Factor Analysis of Well Logs for Total Organic Carbon Estimation in Unconventional Reservoirs

Abstract

Several approaches have been applied for the evaluation of formation organic content. For further developments in the interpretation of organic richness, this research proposes a multivariate statistical method for exploring the interdependencies between the well logs and model parameters. A factor analysis-based approach is presented for the quantitative determination of total organic content of shale formations. Uncorrelated factors are extracted from well logging data using Jöreskog’s algorithm, and then the factor logs are correlated with estimated petrophysical properties. Whereas the first factor holds information on the amount of shaliness, the second is identified as an organic factor. The estimation method is applied both to synthetic and real datasets from different reservoir types and geologic basins, i.e., Derecske Trough in East Hungary (tight gas); Kingak formation in North Slope Alaska, United States of America (shale gas); and shale source rock formations in the Norwegian continental shelf. The estimated total organic content logs are verified by core data and/or results from other indirect estimation methods such as interval inversion, artificial neural networks and cluster analysis. The presented statistical method used for the interpretation of wireline logs offers an effective tool for the evaluation of organic matter content in unconventional reservoirs.

Robust reservoir identification by multi-well cluster analysis of wireline logging data

“…Non-hierarchical cluster analysis was used for assisting permeability prediction with transforming the well logs into electrofacies in dolomite and sandstone intervals in the Ogallah Field, USA [ 10 ], specifying the facies for a well in sandstone formation in West Africa before predicting the formation permeability [ 11 ], and the identification of heterogeneous carbonate reservoirs in a Southern Iraqi oilfield [ 12 ]. Other recent well log applications include improved electrofacies identification and lithology classification [ 13 , 14 ], assisting pseudo-well stochastic seismic inversion [ 15 ], automated layer-thickness determination for inversion procedures and estimation of typical log response values of hydrocarbon formations [ 16 ], clustering of incomplete core laboratory datasets [ 17 ], sweet spot identification and separation of different gas-bearing intervals in unconventional reservoirs [ [18] , [19] , [20] ]. As new alternative, machine learning tools can help to solve geophysical inverse problems.…”

Analysis of Unconventional Oil and Gas Reservoirs using Well Logging, Geochemical and Seismic Data

“…Source rocks are sedimentary rocks that contain fine-grained organic material such as clay or shale and have the potential to produce and become oil and gas reservoirs (Muther et al, 2021). In conducting unconventional oil and gas exploration, several methods can be used, such as the well-logging method (Szabó et al, 2021), seismic method (Harilal & Tandon, 2012), and geochemical analysis (Setyawan et al, 2020). The well-logging method is one of the methods used to obtain subsurface physical information by drilling directly (Fatahillah et al, 2017).…”

Assessment of Three Non-Linear Approaches of Estimating the Shale Volume Over Yewa Field, Niger Delta, Nigeria

“…Reservoir rocks can be said to be clean if the shale volume is less than 10%. If the shale volume is greater than 10% but less than 34%, the rocks can be termed shaly (containing some quantities of shale), Shale volume up to or greater than 34% reveals a pure shale formation (Szabo et al, 2021). Knowing the volume of shale will help the reservoirs analysts to correctly predict other petrophysical parameters which are also important in ranking a reservoir (Pandey et al, 2020).…”