Quantitative Stratigraphic Applications in Mudstones Using a Combined Wavelet and Compositional Data Analysis Approach (CDA-WA): Case Study in Eagle Ford Group Mudstones of South and West Texas With Implications for Sequence Stratigraphy and Cyclostratigraphy

Abstract

A new methodology (called CDA-WA) was developed to process spectral gamma ray logs and x-ray fluorescence (XRF) data from Eagle Ford Group sediments in south and west Texas. It provides an objective method to detect lithologic discontinuities and cyclicity from lithology-sensitive indicators, which can provide information for both sequence stratigraphy and cyclostratigraphy. The method is intended to overcome two long-standing problems in geologic data analysis: the compositional nature of most geologic datasets and the highly discontinuous and non-stationary nature of most geologic time series that renders most Fourier methods impractical. While the methodology was developed for spectral gamma ray data and XRF data from Eagle Ford shale, but should be applicable to other unconventional reservoirs or even conventional carbonates and siliciclastics. The methodology consists of using the 2-D version of the isometric logratio transformation (logit function) for the CDA part of the methodology. The second part is the wavelet analysis (WA part). I use the Haar and real Morlet wavelets to learn about discontinuities and cyclicity, respectively. The Haar wavelet is optimized to detect abrupt shifts that could correspond to sequence boundaries and the real Morlet wavelet is optimized to detect sinusiodal cyclicity that usually occurs at Milankovitch frequency band. The results of the case study (two sections using thorium data) shows that the Haar wavelet is effective at detecting the sequence boundaries observed in an Eagle Ford outcrop in west Texas and core data from the subsurface in south Texas. Likewise the real Morlet wavelet easily identifies ~ 20 ft (6 m) cycle that is present in most of the Eagle Ford sections studied. The ability to objectively correlate based on gamma logs or XRF or ECS data will increase certainty of the correlations made.

AAPG Datapages/Search and Discovery Article #90291 ©2017 AAPG Annual Convention and Exhibition, Houston, Texas, April 2-5, 2017