Type: Poster presentation
Venue: AGU Fall Meeting 2012
Citation:
John Galantowicz; Arindam Samanta; Hanqin Tian (2012) Integration of Daily Inundation Extent Estimates into an Ecosystem-Atmosphere Gas Exchange Model. AGU Fall Meeting 2012, San Francisco, CA.
Soil moisture and the spatial extent of soil saturation, transient inundation, and wetland ecosystems are key determinants of greenhouse gas (GHG, e.g., methane) emissions from the land surface to the atmosphere. We are investigating how near-daily surface water and soil moisture observations from NASA's planned Soil Moisture Active-Passive (SMAP) mission can be integrated into an ecosystem-atmosphere gas exchange model to improve its estimates of GHG fluxes. SMAP, to be launched in 2014, will combine 1- to 3-km resolution synthetic aperture radar (SAR), 40-km-resolution L-band radiometry, and 3-day revisit period to make a novel dataset expected to provide inundation and soil moisture estimates superior to alternative methods at that temporal-spatial scale. We are testing the potential impact of this new data source using the Dynamic Land Surface Ecosystem Model (DLEM). DLEM quantifies regional fluxes of methane (CH4), carbon dioxide (CO2), and nitrous oxide (N2O) given atmospheric forcing data, with soil saturation as a prognostic variable. We have developed a new combined SMAP-DLEM model for North America with ~9-km grid cell size and a time-varying subgrid land cover fraction scheme that allows soil moisture, inundation, and wetlands extent to be externally prescribed from remote sensing observations. The model is designed to use SMAP-like ~9-km inundation fraction inputs as an external forcing term in addition to daily meteorological inputs. We will present results of studies aimed toward validating and optimizing the SMAP-DLEM model when external inundation extent forcing data are available. To emulate SMAP observations, a daily inundation fraction dataset has been derived for 2008 using data from NASA’s Advanced Microwave Scanning Radiometer-EOS (AMSR-E). This dataset was mapped to the DLEM 9-km grid using a downscaling and data-merging algorithm that estimates the spatial extent of water cover at the resolution of digital elevation models (e.g., 30-100 m). The downscaling approach enables subgrid water cover fraction to be classified according to the permanent land cover types being inundated. We present results comparing DLEM methane fluxes from the baseline model configuration (i.e., lacking daily variation in wetlands extent) and new configurations that use daily inundation extent forcing data. We discuss sources of SMAP inundation mapping uncertainties and how errors can be mitigated to minimize their propagation into DLEM GHG flux estimates.