|
Modeling temporal variations in soil moisture across an agricultural landscape using radar-based imagery
Josef
Kellndorfer, WHRC, josefk@whrc.org
(Presenter)
Jill
Derwin, WHRC, jderwin@whrc.org
Rebecca
Phillips, USDA-ARS, rebecca.phillips@ars.usda.gov
Eric
Davidson, WHRC, edavidson@whrc.org
Kathleen
Savage, WHRC, ksavage@whrc.org
Carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) are the most important anthropogenic greenhouse gases (GHG). Soils are the dominant natural source of N2O, and fertilized agricultural soils are a major source of increasing anthropogenic N2O. Wetlands are a source of CH4, while upland soils are a natural CH4 sink. While most anthropogenic CO2 is derived from fossil fuel combustion, a significant fraction is from land use change, including a portion from loss of soil. Variation in soil moisture can be very dynamic, and it is one of the dominant factors controlling soil aeration, and hence the balance between aerobic (CO2 producing) and anaerobic (CH4 producing) respiration. The production and consumption of N2O is also highly dependent on spatial and temporal variation in soil moisture. While soil moisture and GHG fluxes can be measured and logged at high frequency for point measurements, temporal variation across landscapes is difficult to monitor.
At an alfalfa field near Mandan North Dakota, we have collected high frequency GHG fluxes, soil moisture and biomass measurements. We use observed soil moisture measurements to develop a mechanism to model temporal and spatial variations between backscatter data from Synthetic Aperture Radar (SAR) and soil moisture. Since the presence of biomass influences radar backscatter, we account for this using data supplemented by a time series of Landsat image-derived Tasseled Cap Greenness, Tasseled Cap Wetness, and Normalized Difference Vegetation (NDVI) indices. A model of temporal soil moisture variations in alpha-alpha fields is linked with the remote sensing measurements to upscale estimates of GHG fluxes to the landscape scale.
Presentation Type: Poster
Session: Poster Session 1-A
(Tue 11:00 AM)
Associated Project(s):
Poster Location ID: 12
|