Retrieving soil moisture from simulated brightness temperatures by a neural network

We present the retrievals of surface soil moisture (SM) from simulated brig htness temperatures by a newly developed error propagation learning back pr opagation (EPLBP) neural network. The frequencies of interest include 6.9 a nd 10.7 GHz of the advanced microwave scanning radiometer (AMSR) and 1.4 GH z (L-band) of the soil moisture and ocean salinity (SMOS) sensor. The land surface process/radiobrightness (LSP/R) model is used to provide time serie s of both SM and brightness temperatures at 6.9 and 10.7 GHz for AMSRs view ing angle of 55 degrees, and at L-band for SMOS's multiple viewing angles o f 0 degrees, 10 degrees, 20 degrees, 30 degrees, 40 degrees, and 50 degrees for prairie grassland with a column density of 3.7 km/m(2). These multiple frequencies and viewing angles allow us to design a variety of observation modes to examine their sensitivity to SM. For example, L-band brightness t emperature at any single look angle is regarded as an L-band one-dimensiona l (I-D) observation mode. Meanwhile, it can be combined with either the obs ervation at the other angles to become an L-band two-dimensional (2-D) or a multiple dimensional observation mode, or with the observation at 6.9 or 1 0.7 GHz to become a multiple frequency/dimensional observation mode. In thi s paper, it is shown that the sensitivity of radiobrightness at AMSR channe ls to SM is increased by incorporating L-band radiobrightness. In addition, the advantage of an L-band 2-D or a multiple dimensional observation mode over an L-band 1-D observation mode is demonstrated.