[Study on exploring for oil and gas using reflectance spectra of surface soils]

Reflectance spectra in the visible and near-infrared wavelength region provide a rapid and inexpensive means for determining the mineralogy of samples and obtaining information on chemical composition. Hydrocarbon microseepage theory sets up a cause-and-effect relation between oil and gas reservoirs and some special surface alterations. Therefore the authors can explore for oil and gas by determining the reflectance spectra of surface alterations. This determination can be fulfilled by means of field work and hyperspectral remote sensing. In the present paper, firstly a macroscopical feature of reflectance spectra of typical observation points in gas fields is presented. Then a method is proposed in order to provide surface distribution information (e.g., classification) of alterations based on the reflectance spectra determined from the field, and obtain anomaly zones of the special alterations. This method has been applied to the analysis of the reflectance spectra determined in the field of Qinghai X X area, and the classification results tally with the existent gas fields in this area. A robustness analysis of the method shows that good results can be obtained when different combinations of parameters, such as samples, study band regions and thresholds, have been chosen in the process of classification. The valid classification samples and algorithms can be provided for the oil and gas exploration in progress in this area using NASA experimental hyperion hyperspectral satellite.