Landscape freeze/thaw retrievals from soil moisture active passive (SMAP) L-band radar measurements

Abstract:

The NASA Soil Moisture Active Passive (SMAP) mission produces a daily landscape freeze/thaw product (L3_FT_A) which provides categorical (frozen, thawed, or [inverse] transitional) classification of the surface state (for land areas north of 45°N) derived from ascending and descending orbits of SMAP high-resolution L-band radar measurements. The FT retrievals are output to 3 km resolution polar and global grids with temporal revisit of 2 days or better north of ~55°N and 3 days or better north of 45°N. The algorithm classifies the land surface freeze/thaw state based on the time series of L-band radar backscatter compared to frozen and thawed reference states. This presentation will describe pre-launch L3_FT_A algorithm implementation and evaluation using NASA/SAC-D Aquarius L-band radar data, and provide an update on the current status of the SMAP L3_FT_A product.

In advance of SMAP measurements, the L3_FT_A algorithm was configured and evaluated using Aquarius measurements. While the temporal (weekly) and spatial (~100 km) resolution is much coarser than SMAP, Aquarius provides L-band radar measurements at an incidence angle (normalized to 40 degrees) which is close to SMAP. Evaluation of FT retrievals derived using both Aquarius freeze/thaw references and backscatter time series as inputs identified good agreement during the fall freeze-up period with FT flag agreement (Aquarius versus in situ) exceeding the 80% SMAP mission requirement when summarized on a monthly basis. Disagreement was greater during the spring thaw transition due in part to uncertainty in characterizing the surface state from in situ measurements and backscatter sensitivity to the onset of snow melt, independent of the soil temperature beneath the snowpack. Initial challenges for SMAP derived FT retrievals include the scale difference between the Aquarius references (~100 km) and the SMAP measurements (3 km) which is particularly problematic in areas of complex topography and/or mixed land cover, and false freeze detection across the mid-latitudes due to a relatively weak difference in the radar signal between frozen and thawed conditions. It is anticipated that the shift from Aquarius to SMAP derived freeze and thaw references will significantly reduce these challenges.