Spatial simulation of snow and frozen ground using a modified temperature-based model
Date
2018
Authors
Follum, Michael Lee, author
Niemann, Jeffrey, advisor
Fassnacht, Steven, committee member
Julien, Pierre, committee member
Kampf, Stephanie, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
Volume and timing estimates of snowpack and subsequent streamflow are vital for water management and flood forecasting in snow-dominated regions. Numerical models are often employed to estimate the depth of snowpack and presence of frozen ground for assessment of the resulting streamflow. Air temperature based models, such as temperature-index (TI) snow models and degree-day (DD) frozen ground models, are commonly used due to their simplicity and low data requirements. However, because air temperature (a surrogate for available energy) is the main forcing variable, the snowpack and frozen ground in TI and DD models vary spatially based only on elevation. The overall objective of this research is to improve the representation of spatial variations in snowpack and frozen ground within watersheds in order to improve streamflow simulations. To accomplish this goal, this study replaces air temperature in a TI snow model and a DD frozen ground model with a proxy temperature for available energy. The proxy temperature is calculated using a simplified radiation energy balance (requiring precipitation, air temperature, and cloud cover data) that accounts for spatial heterogeneity in both shortwave and longwave radiation due to topography and vegetation. The modified-TI model, referred to as the Radiation-derived Temperature-Index (RTI) snow model, is tested at Senator Beck basin (SBB) in Colorado and at Sleepers River Experimental Watershed (SREW) in Vermont. The RTI model outperforms a pre-existing TI model in simulation of snow water equivalent (SWE) and improves simulation of snow covered area (SCA) at both SBB and SREW. The improvements in snow simulation using the RTI model also improve the streamflow simulation at SBB. The modifications to the DD model, referred to as the modified Continuous Frozen Ground Index (modCFGI) model, also account for insulation of soil by ground cover and simulate frost depth. When tested at SREW, the modCFGI model more accurately captures the variations in frozen ground between the sites, inter-annual variations in frozen ground depths at a given site, and the occurrence of frozen ground than the pre-existing Continuous Frozen Ground Index model. Overall, the modifications made to the snow and frozen ground methods increase the spatial accuracy without requiring much additional data. The RTI and modCFGI methods are also readily transferrable to other hydrologic models.
Description
Rights Access
Subject
frozen ground
radiation-derived
temperature index
GSSHA
degree day
snow