Regional hydrological response to climate change across Sweden - impact modelling and communication

Changes in climatic forcing will change regional hydrological responses, both long-term seasonal dynamics and extreme events. Impact assessments are urgently needed by decision makers for climate adaptation planning, despite the uncertainties in the impact model chain. Therefore, uncertainties in impact assessments need to be evaluated and communicated along with the actual results in a transparent and user-friendly way.

Here, we show results from a hydrological climate impact study for Sweden, which is communicated to regional planners through a web-based assessment tool (smhi.se/en/climate/future-climate/advanced-climate-change-scenario-service/hyd). Climate change impacts are expected to alter seasonal hydrological dynamics as well as hydrological extremes in the region. We use a modelling framework based on a calibrated national hydrological model, S-HYPE, which divides the domain into  ~ 35000 computational sub-basins, for which hydrological states and fluxes are computed using a conceptual process model including lake and river management routines. S-HYPE is forced with an ensemble of Euro-CORDEX regional climate forcing for a 1971 to 2100 assessment period and representative concentration pathways 2.6, 4.5, and 8.5. All forcing is bias-corrected to a gridded reference forcing data set using quantile mapping. Results are communicated in a spatially aggregated form for ~ 250 river basins using climate impact indicators (CII), which highlight expected change patterns for specific parts of the hydrolgical cycle, e.g. change in maximum annual snow water equivalent or summer discharges. Uncertainties are communicated through ensemble spread and robustness measures, allowing users to directly assess at least parts of the uncertainty in the model results.

We also use the modelled results for a comparative analysis of change impacts across river basins in Sweden, taking advantage of the north-south climate gradient across the domain. This allows for a direct comparison of modelled today and future behaviour of similar river basins across that gradient to evaluate the uncertainty in future impacts which orginate in the model representation of interactions and feedbacks between climate and hydrological system within the S-HYPE model framework.