Remote Sensing, Fuzzy Logic and GIS in Petroleum Exploration

Abstract

Abstract Despite the known practical applications of remote sensing in a wide range of industries and situations, it has not been used extensively in petroleum exploration, which has relied mostly on geological and / or geophysical surveys. With the advances made in sensing equipment since the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) became operational, extraction of potential indicators such as: surface emissivity, surface kinetic temperature, brightness temperature and surface radiance needs to be re-evaluated in the context of petroleum exploration. This research will mainly investigate the application of remote sensing techniques to existing petroleum reservoirs with the objective of uncovering diagnostic patterns that could ultimately be used for evaluating virgin territory. Although oil and gas reservoirs are deep below the surface, they have some indicators, which can be detected on the ground. Hazy patches on Landsat MSS images were considered by Collins (1973) as a clue to explore for oil due to their high correlation with existing oil and gas fields, but since the phenomenon was never explained it fell out of favor[1]. This research will try to identify more of these surface phenomena (e.g. surface temperature (TIR) derived from satellite data, vegetation cover, alteration zones, geochemical surface characteristics and any other available surface data) and study their correlation with the presence of oil and gas regardless of whether they can be explained or not. By employing GIS and fuzzy logic a dynamic model will be developed which can be applied to any new petroleum exploration target using the variable input data from that particular exploration target. The selected study area consists of almost 50 existing petroleum oil fields onshore Iran. Following fieldwork in the project area, various valuable layers of information were collected. 20 ASTER scenes over the study area have been received from NASA and data have been retrieved from these satellite images. These data was finally integrated with all the other available data layers to produce the dynamic model. The study is highly significant due to its capability of minimizing exploration costs when remote sensing is combined with other current conventional exploration techniques during the reconnaissance stage. During the course of research quite a good correlation has been observed so far between the analysis results and the presence of oil as a positive indication of the applicability of the proposed model for hydrocarbon exploration.