Direct insertion of NASA airborne snow observatory-derived snow depth time series into the iSnobal energy balance snow model

Authors: Hedrick, Andrew; Marks, Daniel; Havens, Scott; Robertson, Mark; Johnson, Micah; Sandusky, Micah; MARSHALL, HP - Boise State University; Kormos, Patrick; BORMANN, KAT - Jet Propulsion Laboratory; PAINTER, TOM - Jet Propulsion Laboratory

Submitted to: Water Resources Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/24/2018
Publication Date: 10/19/2018
Citation: Hedrick, A.R., Marks, D., Havens, S., Robertson, M., Johnson, M., Sandusky, M., Marshall, H., Kormos, P.R., Bormann, K.J., Painter, T.H. 2018. Direct insertion of NASA airborne snow observatory-derived snow depth time series into the iSnobal energy balance snow model. Water Resources Research. 54:8045-8063. https://doi.org/10.1029/2018WR023190.
DOI: https://doi.org/10.1029/2018WR023190

Interpretive Summary: The historic California Drought of 2012-2015 brought the mountain snow pack into the limelight. This work illustrates the importance of combining remote sensing and model products to accurately predict water storage in mountain basins. This novel approach will likely need to become commonplace in the coming decades as the climate warms, decreasing supply, while population expands, increasing demand. This is the first near-real time assimilation of time series, high-resolution, lidar-derived snow depths into a distributed physics-based snow model over a four-year study period. Findings illustrate the benefits of the Airborne Snow Observatory for the snow modeling community and the importance of accurately defining the spatial distribution of mountain snowpacks for estimating snow water storage.

Technical Abstract: Accurately simulating the spatiotemporal distribution of mountain snow depths and water equivalent (SWE) improves estimates of available melt water and benefits the water resource management community. In this paper we present the first integration of airborne lidar data into a physics-based snow model using direct insertion. Over four winter seasons (2013-2016) the NASA/JPL Airborne Snow Observatory (ASO) has performed near-weekly lidar surveys from the date of peak SWE through melt out to measure high-resolution snow depths over the Tuolumne River Basin above Hetch Hetchy in the Sierra Nevada Mountains of California. The iSnobal model was implemented as part of the ASO program to provide density to convert the ASO-measured depths to SWE, and to provide temporally complete snow cover mass and thermal state between flights. Over the four years considered in this study, snow depths from 39 individual lidar over flights were assimilated into the model to provide updates of snow depth and distribution. Updating the model with the measured values from the ASO program significantly improved the correlation between modeled results before and after updating. Analyses show that for a single update during the melt season, the correlation between modeled results with and without updating were r2 = 0.92 (RMSE=34mm) and r2 = 0.06 (RMSE=113mm), respectively. The precise definition of the snow depth distribution integrated with the iSnobal model illustrate how the ASO program represents a new paradigm for the measurement and modeling of mountain snowpacks, and the potential benefits for managing water in the region.