SEASONAL SNOWPACK RADIOBRIGHTNESS INTERPRETATION USING A SVAT-LINKED EMISSION MODEL

摘要

This study investigates the link between seasonal snowpack radiobrightness dynamics and land-atmosphere energy and moisture fluxes using a 7 month observational record and snowpack simulations. Experimental data are presented from the first Radiobrightness Energy Balance Experiment (REBEX 1) which combined continuous micrometeorological observations with ground-based measurement of terrain radiobrightnesses at Special Sensor Microwave/Imager (SSM/I) frequencies. The REBEX 1 site near Sioux Falls, South Dakota, is characteristic of the northern Great Plains grasslands and climate. The experiment period from October 1992 to April 1993 spanned vegetation senescence, snowpack formation and evolution, and spring thaw. Patterns in the data are analyzed using a snowpack evolution and radiobrightness model driven by REBEX 1 energy and moisture flux measurements. The snowpack model differs from most soil-vegetation-atmosphere transfer (SVAT) schemes in the :level of detail used in modeling the near-surface medium. Snowpack temperature and structural details are necessary in order to model both the microwave thermal source and snowpack interactions and the continuing metamorphism of the snow. Modeled radiobrightness variation at 19, 37, and 85 GHz agree with observations to within 29, 16, and 13% (root mean square), respectively, over a 53 day test period. Both observed and modeled radiobrightness dynamics show sensitivity to (1) unfrozen soil moisture content at 19 and 37 GHz, (2) snow grain size and sky brightness variation (primarily at 85 GHz), and (3) snowpack partial melt and refreeze cycles (all frequencies). Furthermore, we find that snowpack stratification does not drive brightness variation in these data even at 19 GHz where snowpack penetration is greatest. [References: 44]