How Snow Drives the Seasonal Evolution of Land and Sea Surface Albedos in the Alaskan High Arctic

Abstract

Global climate models need to capture the full range of processes governing the albedo, or else the overall strength of the feedback can be grossly over-estimated or under-estimated. Small deviations in aggregate-scale albedos can get magnified into major changes in solar heat input to the Arctic terrestrial-marine system. The proposed work focuses on characterizing the surface albedo during snowmelt at the ARM-NSA site on Alaska’s Arctic coast. Our proposed work has four objectives: (1) to measure concurrently the spatial variability and temporal evolution of the spectral surface albedo and the snow properties on the tundra and the sea ice near NSA over three melt seasons, (2) to place these ground measurements in a larger spatial context using remote sensing products that will encompass the full range of land- and icescapes found around NSA, (3) to model the optical properties and radiation transfer in the surface snow layers to allow us to explore how the measured snow properties produce the observed spectral albedos, and (4) to assess the impact of temporal and spatial variations of the surface albedo on the NSA
radiation budget using an atmospheric radiative transfer model.
Snow blankets the Arctic tundra and sea ice for nearly three-quarters of the year, with a surface albedo that reaches 0.9. In spring, as the weather warms, snowmelt begins, first exposing tundra, then later producing bare ice and melt ponds. This transition results in a reduction in albedo by as much as 6X, taking place at about the time of maximum solar irradiance. The transition also produces extremely heterogeneous surfaces on both land and ice, and these evolve in a linked but asynchronous manner. This complex albedo evolution has a direct impact on the energy balance of the Arctic. Reported albedos and radiation measurements from NSA cannot be fully understood without better knowledge of what is happening during the ssnowmeltperiod.
Specifically, we will make extensive coupled measurements of snow properties and spectral albedos at the NSA site, on the ice of Chukchi Sea, and inland on tundra. These will be embedded in a 30 x 30 km domain over which we will collect MODIS, VIIRS and WorldView images coincident with the melt campaigns. Using these measured and computed area-averaged albedos, we will develop and constrain a model of snow pack optical properties and radiative transfer on ice and tundra. These values will then be used with a high-performance atmospheric radiation code (RRTMGP) for sensitivity studies of how albedo plays into the radiative balance. Combined, our measurements and modeling should (a) produce a unique and highly valuable
data set on snow albedo for an ARM site where snow is extremely important, and (b) provide us with better understanding of key processes involved in Arctic amplification, an important element in the global climate budget.