Towards Large-Scale Daily Snow Density Mapping with Spatiotemporally Aware Model and Multi-Source Data

Abstract. Snow density plays a critical role in estimating water resources and predicting natural disasters such as floods, avalanches, and snowstorms. However, gridded products for snow density are lacking for understanding its spatiotemporal patterns. In this study, considering the multiple influencing variables and the strong spatiotemporal heterogeneity of snow density, the geographically and temporally weighted neural network (GTWNN) model is constructed for estimating daily snow density in China from 2013 to 2020, with the support of satellite, ground, and reanalysis data. The R² and RMSE achieve 0.515 and 0.043 g/cm³, respectively. The constructed GTWNN model is able to improve the estimation of snow density by capturing the weak and nonlinear relationship between snow density and the meteorological, snow, topographic, and vegetation variables. The leaf area index of high vegetation, snow depth, and topographic variables make a relatively great contribution for estimating snow density among the 17 influencing variables. The importance of addressing the spatiotemporal heterogeneity for snow density estimation is further demonstrated by comparing the GTWNN model with other models. The performance of the GTWNN model is closely related to the state and amount of snow, in which more stable and plentiful snow would result in higher snow density estimation accuracy. With the benefit of the daily snow density map, we obtain knowledge of the spatiotemporal pattern and heterogeneity of snow density in different snow periods and snow cover regions in China. The proposed GTWNN model holds the potential for large-scale daily snow density mapping, which will be beneficial for snow parameter estimation and water resource management.