Luit Jan Slooten

Impacts of rainfall spatial variability on hydrogeological response

There is currently no general consensus on how the spatial variability of rainfall impacts and propagates through complex hydrogeological systems. Most studies to date have focused on the effects of rainfall spatial variability (RSV) on river discharge, while paying little attention to other important aspects of system response. Here, we study the impacts of RSV on several responses of a hydrological model of an overexploited system. To this end, we drive a spatially distributed hydrogeological model for the semiarid Upper Guadiana basin in central Spain with stochastic daily rainfall fields defined at three different spatial resolutions (fine → 2.5 km × 2.5 km, medium → 50 km × 50 km, large → lumped). This enables us to investigate how (i) RSV at different spatial resolutions, and (ii) rainfall uncertainty, are propagated through the hydrogeological model of the system. Our results demonstrate that RSV has a significant impact on the modeled response of the system, by specifically affecting groundwater recharge and runoff generation, and thereby propagating through to various other related hydrological responses (river discharge, river-aquifer exchange, groundwater levels). These results call into question the validity of management decisions made using hydrological models calibrated or forced with spatially lumped rainfall.

Modelling of Matrix Diffusion in a Tracer Test in Concrete

A laboratory-scale tracer test has been carried out to improve the characterization of the transport properties of the concrete from the radioactive waste disposal facility at El Cabril (Spain). High entry pressure was employed in order to perform the experiment in a reasonable time span. Lithium, bromide and deuterium were used as tracers. The conceptual model considered matrix diffusion between a mobile pore domain, where water can flow, and an immobile zone without any advective transport. Three geometries have been compared, considering the immobile zone as slabs, spheres or tubes. Porosity of the mobile zone and characteristic time was estimated by calibrating the model results to the measured breakthrough curves of deuterium and bromide. The calculated values showed that the characteristic time depends on the geometry, and similar porosity of the mobile zone was estimated for all geometries. The double-porosity conservative transport model could reproduce the deuterium breakthrough curve. However, the bromide behaviour could not be reproduced even when linear retardation was applied.