Space-time analysis on downscaled soil moisture data and parameters of plant growth

Soil moisture is an important hydrological parameter, and inversion of soil moisture collaborated by multi-source remote sensing data would be the trend in the future. Utilizing passive microwave and optical data, the advantage of spatial and time resolution can be effectively integrated. This paper selected AMSR-E soil moisture product and MODIS data of the study area, which covers Guanting Reservoir and the surrounding areas throughout the year 2010. NDVI, land surface temperature (Ts), and Albedo, multiple regression method has been applied to conduct analysis on AMSR-E soil moisture data having a spatial resolution of 25km. As a result, a group of time series data featuring a cycle of 16 days and average surface soil moisture of 1km was obtained. By taking account into the type of land use and TRMM cumulative rainfall product, relevant data of Spearman and Pearson was selected and the analysis was conducted on the relationship among vegetation growth, cumulative rainfall and soil moisture, separately for different types of land cover during vegetation growth period and two other time periods throughout the year. The conclusion showed that on non-irrigated land, evident correlation between soil moisture and cumulative rainfall could be identified. On irrigated land, changes in soil moisture and that of cumulative rainfall were inconsistent. The average soil moisture of different types of land cover in the study area was closely related to maximum NDVI value; while without distinguishing the vegetation type, the correlation diminished. The research showed the vegetation’s response lag from the surface soil moisture’s change and this lagging characteristic had to do with the vegetation’s delayed response on the root zone’s soil moisture, and the extent of lagging would vary among different types of land cover.