Derivation of surface and TOA direct radiative forcing due to boreal forest fires using satellite retrievals and surface observations

In this paper, we present the results of a collaborative study to assess the direct radiative effect of smoke from a large forest fire outbreak in Manitoba, Canada during July and August of 1989. We report estimated surface and TOA forcings for the affected region during the period of the fires.

Baum and Trepte (1999, J. Atmos. Ocean. Tech., 793ff) developed a grouped threshold approach for Advanced Very High Resolution Radiometer (AVHRR) imagery for the purpose of identifying pixels located over forested surfaces as containing either clouds, fire, smoke, or snow. The method makes use of all 5 AVHRR channels. For this study, new tests were developed to further discriminate burn scars in the imagery. An analysis of the fire region will be presented that shows the relative amounts of cloud and smoke fractional coverage. These statistics will be compared to the cloud fraction from the International Satellite Cloud Climatology Program (ISCCP) for the area and period. The ISCCP data processing scheme makes no allowances for the presence of heavy smoke aerosols.

For those pixels identified as containing smoke, aerosol optical thicknesses are estimated from comparison of the AVHRR channel 1 (0.55-0.68 micron) bidirectional reflectances to values from a look-up table generated for a smoke aerosol. The look-up table incorporates Rayleigh scattering and viewing geometry (such as solar zenith, viewing zenith, and relative azimuth angles). On the basis of the retrieved aerosol optical thickness, the AVHRR pixels are binned subsequently into thin, medium, and thick smoke. By separating the pixels according to the magnitude of the optical thickness, we intend to investigate the effect of the areal extent of each pixel grouping on the top-of-atmosphere and surface radiative forcing caused by the smoke. This also helps to mitigate uncertainties caused by variations in surface albedo and in the composition of the smoke aerosol used to generate the scattering properties used in the radiative transfer calculations.

Finally, radiative transfer calculations will be performed to determine the areal extent and magnitude of the surface and TOA forcings. The results will be compared to calculations from the 1-degree resolution surface insolation estimates from the Global Energy and Water Cycle Experiment (GEWEX) Surface Radiation Budget (SRB) data set. Comparisons will also be made to surface radiometer measurements in the region.