The Analysis of Open Hole Wireline Logs and Core to Predict Grain Size Classes of Western Siberia Jurassic Reservoirs

Abstract

Grain size distribution is one of key sedimentological characteristics used to build fullfield depositional models. This paper is devoted to the integration of conventional and special wireline logs with core data to predict grain size classes of Western Siberia Jurassic formation composed of siliclastic rocks varying from very fine to pebbly sands, silts and shales. Currently core is studied to identify grain size classes, which causes significant uncertainty in uncored intervals. Techniques involving nuclear magnetic resonance (NMR) and borehole image (BHI) logs are used to overcome this shortcoming. These techniques are non-applicable for revisiting of historical and development wells and wells with conventional logs only. We propose an algorithm to estimate grain size classes with conventional wireline logs in each single well calibrated on core and NMR data in reference wells. Five grain size classes (very fine to medium sands, silts and shales) are identified based on conventional wireline logs derived effective porosity tied to the quantity of bound water. Testing of the technique on 50 wells in alluvial and marine depositional environments (Yugra Area, Western Siberia, Russia) is resulted in good match between grain size classes from conventional wireline logs, NMR and core.