HR: 17:15h
AN: B42D-05 [PDF]
TI: Estimation of the Net Ecosystem CO2 Exchange of Chaparral Using Eddy Covariance, remote sensing and
Biome-BGC
AU: * Luo, H
EM: luo@sunstroke.sdsu.edu
AF: Global Change Research Group, Biology Department, San Diego State University, 5500 Campanile Drive,
San Diego, CA 92182 United States
AU: Oechel, W C
EM: oechel@sunstroke.sdsu.edu
AF: Global Change Research Group, Biology Department, San Diego State University, 5500 Campanile Drive,
San Diego, CA 92182 United States
AU: Sims, D
EM: dsims3@calstatela.edu
AF: California State University, 5151 State University Drive, Los Angeles, CA 90032 United States
AU: Heinsch, F
EM: faithann@ntsg.umt.edu
AF: Numerical Terradynamic Simulation Group, School of forestry, University of Montana, Missoula, MT 59812 United States
AU: Kimball, J
EM: johnk@ntsg.umt.edu
AF: Numerical Terradynamic Simulation Group, School of forestry, University of Montana, Missoula, MT 59812 United States
AB:
The chaparral ecosystem ranges from California to Arizona, USA, and into Nuevo Leon and Tamaulipas, eastern Mexico.
Tower-based eddy covariance and measurements above the canopy of a young stand (10 years old) and an old stand ($>$100 years
old) have been conducted at Sky Oaks Biological Field Station, San Diego, USA, for 5 years to quantify the seasonal and
annual variation in the net ecosystem CO2 exchange (NEE) of the chaparral ecosystem. Both stands are a C sink on an annual
basis. Surprisingly, and contrary to the hypothesis by Odum, the old, mature, "scenscent" stand was as large of a C sink as
the young stand. The stands tended to be net sinks during the wet seasons, and net sources during the dry seasons. Seasonal
changes in carbon flux reflected changes of light use efficiency and were well correlated with two remote sensing indices of
leaf pigment composition (NDVI and PRI). This implied the possibility of using remote sensing to estimate ecosystem CO2
balance. Two practical tram systems with optical sensors were set within the footprint of each eddy covariance tower in the
young stand and old stand at Sky Oaks. This comparison between data from the eddy tower and spectral reflectance from remote
sensing presented a good correlation between CO2 flux and NDVI corrected to a constant solar angle, which indicated that
remote sensing is a very promising tool for the estimation of carbon fluxes in a chaparral ecosystem. To develop methods for
scaling eddy flux measurements to the surrounding region, Biome-BGC model and MODIS results were introduced into the
analysis. The eddy covariance data illustrated sink NEE patterns during wet seasons and source patterns during dry seasons,
while the data simulated from Biome-BGC presented a net sink pattern throughout the whole year. The GPP simulated from
Biome-BGC was lower than the estimate from MODIS, and their GPP yearly patterns were also different. Peak GPP from Biome-BGC
was around June, while MODIS showed the peak GPP around March and April. Further data collection, analysis and comparison are
required to explain these differences. Key words: chaparral, carbon dioxide, eddy covariance, remote sensing, global change
DE: 0312 Air/sea constituent fluxes (3339, 4504)
DE: 0315 Biosphere/atmosphere interactions
SC: Biogeosciences [B]
MN: 2003 Fall Meeting