The difficulty of determining effective water saturations in shaly-sand oil reservoirs is an old industry problem. Zones with high water saturation may not be developed even though the resistivity logs show that hydrocarbon exist and production data indicate low water cuts. Reservoirs offshore Southwest Trinidad are examples that show a high degree of mismatch between conventional log analysis and test/production results. These shaly-sandstone reservoirs have water saturations of 50-60% and produce water-free oil or have small water cut values (<5%) based on production reports. The benefits of Nuclear Magnetic Resonance (NMR) logging data to improve shaly-sand analysis and the ability to separately predict mobile water and bound water have been demonstrated by many practitioners. However, the methods of integrating NMR data can be standalone, deterministic or statistical and the type of petrophysical outputs can vary widely.
This study assesses the impact of integration methods of NMR/Open-hole data on petrophysical outputs by carrying out an extensive study on log data from five (5) wells offshore Southwest Trinidad. Methodologies for quality control of NMR data and deterministic and statistical workflows for integrating NMR and Conventional Logging data were demonstrated. An effective baseline for comparison was established by using the same environmentally corrected log data, shale/ clay parameters, water resistivity (Rw) and saturation exponents for all techniques. A Dual Water (DWM) and Wet Shale Model (WSM) workflows were developed and three sets of analyses were conducted. In the first analysis Conventional Logging data only were applied in a typical deterministic approach. In the second analysis NMR data and Conventional data were applied using a modified deterministic approach. In the third analysis NMR and Conventional log data were applied in a statistical approach via a system of simultaneous equations. All of the corresponding petrophysical outputs from these three methods of analyses were then compared with core and production data.
NMR-derived total porosities were found to match core porosities regardless of shale/clay content, whereas density log porosities match only in clean reservoir sections. In shaly intervals, the Neutron-Density porosities were 5-7% higher and Density porosities were 3-5% lower than core porosities. The results from this study also show that total water saturations (Swt DWM) using NMR- derived porosities (total and bound) were similar to core data. In shaly reservoir sections, Swt-DWM and Swt-WSM using conventional logging data were 10-15% and 15-20% higher than core water saturations respectively. The integration of NMR data using a statistical approach gives the most reliable results for computing irreducible water saturations for shaly sand reservoirs with high water sturations and low water-cut.
This case study illustrates how to undertake shaly-sand analysis using NMR data, including the quality control process, deterministic methods, statistical approaches and petrophysical outputs that are obtained.