基于SWIR-Red光谱特征空间的农田干旱监测新方法

干旱是一种频繁发生的自然灾害,遥感监测干旱已成为重要的研究方向。可从农田遥感干旱监测最主要的两种地物类型(植被和土壤)的光谱特性分析入手,选择了对水分变化敏感的红光、短波红外波段来监测干旱状况,以短波红外与红光的差值和短波红外与红光的和构建新的光谱空间特征,提出了干旱监测的新方法——归一化的干旱监测指数NPDI。用野外实测的土壤含水量对NPDI模型进行验证,结果表明:NPDI,MPDI与10cm处的土壤含水量模型都具有较高的相关性,其R2分别为0.583和0.438,NPDI模型的监测效果要优于MPDI。此模型是对PDI,MPDI和SPSI等模型的进一步改进,可实现对不同植被覆盖度的、整个生长季的农田干旱监测,在实际的农田干旱监测中具有较高的应用潜力和推广价值。

The new method monitoring agricultural drought based on SWIR-Red spectrum feature space

Drought was a chronic, natural disaster, and Remote sensing drought monitoring had become a potential research field. In the present, short-wave infrared and red bands which sensitive to moisture variation were selected to monitor farmland drought conditions by analyzing the spectral characteristics of vegetation and soil. The goal of this paper was to provide a new method of drought monitoring--normalized drought monitoring index (NPDI), based on new constructed spectrum feature space by the difference of SWIR and Red and the sum of SWIR and Red. Field surveyed soil moisture verified NPDI model, and the result showed that NDPI and MPDI model could effectively monitor agricultural drought, and that had high correlation with soil moisture. The R2 was 0.583 and 0.438 with soil water of 10 cm. The monitoring effect of NPDI model was better than the MPDL. This model was further improvement to PDI and MPDI, and it could monitor the drought condition of different vegetation coverage and whole growing season. It has high application potential and popularization value.