--> Multigeophysical Exploration

AAPG European Region, Geothermal Cross Over Technology Workshop, Part II

Datapages, Inc.Print this page

Multigeophysical Exploration

Abstract

Exploration for natural resources is based on the use of geophysical data and advanced mapping tools. Petroleum exploration and geothermal exploration has many fundamental things in common, and a few important details that are different. In both cases, exploration is carried out within the a play concept. The purpose of geophysical studies in the predrill exploration phase is to contribute to prospect evaluation, and to understand the risk factors defining the play. In petroleum, the risk factors are source rock, reservoir and trap/seal. In geothermal exploration, the corresponding factors are heat source, container and recharge. The most important geophysical technique in petroleum exploration is seismic imaging. This is used to define the geological framework, and to identify reservoirs and closures. Non-seismic data is heavily used only when seismic methods fail. Examples are subsalt imaging, and fluid prediction, when CSEM data is used together with AVO inversion to obtain higher confidence (Nordskag et al., 2013). The most useful geophysical parameter for geothermal exploration is resistivity, due to its direct sensitivity to temperature. Seismic information is often obtained only from passive methods. Reflection seismic should probably be used more often, to obtain better structural information. In complex exploration problems, on the regional as well as the prospect scale, it’s beneficial and necessary to systematically utilize all the data available, together with general geological knowledge. We have experienced, in both petroleum and geothermal exploration, that the best results are obtained when combining at least one mechanical parameter (P-wave velocity, S-wave velocity, density) and one electromagnetic parameter (resistivity, susceptibility, magnetization). A general procedure for multigeophysical inversion can be presented as a Bayesian network. We assume that geophysical model parameters are conditionally independent, but depend on common parent parameters of interest. Multigeophysical inversion can then be performed in two steps: (1) Inversion for model geophysical models given data. (2) Inversion for properties given model parameters. The first step is conventional geophysical imaging and inversion. The second step is typically rock physics inversion. We do the multigeophysical inversion in a pragmatic way. Only the second inversion step is fully statistical. Then we can utilize geophysical models computed with different methods, by various service providers, and produce results with fast turnaround. From the geological models, we may infer the properties of key interest, such as subsurface temperature in geothermal explorationr heat flow for basin modeling (Hokstad et al., 2017). Also, geochemical data and remote-sensing imagery can be utilized in the inversion This research was partly funded by the EC Horizon 2020 project DEEPGS.