Modelling Seasonal and Decadal Variability of Snow Conditions across Finland using SnowModel

Snow, including both seasonal and perennial snow cover plays an important role in regulating surface energy balance, water availability, and hydrological processes in high-latitude and mountainous regions. Finland, located in the boreal and subarctic zone, experiences long winters with extensive snow cover and exhibits substantial spatial and temporal variability in snow accumulation, structure, and melt dynamics. While Finland has extensive in situ snow observations, a nationally consistent, high-resolution (e.g., 250 m), multi-decadal reconstruction of snow water equivalent (SWE), snow depth, and snow density that integrates distributed physical modelling with observational data assimilation is still lacking. Understanding how these snow processes evolve seasonally and over decadal timescales is essential for assessing climate change impacts on Nordic cryosphere–hydrology systems and for improving predictive capabilities at regional to national scales.

This study uses SnowModel (Liston & Elder, 2006), a physically-based, distributed snow evolution model, to simulate snowpack processes across Finland over the past three decades (1990-2024). The model runs at a 3-hourly internal time step, with analysis conducted at a daily temporal scale at 250 m spatial resolution. The meteorological forcing is derived from a high-resolution (10 km) NORA10 reanalysis dataset, ensuring consistent spatial coverage for the study domain. The static surface inputs includes a 10 m digital elevation model from the National Land Survey of Finland and CORINE land cover data from Finnish Environment Institute ( SYKE; 25 m), resampled to the model grid enabling a detailed representation of snow processes across diverse Finnish landscapes.

The modelling framework in our study explicitly focuses key on snow variables, including snow depth, snow density, snow water equivalent (SWE), snowmelt timing, and snowmelt runoff. To improve realism and robustness, the simulations incorporate data assimilation of long-term in situ snow observations, primarily snow depth and SWE, from a nationwide network of stations. Model evaluation is conducted using independent observational datasets helping SnowModel accurately capture Finland’s spatial and temporal snow variability.

The analysis is designed to characterise both seasonal snow dynamics and long-term snow variability across Finland. Seasonal behaviour is examined through time series analysis and interannual variability of snow accumulation and melt, while long-term changes are assessed using spatial trend analyses and decadal comparisons. Emphasis is placed on contrasting snow regimes across southern, western, eastern, and northern Finland, reflecting the influence of maritime effects from the Baltic Sea, continental climate in eastern regions, and persistently cold conditions in northern Lapland.

By combining the high-resolution modelling with extensive snow observation datasets, our study aims to establish a multi-decadal snow reconstruction using a physically distributed model and provide a national-scale baseline of snowpack variability under changing climate conditions. The results are expected to provide valuable insights into evolving snow processes in boreal and sub-Arctic environments, supporting hydrological and climate impact studies, and contribute to improved cryospheric modelling frameworks for northern Europe.