STUDY AREA AND DATA SOURCES

The study area is Reynolds Mountain in the Owyhee Mountains in Idaho, USA. Reynolds Mountain is a zero-order basin that drains to the Snake River (Figure 1). Reynolds Mountain has a small drainage area (0.38 km2) and it is characterized by large patches of steep north and west facing slopes (Figure 2c). Elevation ranges in Reynolds Mountain vary between 2028 m and 2137 m. The vegetation is dominated by low, big, and Wyoming sagebrush (Artemisia arbuscula Nutt., Artemisia tridentata Nutt. subsp. vaseyana [Rydb.] Beetle and subsp. Wyomingensis, respectively), bitterbrush (Purshia tridentata [Pursh] DC), and native and non-native grasses, including cheatgrass (Bromus tectorum ). Aspen (Populus tremuloides ), subalpine fir (Abies lasiocarpa ), and Douglas-fir (Pseudotsuga menziesii ) communities are found in the water-rich areas such as below drifts and riparian zones. Figure 3 shows a distribution of the sagebrush and forest communities. Soils in Reynolds Mountain are derived from igneous granitic and volcanic rocks and lake sediments.
Reynolds Mountain has been monitored since 1983. The instrumentation includes a streamflow gauge at the basin outlet and two sheltered and exposed weather stations (Figure 1). The meteorological data include hourly air temperature, relative humidity, wind speed, precipitation (snow undercatch-corrected), shortwave radiation, and longwave radiation, which are used to force the hydrological model. The hydrological data include hourly snow water equivalent (SWE) measurements at a snow pillow site and hourly streamflow observations at the basin outlet. A snow pillow was used near the sheltered station to estimate SWE. Reba et al. (2011b) suggest that the snow accumulation at this site is enhanced by the impact of topographic and vegetation sheltering on wind redistribution. On the contrary, Winstral and Marks (2014) showed that SWE measurements at the snow pillow site are representative of the basin averages. The snow pillow SWE measurements and a lidar-derived snow depth product were used in this study for assessing the model performance in capturing the spatial variability of snow accumulation (Shrestha & Glenn, 2016). Lidar-derived snow depths were obtained from processing and then subtracting a snow-free lidar digital elevation model (DEM) from a snow covered DEM in Reynolds Mountain on March 19, 2009.