|
|
|
Poster Session R
|
A Remote Sensing Basic Study on Relation Between the Spectral properties (0.40 – 1.10um) of the Oil and Gas Microseepage in East Hebei Province
Tian Qingjiu,Liu Yi,Zhang Lei
Institute of Remote Sensing Application,Chinese Academy of Science.
Abstract
By the field spectral measurement and analysis of the typical substance between the known oil sites and the surrounding background areas in East Hebei Province, the optimum bands were determined for the oil and gas micro seepage Remote Sensing ion the range of 0.40 – 1.10um. The feasibility was further verified for the oil and gas detection by the use of the Remote Sensing data.
Introduction
Searching for the oil and gas resources through the high spectral resolution Remote Sensing technology is a newly developed method. The oil and gas microseepage of hydrocarbon material over oil and gas deposites causes the surface material variation, for example, the surface environmental acidification, and results in geobotanic ecocline and the restrain of geobotanic growth. In addition, That Fe8+ in the surface rocks and soil reduces to Fe2+ will creat the natural landscape feature “Red rock formation fades”, clay grouting and carbonate content anomaly and so on. The variation of surface substance composition will lead to the anomaly of spectral features. This spectral anomaly can be identificated by the multi-spectral features. This spectral anomaly can be identification by the multi-spectral Remote Sensing technology. Determining the areas with the spectral anomaly of the surface material, the oil and gas resources can be inferred and exploited.
Spectral Measurement and Processing
SE-590 spectrometer (made in America) was selected during the field spectral measurement. The spectral range is 0.38 – 1. 10.um, spectral resolution 3 -5 nm, spectral interval 10nm. As the spectral range is visible and near-infrared, the mainly selective subjects are geobotany with the sensitive spectral property, such as locust, reed, rich poplar and soil water were selected as well. The spectra of every sampling was measured in the known oil sites and in background areas, and the
data was recorded n tapes, finally were transmitted to PC computer and processed : curve drawing, statistical analysis, ratio and subtraction calculation and so on.
Spectral Feature Analysis
In this study area, the spectral analysis of the earth’s substance was interfered seriously by a lot of artificial factors. Only the natural perennial plant and soil was relatively stable, and the study their spectral features was representative. Therefore, the spectra of locust and soil were selected to discussed as following in this paper.
Figure 1 showed us the comparison of relfectivities of the leaves of locust between the known oil sites and the background areas, and the reflectivity in oil sites is lower than that in background, and red shift appeared for the absorption edge of chlorophyll. Figure 2 showed that a lot of metal element content of plant in oil site were more than that in background area, special for the nobel metal element Ni, Zn, Li, Cr, Mo, Cu and Co. The spectral feature of locust was just due to the effect of the nobel metal elements.
Figure 3 showed the soil spectral features curves. Curve b (soil in oil site) is more smooth and lower than curve a (soil in background) totally. The Fe3 + absorption valley of curve a at 0.96um is not obvious, because soil in oil site contains much more organic material and the Fe3 + is reduced to Fe2+ (See table 1).
Optimum – band Selection
In order to acquire and process multi – spectral microspeepage Remote Sensing data, the optimum band selection was carried out by the field spectral analysis. The methods of “subtraction” and “ratio” were adopted. The spectral features of the earth’s substance (popular, locust, reed, soil, rich) were selected to be researched. At first, the field spectral internal of the these samples imitated the multi-spectral scanner (made in Chinese Academy of Sciences) with 16 channels (see table 2) in 0.40 – 1.10um range. Figure 4 and Figure 5 showed us these optimum bands for micro seepage Remote Sensing :
0.54 – 0.65 um -.77 – 0.85um 0.89 – 1.01um
Result Verification
In the multi-spectral scanning image with radiation correction and reflectivity - transform for this study area, every channel digital number (DN) were exacted for the locust in the oil site and in background area. Figure 6 appeared the optimum band s: 0.85 – 0.89um, 0.89 – 0.93um. 0.97- 1.01um. This results was corresponding with the result discussed above.
In this study, the information extracted from image by the radio method, and geobotany – locust spectral anomaly was very clear (figure 7 photo omitted). This radio method is basic on following formation :
DDN = [(band11 – band4) – c] / [band4-c]
where
DDN - - gradient value C – dataum value
band11, band4 represent DN value of channel 11 and channel 4
|
|
|
|
|
|
|
