Final Report: Improved Site Characterization And Storage Prediction Through Stochastic Inversion Of Time-Lapse Geophysical And Geochemical Data

Ramirez, A ; Mcnab, W ; Hao, Y ; White, D ; Johnson, J

Lawrence Livermore National Laboratory

关键词: Storage;    Forecasting;    Progress Report;    Site Characterization;    58 Geosciences;   

DOI  :  10.2172/1022880
RP-ID  :  LLNL-TR-480694
RP-ID  :  W-7405-ENG-48
RP-ID  :  1022880

美国|英语

来源: UNT Digital Library

【 摘 要 】

During the last months of this project, our project activities have concentrated on four areas: (1) performing a stochastic inversion of pattern 16 seismic data to deduce reservoir bulk/shear moduli and density; the need for this inversion was not anticipated in the original scope of work, (2) performing a stochastic inversion of pattern 16 seismic data to deduce reservoir porosity and permeability, (3) complete the software needed to perform geochemical inversions and (4) use the software to perform stochastic inversion of aqueous chemistry data to deduce mineral volume fractions. This report builds on work described in progress reports previously submitted (Ramirez et al., 2009, 2010, 2011 - reports fulfilled the requirements of deliverables D1-D4) and fulfills deliverable D5: Field-based single-pattern simulations work product. The main challenge with our stochastic inversion approach is its large computational expense, even for single reservoir patterns. We dedicated a significant level of effort to improve computational efficiency but inversions involving multiple patterns were still intractable by project's end. As a result, we were unable to fulfill Deliverable D6: Field-based multi-pattern simulations work product.