Abstract
Soil moisture is one of the most important hydrological values for a long-term water cycle variation as well as a rain-fed agriculture in developing countries. This research uses the data from AMSR2 boarded GCOM-W satellite, and examined an estimation approach to eliminate the effect of brightness temperature from water and irrigation areas, to avoid the heterogeneity in the footprint area of 6.9GHz. The targeted area is in the Pursat observation site in the west side of Tonlesap lake in Cambodia. Its target period is one year from August 2013 to July 2014. Firstly we calculated the brightness temperature after removing the effect from water and irrigation area, then input this brightness temperature into the LDAS-UT, and finally estimated the soil moisture. Validation was made by the 10cm depth soil moisture observation data in the Pursat site, and the results showed that the estimation accuracy was much improved.
