Velocity and celerity in a forested headwater catchment : a combined experimental and modelling approach

One of the most important issues in modern hydrology is to improve our understanding of the release of old water during rainfall events. This thesis approaches the problem of estimating velocities, a measure of water transport, and celerities, a measure of the hydrograph response. We aim at measuring and interpreting estimates of velocity and celerity in a consistent way using in situ data. For this purpose, we performed multi-tracer irrigation experiments at different scales, from soil column to hillslope, to a sub-catchment scale analysis in a forested headwater catchment characterised by fractured bedrock. Our field experiments proved the importance of bedrock cleavage orientation in controlling subsurface flow direction and demonstrated the importance of quantifying the extent of fractures as well as their orientation relative to dominant topographically related flowpaths. An undisturbed soil column experiment was used in a hypothesis-testing framework in combination with a Multiple Interacting Pathways (MIPs) model. The use of a transition probability matrix (TPM) in combination with immobile water and variable field capacity parameters allowed the representation both volume and tracer dynamics. A framework to estimate both velocities and celerities using commonlyavailable hydrometric and tracer data is presented, emphasising the importance of choosing appropriate distance information as it can strongly influence the estimates and thereby the interpretation of controls on catchment function. The analysis of velocities and celerities at different spatial and temporal scales showed that the relationship between velocity and celerity shows a positive relation at the stream outlet. The experimentally-derived velocities and celerities metrics hereby explored have the potential to contribute to the evaluation hypothesis regarding catchment storage and release of water, by providing a direct comparison of what controls the old-water paradox.