Other Research Projects

Future resiLient forest in a chAnging cliMatE (FLAME): isotope observations and mechanistic modeling of soil water residence time and vegetation water uptake dynamics

It is unclear how the dynamics of subsurface water storage and release, the seasonal origins and turnover time of water used by plants, and plant water uptake depths will change when environmental conditions change (e.g. receding groundwater, more frequent droughts). Yet, they are the most crucial in predicting vegetation resilience in response to drought. Previous studies have attempted to improve the mechanistic understanding of ecosystem response to dry conditions or climate change by focusing either on vegetation water availability or plant physiological adaptation strategies, but the combined effects of shifting terrestrial water availability and atmospheric demand have not been mechanistically investigated. FLAME will use a newly developed high-frequency in-situ measurements of stable water isotopes (18O and 2H) in soil and xylem as a unique natural signature to trace the origin of vegetation water uptake and its residence time in subsurface. It will combine these observations with a high resolution physically based water and vegetation uptake model to track water flow paths and constrain the spatiotemporal heterogeneities in terrestrial ecosystems at the soil-vegetation interface. 

Contact: Elham Freund

Contributions of snow and ice to streamflow of glacierized headwater catchments

The aims of this project are to quantify the past and present streamflow components in all Swiss glacierized headwater catchments, and to predict changes for the future using CH2018 climate scenarios. This will contribute to a better understanding of the potential risks of changed streamflow dynamics, especially for extreme events, due to climatic variations and changes.

Collaboration: University of Freiburg (DE)

Contact: Daphné Freudiger

Assessing the value of groundwater and phenological data to improve low flow simulations

Aim: Drought is a complex natural hazard that impacts ecosystems and society in multiple ways. Many of these impacts are associated with hydrological drought, visible as below average streamflow, lake or groundwater levels. Hydrological droughts impact ecology, agriculture, power generation, drinking water supply, river navigation, but also water-based tourism. However, hydrological models are traditionally designed to simulate peak flows as good as possible - often at the cost of the quality of low flow simulations.
The aim of this project is to systematically investigate the value of phenological and groundwater data to improve low flow simulation.

Contact: Maria Staudinger

Extreme floods in Switzerland 

Aim: Provide a consistent basis for the assessment of flood hazards in all of Switzerland. The corresponding hydrometeorological scenarios are developed for a wide range of scales and return periods, using a simulation chain consisting of weather generators, hydrological modelling and hydrological routing.

Duration: 2020–2024

Collaborations: University of Bern, University of Graz, University of Grenoble

Sponsors: Swiss Federal Office for the Environment (FOEN), Swiss Federal Office of Energy (SFOE)

Contact: Daniel Viviroli, Maria Staudinger, Eleni Kritidou, Jan Seibert

Intelligent parameter sampling (IntelSamp)

Aim: Among different uncertainty sources, the parameter uncertainty is most often considered in environmental modeling. This parameter uncertainty is usually estimated by means of resampling from the possible parameter space and the resulting uncertainty estimates are communicated in terms of ensembles. This project aims at developing an innovative method to select a representative sample of ensemble members and model parameters to be used within a complex model chain (i.e., with numerous input or future scenarios). In this way, computational and time requirements of estimating uncertainty ensembles should be optimized.

Value of radar-based data for flood modelling

Aim: This project explores the value of radar data for supporting flood modelling and flood predictions in Switzerland, as compared to the information acquired from a traditional rain gauge network. In particular, we analyse how radar-based precipitation products contribute to input and predictive uncertainty in flood estimates in a small fast reacting catchment.

Collaboration: Anna Sikorska (H2K), Jan Seibert (H2K), Ioannis Sideris (MeteoSwiss), Urs Germann (MeteoSwiss)

Re-thinking the snow routine of HBV-light
Aim: Review and assess potential changes to the snow routine of the HBV-light hydrological model in order to increase its realism but preserving the model’s characteristic simplicity. Potential modifications include variations of the current degree-day method for snowmelt estimation or the use of additional data sources such as radiation or relative humidity data.

Collaboration: Charles University in Prague (CUNI), WSL Institute for Snow and Avalanche Research (SLF)