HR: 0830h
AN: B51E-1004    [PDF]
TI: Towards Constraining Regional Carbon Exchange In The Southern Great Plains With DOE ARM and COBRA Data And Tools
AU: * Jin, L
EM: ljin@lbl.gov
AF: Earth Science Division Lawrence Berkeley National Laboratory, 1 Cyclotron Rd., Berkeley, CA 94720 United States
AU: Fischer, M L
EM: mlfischer@lbl.gov
AF: Atmospheric Science Department Environmental Energy Technologies Division Lawrence Berkeley National Laboratory, 1 Cyclotron Rd., Berkeley, CA 94720 United States
AU: Lin, J C
EM: johnlin@fas.harvard.edu
AF: Hoffman Laboratory Harvard University, 20 Oxford Street Hoffman Laboratory Harvard University, Cambridge, MA 02138 United States
AU: Gerbig, C
EM: gerbig@fas.harvard.edu
AF: Hoffman Laboratory Harvard University, 20 Oxford Street Hoffman Laboratory Harvard University, Cambridge, MA 02138 United States
AU: Wofsy, S
EM: Steven_Wofsy@harvard.edu
AF: Hoffman Laboratory Harvard University, 20 Oxford Street Hoffman Laboratory Harvard University, Cambridge, MA 02138 United States
AU: Torn, M
EM: MSTorn@lbl.gov
AF: Earth Science Division Lawrence Berkeley National Laboratory, 1 Cyclotron Rd., Berkeley, CA 94720 United States
AB: Long term measurements of CO2 concentration and flux from tower sites may be effective for use in inversion estimates of regional land surface-atmosphere CO2 exchange. To explore the potential for application of tower data we have combined measurements from the 60m tower at the Atmospheric Radiation Measurement Program (ARM) Central Facility (36.61N, 97.49W) in the U.S. Southern Great Plains with a Stochastic Time-Inverted Lagrangian Transport (STILT) model [Lin et al., 2003] using a receptor oriented modeling (ROM) framework (Gerbig et al., 2003) developed for CO2 Budget and Regional Airborne Study (COBRA). We first investigate the uncertainties in the influence functions obtained from STILT: 1) STILT estimates of PBL heights show good correlation with those estimated from the radio sound profiles in May-June 2002 and 2003; 2) The measured mean concentrations on the 60 meter tower are independent of the flow directions patterns summarized from STILT particle trajectory statistics. Then we regress the observed CO2 concentration on the STILT/ROM simulated CO2 signals of different vegetation and emission sources in the growing season 2002 to obtain the scaling factors. The residuals are modeled as an autoregressive process to resolve the temporal correlation and calculate the uncertainties associated with the estimated coefficients. We find that: 1) STILT /ROM captures 60-80% of the variance in the observations; 2) Crop and shrub are the most important vegetation types that affect the regional carbon exchange; 3) Phenology and land use management cause significant seasonal variation in the scaling factors. Progress toward Bayesian inversion of the annual cycle of regional ecosystem-atmosphere CO2 exchange constrained by measurements from the 60m tower and aircraft flights, incorporating uncertainties in PBL heights, will be discussed.
DE: 0315 Biosphere/atmosphere interactions
DE: 0322 Constituent sources and sinks
DE: 0400 Biogeosciences
SC: Biogeosciences [B]
MN: 2003 Fall Meeting