Publication

Back to overview Show all

Other publication (non peer-review)

Book Impact of snow on drought Hydrological modelling of two mesoscale catchments and evaluation against field observations of runoff and environmental isotopes
Publisher ETH Zürich, Zürich

Abstract

Snow resources have a remarkable impact on summer low flows in mountainous basins. Climate change will very likely decrease winter snow accumulation and might thereby increase the risk of droughts in the summer season. Hence, it is important to improve the forecast of runoff during dry periods and to quantify the effect of temperature changes on snow resources and on the seasonal runoff. In order to gain more insight about the impact of snow on drought, a Tri-National pro- gramm called SREP-DROUGHT (Snow Resources and the Early Prediction of hydro- logical DROUGHT in mountainous streams) was initiated in 2015. Switzerland, Czech Republic and Georgia are the countries involved. This Master Thesis focuses on the catchments Alp (Switzerland) and Kamenice (Czech Republic) and their subcatchments Erlenbach and Uhlirska. The distributed hydrological model PREVAH is used to do model simulations of the catchment’s water balance including the interactions between snow hydrology and runoff generation. Automatic calibration as well as Monte-Carlo simulations (MCS) are com- pleted to address the problem of parameter estimation of low flow periods. Simplified climate change simulations are investigated, for which the temperature is raised by +2◦C and +4◦C as well as there is no snow available in the catchment (rain-not-snow simula- tion). Furthermore, environmental isotopes δ2H and δ18O are investigated, in order to improve the prediction of low flow periods. The MCS shows higher added value in the Czech catchments as compared to the standard calibration procedure integrated in PREVAH. This might be linked to the poor a priori knowledge of the local situation and also to the rather scarce availability of meteorological information as compared to the Swiss target areas. In the Swiss catchments, the MCS does not perform much better than the standard calibration, due to longterm data avail- ability and good knowledge of the catchment. The model simulations with higher mean temperatures of +2◦C and +4◦C and also the rain-not-snow simulation had a remarkable effect on the winter and spring discharge. The hydrographs show an increase in winter runoff and a strong decrease in spring runoff during the melting period. The hydrographs of +4◦C and the rain-not-snow simulation are very similar, whereas the +2◦C simulations display a higher runoff at the beginning of the melting period. The calibrated and validated simulations of discharge and its components are used to explore possible correlations between the hydrological regime and the observed value of enivronmental isotopes in the streams. ”Strong” correlations are found between the plant available soil moisture (SSM), the runoff generation storage (saturated zone) (SLZ) and δ2H vs. δ18O. When linear regression of two variables are evaluated, strong correlations are found between δ18O and ”SSM+SLZ”. For further studies, one could simulate Kamenice and Uhlirska with complete input vari- ables and compare the increase of skill. Strong correlations between environmental iso- topes and state variables could be implemented in PREVAH in order to predict low flow periods more precisely.
-