The water quality of the River Kennet: initial observations on a lowland chalk stream impacted by sewage inputs and phosphorus remediation.

The water quality of seven sites on the upper reaches of the River Kennet round the market town of Marlborough is described and related to the introduction of phosphorus treatment of effluent from Marlborough sewage treatment works (STW). The River Kennet is mainly groundwater-fed from a Cretaceous chalk aquifer and hence the river water is calcium- and bicarbonate-bearing and has a relatively constant composition of many major water quality determinants. In-stream biological activity gives rise to marked diurnal fluctuations in pH (of approx. 0.8 units). Dissolved carbon dioxide and dissolved oxygen also show marked diurnal fluctuations. Dissolved carbon dioxide varies from approximately 10 to 70 times atmospheric pressure, indicating net release of carbon dioxide and the dominance of heterotrophic (respiratory) processes over autotrophic processes (photosynthesis). Much of the excess carbon dioxide is probably associated with carbon dioxide laden groundwater inputs and the relatively short within-stream residence times ensures only limited degassing to the atmosphere. Diurnal fluctuations in dissolved oxygen vary from approximately 20% to 200% saturation. For both dissolved carbon dioxide and dissolved oxygen, the amplitude of fluctuations is much lower during the winter period, when biological activity is at its lowest. The concentrations of soluble reactive phosphorus (SRP), total phosphorus (TP) and boron increase markedly just downstream of the sewage works as a result of this point source input. These concentrations slowly decline further downstream as additional groundwater inputs dilute the effluent further. The introduction of chemical treatment of sewage effluent for phosphorus reduction at Marlborough STW resulted in a marked decrease in within-river SRP and TP concentrations to levels approximately the same as those upstream of the STW. A comparison of SRP and boron concentrations reveals a reduction in in-stream SRP concentrations by approximately 75% following effluent treatment. In terms of within-river processes controlling in-stream phosphorus concentrations, previous studies have indicated that one potentially important mechanism within calcium bicarbonate bearing rivers may be related to co-precipitation of phosphorus with calcium carbonate (calcite). The present study shows that the waters are oversaturated with respect to calcium carbonate, that no equilibrium conditions exist and that phosphorus removal has led to undetectable changes in calcium carbonate oversaturation. Hence, it is concluded that the primary changes in phosphorus levels within the river is directly associated with changing point source contributions from the STW and physical dilution within the river. However (1) the results relate to only the first year of study and subsequent differences may become apparent and (2) reactions between the water column and plant and bottom sediment interfaces may be important in regulating phosphorus fluxes within the river. The results presented in this paper mark a pilot phase of a longer-term initiative and this paper provides a background setting. The paper discusses the longer-term objectives and important gaps in knowledge of the system that requires further address.