Terrain Induced Biases in Clear-Sky Shortwave Radiation Due to Digital Elevation Model Resolution for Glaciers in Complex Terrain


Advancements in remote sensing, along with greater access to high spatial and temporal resolution imagery, have improved our ability to model glacier surface energy and mass balance in remote regions of complex terrain, such as High-mountain Asia (HMA). In general, net shortwave (SW) radiation accounts for the majority of energy available on a glacier surface during the summer months, suggesting that SW modeling errors can critically impact surface energy balance estimates. In this study, we model the clear-sky SW irradiance for a group of glaciers in the Everest region of HMA using a high-resolution (8-meter) digital elevation model (DEM) composite derived from commercial stereo satellite imagery. We then systematically downsample this DEM and considered the effect on incoming SW irradiance, with a sensitivity analysis for standard terrain attributes. The slope and aspect (combined) and topographic shading have the greatest impact on daily SW irradiance and also introduce a larger SW bias when DEM resolution is downsampled. Our results show that modeled incident SW is overestimated as resolution becomes coarser. For 10 selected glaciers in the Everest region, decreasing spatial resolution from 8 to 30 meters results in a range of average daily biases between +20 to +60 Wm-2 (or ~7 to 20%) at some high and low elevations, and an average bias of more than +100 Wm-2 (~33%) as resolution is coarsened to 500 meters. In order to determine the bearing these results have on surface melt, we explore the diurnal variability of this bias. Additionally, we compare our results with modeled incident SW using several global DEM products (ASTER, SRTM, and ALOS) to evaluate error introduced by lower resolution. Models using the 30-meter products show an overall average daily SW bias of +24 Wm-2 (or ~8%) across elevation with some elevations showing a bias up to +60 Wm-2 (~20%) on multiple glaciers. Taken together, our results demonstrate the value of high-resolution data to correct biases in modeled SW radiation and constrain uncertainties for glacier energy balance modeling in regions of complex terrain.

Frontiers in Earth Science