Abstract: In order to improve the instability of dry and wet edges of Ts-NDVI space in drought monitoring, a general Ts-NDVI space method was presented in the paper. Firstly, the single Ts-NDVI space of every year was constructed separately. Then the maximum and minimum temperatures were extracted respectively from the dry and wet edges of the single spaces to combine the general space suitable to all years. The experiment and validation were conducted with MODIS data, precipitation and SPI. The results show that the general space without changing the actual observation of every pixel is free from the any in-situ ancillary data. The general space method improves the stability of Ts-NDVI space, and can give more accurate surface moisture information than the primary single space. The general space is not only related to the recent rainfall but also related to the past cumulative rainfall.