--> --> A New Method to Identify Lithofacies of Complex Lacustrine Carbonate Reservoirs via Log Data – A Case Study From Yingxi Oilfield of Qaidam Basin (Northwest China)

2018 AAPG International Conference and Exhibition

Datapages, Inc.Print this page

A New Method to Identify Lithofacies of Complex Lacustrine Carbonate Reservoirs via Log Data – A Case Study From Yingxi Oilfield of Qaidam Basin (Northwest China)

Abstract

Lacustrine carbonates are the most widely distributed continental carbonates, as historically they are not common as major carbonate reservoirs and therefore have been less studied in terms of reservoir evaluation. In recent years, lacustrine carbonate reservoirs have been found in a lot of countries and regions such as Brazil, West Africa and China. We present here a case study from Yingxi oilfield of Qaidam Basin(NW China). Because of adequate sediment supply and high-salinity water, the mineral components in study area are more complex than open marine carbonates. Logging characteristics between different lithofacies are similar, which presents challenges in lithofacies identification. Therefore, we propose a parameter called Rock Fabric Factor(RFF) to identify lithofacies of lacustrine carbonates, helped with spectroscopy data(obtained from Litho Scanner) and image data(obtained from FMI). After observing more than 300 thin sections from 12 wells, we find that there are mainly five types of lithofacies in study area, they are grainstone, mudstone, laminated carbonate, shale and evaporite. First, we analyze log characteristics of each lithofacies and find that natural gamma ray, bulk density and matrix density (obtained from Litho Scanner) are the most sensitive to changes in lithofacies. Based on this, we propose RFF which can be calculated by a formula we build to identify lithofacies. Second, Litho Scanner can provide the key measurements of elements and mineralogy, and from these data we find that the five types of lithofacies have differences in mineral compositions and contents, which can help to identify lithofacies. Furthermore, FMI data related to lithological variations are interpreted on the image in terms of rock texture, stratigraphic and structural features, and fractures. Therefore, core-calibrated FMI data are used to aid lithofacies identification. It is concluded that we integrated natural gamma ray, bulk density, matrix density, Litho Scanner and FMI data in order to identify lithofacies of complex lacustrine carbonate reservoirs. Using this approach, we interpreted data from 48 wells and the average accuracy is 80%. This new method can support sedimentology research and may help in the prediction of high-quality reservoirs.