Snow Property Inversion from Remote Sensing (SPIReS): a Generalized Multispectral Unmixing Approach with Examples from MODIS and Landsat 8

Abstract:

Spectral mixture analysis has a history in mapping snow, especially in the mountains where mixed pixels prevail. Accuracy over older techniques that use band ratios has been demonstrated. Similarly, retrievals of properties that affect snow albedo lead to more accurate estimates than widely used age-based models. Nevertheless, there is substantial room for improvement. To this end, we present the Snow Property Inversion from Remote Sensing (SPIReS) approach, offering the following improvements: (1) Solutions for clean and dirty snow are computed together; (2) Only snow and snow-free endmembers are employed; (3) Cloud-masking and smoothing are integrated; (4) Similar spectra are grouped together and interpolants are used to reduce computation time. The code is available in version-controlled online repositories. Computation is fast enough that users can process imagery on demand. Validation against WorldView-3 imagery and the Airborne Snow Observatory show accurate detection and estimates of the fractional snow cover for Landsat OLI and MODIS. Validation of albedo uses terrain-corrected in situ measurements. We conclude by discussing the applicability of this approach to any airborne or spaceborne multispectral sensor and options to further improve retrievals.