Presented within this paper is an analysis of data from five flights of the Facility for Airborne Atmospheric Measurement BAe146 during the February 2008 Cold Land Processes II campaign over the North Slope of Alaska. Snow pits were dug under the aircraft tracks which provided information about snow pack stratigraphy and grain type, density, and temperature profiles. Lambertian emissivity retrievals were carried out yielding an emissivity and effective temperature time series for each of the five flights discussed. These emissivities were used as the truth for the validation of the Microwave Emission Model of Layered Snowpacks (MEMLS) appropriate for use at Microwave Humidity Sounder and Advanced Microwave Sounding Unit B frequencies. As there are uncertainties in the model input parameters which were determined from the snow pit measurements, an efficient global optimization routine was used to determine if MEMLS could adequately simulate the observed emissivities within the uncertainties of the input parameters. In order to reduce model errors, various optimization experiments were carried out. The most successful of these were obtained in the following two cases: 1) A parameterization of surface roughness was introduced, and 2) a limit on the scattering coefficient at high frequency was introduced. In both of these cases, MEMLS is able to reproduce the emissivity spectra over the sites studied within the uncertainties in the measurements.
|Number of pages||17|
|Journal||IEEE Transactions on Geoscience and Remote Sensing|
|Publication status||Published - Nov 2012|