Detection of hydrocarbon reservoir boundaries using neural network analysis of surface geochemical data

Surface geochemical surveys could become important tools for defining the b oundaries of a hydrocarbon reservoir. Conventional statistical analysis has shown that a correlation can indeed be found between surface geochemical d ata and the location of a sample site with respect to the boundaries of a k nown reservoir. However, such analysis methods cannot be used directly as p redictive tools. This article describes the successful application of artif icial intelligence in the form of neural network analysis to determine whet her a specific sample site, given the ethane concentration in the soil and certain environmental data, is within the surface trace of the reservoir bo undaries. Data from a previous study over a known gas storage reservoir were used to train a back-propagation neural network. No attempt was made to optimize th e structure of the network. We used 85% of the data to train the network an d withheld 15% to act as unknowns. The input variables consisted of adsorbe d ethane concentration and a series of soil description and environmental p arameters. The output variable was a simple binary reflecting whether the s ample site was directly over the reservoir. The final network was able to p redict 95% of unknown sample sites. We found it necessary to include in the input data the ethane concentrations far sites on either side of each site studied. This is consistent with previous observations that a series of ad jacent sites having anomalous concentrations hold more significance than do isolated sites. We also found that the use of the land (probably reflectin g the degree of disturbance) and soil moisture are the most important envir onmental variables. This is consistent with previous conventional statistic al studies of the same data. We conclude that application of neural network s to properly designed surface geochemical studies holds promise for use in defining the boundaries of known reservoirs.