Application of two shortwave infrared water stress indices to drought monitoring over northwestern China

Drought is a harmful and headachy natural disaster in the world, which has caused considerable loss to agricultural production and economy per year with high frequency. It is therefore very important and necessary for drought monitoring over large scales by remotely sensed techniques. In order to establish physically meaningful water stress index, the SWIR bands with strong water absorption features are used in this study, then the shortwave infrared water stress indices (SIWSI) are constructed using NIR and SWIR bands. A representative arid and semi-arid region over northwestern China, Ningxia region, where droughts occur frequently, is selected to assess the drought status using two indices of SIWSI_{6, 2} and SIWSI_{7, 2} derived from NIR channel 2 and SWIR channel 6 or 7 of Moderate Resolution Imaging Spectroradiometer (MODIS) sensor, combining with ground truth data of soil moisture observed by the meteorological stations across the whole study area. The fitted regressions of negative exponential curves indicate that both indices have significantly correlated with the in-situ measurements (P<0.01), the greater of the indices, the drier of the land. The SIWSI_{6, 2} performs slightly better than the SIWSI_{7, 2} (r² =3D0.56 and 0.48).