Macrophyte beds in a subtropical reservoir shifted from a nutrient sink to a source after drying then rewetting

Abstract

1. Macrophytes play a key role in assimilating and storing nutrients in shallow aquatic ecosystems, but their capacity to act as a long-term nutrient sink can be affected by water level fluctuations. Water level drawdown in reservoirs followed by rewetting may mobilise a significant nutrient pool. These nutrients can be stored in the littoral zone in dead or dormant macrophytes, and in the desiccated sediments within the macrophyte beds. However, the contribution of desiccated macrophyte beds to nutrient release upon rewetting has not been well quantified. 2.Our study examined the effect of rewetting the previously desiccated waterlily Nymphoides indica (Menyanthaceae) in treatments (1) without sediments (N.i.−Sed), and (2) with sediments in N. indica beds (N.i.+Sed) on water quality. 3. We found that longer drying duration increased dissolved nutrients (nitrate/nitrite, ammonium and total dissolved organic nitrogen/phosphorus) and organic carbon release from N.i.+Sed and N.i.−Sed treatments after rewetting. In the N.i.+Sed treatment with <4 weeks of desiccation, all N. indica plants regenerated from roots after subsequent rewetting. In addition, the resulting nutrient/carbon release was not significantly different to the control treatment which did not have desiccation. A significant increase in dissolved nutrient and carbon concentrations in the water column was found in treatments with more than 10 weeks of desiccation followed by rewetting. This coincided with the sediment reaching its minimum moisture content. Furthermore, chlorophyll a (Chl-a) concentrations in the overlying water also increased with more than 10 weeks of desiccation, presumably in response to the increased nutrient availability and the removal of competition for nutrients from macrophytes. 4. On the basis of our laboratory experiments, we calculated the potential effect of desiccated and rewetted N. indica beds on water quality in a local N. indica-dominated reservoir after water level drawdown. We also separated the contribution of N. indica plants from their macrophyte beds on water quality changes. Fourteen days after rewetting, the total dissolved nutrients released from N. indica alone (N.i.−Sed) could contribute 0.5% of the total nitrogen and 29% of total phosphorus to the water column concentrations in the whole reservoir. In contrast, N. indica beds (N.i.+Sed) contributed more total dissolved nitrogen (4.3%) but less total dissolved phosphorus (0.3%) release into the water column. The higher nitrogen release for the N.i.+Sed treatment was likely due to the organic matter decomposition in the sediment in macrophyte beds. In contrast, the less dissolved phosphorus release, compared with the N. indica alone, was likely the result of phosphate adsorption by previously desiccated soil particles and/or assimilation by phytoplankton, since the phytoplankton biomass (as measured by Chl-a concentrations) was significantly higher in the N.i.+Sed treatment. 5. This study highlights the importance of managing both the duration and rate of water level drawdown in reservoirs to prevent rooted macrophytes, like N. indica, from becoming a source of nutrients, which may cause deterioration in water quality.