Submitted to: Specialist Meeting on Microwave Remote Sensing
Publication Type: Abstract Only
Publication Acceptance Date: 8/30/2008
Publication Date: 10/20/2008
Citation: Cosh, M.H., Jackson, T.J. 2008. Large-scale surface soil moisture estimation using in situ networks [abstract]. International Workshop on Microwave Remote Sensing for Land Hydrology. 2008 CDROM. Interpretive Summary:
Technical Abstract: Surface soil moisture estimation impacts a wide range of concerns, including agricultural management, climate, and weather modeling. New satellite technologies have been developed which allow for the estimation of surface soil moisture with moderate accuracy for the agricultural heartland of the U.S. Current remote sensing platforms have a scale of approximately 60 km (Advanced Microwave Scanning Radiometer-AMSR). Future sensors will provide improved accuracy (Soil Moisture Ocean Salinity-SMOS) and resolution that could exceed 10 km (Soil Moisture Active Passive-SMAP). These sensors measure a surface layer on the order of several centimeters. Validation of these retrievals using in situ sensors requires a strategy that considers both theoretical as well as logistical issues. Chief among these is scaling. As part of the calibration and validation program for the AMSR-E instrument, an approach was developed that provides accurate surface estimates. Existing instrumentation in four watersheds was adapted to include near surface soil moisture. These watersheds ranged in size from 150 km2 (Walnut Gulch Experimental Watershed in Tombstone, AZ) to 625 km2 (Little Washita River Experimental Watershed in Chickasha, OK) in a variety of hydrologic regions including semi-arid to sub-humid. Each watershed network has been in operation since 2002 providing an hourly record of soil moisture data at the near surface. Establishing the calibration and scaling of these networks has been the focus of field experiments over several years. These studies have revealed much about the temporal stability and spatial variability of soil moisture at large scales, providing a basis for future in situ network deployment and operation with regards to remote sensing validation. High density sampling of the watershed domains during field experimentation provided ‘true’ estimates of large scale soil moisture and the networks were shown to have high accuracy for estimating the watershed average.. Each watershed presented individual challenges in monitoring and adaptive strategies have been developed.