Parameterisation and application of a hillslope model to assess hydrological impacts of forest harvesting

This paper combines an empirical paired-catchment method with hydrological simulation techniques in quantifying and explaining the influence of forest removal on streamflow generation. The pair of catchments (56 and 29 ha) is located in a northern boreal zone in eastern Finland. As the first part of the study, measured streamflows from the two catchments, of which one has been partially clear-cut (35%) and one has not been treated (control), were analysed to detect differences in streamflows preceding and following the logging. Subsequently, canopy and snow process models were calibrated to characterise land surface processes in the forested areas of the study site. In clear-felled areas the canopy routine was simply not active in simulation runs. Output from the canopy and snow models provided an input for a hillslope hydrological model, which was parameterised for the two catchments. Parameterisation was based on spatial data on topography, soil types, soil depths, and location of treated areas with respect to the stream network draining the catchment. The hillslope hydrological model was calibrated against streamflow measurements, and the paired-catchment analysis was repeated using the simulated streamflows. Results inferred from a direct data analysis and model simulations were compared to address mechanisms that could explain the observed difference in the hydrological behaviour of the two catchments. According to the model, decreased interception and increased snowmelt were largely responsible for the observed differences in timing of the spring flood and volumes of total streamflow.