Spatial Distribution and Scaling Properties of Lidar‐Derived Snow Depth in the Extratropical Andes

Abstract

We characterize elevational gradients, probability distributions, and scaling patterns of lidar-derived snow depth at the hillslope scale along the extratropical Andes. Specifically, we analyze snow depth maps acquired near the date of maximum accumulation in 2018 at three experimental sites: (i) the Tascadero catchment (31.26°S, 3,270-3,790 m), (ii) the Las Bayas catchment (33.31°S, 3,218-4,022 m); and (iii) the Valle Hermoso (VH) catchment (36.91°S, 1,449-2,563 m). We examine two subdomains in the latter site: one with (VH West) and one without (VH East) shrub cover. The comparison across sites reveals that elevational gradients are site-dependent, and that the gamma and normal distributions are more robust than the lognormal function to characterize the spatial variability of snow depth. Multiscale behavior in snow depth is obtained in all sites, with up to three fractal regimes, and the magnitude of primary scale breaks is found to be related to the mean separation distance between local snow depth peaks. The differences in snow depth fractal parameters between VH West-the only vegetated subdomain-and the remaining sites suggest that local topographic and land cover properties are dominant controls on the spatial structure of snow, rather than average hydroclimatic conditions. Overall, the results presented here provide, for the first time, insights into the spatial structure of snow depth along the extratropical Andes Cordillera, showing notable similarities with other mountain regions in the Northern Hemisphere and providing guidance for future snow studies.

A conceptual hydrological model of semiarid Andean headwater systems in Chile

“…As is seen in all mountain systems, seasonal snow in the SA varies greatly through both space and time (Schauwecker et al, 2022), and understanding snow distribution and its relationship to climate patterns is needed to better predict water availability seasonally (Sproles et al, 2016). However, spatial distribution and depth of the accumulated snow is not only the product of the frequency and intensity of the fronts (Masiokas et al, 2006(Masiokas et al, , 2012, is further complicated by post-depositional processes, including local topography, wind redistribution, avalanching, and ablation processes such as sublimation and melting (Gascoin et al, 2013;Mendoza et al, 2020).…”

“…The temporal and spatial distribution of snow depth and density are important parameters that impact the initial conditions for snowmelt generation and its consequent streamflow contribution calculation (Shaw et al, 2020a and references therein). However, understanding the snow cover evolution remains limited due to high spatial variability in snow depth, cover and duration (Malmros et al, 2018; Mendoza et al, 2020) that is difficult to represent in snow models (Réveillet et al, 2020). Additionally, due to arid conditions, sublimation rates are relatively high (Réveillet et al, 2020; Voordendag et al, 2021), and yet difficult to model due to high variability in wind speeds that is difficult to spatially represent due to severe topography (Gascoin et al, 2013; Réveillet et al, 2020).…”

Impacts of Subgrid Temperature Distribution Along Elevation Bands in Snowpack Modeling: Insights From a Suite of Andean Catchments

“…Because water resources applications in mountainous areas require model simulations at the watershed or regional scales (Mendoza et al., 2020), spatial discretization strategies are needed to address heterogeneities within the domain of interest. Common choices involve the delineation of grid cells (e.g., Beck et al., 2020; Liang et al., 1996), subcatchments (e.g., Bandaragoda et al., 2004; Tesfa et al., 2014) and, more generally, hydrologic response units (e.g., Markstrom et al., 2008; Newman et al., 2014) as spatial modeling units.…”