This paper presents a method to generate a >70% accurate predictive map of sweet spots in shale plays prior to drilling. It indicates where to drill, and where not. The approach uses DNA analysis of surface soil samples, to derive information on the mix of microbial species in the samples. Using our database to correlate DNA in soil samples and production data of earlier drilled areas, the new DNA fingerprint is an indicator of the presence of vertical micro-seepage to the surface from hydrocarbon accumulations in the subsurface - including sweet spots in shale plays. In times of low oil and gas prices, stepping away from grid drilling and implementing an iterative procedure of prioritized development of higher profitable areas of a play, could prove a game changing strategy.
First technological break-through: DNA ‘fingerprinting’, biotechnology
The occurrence of vertical upward micro-seepage has been known for decades and is extensively described in the literature. But the microbial life is much more complicated than the few species that were known as hydrocarbon-oxidizing bacteria. It is necessary to determine the complex composition of microbes -not only those that flourish at micro-seepage sites, but also those that are eliminated under such conditions and are therefore found in reduced concentrations above sweet spots. Recent developments in DNA analysis techniques have made this complex and previously expensive problem efficiently and economically solvable.
Second technological break-through: Big Data, Machine Learning, super computing
The millions of microbes counted in thousands of soil samples by applying 16SrDNA ‘fingerprinting’ techniques create terabytes of data that must be correlated with the presence of hydrocarbons. This is a huge mathematical and computational big data problem. Advancements in machine learning applications together with parallel computing (Hadoop in the cloud, GPU) have made it possible to construct robust and reliable predictive DNA based models for sweet spot locations.
The combination of both technologies will be illustrated with two case studies:
a validation case in the Haynesville shale, an area with known production data, and
two areas in The Netherlands where the prospectivity of two shale formations was estimated.
