Understanding Frequency Decomposition Results Using Forward Modeling to Correlate Geology With Frequency Responses

Abstract

Frequency decomposition techniques have been used with high success rates in siliciclastic environments, from exploration to reservoir characterization for revealing depositional environments and gaining a better understanding of the geology. Frequency decomposition techniques combined with RGB blending allow for rapid screening of large volumes of data, optimizing prospect screening and lead generation. At the reservoir level these two techniques deliver superior information, in quantity and quality in relation to standard amplitude extraction. Despite the wealth of the information in the frequency decomposition color blends its correlation with geological information is not always easy to determine. Differences in frequency responses might be due to changes in thickness, changes in lithology or pore fill or a combination of all three. Isolating one is, in most cases impossible using reflectivity data alone.

In this case study we propose the use of forward modeling in order to correlate the frequency responses of a synthetic model where the geological properties are known with the frequency responses of the real data. Several forward models can be tested to determine which geologic factors are controlling the frequency responses. Wedge models using the wells will determine tuning thickness and the frequency responses of thicker sections. Models can be created from the facies classification, and populated with rock properties to generate different scenarios. These models are then convolved with a wavelet extracted from the seismic data to produce a synthetic volume. Frequency decomposition is performed with the same parameters as for the real data and a synthetic color blend is calculated. A comparison of the two blends is made to determine how closely the rock properties in the model match the real geology. The properties of the forward model can be changed until a correlation between the synthetic blend and the real blend is achieved. This method enables us to determine how rock properties influence the frequency decomposition results and correlate geological information with the responses visible in the color blends. Knowing the frequency responses at the well locations, assumptions can be made away from the wells to help with reducing the risk when drilling.

AAPG Datapages/Search and Discovery Article #90323 ©2018 AAPG Annual Convention and Exhibition, Salt Lake City, Utah, May 20-23, 2018