Abstract Rivers in karstic environments are known to be greatly influenced by surface water–groundwater interactions, with significant localized inflows during floods from springs, or with losses that can dry up rivers. The Middle Risle River is frequently affected by the development of sinkholes in a chalk karst area (Normandy, France). In the 2010s, two new major sinkholes in the riverbed caused a complete loss of water into the underlying phreatic aquifer, causing the river to dry up over several kilometres. The resulting changes in hydrogeological processes and surface water–groundwater interaction greatly affected water quality, water use and water-dependent ecosystems, causing a political crisis in this river-dependent touristic valley. To understand these phenomena and improve crisis management, the Middle Risle Critical Zone Observatory was set up to enhance monitoring, surveying and/or modelling of groundwater and river levels, river and spring flow, water temperature and conductivity, and ecosystem characteristics (fish, macro-invertebrates and vegetation). The results showed notable impacts on fish, macro-invertebrates and vegetation, some plants proving to be reliable indicators of surface-water–groundwater interaction. The dynamics of local hydrogeological processes were assessed and linked to the measured effects on ecosystems and water supply. Inverse modelling based on an analytical solution of the diffusive wave equation assessed lateral flow during floods, quantifying the spatial–temporal variability of surface-water and groundwater exchanges. It also highlighted the important role of karst zones in both storage and flood-peak attenuation processes, thereby protecting downstream villages against floods.

Abstract The study of the temperature of two rivers in Normandy (France), the Orne and the Touques, between 2013 and 2018 allowed the main controlling factors regulating their thermal regime to be determined. The analysis was conducted by coupling different statistical treatments: linear regression between water and air temperatures, independent component analysis, principal component analysis and a multiple linear regression model. The temperature of the two rivers is mainly controlled by climatic factors but secondary regulation factors are demonstrated to play important roles: runoff for the two rivers and groundwater for the Touques. The influence of the chalk aquifer on river temperature appears to vary seasonally throughout the year, reaching its maximum in the early spring and increasing from upstream to downstream. The coupled use of the different statistical complex methods showed its validity in understanding both the temporal and spatial variations in water temperature and its correlation with the secondary factors, that could not be inferred from a simpler approach based on linear regression. These techniques could be valuable in other areas with rivers sufficiently monitored to determine the controlling water temperature factors and thus their sensitivity to climate change.