A multi-year assessment of the Synthetic Aperture Radar (SAR) signature of supraglacial lake evolution on the George VI ice shelf, Antarctic Peninsula.
Supraglacial lakes (SGLs) that form annually on the ice shelves fringing Antarctica exert a significant influence on the stability and mass balance of the entire ice sheet, yet their continent-wide extent and evolution is poorly constrained. This study aims to overcome the limitations of optical sat...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Master Thesis |
| Language: | English |
| Published: |
The University of Edinburgh
2020
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/1842/37630 https://doi.org/10.7488/era/911 |
| Summary: | Supraglacial lakes (SGLs) that form annually on the ice shelves fringing Antarctica exert a significant influence on the stability and mass balance of the entire ice sheet, yet their continent-wide extent and evolution is poorly constrained. This study aims to overcome the limitations of optical satellite imagery by utilising the all-weather, day-and-night capabilities of Synthetic Aperture Radar (SAR) to examine the evolution of SGLs on the George VI ice shelf, Antarctic Peninsula. Sentinel-1 SAR data are analysed across four melt seasons on George VI (2016-2020) in an effort to understand the backscatter signature of SGLs at different stages of development. A distinct cyclical pattern is identified: a sharp drop in backscatter (~10dB) at the beginning of austral summer indicates melt onset, followed by fluctuations reflecting meteorologically-driven melt and re-freeze events and a return to pre-melt values into winter. The C-band SAR proves effective at discriminating between open water and ice, but the interim melt and re-freeze stages present more complex signatures that vary across the ice shelf, limiting the ability to isolate the SGL backscatter signature statistically. Alongside an assessment of concurrent meteorological data and optical imagery, the findings have informed the production of a novel detailed schematic of the C-band SAR backscatter signature of SGLs. Finally, we present promising results from the first attempt at a threshold-based classification of Antarctic SGLs directly from SAR imagery; an exciting preliminary step in the effort to develop automated SAR-based SGL detection techniques across Antarctica. |
|---|