基于高分一号的农田植被区土壤湿度反演研究

Study on soil moisture inversion in farmland vegetation area based on GF-1

以国产高分一号(GF-1)遥感数据为数据源,以荆州市公安县为实验区,选择垂直干旱指数(PDI)、改进型的垂直干旱指数(MPDI)和植被调整垂直干旱指数(VAPDI),对土壤湿度指数反演的效果进行比较和验证,同时为减少水体对地表反射率的干扰,采用归一化差异水体指数法(NDWI)对水体进行掩膜处理.研究结果表明,PDI、MPDI和VAPDI分别反演得到的土壤湿度值与实测值的相关系数为0. 520 5、0. 697 2、0. 687 2,实测土壤含水量验证精度评价也表明各模型均能满足反演的精度要求;在植被覆盖较好的农区,MPDI和VAPDI能够在一定程度上克服混合像元对土壤湿度光谱信息的影响,模型拟合效果优于PDI.研究结果可为江汉平原大范围和动态监测表层土壤湿度指数提供理论基础和实践参考.

In order to explore the applicability of soil moisture inversion model in farmland vegetation area,three models of soil moisture inversion based on PDI,MPDI and VAPDI were built by using the domestic GF-1 remote sensing data with high spatial and temporal resolution as the data source,and Gongan,Jingzhou County as the study area.After the mask processing to water body using normalized difference water index(NDWI)method,the interference of water to the acquisition of surface reflectance is reduced to some extent.The results show that the correlation coefficients between the measured soil moisture values and the inversed values obtained from three indexes are 0.520 5,0.697 2 and 0.687 2.Among them,MPDI and VAPDI have similar fitting effects,both better than PDI.The values inversed from MPDI and VAPDI also have higher discrimination in the farmland vegetation area,and show better overall effects than PDI,which can overcome the influence of vegetation coverage on soil moisture inversion to a certain extent.