基于高光谱遥感的天然气微泄露植被胁迫识别研究OA

The leakage of natural gas storage facilities can lead to issues related to economics,ecology,and safety.Hyperspectral remote sensing can indirectly detect leakage through the spectral changes of vegetation in the leakage area.We performed a field experiment using beans as test plants,to study the bean canopy spectral characteristics under natural gas leakage stress.The results showed that the bean canopy spectra during the medium-term stress period exhibited significant changes compared to those of normally growing beans.By performing the one way analysis of variance on the spectra of beans in the control and experimental plots within the 550～1 000 nm wavelength range,the bands with P<0.05 and F>Fmean were selected as sensitive bands.The natural gas stress bean index(NGSBI)was constructed using the selected bands.We used the Jeffries-Matusita distance method to compare this result with LIC,NDVI705 and NPCI.The result shows that NGSBI can well identify the stress of natural gas micro-leakage on beans during the medium-term stress period.