Лëд и снег (May 2021)
Estimates of recent changes in snow storage in the river Northern Dvina basin from observations and modeling
Abstract
The variability of snow accumulation in the Northern Dvina River basin at the end of March 1980-2016 was studied using data on the snow water equivalent of (SWE) obtained from archives of the Russian Institute of HydroMeteorological Information-World Data Center (RIHMI-WCD) as well as calculated by models of the local heat and moisture exchange SWAP and SPONSOR using the WATCH reanalysis (WFDEI) as input data. A possibility to use the SWE data from these sources to describe long-term variability of the SWE values, including trend, high-frequency component, quasi-decadal fluctuations, and spatial distribution, is evaluated. When describing the structure of the SWE variability, in particular, the contribution of trend and quasi-decadal fluctuations, as well as spatial characteristics, uncertainty remains associated with both the capabilities of the models under consideration and the imperfection of the observation network (insufficient density, measurement errors, etc.). Taking into account these uncertainties, the following conclusions can be made: the SWE variability in the Northern Dvina basin at the end of March has a low-frequency component (trend), as well as high-frequency, two- and five-year quasi-periodicities and quasi-decadal fluctuations. Long-lasting SWE anomalies in 1989–1995 and 1999–2005 and the absolute minimum in 1996 associated with quasi-decadal fluctuations are almost synchronously reflected in spring runoff anomalies. The informativeness of the considered data was also investigated from the point of view of the influence of SWE on the anomalies of the spring runoff of the Northern Dvina. The results of regression estimates and calculations of predictive values point to the advantage of the model SWE data for describing anomalies of spring river discharge compared to observations, which is primarily due to the high resolution of the model data. All the considered data sources indicate a long period of SWE deficits, starting from 2005 – 15-20%. Estimates of trend parameters are in a wide range. Depending on the data source, the rate of the SWE decrease over the basin, can vary from 4 mm per 10 years according to observations and up to 10 mm per 10 years according to calculations using the SPONSOR model.
Keywords
