Snowpack, soil and forest energy budget and flux partitioning in boreal ecosystems

The snow cover has a major influence on the wintertime surface energy budget. Accurate simulation of the snowpack energy fluxes is difficult due to limitations in the parameterization of turbulent fluxes under stable conditions and landscape properties (e.g. canopy and topography) that complicate the radiation budget. In fact, description of turbulent fluxes is subject to major uncertainties in snow modelling, and simulating snow in forests is critical for hydrological and climate modelling. Yet, detailed studies that evaluate the models with surface energy flux observations at high latitudes are rare. In this study, we evaluate components of the SURFEX land surface model on four eddy covariance sites in Finland. These sites cover two different climate and snow conditions, the southern and northern subarctic zone, and two different boreal landscape types, peatland and forest. On the peatland sites, we evaluate the sensitivity of simulated surface energy fluxes and snow conditions to different process parameterizations (e.g. snow processes and turbulent exchange) implemented in the detailed snowpack model Crocus. On the forest sites, we examine alternative approaches to represent the energy and mass budgets of the soil and vegetation with the ISBA and MEB models, and assess their performance in simulating energy fluxes, snow conditions and soil thermal regimes. We show that the turbulent fluxes under stable conditions simulated by the default stability correction function do not match the observed values, and thus, it is necessary to increase the simulated turbulent exchange under stable conditions. Moreover, we demonstrate that explicit vegetation is required to concurrently simulate accurate surface heat fluxes and snow/soil conditions in forests. Our results have larger implications for choosing suitable model parameterizations and structures depending on the use case of interest.