Soil frost effects on streamflow recessions in a subarctic catchment

The Arctic is warming rapidly. Changing seasonal freezing and thawing cycles of the soil are expected to affect river run-off substantially, but how soil frost influences river run-off at catchment scales is still largely unknown. We hypothesize that soil frost alters flow paths and therefore affects storage-discharge relations in subarctic catchments. To test this hypothesis, we used an approach that combines meteorological records and recession analysis. We studied streamflow data (1986-2015) of Abiskojokka, a river that drains a mountainous catchment (560 km(2)) in the north of Sweden (68 degrees latitude). Recessions were separated into frost periods (spring) and nofrost periods (summer) and then compared. We observed a significant difference between recessions of the two periods: During spring, discharge was linearly related to storage, whereas storage-discharge relationships in summer were less linear. An analysis of explanatory factors showed that after winters with cold soil temperatures and low snowpack, storage-discharge relations approached linearity. On the other hand, relatively warm winter soil conditions resulted in storage-discharge relationships that were less linear. Even in summer, relatively cold antecedent winter soils and low snowpack levels had a propagating effect on streamflow. This could be an indication that soil frost controls recharge of deep groundwater flow paths, which affects storage-discharge relationships in summer. We interpret these findings as evidence for soil frost to have an important control over river run-off dynamics. To our knowledge, this is the first study showing significant catchment-integrated effects of soil frost on this spatiotemporal scale.