seismic interpretation and reservoir characteriza- tion communities. In many cases with picking weak re- flections associated with shale reservoir formations, the waveform-based tracker often encounters two common problems (Herron, 2015). First, the tracked horizon pi- lots the picking to the nearest strong reflection that de- viates from the right track of the intended horizon. Second, the tracking terminates unexpectedly at dis- continuities at which the reflection is either poorly im- aged or changes significantly, causing the local waveform similarity to fall below the threshold. Both 1 Georgia Institute of Technology, School of Electrical and Computer Engineering, Center for Energy and Geo Processing (CeGP) at Georgia Tech and KFUPM, Atlanta, Georgia, USA and Schlumberger, Houston, Texas, USA. E-mail: hdi7@gatech.edu. 2 SINOPEC, State Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Effective Development, Beijing, China and West Virginia University, Department of Geology and Geography, Morgantown, West Virginia, USA. E-mail: dengliang.gao@mail.wvu.edu. 3 Georgia Institute of Technology, School of Electrical and Computer Engineering, Center for Energy and Geo Processing (CeGP) at Georgia Tech and KFUPM, Atlanta, Georgia, USA. E-mail: alregib@gatech.edu. Manuscript received by the Editor 22 February 2018; revised manuscript received 18 May 2018; published ahead of production 13 August 2018; published online 25 September 2018. This paper appears in Interpretation, Vol. 6, No. 4 (November 2018); p. SN47–SN56, 15 FIGS. http://dx.doi.org/10.1190/INT-2018-0053.1. © 2018 Society of Exploration Geophysicists and American Association of Petroleum Geologists. All rights reserved. t Special section: Shale oil and gas enrichment mechanisms and effective development Interpretation / November 2018 SN47 Interpretation / November 2018 SN47 Downloaded 11/08/18 to 192.23.22.190. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/