The aim of this study was to investigate whether a controlled upstream release of water could be delivered to the Torrens Lake to effectively dilute the population and control cyanobacteria numbers below a threshold that would force lake closure. The relatively high nutrient loads to the Torrens lake and the shallow morphometry means it is unlikely that physical and chemical conditions can be manipulated sufficiently to prohibit cyanobacterial growth. However, it was postulated that rather than control growth it might be possible to control population size with continual dilution of the population with flow.
For a growth rate of 0.4/day, which is a typical exponential growth rate of cyanobacteria in the Torrens Lake, it was concluded that a diluting flow of at least 10% per day would be required to have noticeable impact on the cyanobacteria population. With a starting cell concentration of 100 cells/mL, a growth rate of 0.4/dayand a diluting flow of 10%, the cell concentration after 20 days would be 74,420 cells/mL, which is below the critical threshold for cell numbers that would force lake closure.
The 2011/12 trial failed to maintain concentrations sufficiently low to avoid lake closure. However, the rainfall was also incredibly low during this period and so the lake population was not reset with elevated rain-event inflows. Sufficient questions remained on the viability of the trial for it to be repeated in the summer of 2012/13.
The results of the trial suggest that if dilution flows are released early enough, the size of the cyanobacterial population can be controlled. However, there is a reliance on rain events to flush the system and dilute the resident cyanobacterial population. Although on average, the flow return interval analysis suggests that rain events occur frequently enough in summer for this strategy to be effective this did not occur in 2011/12 or in 2012/13. In a variable climate like that observed in Adelaide, there may be very long periods between significant rainfall events. This may reduce the confidence in rain events to reset the population. Having even higher flows may dilute the population further and so control cyanobacterial abundance.
Cyanobacteria population showed explosive growth between 24 Dec and 31 Dec 2012. The rate of growth was 1.33 doublings/day, which is a rate four times higher than the long-term average. Flows were released in response to visual observations that cyanobacteria were present but these were unable to significantly reduce the rapid accumulation of biomass. Consequently it can be concluded that the amenity flow released from Hope valley was ineffective at reducing the cyanobacterial abundance during the highest growth period. Furthermore, commencing flows at the first detection of Microcystis aeruginosa would not have prevented lake closure because growth rate was too rapid and population increases could not be offset by dilution. The amenity flow alone was insufficient to prevent lake closure.
There was some concerns that the flow releases are having a detrimental impact on water quality downstream of the Torrens Lake. Sampling for cyanobacteria and Enterococci was initiated at a number of sites downstream and at West Lakes where some of the water was discharged. Cyanobacteria and Enterococci were transported downstream from the lake as a result of the amenity flow releases however, the abundance decreased significantly during river passage resulting in a two log removal (100 times dilution).
In consultation with the Department of Health a revision of the recreational guideline was undertaken. This resulted in an increase in the recreational guideline to a biovolume of 40 mm3/L effectively doubling the tolerable concentrations of cyanobacteria before the lake is closed based on health and exposure risk (biovolume = cell/mL multiplied by voume of cells). This effectively increases the length of time the lake can remain open. Depending on the criteria used to determine the whole lake concentration, the lake could remain open even with a dense bloom. For example if the mean cell concentration in the main lake site (sites 1-6) is used then a mean concentration of 230,000 cells/mL of Microcystis aeruginosa would be permissible and the lake would have remained open all summer. A cautionary note is that at these concentrations dense surface blooms would be present which would be unsightly and may be odorous. It is recommended that the concentration of cyanobacteria at the most upstream site (site 7) be excluded from whole lake determination of mean cell concentration. Cells often accumulate at the upstream site but persist there until they are washed downstream during storm events. The persistent upstream biomass presents a very low risk to lake users and so may unnecessarily bias the value used to determine lake closure.
The amenity flows achieve environmental benefits other than dilution. They serve to maintain water in a river that would have had base flow during summer which has been largely lost by river regulation and catchment development; they provide habitat for native fish; and the water is used to provide water to the Barker Inlet wetlands during long dry summer and there is freshening of water in the entire reach from Hope Valley reservoir to the coast. Amenity flow releases appear to have had a positive effect on native fish communities in the lower Torrens. This translated to increased endemic fish abundance and diversity and a decrease in alien and translocated fish abundance. No immediate negative effects resulting from changes in water quality due to summer flow releases were apparent and, although spawning was stimulated in the alien common carp (Cyprinus carpio), idiosyncrasies in the Torrens morphology meant that these events were of little consequence to fish diversity downstream of the Torrens weir.
On balance the amenity flows are insufficient on their own to provide relief against high cyanobacterial growth, however, if coupled with algicidal technology such as hydrogen peroxide they may attain the upstream and downstream benefits and cyanobacterial control in-lake. A larger flow volume released from Hope Valley would have greater dilution effects and so slow the increase in population expansion in the lake. In the absence of rain events to reset the cyanobacterial populations in the lake, large water releases from upstream reservoirs could fulfil a similar role.
This project was an experimental trial but was also a high-profile strategy to improve water quality in the River Torrens. The results were used immediately for setting flow targets maintain cyanobacterial populations below concentrations that would force lake closure. The trial 2011/12 had unanswered questions and so was repeated to determine whether this is an appropriate long-term strategy for controlling cyanobacteria in the River Torrens.
