Improving SWE Estimation by Fusion of Snow Models with Topographic and Remotely Sensed Data

Ludovica De Gregorio,Lorenzo Bruzzone,Mattia Callegari,Ulrich Strasser,Daniel Günther,Marc Zebisch,Claudia Notarnicola

doi:10.3390/rs11172033

Ludovica De Gregorio, Lorenzo Bruzzone + Show 5 more

Open Access

https://doi.org/10.3390/rs11172033

Copy DOI

Abstract

This paper presents a new concept to derive the snow water equivalent (SWE) based on the joint use of snow model (AMUNDSEN) simulation, ground data, and auxiliary products derived from remote sensing. The main objective is to characterize the spatial-temporal distribution of the model-derived SWE deviation with respect to the real SWE values derived from ground measurements. This deviation is due to the intrinsic uncertainty of any theoretical model, related to the approximations in the analytical formulation. The method, based on the k-NN algorithm, computes the deviation for some labeled samples, i.e., samples for which ground measurements are available, in order to characterize and model the deviations associated to unlabeled samples (no ground measurements available), by assuming that the deviations of samples vary depending on the location within the feature space. Obtained results indicate an improved performance with respect to AMUNDSEN model, by decreasing the RMSE and the MAE with ground data, on average, from 154 to 75 mm and from 99 to 45 mm, respectively. Furthermore, the slope of regression line between estimated SWE and ground reference samples reaches 0.9 from 0.6 of AMUNDSEN simulations, by reducing the data spread and the number of outliers.

Highlights

Melt water from snow and glaciers plays a key role in the hydrological cycle by contributing to the river flow and water resources in many parts of the world
The analysis of the AMUNDSEN simulations helps to understand how the model results vary with respect to the period of the year, the altitude and the different regions included in the study area
The lower maximum value of snow water equivalent (SWE) observed in the hydrological year 2005–2006 is due to lower values of snow depth recorded in this year with respect to the other studied years

Summary

Introduction

Melt water from snow and glaciers plays a key role in the hydrological cycle by contributing to the river flow and water resources in many parts of the world. It is estimated that about one-sixth of the world’s population depends on snow- and ice-melt for the supply with drinking water [1]. For hydrological assessments in these regions, knowledge about the spatial and temporal distribution of the snow water equivalent (SWE) is of uttermost importance. A spatialized estimation of SWE in mountain areas, which are typically complex terrains with high topographic heterogeneity, is currently one of the most important challenges of snow hydrology [3]. An improved knowledge of the spatial distribution of SWE and its evolution over time would allow a better management of mountain water resources for drinking water supply, agriculture and hydropower, as well as for flood protection

Objectives

Methods

Results

Conclusion