Abstract:
The Soil Moisture Active Passive (SMAP) satellite, to be launched in 2013, will use both radiometer and radar data to estimate soil moisture. Improved soil moisture knowledge has many applications in hydroclimatology, numerical weather prediction, flood forecasting, and human health. In this thesis, an observing system simulation experiment (OSSE) was used to study the error structure of radiometer measurements using two different retrieval algorithms. In an OSSE, geophysical fields are used to create a model of surface emission, which is coupled to an orbital sampling module and proposed retrieval algorithms. Comparing output from the retrieval algorithm to the starting soil moisture values demonstrates retrieval error. Significant uncertainty remains about the optimal representation of the effect of dielectric mixing, soil roughness, and vegetation opacity on radiometric emissions at a given soil moisture. The effect of this uncertainty on retrieval algorithms is studied by using different representations for each term in the forward and retrieval modules of the OSSE. Uncertainty due to roughness causes less error than errors in dielectric mixing and vegetation opacity treatment. In both algorithms, the retrieval shows a spatially variable bias, which is particularly large when using a single-polarization retrieval algorithm. The spatial and temporal variation of the bias, and the implications for characterization and removal of this bias as a possible error reduction strategy, are discussed.
Description:
Thesis (S.B. in Environmental Engineering Science)--Massachusetts Institute of Technology, Dept. of Civil and Environmental Engineering, 2009.Cataloged from PDF version of thesis.Includes bibliographical references (p. 57-61).
