Spatial and temporal variations of surface albedo and snowmelt in Northern Alaska using AVHRR Polar Pathfinder datasets

Abstract

Snowmelt in the Arctic is a rapid process. Within a period of one to two weeks the land surface may go from completely snow-covered to completely snow-free, causing a substantial decrease in its albedo. The timing of snow disappearance in the Arctic is an indication of the onset of active layer thawing over permafrost and subsequently the start of the growth season. The timing of snow melt strongly influences summer biologic processes and the ecosystem as a whole. The pace and timing of snowmelt is also important for hydrological studies, such as water supply, river runoff, and flood forecasting. In some catchments, snow meltwater accounts for as much as 80% of the annual runoff. The objective of this paper is to investigate the spatial and temporal variations in surface albedo and snowmelt in northern Alaska. To do this, the authors use ground-based measurements and 1.25 km Advanced Very High Resolution Radiometer (AVHRR) Polar Pathfinder data products. The ground-based data were acquired at nine tower sites over several years in northern Alaska by the Water Research Center, University of Alaska Fairbanks, archived at the National Snow and Ice Data Center (NSIDC), University of Colorado at Boulder. The algorithm the authors develop is based on matching the temporal pattern of surface albedo change measured at the tower sites with the albedo variations observed at each grid cell in the AVHRR product data. The pattern from the tower sites, which they call the 'kernel', is a merged pattern from all sites, with all nine data records shifted to co-align the last day of snow-cover: i.e., a template of how albedo should change during snowmelt in all areas in the Alaskan North Slope.