Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Noah P. Molotch

Second Advisor

Ben Livneh

Third Advisor

Mark Raleigh

Fourth Advisor

Peter Blanken

Fifth Advisor

Thomas Painter


Snow is indispensible to the water resources and economy of the western United States, making it essential to accurately predict snowmelt volume, timing, and rate. However, uncertainties in snowpack processes, the effects of climate change, and spatial variability in precipitation phase partitioning all complicate efforts to simulate snow accumulation and melt. With those three issues in mind, this work clarifies seasonal snow cover evolution in a changing climate by utilizing ground observations and validated output from a physics-based snow model.

The first project focuses on how snowpacks develop cold content, the internal energy deficit that must be satisfied before snowmelt can begin. Previously it was unknown whether cold content developed primarily through meteorological or energy balance processes. Using snow pit data and model output, I show that new snowfall exerts the primary control on cold content development in the snowpacks at an alpine and subalpine site in the Colorado Rocky Mountains. Additionally, model output indicates that cold content damps snowmelt rate and delays snowmelt onset at time scales one month and shorter, but has little correlation to those quantities at seasonal time scales.

The second project evaluates the physical processes controlling the response of the alpine and subalpine snowpacks to increases in air temperature and changes to precipitation total and seasonality. The increased sensitivity of the subalpine snowpack to climate warming is primarily a result of decreases to snowpack cold content and increases in positive energy fluxes. As opposed to the differential response to warming, the two snowpacks exhibited fairly consistent responses to changes in total precipitation with later melt onset and faster snowmelt rates being associated with increased precipitation. Changes to precipitation seasonality had a near-negligible impact on snow cover properties at both sites.

The final project expands on the spatial scope of the first two by simulating snow accumulation and melt at sites in the western United States that span a climatic gradient from warm maritime to cold continental. Previous research had shown spatial variability in rain-snow partitioning, but little was known about how this variability affected snow model simulations. The results from this project indicate that the selection of a method to partition rain and snow leads to the greatest divergence in seasonal snow cover evolution at the lower elevation maritime sites. Peak snow water equivalent and snowmelt timing simulated by the different methods varied by several hundred millimeters and over one month, respectively, at the warmest sites, and typically less than 20 mm and one week at the two coldest sites. Overall, this dissertation highlights how snow models and ground observations can be used to better understand snow accumulation and melt processes in a changing climate.

Available for download on Thursday, January 27, 2022