Estimation of the spatiotemporal dynamic of snow water equivalent at mountain range scale under data scarcity.

The snow dynamics in alpine systems play a significant role in the hydrosphere, biosphere, and anthroposphere interfaces of these regions. The storage of water resources as snow is essential for ecosystems, human consumption, tourism, and hydropower in many areas. However, snow data are usually scarce due to poor accessibility, difficulties to maintain monitoring system under harsh climatic conditions and limited economic funds. Most of the scientific studies aimed to quantify water stored as snow are carried out at small or medium spatial scales, but few analyses are done for the whole mountain ranges. The main goal of this work is to propose a general parsimonious methodology to estimate snow water equivalent under data scarcity for the Sierra Nevada mountain range (Spain). The methodology is easily transferable to any other study areas. It combines a dynamic regression approach of snow depth from punctual data, snow cover area data from the MODIS satellite and simulations of snow density from a coupled mass and energy balance model. The regression model includes two kinds of explanatory variables (steady and non-steady) to assess the snow depth dynamics. The dynamic of the snow density in the mountain range has been obtained using a physically based simulation driven by climate model data for the Iberian Peninsula. These three variables (snow depth, snow cover area and snow density) have been used to obtain spatially distributed series of snow water equivalent for the whole mountain range. The proposed solution allows studying the snow water equivalent distribution, duration of the snow cover and number of accumulation and melting days for different snow seasons. The mean accumulated snow water equivalent per season in the historical period is 330 Hm3 and the maximum of 480 Hm3, which is a significant amount of resources in an area characterized by limited water availability.