Determining treatment requirements for turbid river water to avoid clogging of aquifer storage and recovery wells in siliceous alluvium

摘要

The success of Aquifer Storage and Recovery (ASR) schemes relies on defining appropriate design and operational parameters in order to maintain high rates of recharge over the long term. The main contribution of this study was to define the water quality criteria and hence minimum pre-treatment requirements to allow sustained recharge at an acceptable rate in a medium-coarse sand aquifer. The source water was turbid, natural water from the River Darling, Australia. Three treatments were evaluated: bank filtration; coagulation and chlorine disinfection; and coagulation plus granular activated carbon and chlorine disinfection (GAC). Raw source water and the three treated waters were used in laboratory columns packed with aquifer material in replicate experiments in saturated conditions at constant temperature (19 ℃) with light excluded for 37 days. Declines in hydraulic conductivity from a mean of 2.17 m/d occurred over the 37 days of the experiment. The GAC-treated water gave an 8% decline in hydraulic conductivity over the 16 cm length of columns, which was significantly different from the other three source waters, which had mean declines of 26-29%. Within the first 3 cm of column length, where most clogging occurred in each column, the mean hydraulic conductivity declined by 10% for GAC-treated water compared with 40-50% for the other source waters. There was very little difference between the columns until day 21, despite high turbidity (78 NTU) in the source water. Reducing turbidity by treatment was not sufficient to offset the reductions in hydraulic conductivity. Biological clogging was found to be most important as revealed by the accumulation of polysaccharides and bacterial numbers in columns when they were dissected and analysed at the end of the experiment. Further chemical clogging through precipitation of minerals was found not to occur within the laboratory columns, and dispersion of clay was also found to be negligible. Due to the low reduction in hydraulic conductivity, GAC-treated water quality was used to set pre-treatment targets for ASR injection of turbidity ＜0.6 NTU, membrane filtration index (MFI) ＜ 2 s/L~2, biodegradable dissolved organic carbon (BDOC) ＜ 0.2 mg/L, total nitrogen ＜ 0.3 mg/L and residual chlorine ＞ 0.2 mg/L.