UCLA

Seasonal snow water is a key component of the food-energy-water nexus in many regions, supporting one sixth of the global population. Given its importance, characterizing seasonal snow is critical to close terrestrial water budgets, especially for mountainous regions where as much as 70% of the water supply for humans originates from snowmelt. However, it is an ongoing challenge to characterize snow water equivalent (SWE) from existing snow products in snow-dominated mountain regions.

In Chapter 2 of this thesis, a novel Western U.S. (WUS) snow reanalysis dataset (WUS-SR) that is continuous in space and time was developed at high-resolution (~ 500 m) over water years (WYs) 1985 to 2021. The snow dataset has been significantly verified with > 25,000 station-years of independent in situ and airborne data. Overall, WUS-SR peak SWE is well correlated against in situ peak SWE with correlation coefficient of 0.77, and against lidar-derived SWE taken near April 1st with correlation coefficients ranging from 0.75 to 0.91.

In Chapter 3, the newly derived WUS-SR dataset wass used for examining the role of snow on streamflow drought. The analysis in this thesis shows that WY 2021 stands out as an unpredictable year with extremely low streamflow in the WUS, with only moderately low upstream snow conditions. Although snowmelt played a key role in the streamflow drought, the 2021 streamflow drought was a compound event modulated by contributors linking snow, soil moisture, and streamflow.

In Chapter 4, the WUS-SR along with a previously derived Andes snow reanalysis were used as reference datasets in an intercomparison of other global products. Climatological snow storage is quantified as 269 km3 in the WUS and 29 km3 in the Andes from the reanalysis datasets. Existing high- and moderate-resolution products agree with the WUS-SR, whereas coarse-resolution products generally underestimate snow with large uncertainty in both WUS and Andes. Snow products with resolutions greater than 5 km did not resolve the orographic-rainshadow patterns that are important to downstream water resources. In addition to precipitation as the main driver of snow uncertainty, product spatial resolution, and LSM mechanisms such as rain-snow partitioning and snowmelt generation play important roles.