Research on Oil Sands Spectral Characteristics and Oil Content by Remote Sensing Estimation
YOU Jin-feng1, XING Li-xin1*, PAN Jun1, SHAN Xuan-long2, LIANG Li-heng3, FAN Rui-xue1
1. College of Geo-exploration Science and Technology, Jilin University, Changchun 130026, China 2. College of Earth Sciences, Jilin University, Changchun 130061, China 3. College of Urban and Environmental Sciences, Changchun Normal University, Changchun 130032, China
Abstract：Visible and near infrared spectroscopy is a proven technology to be widely used in identification and exploration of hydrocarbon energy sources with high spectral resolution for detail diagnostic absorption characteristics of hydrocarbon groups. The most prominent regions for hydrocarbon absorption bands are 1 740～1 780, 2 300～2 340 and 2 340～2 360 nm by the reflectance of oil sands samples. These spectral ranges are dominated by various C—H overlapping overtones and combination bands. Meanwhile, there is relatively weak even or no absorption characteristics in the region from 1 700 to 1 730 nm in the spectra of oil sands samples with low bitumen content. With the increase in oil content, in the spectral range of 1 700~1 730 nm the obvious hydrocarbon absorption begins to appear. The bitumen content is the critical parameter for oil sands reserves estimation. The absorption depth was used to depict the response intensity of the absorption bands controlled by first-order overtones and combinations of the various C—H stretching and bending fundamentals. According to the Pearson and partial correlation relationships of oil content and absorption depth dominated by hydrocarbon groups in 1 740～1 780, 2 300～2 340 and 2 340～2 360 nm wavelength range, the scheme of association mode was established between the intensity of spectral response and bitumen content, and then unary linear regression(ULR) and partial least squares regression (PLSR) methods were employed to model the equation between absorption depth attributed to various C—H bond and bitumen content. There were two calibration equations in which ULR method was employed to model the relationship between absorption depth near 2 350 nm region and bitumen content and PLSR method was developed to model the relationship between absorption depth of 1 758, 2 310, 2 350 nm regions and oil content. It turned out that the calibration models had good predictive ability and high robustness and they could provide the scientific basis for rapid estimation of oil content in oil sands in future.
[1] TONG Xiao-guang, ZHANG Guang-ya, WANG Zhao-ming, et al(童晓光, 张光亚, 王兆明, 等). Earth Science Frontiers(地学前缘), 2014, 3: 1. [2] ZOU Cai-neng, ZHANG Guo-sheng, YANG Zhi, et al(邹才能, 张国生, 杨 智, 等). Petroleum Exploration and Development(石油勘探与开发), 2013, 4: 385. [3] SHAN Xuan-long, CHE Chang-bo, LI Jian, et al(单玄龙, 车长波, 李 剑, 等). Global Geology(世界地质), 2008, 26(4): 459. [4] JIA Cheng-zao, LIU Xi-jian, LEI Qun, et al(贾承造, 刘希俭，雷 群，等). Oil Sands Resources and Evaluation Methods of Reserves. Beijing: Petroleum Industry Press(北京: 石油工业出版社), 2007. [5] Cloutis E A. Science, 1989, 245(4914): 165. [6] WU Chang-yong(吴常泳). Journal of Infrared and Millimeter Waves(红外与毫米波学报), 1992, 11(3): 189. [7] WANG Jin-yi, SHU Wen-pei(王津义, 舒文培). Earth Science·Journal of China University of Geosciences(地球科学·中国地质大学学报), 1996, 21(2): 215. [8] Lammoglia T, Filho C R de Souza. Remote Sensing of Environment, 2011, 115(10): 2525. [9] Cambule A H, Rossiter D G, Stoorvogel J J, et al. Geoderma, 2011, 183-184(8): 41. [10] ZENG Hua-jin, LIANG Hui-li, YOU Jing, et al(曾华金, 梁会丽, 游 静, 等). Chinese Journal of Luminescence(发光学报), 2013, 34(3): 369. [11] LIU Bin, LIU Jian-zhong, ZHANG Guang-liang, et al. Chinese Journal of Geochemistry, 2014, 33(1): 86.
