HR: 1340h
AN: B33A-0248    [Abstracts]
TI: Reconciling Top Down and Bottom Up Approaches to Understand Land Carbon Cycle Variability
AU: * Collatz, G J
EM: jim.collatz@nasa.gov
AF: NASA's Goddard Space Flight Center, Biospheric Sciences, Code 923, Greenbelt, md 20771 United States
AU: Gurney, K R
AF: Department of Atmospheric Sciences, Colorado State University, Department of Atmospheric Sciences, Fort Collins, CO 80523
AU: Denning, A S
AF: Department of Atmospheric Sciences, Colorado State University, Department of Atmospheric Sciences, Fort Collins, CO 80523
AU: Randerson, J T
AF: Department of Earth Science, University of California Irivine, 3212 Croul Hall, Irvine, CA 92697
AU: van der Werf, G R
AF: USDA-FAS, NASA/GSFC, Goddard Space Flight Center, Code 923, Greenbelt, MD 20771
AB: Cycle Variability Two fundamentally different approaches for estimating global carbon sources and sinks have been used over the past 15 years. The so-called "Top-down" approach involves analysis of atmospheric composition and often includes inversions of atmospheric transport. Bottom-up approaches, on the other hand, involve using carbon cycle process models driven by various observational data. Reconciling the results of these two approaches can provide powerful constraints on each but is challenging because of the large uncertainties in atmospheric measurements and transport and in our understanding of the processes controlling biogeochemical cycling of carbon. Recently, the Atmospheric Carbon Inversion Intercomparison (TransCom 3) completed mean seasonal cycle and interannual variability inversions using 12 transport models. Their results include predictions of biogeochemically driven net carbon fluxes with associated uncertainties for the globe divided into 22 regions, half of which are land regions. The cyclo-stationary inversions predicted the mean seasonal cycle as well as the mean sink/source of each region. The interannual inversions predicted the interannual variability in the sources and sinks for each region between 1980 and 2000. This study describes an analysis of the processes controlling biogeochemically driven net carbon fluxes over the seasonal cycle for each of the Transcom land regions. The processes considered are those included in the CASA biogeochemical model. The seasonally variable model inputs include NDVI, temperature, precipitation and solar radiation and burned area. The contributions of NPP, heterotrophic respiration and fire season to the seasonal cycle are evaluated for each of the 11 TransCom 3 land regions. We prescribed plausible scenarios in the biogeochemical model to evaluate the mechanisms responsible for the size and seasonality of the mean annual carbon sinks reported by TransCom 3. Initial results will also be presented for a top-down/bottom-up analysis of interannual variability in land biogeochemical carbon fluxes.
DE: 1615 Biogeochemical processes (4805)
DE: 0315 Biosphere/atmosphere interactions
DE: 0400 Biogeosciences
SC: Biogeosciences [B]
MN: 2004 AGU Fall Meeting