Flexible Operation of a Deep Multi-Well System in a Coastal Aquifer

PWN is the drinking water supplier for the Province of North Holland in the Netherlands. They manage a deep infiltration system (DWAT) since 1990 with wells between NAP -55 and NAP -90 meters, which are used to inject water and to extract it after at least 60 days to produce drinking water. Renovation of these wells will lead to an increased pumping capacity (from 50 to 90 m3/hour/well) of the extraction wells and creates opportunities for a more flexible operation. Seasonal fresh water storage can be applied to deal with the difference in water demand between summer and winter. PWN applies a continuous injection (30 m3/hour/well) and aims to get an annual injection volume which is 5% larger than the extraction volume. A period of maximum extraction is compensated by lower extraction rates during the remaining months of that year so that the total extraction volume remains the same. The objective of this research is to examine the effects of a more flexible approach on the extracted chloride concentrations and the salinity around the system. A new combined flow and transport model for the DWAT area is developed. Two programs are used in order to derive a stable transport model without large initial fluxes. First, MODFLOW's SWI package was used to develop a large groundwater model with a low computational effort. SWI was used to simulate the interface position and to find boundary conditions that matched with measured interface depths and groundwater levels. Second, the boundary conditions, interface depths, and groundwater levels were used to set up a smaller flow and transport model with SEAWAT. PWN constrains the application of seasonal storage by requiring a minimum residence time of 60 days which limits the duration of maximum extraction. Borst (2015) has shown that the maximum extraction rate of 90 m3 /hour/well can be applied up to 45 consecutive days with standard injection. No significant effects on the fresh-salt interface can be seen after this increased extraction period. Model results show that the past 26 years of DWAT operation have formed brackish cones just outside the DWAT system, particularly at the sea side. Measurements confirm the occurrence of upconing around the system. Seasonal storage has a small influence on the brackish cones: it influences the salinity around the system by a slight expansion of the brackish groundwater volume, but the extracted chloride concentrations do not increase over the years since PWN requires an annual infiltration surplus of 5%. This research indicates that the limitation on the extraction of groundwater is influenced more by the residence time restriction (as computed by Borst (2015)) than by the chloride concentration restriction (