Assessing Climate Change Effects on Hydrology in the Ave catchment, Portugal: A Comparative Analysis of Various Shared Socioeconomic Pathways (SSPs)

Climate change can pose a significant threat to water fluxes on terrestrial surfaces, impacting water availability, and increasing the risk for human-environment systems to floods and droughts. Understanding the repercussions of climate change on future water resources is imperative for effective integrated water resources management. As part of the DISTENDER project (EU Horizon-ID 101056836), we scrutinize the effects of climate change on diverse watersheds in Europe to develop strategies for climate change adaptation.
To simulate the impacts of climate change on water resources, we chose MIKE SHE for its spatially distributed and physically based modeling concept. Here we present the results of different climate models vs. SSPs on water balance components and runoff for the Ave River Basin in Northern Portugal.  MIKE SHE was calibrated and validated utilizing measured gauge runoff data from 1980 to 1986 and 1986 to 1990, respectively. For the various gauges, Nash-Sutcliffe efficiencies between 0.59 and 0.81 were achieved.
Statistically downscaled climate change projections for the period (2021-2050) from the Coupled Model Intercomparison Project Phase 6 (CMIP6) were used as input to MIKE SHE. We used three different climate models (CanESM5, EC-EARTH3, MPI-ESM1-2-HR) and four shared socioeconomic pathways (SSPs 1-2.6, 2-4.5, 3-7.0, 5-8.5) each. Hydrological variables were evaluated for each of the twelve-climate model runs in comparison to the reference period (1980-2010).  
All climate simulations show an increase in annual precipitation, except for CanESM5 SSP 3-7.0, MPI-ESM1-2-HR SSP 2-4.5, and MPI-ESM1-2-HR SSP 3-7.0. The precipitation increases range from 1 % to 24 %. This underscores the impacts of different SSPs and climate models on projected regional precipitation patterns and emphasizes their importance in comprehensive climate change assessments. 
In all scenarios, the projections indicate an increase in flood for different durations (1-day, 3-days) at all gauges across different return periods. The flood increase calculated for the three different climate models exhibits greater differences than the flood increase calculated for different SSPs across climate models. For example, in the Ave River, the range of the 100-year flood across SSPs varies from 81 m³/s (Min: 432m³/s, Max: 513 m³/s) for MPI-ESM1-2-HR to 225 m³/s (Min: 496 m³/s, Max: 721 m³/s) for EC-EARTH3. The corresponding range across models spans from 71 m³/s (Min: 425 m³/s, Max: 496 m³/s) for SSPs 3-7.0 to 213 m³/s (Min: 508 m³/s, Max: 721 m³/s) for SSPs 5-8.5. The 100-year flood (1-day duration) in the reference period value is 372 m³/s. In addition, the duration of low-flow events increases significantly for most climate scenarios. This increase in extreme events, which includes both, an increase in the volume of floods and an increase in the duration of droughts, emphasizes the need for proactive measures to address and adapt to the anticipated changes in hydrological patterns due to climate change.
However, our findings show that the selection of the climate model has a great impact on the hydrological variables. Decision-makers should carefully choose a climate model aligned with their planning objectives, considering the potential risk for robust planning.

Keywords: Climate change, CIMP6 Climate Model, MIKE-SHE, Ave catchment