Publication Date

5-2021

Date of Final Oral Examination (Defense)

3-3-2021

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Geophysics

Department

Geosciences

Supervisory Committee Chair

Ellyn M. Enderlin, Ph.D.

Supervisory Committee Member

HP Marshall, Ph.D.

Supervisory Committee Member

Dylan Mikesell, Ph.D.

Abstract

Seasonal snowpack accounts for ~70% of the water supply in the western United States, and measuring snow accumulation and ablation remotely has long been a stated goal of NASA. The 2018 launch of ICESat-2, a spaceborne Lidar system, has offered unparalleled spatial and temporal coverage of mountainous terrain with the potential for unprecedented vertical accuracy. Data from ICESat-2 are used to measure seasonal snow depths using the level-3A ATL08 (land and canopy elevation) product for the Reynolds Creek Experimental Watershed in southwest Idaho and the ATL06 (land ice elevation) product for Wolverine Creek in the Kenai Mountains of Alaska. The methodology for coregistering ICESat-2 transects to reference digital terrain models then estimating snow depths as the difference between the ICESat-2 and reference elevations is described. Median and MAD snow depths for transects from 2019 and 2020 are 3.1 +/- 6.7m at Reynolds Creek EW and are 5.5 +/- 2.1m at Wolverine glacier. Here we find that measuring snow depths using ICESat-2 is crude in variable, vegetated terrain covered by the ATL08 data product, and that there is not a strong relationship between the residual values reported at Reynolds Creek EW and terrain parameters such as slope, aspect, vegetative coverage, and elevation. We do find that the ATL06 analysis results in reasonable first-order estimates of snow depth but that the evolution of the glacier surface elevations must be more accurately constrained in order to ensure the snow depth estimates are unbiased.

DOI

10.18122/td.1802.boisestate

Download

Share

COinS