Authors:
Koen Zuurbier | KWR Watercycle Research Institute | Netherlands
Dr. Niels Hartog | KWR Watercycle Research Institute | Netherlands
Prof. Pieter Stuyfzand | KWR Watercyle Research Institute | Netherlands
The recovery of freshwater during conventional aquifer storage and recovery (ASR) in coastal (i.e. brackish or saline) aquifers is impeded by buoyancy effects, causing salinization before the injected volume is recovered. Recently, the use of multiple partially penetrating wells (MPPW) in such aquifers was shown to significantly improve the recovery of freshwater (i.e., water with low Cl). However, this study shows that the salinity contrast and the deviating flow paths in the MPPW-ASR system does not result in progressively improving water quality with subsequent cycles observed at conventional ASR systems. Moreover, other parameters than chloride may limit the recovery efficiency, depending on the intended use of the recovered water.
Due to cation exchange processes, the first freshwater injected at the base is enriched with sodium (Na) concentrations and subsequently recovered at the top of the aquifer, whereas the arrival of Na is retarded during recovery in the lower and central parts of the aquifer. Depending on the maximum limits applied for Na and the cation exchange capacity and native groundwater composition of the aquifer, this can either increase or decrease the recovery efficiency (RE) of ASR operation compared to the RE based on conservative Cl.
Similar to conventional ASR in anoxic aquifers, oxygen-containing, fresh injection water cause oxidation and/or dissolution of reactive minerals like pyrite, organic matter, and (Fe,Mn)CO3, which can lead to undesired mobilization of Fe, Mn, and As. The Nootdorp MPPW-ASR pilot demonstrated that the depth position of reactive layers is relevant for the mobilization of these undesired elements, rather than the average geochemical composition of the target aquifer. Additionally, the net extraction at the shallow wells of a MPPW-ASR system permit mobilized elements to travel relatively unhampered vertically through the aquifer towards the shallow recovery wells. Recovery of Fe and Mn may be prevented or postponed, however, by frequent injection of small volumes of oxygen-rich water at the shallowest well during recovery, triggering subsurface Fe and Mn removal.
The findings indicate that the reactive transport impacts clearly deviate at the MPPW-ASR set-up compared with conventional, bi-directional ASR, where the effect of cation-exchange reduces cycle-after-cycle and a smaller portion of the injected water flushes reactive intervals. This may require a more detailed geochemical characterization of target aquifers for MPPW-ASR, as well as an optimized operation of its injection and recovery wells.
The results of this study were obtained at a small-scale ASR field pilot in Nootdorp, The Netherlands. Here, rainwater collected via a greenhous roof was stored in a shallow brackish aquifer for later use as irrigation water. This way, the greenhouse owner became self-sufficient in its freshwater supply and acts as a showcase for future freshwater management in The Netherlands.