High-Resolution Reconstruction of the Maximum Snow Water Equivalent Based on Remote Sensing Data in a Mountainous Area

Abstract

Currently, the accurate estimation of the maximum snow water equivalent (SWE) in mountainous areas is an important topic. In this study, in order to improve the accuracy and spatial resolution of SWE reconstruction in alpine regions, the Sentinel-2(MSI) and Landsat 8(OLI) satellite data with the spatial resolution of tens of meters are used instead of the Moderate-resolution Imaging Spectroradiometer (MODIS) data so that the pixel mixing problem is avoided. Meanwhile, geostationary satellite-based and topographic-corrected incoming shortwave radiation is used in the restricted degree-day model to improve the accuracy of radiation inputs. The seasonal maximum SWE accumulation of a river basin in the winter season of 2017-2018 is estimated. The spatial and temporal characteristics of SWE at a fine spatial and temporal resolution are then analyzed. And the results of reconstruction model with different input parameters are compared. The results showed that the average maximum SWE of the study area in 2017-2018 was 377.83mm and the accuracy of snow cover, air temperature and the radiation parameters all affects the maximum SWE distribution on magnitude, elevation and aspect. Although the accuracy of other forcing parameters still needs to be improved, the estimation of the local maximum snow water equivalent in mountainous areas benefits from the application of high-resolution Sentinel-2 and Landsat 8 data. The joint usage of high-resolution remote sensing data from different satellites can greatly improve the temporal and spatial resolution of snow cover and the spatial resolution of SWE estimation. This method can provide more accurate and detailed SWE for hydrological models, which is of great significance to hydrology and water resources research.Remote Sens. 2020, 12, 460 2 of 20 interior, the contribution of runoff from mountain snowmelt reaches more than 80% [3,4]. In addition, in the context of the climate transition from warm and dry to warm and wet in northwest China, the hydrological process of rivers in northern Xinjiang has a significant response to climate change and increased snow cover, leading to an earlier flood season and an increase in peak flow. The northern Xinjiang is one of the areas exhibiting frequent snowmelt flood disasters in Xinjiang [5]. Therefore, the assessment of maximum snowpack volumes plays an important role in regional water resource management and flood prevention [6][7][8].The snow water equivalent (SWE), as an important variable in the earth system, reflects the amount of water resources in the form of snow. The SWE is also the main factor influencing river runoff, the regional water resources supply, and flood safety during the snowmelt period. Therefore, accurate acquisition of the snow water equivalent is of great significance for studies on hydrology, meteorology, the water cycle, and global climate change [8,9]. However, accurate estimation of the watershed SWE is the major problem to be solved in the study of mountain hydrology [10]. There are four commonly used methods ...

Downscaling Snow Depth Mapping by Fusion of Microwave and Optical Remote-Sensing Data Based on Deep Learning

“…The change of temperature will influence the content of liquid water in snow, then changes the grain size and density of snow, which will affect the detection of snow depth by satellite microwave data [53]. The air temperature used in the downscaling snow depth retrieval model was from the Global Land Data Assimilation System (GLDAS) data [54,55], GLDAS-Noah is the Noah model driven by the Global Land Data Assimilation System, version 2.1 (GLDAS-2.1) (https://disc.gsfc.nasa.gov/datasets/ (accessed on 1 November 2020)), which was developed jointly by scientists at the National Aeronautics and Space Administration (NASA) Goddard Space Flight Center (GSFC) and the National Oceanic and Atmospheric Administration (NOAA) NCEP [56]. GLDAS provides temperature data with a maximum spatial resolution of 0.25 • × 0.25 • and a temporal resolution of 3 h. These data are resampled and calculated into daily mean temperature of 500 m resolution to match other datasets and were fed to the model.…”

Modeling Snow Depth and Snow Water Equivalent Distribution and Variation Characteristics in the Irtysh River Basin, China

“…The proportion of snow melting water is the largest, accounting for 45%, while rainfall and glacier melting water account for 26% and 7.7%, respectively. The snow cover period lasts from November to April of the next year, and the snow cover period is longer in the areas with higher elevations [39]. The snow cover is thick, and the maximum snow cover thickness can even reach more than 1 m in some years.…”

Water Balance in Alpine Catchments by Sentinel Data

“…This method is physically realistic but is affected by a high uncertainty in the unrepresented areas and influenced by the location of the monitoring sites on flat terrain (Bavera & De Michele, 2009; Rice et al., 2011). Another method is the backward reconstruction of the SWE accumulation time series, with a high spatial resolution (10–30 m) and based on daily snowmelt and SCA changes from the last significant snowfall (Jonas et al., 2009; Liu et al., 2020). However, this method is still impracticable on extensive catchments for its computational load (Dozier et al., 2016).…”

“…The spatially‐distributed quantification of snow depth and snow water equivalent (SWE) is still challenging in mountain hydrology (Dozier et al., 2016; Liu et al., 2020). One of the most consistent methods is the spatial interpolation of local measurements of SWE, constrained by remotely sensed snow cover area (SCA) (Dozier et al., 2016).…”