Water-level drawdowns can improve surface water quality and alleviate bottom hypoxia in shallow, eutrophic water bodies

Water-level drawdowns are a management option for improving water quality in shallow, eutrophic lakes, but how much water levels should be reduced and what abiotic and biotic mechanisms can be expected to reduce the adverse effects of eutrophication are unclear. We conducted a study with two experimental pools (10 m wide, 30 m long, and approx. 2.5 m water depth) in a before-after-control-impact design to examine whether water-level drawdowns could influence surface water quality and bottom-water conditions, as well as how physicochemical and biological variables contributed to improvements. In the experiment, we fertilised the pools to increase productivity and induce hypoxia, and then reduced water levels four times in increments of 0.5 m. Our experiment using high-frequency sensors showed that water-level drawdowns could decrease cyanobacterial blooms and alleviate hypoxia relatively quickly by changing physicochemical conditions. Water-level reductions quickly increased bottom light illuminance, increased bottom-water temperatures, and weakened stratification. The result was a dramatic increase in bottom-water dissolved oxygen concentrations. Water-level drawdowns also decreased the relative fluorescence of chlorophyll and phycocyanin, probably because of a decrease of internal nutrient loadings from sediment. Total nitrogen and NH4-N concentrations in the water column decreased after water-level reductions, and NH4-N and PO4-P concentrations in sediment pore waters were reduced in the water-level treatment. Periphyton biomass did not change after the water-level drawdowns. Water-level manipulations increased the abundance of cyclopoids and calanoids but not large cladocerans. These biological processes responded slowly to water-level drawdowns and are unlikely to have contributed to water quality improvement. Overall, the size of the water-level reduction required to start changing water quality was 0.5-1.0 m. Our results suggested that a temporary reduction in water-level of 20%-40% of the depth might help to suppress cyanobacterial blooms and hypoxia in shallow lakes, reservoirs, and agricultural ponds. Such water-level management may help resolve conflicting demands for water and may be incorporated into management plans for flood control.