Refine
Year of publication
- 2017 (2) (remove)
Document Type
Language
- English (2)
Keywords
- Change points (1)
- Global dimming (1)
- Regional climate models (1)
- SWIM (1)
- Stakeholder outreach (1)
- Trends analysis (1)
- Water management (1)
- climate change (1)
- model comparison (1)
- uncertainty (1)
Institute
- FG Hydrologie (2)
Climate change impact studies are associated with error propagation and amplification of uncertainties through model chains from global climate models down to impact (e.g. hydrological) models. The effect of water management, which reduces discharge variability, is often not considered in climate change impact studies. Here, we investigated how water resources management influences discharge variability and uncertainty propagation of climate change scenarios by combining the analyses of observed flow records and model-based climate change impact simulations. Two neighbouring catchments, the Schwarze Elster River (Germany) and the Spree River (Germany and Czech Republic) which are similar in climate, topography and land use, but different in terms of water resources management were chosen as study area. The intense water resources management in the Spree River catchment includes a high reservoir capacity, water use in terms of mining discharges and water withdrawals by power plants as well as water transfers.
The analysis of historical flow records focusses on variability indices (Parde index, Richards-Baker-Flashiness Index, Interquartile Ratio and Baseflow Index). The climate change impact simulations were carried out using a model cascade of (i) the statistical regional model STAR (100 stochastically generated realizations each for 3 scenarios with different prescribed temperature trend), (ii) the hydrological models SWIM and EGMO, and (iii) the water resources management model WBalMo.
The analysis of the observed discharges reveals that the annual discharge variability in the Spree catchment is dominated by mining activities rather than natural rainfall-runoff processes. Due to the high reservoir capacity in the Spree catchment its discharge is characterised by less seasonality and short-term variability compared to the Schwarze Elster. Simulations with climate change scenarios assuming increasing temperature and decreasing precipitation result in pronounced reductions of discharge in both catchments. The differences in potential natural discharges between the Schwarze Elster and the Spree catchments as projected by the hydrological models SWIM and EGMO are marginal. The uncertainties related to the climate projection are propagated through the hydrological models. In the Schwarze Elster catchment, the managed discharges simulated by WBalMo are comparable to the potential natural discharges. In the Spree River however, the short-term variability is moderated by water resources management and managed discharge under climate change is less affected by amplification of uncertainties through model chains.
The results of the study, which combines the analyses of observed flow records and model-based climate change impact simulations, imply that generally, effective water resources management reducing discharge variability hence also reduces uncertainty related to climate change impacts on river discharge. Catchments with a high storage ratio are thus less vulnerable to changing climate conditions. This underlines the role of water resources management in coping with climate change impacts. Yet, due to decreasing reservoir volumes in drought periods, reservoir management alone cannot compensate strong changes in climate conditions over long time periods.
Trend analysis on observations and model-based climate change simulations are two commonly used methods for climate change detection and impact analysis. Here we propose an integrated assessment and interpretation of climate change impacts as a prerequisite for stakeholder outreach and planning of suitable climate change adaptation measures. The assessment includes (i) identifying trends in mete-orological and hydrological observations and their nature, (ii) analysing the relation between the mete-orological drivers and generated run-off as an integrated catchment response and (iii) analysing how hitherto changes agree with the simulations by regional climate models (RCMs). The Lusatian river catchments of Spree and Schwarze Elster, characterised by high anthropogenic impact (e.g. mining activities) and low natural water yield, serve as study areas. The results of this study suggest that in-creases in observed temperature and potential evapotranspiration are robust while observed precipitation remained nearly unchanged (1963–2006). The RCMs agree on simulating a temperature increase but simulate opposing trends for precipitation for both past (1963–2006) and future (2018–2060) periods, the latter inducing differences in the hydrological response (actual evapotranspiration and run-off). For stakeholder outreach, we communicated a range of potential future climates and identified the statistical RCMs (STAR, WettReg) as warm and dry scenarios, and the dynamical RCMs (REMO, CCLM) as wet scenarios. Ultimately, the combined analysis of trends in observations and simulation models can be beneficial for stakeholder outreach and may increase their willingness to plan and implement suitable climate change adaptation strategies which are urgently needed within the Lusatian river catchments.