Satellite Microwave Retrieval of Arctic and Boreal
Soil Temperature and Moisture for Biophysical Monitoring
Lucas
A.
Jones, University of Montana, lucas@ntsg.umt.edu
John
S.
Kimball, University of Montana, johnk@ntsg.umt.edu
(Presenting)
Kyle
C.
McDonald, Jet Propulsion Laboratory, California Institute of Technology, kyle.mcdonald@jpl.nasa.gov
Eni
G.
Njoku, Jet Propulsion Laboratory, California Institute of Technology, eni.g.njoku@jpl.nasa.gov
Walt
Oechel, San Diego State University, oechel@sunstroke.sdsu.edu
Soil temperature and moisture are important drivers of plant and soil biophysical processes and are strongly coupled to a changing climate in the northern latitudes. Satellite passive microwave sensors such as AMSR-E on Aqua are sensitive to soil temperature, surface moisture and vegetation, and provide a potentially useful means for quantifying spatial patterns and regional monitoring of these variables in arctic and boreal regions where frequent cloud cover, low solar illumination and sparse surface station networks constrain regional monitoring from other means. We applied AMSR-E time-series multi-channel, dual-polarization brightness temperatures with surface biophsysical information from intensive monitoring sites across Alaska and Canada to develop and verify retrieval algorithms for soil temperature and surface moisture. Microwave V-polarization brightness temperatures were well correlated (r >0.70; p<0.001) with near surface (< 8 cm) soil and above and below forest canopy air temperatures. We applied a microwave polarization ratio to account for land cover heterogeneity effects on microwave emissivity, which also corresponded closely with vegetation seasonal phenology inferred from the MODIS LAI time series. We then applied multi-band regression and semi-empirical polarization ratio algorithms for near surface soil temperatures, which corresponded closely (RMSE < 3.8 K; R2 > 0.80) to soil temperature data from surface network sites. The AMSR-E data were also capable of extracting meaningful soil information at even greater (up to 0.5m) soil depths. The algorithms were found to be relatively robust for regional application, while retrieval accuracy may be improved using relatively simple approaches for mitigating snow cover and freeze-thaw effects. We found a generally low level of agreement between surface soil moisture and AMSR-E operational L3 and back-up algorithm soil moisture products for boreal-Arctic monitoring sites. Alternative approaches for improved soil moisture retrievals specific to boreal and arctic biomes are presented.
This work was carried out at the Jet Propulsion Laboratory, California Institute of Technology, and at the University of Montana under contract to the National Aeronautics and Space Administration.