Terrestrial snow remote sensing has been of major interest in understanding the dynamics and distribution of water storage in solid form on land. Different remote sensing techniques have been proposed…

This is a research grant awarded by the National Aeronautics and Space Administration (NASA) to the University of Maryland, College Park (UMD) to evaluate a snow water equivalent (SWE) retrieval algorithm using combined active and passive microwave observations. The goal is to develop an improved SWE retrieval approach that leverages the high sensitivity of radiometry to SWE and the higher spatial resolution of radar. The research utilizes data from multiple snow remote sensing campaigns, including NASA's SnowEx, ESA's NoSREx, and ESA-CSA's SnowSAR. The $366,180.64 grant does not have a set-aside designation. UMD, as the prime awardee, is a state-controlled institution of higher learning that has received numerous federal contracts and grants for research, academic, and administrative services supporting agencies like the Department of Defense, Department of Transportation, and NASA.

Terrestrial snow remote sensing has been of major interest in understanding the dynamics and distribution of water storage in solid form on land. Different remote sensing techniques have been proposed and evaluated through dedicated snow campaigns such as the NASA sclp, the ESA nosrex, and ESA-CSA snowsar campaigns. Most recently, the NASA snowex campaign, multiple airborne and ground based sensors were deployed and extensive field observations were made, aiming to determine the optimal combination of sensors that is most suitable for a spaceborne mission dedicated for terrestrial snow remote sensing. The goal of this proposal is to evaluate the performance of a snow water equivalent (SWE) retrieval algorithm by using an active/passive combined approach with xand KU- band radar backscatter and KU- and KA- band brightness temperatures available from the snowex, nosrex, and snowsar campaigns. The retrieval algorithm is based on the state-of-art physically-based scattering model of bicontinuous media and the dense media radiative transfer (dmrt) theory. This model has been validated by using active and passive microwave observations at multiple frequency and polarization channels combined with in-situ snowpack measurements. In the active portion of the algorithm, we first subtract out the background scattering arising from the underlying snow/ soil interface, and then make use of vertically copolarized backscatter at x- and KU- band to retrieve the snow water equivalent (SWE) and an equivalent grain size of the snowpack simultaneously. The removal of the inhomogeneous background scattering from the total backscatter and the recovery of the equivalent grain size from the microwave observations significantly enhances the sensitivity to SWE, addressing a significant limitation of previous retrievals. At the snowex primary site of grand mesa, CO during february 2017, the dual frequency (x- and KU- band) and quad-polarized snowsar airborne radar was deployed. The university of waterloo ground-based x- and KU- band fully polarized scatterometer (hereafter uwscat) collected local scale backscatter at 8 sites with in-situ snowpit ground truth. In the central station, a passive microwave measurement, university of michigan boom truck having a wideband radiometer collecting brightness temperatures at l-, KU- and KA- bands, also provides another controlled dataset. These two provide the combined active and passive microwave observations. Coincident airborne acquisitions from besides snowex, the ESA nosrex campaign in finland includes coincident time series of x- (10.2GHZ) and KU- (13.3GHZ and 16.7GHZ) band fully polarized backscatter and l-, x-, KU- and KA- band brightness temperatures using tower based active scatterometer (snowscat) and passive radiometers from 2009 to 2013. We thus propose to validate and improve the radar retrieval algorithm using the new dataset from snowex campaign and evaluate its performance over different terrain types with varying vegetation coverage. We will improve the active algorithm by better consideration of background scattering, and extend the radar retrieval algorithm by including the lower KU- band at ~13GHZ using the nosrex dataset and possible wism data. We will develop a combined active - passive SWE retrieval algorithm using the snowex and the nosrex datasets and to demonstrate improved retrieval capability VS. The existing active-only algorithm. The combined active/passive algorithm will thus enable a new satellite retrieval approach that leverages the high sensitivity of radiometry (brightness temperatures) to SWE and the higher spatial resolution of radar. We will test the SWE retrieval performance using the snowsar acquisitions in finland and canada and the coincident amsr-e or amsr-2 satellite brightness temperatures. We will compare and quantify the retrieval performance from the active only, the passive only, and the combined active/passive retrieval algorithms.