MODIS-based climate monitoring and snow cover modeling in the area of Deadhorse/Prudhoe Bay, Alaska

• Main Author: Kemper, Tarek
• Format: Thesis
• Language: unknown
• Published: 2019
• Subjects: Active layer thickness; albedo; permafrost; Prudhoe Bay; Alaska
• Online Access: https://epic.awi.de/id/eprint/49128/; https://epic.awi.de/id/eprint/49128/1/Master_Thesis_Tarek_Kemper.pdf; https://hdl.handle.net/10013/epic.4db47ff4-4ce1-4323-8736-d53eaf17e640

Detailed knowledge about ground surface conditions is crucial for monitoring the thermal state of permafrost and dynamics in active layer thickness. In vast areas such as the high arctic latitudes of northern Alaska, where observational data are sparse, remote sensing has proved to afford opportunities in detecting surface conditions. In order to detect peak snow water equivalents (SWE), this study presents an approach that makes use of Moderate Resolution Imaging Spectroradiometer (MODIS) land surface temperatures (LST), NDSI snow cover and albedo for downscaling ERA reanalysis data by the means of an SEB-scheme. Furthermore, MODIS-derived data sets are used to give an overview on the general climatic conditions in the area of Deadhorse/Prudhoe Bay, Alaska. Over the period from 2003 till 2016 LST show an increasing trend for every pixel in the about 6500km² large study area. Major warming trends of up to 0.1K per year appear next to the coast and in flat inland regions. The mean annual snow-offset date discovered by MODIS happens about 9 days erlier at the end of the study period. The SEB-scheme provides feasible results of annual peak SWE with a mean value of 0.31 mSWE for the entire studyarea(over the ERA period 2010-2016).The resulting peak-SWE maps can be utilized to detect specific prevailing conditions in the study area as well as they demonstrate the high spatial variability of snow cover in 500m spatial resolution. The two reanalysis products ERA-5 and ERA-Interim have been found to act differently due to the large difference in grid spacing, especially in coastal regions. Since the scheme is strongly dependent on the length of the snowmelt period, especially inaccuracies in the MODIS-based complete snowmelt date can lead to uncertainties.