Experts are calling for more sensitive water-quality testing in the Darling-Baaka River amid concerns that pesticides could be contributing to poor conditions, blue-green algae blooms and fish deaths.
It follows two of the state’s top scientific bodies publishing test results from water samples taken near Menindee in far western New South Wales which contained inconsistent results.
Testing conducted on behalf of the NSW chief scientist, on samples collected in August 2023 as part of an independent review into the deadly fish kill event in March in which 30m fish died, found low levels of some pesticides. Separate testing led by the NSW Environment Protection Authority did not detect any pesticides, despite collecting samples at similar locations and during the same month as the chief scientist’s testing.
The testing for the chief scientist was conducted by Charles Sturt University. It detected several herbicides including atrazine, simazine, terbuthylazine, tebuthiuron, metolachlor, clopyralid and fluroxypyr.
Atrazine, simazine, tebuthiuron and metolachlor are no longer approved for use in the European Union and have been classified by the European Chemicals Agency as being “very toxic to aquatic life with long lasting effects”, but they are still approved for use in Australia.
The NSW deputy chief scientist and engineer, Dr Darren Saunders, said the CSU testing was commissioned due to requests from the community and the independent review’s need for further data.
The CSU report said the herbicide levels were “consistent with agricultural land-use in the area” and recommended ongoing monitoring. Two pesticides, tebuthiuron and metolachlor, were found to have exceeded the 99% species protection level in Australian and New Zealand guidelines for fresh and marine water quality, which signifies a concentration estimated to be toxic to 1% of organisms in that aquatic environment. Both were below the 95% value, which the EPA tests for.
In contrast, the NSW EPA said it had not detected any pesticides of a reportable level in any routine testing of the river which had been conducted since the fish kill in March. This included its results from August, when the CSU samples were also collected.
‘A cocktail of pesticides’
Dr Matt Landos, an adjunct associate professor at the University of Queensland and director of Future Fisheries Veterinary Service, said the results showed “seriously concerning” levels of herbicides, which could produce “an elevated risk of ecological harm occurring”.
Landos said the guidelines did not consider the cumulative impact of “a cocktail of multiple pesticides” on the aquatic ecosystem.
He said that while the mass fish deaths at Menindee were attributed to low oxygen levels, the potential contribution of pesticides should also be investigated, alongside other issues like excessive nutrients from agricultural run-off.
“Pesticides were likely one of several contributors harming the ecosystem, particularly upstream, making the water that flowed to Menindee vulnerable to produce near-zero oxygen conditions,” he said.
Landos said pesticides, even at low levels, could modify the aquatic food web by “selectively harming sensitive organisms” such as algae and allowing more tolerant species to predominate.
“This is what’s happening in the Murray-Darling,” Landos said. “And the most tolerant organisms, as it turns out, are cyanobacteria, or blue-green algae.”
He said blue-green algae blooms could impair the ecological function of the river by creating volatile dissolved oxygen conditions, where oxygen levels rapidly rise during the day “due to extreme photosynthetic activity” and then plummet overnight.
“The bigger the biomass of algae, the more oxygen they’re going to require in that water body to keep everything alive,” he said. “If the bloom is too dense, it will suck the oxygen level down far too low and it can in fact cause mortality even to the algae.”
after newsletter promotion
Landos said it was concerning that the EPA testing and CSU testing had produced such different results, and called for a review into the “sensitivity” of the testing methods.
The EPA told Guardian Australia its results were valid and its testing did not show any concentrations which exceeded the reporting limits. It said those limits were “well below the 95% guideline values” and consistent with best practice under the laboratory standard used by the Department of Climate Change, Environment, Energy and Water.
In April, the EPA announced its testing had “ruled out a pesticide pollution event” after the results came back negative for more than 600 pesticides. It has since reduced the frequency of pesticide sampling “following consistent zero results” from multiple testing rounds, including its results from August.
A spokesperson for NSW EPA said the organisation’s testing methods “are sensitive enough to detect concentrations of ecological significance”. They added that results below that limit “have a high uncertainty, are not meaningful and reporting them is not appropriate nor consistent with analytical best practice”. They also suggested the CSU results were inconsistent, as the published report was amended in December.
“The reporting limit is the lowest concentration that can be reliably measured,” the EPA said.
They also said that exceeding the 99% or 95% default guideline values “does not indicate an actual impact, rather that further investigations are warranted”.
Vincent Pettigrove, a professor of aquatic pollution at RMIT, said while the guideline values provided an indication of whether the chemicals of interest were at concentrations which may cause environmental harm, there were no national guidelines for “the majority of pesticides registered for use in Australia”.
He added that the ecotoxicological tests used to derive the guideline values for pesticides “are often short-term acute tests and do not account well for chronic long-term exposures”.
“I believe that the impact of pesticides on fish is more likely to be more subtle from chronic exposure that may reduce the viability of populations.”
Pettigrove said the guideline values were for one chemical and therefore “unable to consider the total effects of a mixture of chemicals”. For example, the triazine herbicides detected in the CSU study had a similar mode of toxicity, “so their combined effect should be considered”, he said.