Climate change impact on water balance and export of dissolved organic carbon : a sub-catchment modelling approach

Abstract: Climate change will alter the hydrological cycle in the 21st century with implications for the water balance and water quality. As characteristics of the landscape may vary significantly between nearby locations, hydrological models need to be able to delineate responses attributed to specific landscape characteristics, to estimate their responses to altered climatic drivers. Small streams are the main carriers of Dissolved Organic Carbon (DOC) originating from the terrestrial landscape, however the mechanisms controlling production and mobilisation are not yet fully scientifically understood. The Skogaryd research catchment in Sweden was established in 2013 and includes characteristically different sub-catchments where discharge and other abiotic measurements take place at sub-catchment level and at the main catchment outlet. Covering the dynamics of the constituent sub-catchments would facilitate environmental assessments of the large catchments as the respective contribution of runoff and solutes are known. This study’s aims were to 1) model and compare the water balance of two characteristically different sub-catchments; a forest on mineral soil-, and a mire in the Skogaryd research catchment, by means of the hydrological model HYPE, 2) investigate the impact of climate change on the water balance, especially discharge, and 3) quantify DOC export by studying the control on the temporal DOC aquatic-terrestrial connectivity in the two sub-catchments. The model HYPE (HYdrological Predictions for the Environment) was set-up and calibrated for both catchments. Downscaled regional bias corrected climate data following the climate scenarios RCP2.6, RCP4.5 and RCP8.5 was then applied to the HYPE model. DOC was sampled and then modelled using catchment specific linear regression derived from an Automatic Linear Modelling regression analysis. The model adequately captured seasonal dynamics in the hydrological regime in spring, winter and autumn but not in summer. The HYPE model simulated a decrease in discharge in the long-term future, which is inconsistent with other model studies. The running mean temperature 60 days prior to sampling could explain 57% and 65% of DOC concentrations for the mire and forest sub-catchment, respectively. Likely, discharge will alter in the long-term future, there will be a shift towards autumn and winter discharge, whereas snowmelt driven discharge will diminish. Both catchments proved a very strong temperature control to DOC concentrations; however, the model likely captured the production of DOC rather than the mobilisation. Due to the complex interactions, governing DOC production and mobilisation, it was not possible to estimate the effect of climate change on DOC export.