Master thesis defense by Diana Spurzem

Title: Representativity of albedo measurements of Automated Weather Stations in Greenland

Abstract: The Greenland Ice Sheet is a key component of Earth’s climate system that has been evidently influenced by global warming. Determined by an interconnected climate system, the changes of the ice sheet are involved in feedback mechanisms that can enhance further warming of the Earth. The surface’s albedo marks one essential feedback mechanism. In order to better emphasize scientific research of climate change impacts, it is essential to constantly monitor the Greenland ice sheet. Two predominant sources of climate-related data about the Greenland ice sheet are automated weather stations and remote sensing products. An ongoing concern in the field is the comparability of the differently retrieved data. Weather stations have a limited footprint on what they measure in situ, which might not correspond to the actual circumstances of the site. This makes it challenging to process the data in terms of representativity. Because satellites measure way larger footprints, a point-to-pixel comparison can disagree. A homogenous ground surface indicates similar conditions within a certain area. A heterogeneous surface however might distort the results. 
Global warming can enhance this matter by increasing the surface melt in Greenland and therefore decreasing the amount of homogenous snow surfaces. To confront the issue, this study evaluated the representativity of 40 automated weather stations in Greenland to detect whether they represent their local environment well. It is expected to give further access to the aspects on (1) how representative the individual weather stations are for their surrounding area; (2) which local conditions occur when the station is representative or not representative and (3) how this is influenced by the melt season. The analysis was done using Landsat 8 high resolution data in an area corresponding to a 1km2 MODIS grid pixel to approach the viewpoint of point-to-pixel comparison. The 40 weather stations were categorized into three groups, characterizing how representative they are in terms of albedo. The results showed that stations located on a homogenous snow cover all year around are highly representative. This occurred for all weather stations located at an altitude above 1500m. Some slightly lower elevated stations, placed further inwards on the ice sheet and thereby characterized by stable weather conditions, revealed comparable representativity. Stations in the ablation zone, primarily near the coast and in the Southern half of the island, were stronger affected by changing conditions during melt season. For weather stations at sites with varying elevations and different surfaces, a high representativity was not a given anymore. It mainly appeared during the shift between snow-covered and snow-free land as the area partly covered in snow showed differences in reflectance. Meltwater lakes close by influenced the albedo representativity markedly. Meltwater streams and crevasses revealed deviations as well but not as strong as lakes.

Supervisors: Baptiste Vandecrux (GEUS) and Aslak Grinsted (NBI)

Participating by using the following zoom link: https://ucph-ku.zoom.us/j/62497737884?pwd=eHpiektXVnh1SjBwN29TSmFQNFRjQT09