Surface soil moisture estimates from AMSR-E observations over an arid area, Northwest China

Abstract. In this paper, 7 years (during the growing season (April–October) of 2002–2008) of the Advanced Microwave Scanning Radiometer (AMSR-E) data taken at a frequency of 6.9 GHz for night observations at both polarizations are processed and used to conduct 7 years of surface soil moisture dataset for an arid area, in Northwestern China. This soil moisture dataset can contribute to better understanding the climate change and forecast modeling for this area. Based on the first-order radiative transfer model calculation, a unique method for estimating surface soil moisture over the study area is developed. Considering extremely complex topography over the study area, we have to present a parameterization of surface roughness at the 6.9 GHz and spatial resolution of the AMSR-E using the annual minimum MPDI (Microwave Polarization Difference Index). For validation purpose, the comparisons of soil moisture patterns with precipitation fields are made. The results indicate the evolutions of soil moisture estimated from the AMSR-E and antecedent ground daily precipitations are in a good agreement. Furthermore, a series of rainfall traces is captured over the Taklimakan Desert. Comparisons of the estimated values of soil moisture with the ground observations are also made for two reprehensive sites over 2002–2004. The results indicate there has a good agreement between them, with higher correlation coefficients (R=0.649, 0.604) and RMSE (3.5, 5.4%), and the soil moisture product derived from the AMSR-E is realistic and acceptable. This new long time series of estimated soil moisture will prove valuable for other studies of climate change and model evaluation.