基于连续小波变换的地下储存CO2泄漏高光谱遥感监测

为尝试应用高光谱遥感监测地表植被变化而间接探测CO2轻微泄漏信息。试验于2008年5-9月在英国诺丁汉大学Sutton Bonington校区完成(52.8°N，1.2°W)，以多年生黑麦草（cv Long Ley）为研究对象，测量了6次黑麦草冠层光谱，利用连续小波变换对其冠层光谱进行处理分析，寻找对CO2泄漏胁迫较为敏感的波段，从而实现CO2泄漏信息的识别。研究结果表明：随着CO2胁迫程度增大，黑麦草冠层光谱小波系数在680~760 nm范围内逐渐增大（负值），在其他区域无明显变化；而在720~745 nm范围内黑麦草冠层光谱小波系数能量总和逐渐减小。黑麦草小波系数能量总和最大值波段位于725 nm附近，且725 nm处小波系数能量总和可以在整个生育期识别CO2泄漏胁迫下的黑麦草。因此，可以通过高光谱遥感监测地表植被而间接探测地下储存CO2轻微泄漏点，结果可为将来利用机载或星载高光谱遥感监测地下封存CO2泄漏点提供参考。

Monitoring the micro-seepage of underground stored CO2 by using hyperspectral remote sensing based on continuous wavelet transform

This paper presents a study on detecting CO2 leakage spots indirectly through monitoring the responses of plants which grew on the sequestration field by using hyperspectral remote sensing.The experiment was performed from May to September 2008 at the Sutton Bonington Campus of the University of Nottingham (52.8°N，1.2°W)，UK and a perennial rye grass (cv Long Ley) was chosen and planted in the field.Canopy spectra were collected six times during the experimental period and were processed by using continuous wavelet transform (CWT) in six scales，respectively，the sensitive bands were searched for detecting CO2 leaking information.The results showed that the wavelet coefficients (negative) of the grass gradually increased in 680-760 nm with the elevation of CO2 concentration in the soil，however，there was no significant responses in other wavelengths.Moreover，in 720-745 nm，the sum of wavelet energy of the grass decreased while the levels of CO2 leakage stress increased.With CO2 leakage stress lasting，the position of maximum values of sum of wavelet energy would move toward the long bands，however，the maximum values were always located nearby 725 nm.Thus，the sum of wavelet energy in 725 nm was used to identify the grass under CO2 leakage stress，and the results suggested it can completely distinguish the grass under CO2 leakage stress from the control during the whole growth period.Therefore，it is feasible for indirectly detecting CO2 leakage spots through monitoring the responses of plants growth on the sequestration field by using hyperspectral remote sensing.Moreover，the results can provide a theoretical and methodological support for monitoring micro-seepage spots of CO2 in geological sequestration by utilizing airborne or satellite hyperspectral data in the future.