HR: 17:30h
AN: B44B-07    [Abstracts]
TI: Estimation of regional CO2 fluxes using concentration measurements from the ring of towers in northern Wisconsin
AU: * Uliasz, M
EM: marek@atmos.colostate.edu
AF: Department of Atmospheric Science, Colorado State University, Fort Collins, CO 80523 United States
AU: Denning, A
EM: denning@atmos.colostate.edu
AF: Department of Atmospheric Science, Colorado State University, Fort Collins, CO 80523 United States
AU: Schuh, A
EM: aschuh@atmos.colostate.edu
AF: Department of Atmospheric Science, Colorado State University, Fort Collins, CO 80523 United States
AU: Richardson, S J
EM: srichardson@psu.edu
AF: Department of Meteorology, The Pennsylvania State University, University Park, PA 16802 United States
AU: Miles, N
EM: nmiles@met.psu.edu
AF: Department of Meteorology, The Pennsylvania State University, University Park, PA 16802 United States
AU: Davis, K J
EM: davis@met.psu.edu
AF: Department of Meteorology, The Pennsylvania State University, University Park, PA 16802 United States
AU: Zupanski, D
EM: Zupanski@cira.colostate.edu
AF: Cooperative Institute for Research in the Atmosphere, Colorado State University, Fort Collins, CO 80523 United States
AB: The WLEF TV tower in northern Wisconsin is instrumented to take continuous measurements of CO2 mixing ratio at 6 levels from 11 to 396m. During the spring and summer of 2004 additional CO2 measurements were deployed on five 76 m communication towers forming a ring around the WLEF tower with a 100-150 km radius. The data from the ring of towers are being used to estimate regional fluxes of CO2. The modeling framework developed for this purpose is based on SiB-RAMS: Regional Atmospheric Modeling System linked to Simple Biosphere model. This model system is capable to realistically reproduce diurnal cycle of CO2 fluxes as well as their spatial patterns in regional scale related to different vegetation types. However, there is still significant uncertainty in simulating atmospheric transport of CO2 due to synoptic and mesoscale circulations. We are attempting to assimilate available CO2 tower data into our modeling system in order to provide corrections for fluxes simulated by the SiB-RAMS. These corrections applied separately to respiration and assimilation fluxes have spatial patterns but are assumed constant in time during a period of 5 to 10 days. The CO2 data assimilation is based the Lagrangian Particle Dispersion (LPD) model and two different inversion techniques. The LPD model is driven by meteorological fields from the SiB-RAMS and is used for a regional domain in its adjoint mode to trace particles backward in time to derive influence functions for each concentration sample. The influence functions provide information on potential contributions both from surface sources and inflow fluxes that make their way through the modeling domain boundaries into the CO2 concentration sample. Then the Bayesian inversion technique is applied to estimate unknown corrections for the CO2 fluxes. Several tests of the modeling framework were performed with the aid of model generated concentration pseudo-data. Different configurations of source areas within 500 km radius from the WLEF tower and different assumptions concerning expected model-data mismatch error were investigated. Finally, real CO2 data observed at the towers were used to provide preliminary estimates of CO2 regional fluxes. The optimal time period within which the flux corrections are assumed to be constant in time is still being investigated. It depends on resolution and configuration of source areas, atmospheric transport patterns and available data including gaps in the observed time series. The Bayesian inversion technique is limited to a rather small number of source areas when applied to time series of CO2 concentration from several towers. As an alternative approach potentially capable of handling CO2 flux resolution comparable to the horizontal resolution of SiB-RAMS we are exploring the ensemble data assimilation method based on MLEF (maximum likelihood ensemble filter).
DE: 0428 Carbon cycling (4806)
DE: 0430 Computational methods and data processing
DE: 1637 Regional climate change
DE: 3315 Data assimilation
DE: 3329 Mesoscale meteorology
SC: Biogeosciences [B]
MN: Fall Meeting 2005