[Zurück]

K. Freund:
"Time series analyses of Ku-band scatterometer data over Greenland for retrieval of spatial and temporal patterns of winter thaw events";
Betreuer/in(nen): A. Bartsch; Department für Geodäsie und Geoinformation, 2017; Abschlussprüfung: 27.11.2017.

The ongoing global warming has high effects on the polar regions. Greenland and its ice shield has been object of numerous investigations, especially when considering spring and summer snow melt trends. Scatterometer data acquired in Ku-band from the QuikSCAT satellite have been proven to be highly applicable because of the sensitivity to snow cover thaw-freeze dynamics. Due to the higher average temperatures, the influence of the North Atlantic Oscillation (NAO), and prevailing wind systems over Greenland winter thaw is of growing importance. It affects the biological, ecological, and hydrological processes by causing ungulate mortality, increasing NO3-N export, and leading to severe flooding. In this study, the focus was set on winter months (November-March) for the time frame from 2000 to 2008. Four different approaches were examined due to their capability of detecting thaw events. In the first case the difference of the backscatter coefficient sigma0 based on a three-day moving window was used. In the second case the difference between two consecutive days was used. Two types of parameterization have been investigated: a constant threshold was set to 1.5 dB and a location speciffc noise level. For all winters thaw events were detected, however the spatially largest extent was found during late November, 2005. Compared to in-situ data from the Greenland Climate Network (temperature) and National Oceanic and Atmospheric Administration(precipitation), the results were investigated to validate if the temperature and/or precipitation were an indicator for the ongoing melt. The different approaches not only varied for date and size when detecting thaw, even spatial irregularities were found.