Spatial variations in atmospheric CO2 concentrations during the ARCTAS-CARB 2008 Summer Campaign

The Arctic Research of the Composition of the Troposphere from Aircraft and Satellites (ARCTAS) was a major NASA field campaign designed to understand the transport and transformation of trace gases and aerosols on transcontinental and intercontinental scales and their impact on the composition of the arctic atmosphere and climate. Preceding the summer ARCTAS deployment, measurements were conducted over the state of California in collaboration with the California Air Resources Board (CARB) utilizing the airborne chemistry payload already integrated on the NASA DC-8. In situ CO2 measurements were made using a modified infrared CO2 gas analyzer having a precision of 0.1 ppmv and accuracy of ±0.25 ppmv traceable to the WMO scale. This analysis focuses on the atmospheric CO2 variability and biospheric/atmospheric exchange over California. We used multi-satellite remote sensing datasets to relate airborne observations of CO2 to infer sources and sinks. Georeferencing the airborne CO2 transect data with the LANDSAT derived land cover datasets over California suggested significant spatial variations. The airborne CO2 concentrations were found to be 375-380ppm over the Pacific ocean, 385-391ppm in the highly vegetated agricultural areas, 400-420 in the near coastal areas and greater than 425ppmv in the urban areas. Analysis from MODIS fire products suggested significant fires in northern California. CO2 emissions exceeded 425ppmv in the fire affected regions, where mostly Douglas and White Fir conifers and mixed Chaparral vegetation was burnt. Analysis from GOES-East and GOES-West visible satellite imagery suggested significant smoke plumes moving from northern California towards Nevada and Idaho. To infer the biospheric uptake of CO2, we tested the potential correlations between airborne CO2 data and MODIS normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI). Results suggested significant anti-correlations between the airborne CO2 data and vegetation indices. The correlations were stronger in agriculture than forested regions. We also tested the correlations between the MODIS leaf area index (LAI) and airborne CO2 data. Although, the correlation strength was low, higher LAI values were associated with low CO2 concentrations suggesting biospheric uptake. Particulate matter (PM2.5) concentration records obtained from the US EPA corresponding to the airborne sampling domain in the peak forest fire areas suggested concentrations exceeded 42.0ug/m3. Furthermore, analysis of MODIS aerosol optical depth values at 500nm exceeded 0.70 at several places, suggesting intense fire activity. Using the airborne measurements as well as multi-satellite remote sensing datasets, the implications of fire and other land use activities on atmospheric CO2 concentrations in California during the summer of 2008 will be presented.