Development of a novel approach to assess the risk of physical clogging at managed aquifer recharge sites

Increasing aquifers' recharge and storage is of great importance in addressing challenges posed by climate change and growing water demand. Managed Aquifer Recharge (MAR) technologies may ensure water supply for agriculture and diminish impacts from groundwater overexploitation. The expansion of MAR solutions in Europe still requires the implementation of these waterworks at their maximum efficiency. Physical clogging is one of the main bottlenecks for these technologies. In spreading methods, during water recharge, eroded clays from surface runoff reach the infiltrating surface and intrude into the soil matrix, decreasing the basin infiltration capacity over time. The resulting loss in performance increases the operation and maintenance (O&M) costs and, in extreme cases, can lead to the MAR site's abandonment. Thus, it is vital to assess the risk of physical clogging during the MAR planning phase, extending the MAR scheme lifespan and minimising O&M costs. Our study aims to develop a comprehensive model for physical clogging transferable to multiple MAR sites, based on the characterisation of the sediment matrix and MAR operations. To achieve this, we built a semi-empirical 1D numerical model for physical clogging. Evolution in soil permeability via the Kozeny-Carman equation is computed in function of depth based on the input of fines into the soil matrix and the porous media characteristics. The vertical distribution of fines is derived through a general relationship from a systematic review of multiple studies in the literature. The model allows computing the evolution in infiltration rates over time for the MAR site and the depth of soil to be treated to restore infiltration efficiency. Preliminary validation at the field scale is conducted at a MAR infiltration basin in Suvereto, Italy. To spatially apply the model, zoning is performed through an electromagnetic induction (EMI) survey, defining areas with similar soil properties. Values of hydraulic conductivity near saturation and soil samples were collected to characterise the sediment matrix and fines content for the entire basin. Predictions of the expected decrease in infiltration capacity for spreading methods assists maintenance scheduling and reduce O&M costs for the specific site. The proposed model for physical clogging can serve as a tool for decision support when exploring a set of design alternatives prior to MAR construction.