Premium
Initial soil moisture retrievals from AMSR‐E: Multiscale comparison using in situ data and rainfall patterns over Iowa
Author(s) -
McCabe M. F.,
Wood E. F.,
Gao H.
Publication year - 2005
Publication title -
geophysical research letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.007
H-Index - 273
eISSN - 1944-8007
pISSN - 0094-8276
DOI - 10.1029/2004gl021222
Subject(s) - environmental science , hydrometeorology , data assimilation , water content , watershed , precipitation , vegetation (pathology) , in situ , moisture , remote sensing , atmospheric sciences , soil science , meteorology , geology , geography , medicine , geotechnical engineering , pathology , machine learning , computer science
Coupled with information from the North American Land Data Assimilation System (NLDAS), standard soil datasets and vegetation and land surface parameters, a land surface microwave emission model (LSMEM) is employed using AMSR‐E brightness temperatures at X‐band (10.7 GHz) to determine soil moisture over Iowa for June and July 2002. Comparisons of calculated soil moisture with in situ validation data collected from a densely monitored watershed as part of the SMEX02 campaign, indicate that accuracies in the order of 3% vol./vol. are achievable, even where agricultural surfaces such as corn and soybean dominate. Additionally, to qualitatively evaluate the derived product and identify the level of coherence between related hydrometeorological data, a comparison of soil moisture retrievals with precipitation patterns is undertaken. Consistent spatial correlation is observed between these two fields, illustrating not only that remotely sensed soil moisture has potential to provide improved characterisation of large scale precipitation patterns, but that such patterns may also offer a pathway towards enhanced assessment of soil moisture retrievals.