Date of Award

5-2023

Document Type

Honors Thesis

Degree Name

Bachelor of Science

Department

Atmospheric and Environmental Sciences

Advisor/Committee Chair

Justin Minder

Abstract

When snow melts, the water drains into nearby streams and rivers which can impact water supply and flood hazards. The National Water Model (NWM) provides high quality forecast data for streamflow in the continental United States using mathematical representations of hydrologic processes. This research evaluates how the land surface model (LSM), Noah-MP, in the NWM simulates snowpack and snowmelt in New York State. To evaluate the representation of snow melt in Noah-MP, we examine point simulations at the New York State Mesonet (NYSM) sites where Noah-MP is forced by NYSM meteorological observations to avoid biases in the meteorological input and isolate the LSM formulation as the source of biases. Additionally, we examine distributed runs across the whole state forced by a gridded meteorological analysis. Compared to the NYSM snow depth observations averaged at all sites, the snow depth output in the ablation period of point-simulations is too high. Through controlled sensitivity experiments where an aspect of the model is altered in isolation, it is revealed that the snowpack in Noah-MP is strongly sensitive to a parameter of snow albedo decay, 𝛕0. There is a bias in the melt rate in the model causing slow snow melt which can be changed by altering the 𝛕0 value. Decreasing the value of 𝛕0 causes the snow to melt faster and the predicted snow depth and snow water equivalent to be more similar to observations. Decreasing the value of 𝛕0 and increasing the rate of albedo decay also causes the model albedo decay rate to be more similar to observations. Through evaluation of how albedo changes in the model for snow melt periods at NYSM sites, the NWM biases in the northeast United States can be more deeply understood.

Additional Files
Sierra Liotta_Completion Form.pdf (336 kB)
Download

Share

COinS