HR: 17:45h
AN: B54B-08    [Abstracts]
TI: Regional Scale Inverse Modeling of North American Carbon Fluxes
AU: * Hirsch, A I
EM: Adam.Hirsch@noaa.gov
AF: Cooperative Institute for Research in Environmental Sciences, University of Colorado at Boulder, Boulder, CO 80309 United States
AU: * Hirsch, A I
EM: Adam.Hirsch@noaa.gov
AF: NOAA/CMDL, 325 Broadway Ave., Boulder, CO 80305 United States
AU: Michalak, A M
EM: amichala@umich.edu
AF: Department of Civil and Environmental Engineering, The University of Michigan, Ann Arbor, MI 48109 United States
AU: Gourdji, S
EM: sgourdji@umich.edu
AF: Department of Civil and Environmental Engineering, The University of Michigan, Ann Arbor, MI 48109 United States
AU: Peters, W
EM: Wouter.Peters@noaa.gov
AF: Cooperative Institute for Research in Environmental Sciences, University of Colorado at Boulder, Boulder, CO 80309 United States
AU: Peters, W
EM: Wouter.Peters@noaa.gov
AF: NOAA/CMDL, 325 Broadway Ave., Boulder, CO 80305 United States
AU: Schaefer, K
EM: Kevin.Schaefer@noaa.gov
AF: NOAA/CMDL, 325 Broadway Ave., Boulder, CO 80305 United States
AU: Andrews, A E
EM: Arlyn.Andrews@noaa.gov
AF: NOAA/CMDL, 325 Broadway Ave., Boulder, CO 80305 United States
AU: Lin, J
EM: jcl@atmos.colostate.edu
AF: Department of Atmospheric Science, Colorado State University, Fort Collins, CO 80523 United States
AB: Geostatistical inverse modeling is a numerical technique that can be used to infer finely-resolved patterns of biosphere-atmosphere exchange from atmospheric mole fraction data (see Michalak et al., JGR 2004). In this study we employed geostatistical inverse modeling in a regional framework to characterize the spatial and temporal information that can be extracted from simulated mole fraction data at the location of the WLEF tall tower in Wisconsin (45.93°N, 90.27°W, 396m height). We retrieved known six-hour averages of net ecosystem exchange (NEE) generated by the Simple Biosphere (SiB) model for one month at the spatial scale of 1x1 degrees using four measurements per day. The Stochastic Time-Inverted Lagrangian Transport (STILT) model was used to generate the pseudo-data time series of CO2 at the tower and the transport matrix used in the inversion. We found that while the spatial pattern of the six-hour fluxes from individual days could not be accurately retrieved, there was good agreement between the real (from SiB) and retrieved values for the spatially-averaged six-hour and daily averaged fluxes. This agreement between spatially-averaged real and retrieved fluxes was the consequence of the geostatistical inverse modeling framework, in which poorly constrained regions tend to the optimized mean flux rather than to an explicit prior value. In contrast, the spatial patterns of the real and retrieved monthly averaged diurnal cycle of NEE were similar. Thus, there appears to be a trade-off between the spatial and temporal resolution that can be reproduced with the inverse modeling setup used in this study.
DE: 0315 Biosphere/atmosphere interactions (0426, 1610)
DE: 0429 Climate dynamics (1620)
SC: Biogeosciences [B]
MN: Fall Meeting 2005