Toxicity of the herbicide Tebuthiuron to Australian tropical freshwater organisms: Towards an ecological risk assessment
Supervising Scientist Report 131
Camilleri C, Markich S, van Dam R and Pfeifle V
Supervising Scientist, 1998
ISBN 0 642 24334 4
- SSR131 - Toxicity of the herbicide Tebuthiuron to Australian tropical freshwater organisms: Towards an ecological risk assessment (PDF - 838 KB)
One of the major recognised causes of wetland degradation and loss in Australia is the invasion of exotic species. Nowhere is the threat of weeds currently greater than in the wetland habitats of the wet/dry tropics of northern Australia. One of the major weed threats, Mimosa pigra (mimosa), grows as a leguminous shrub up to 6 m tall, forming dense, impenetrable monospecific stands in floodplain environments. It is known as the 'giant sensitive plant' due to its bipinnate leaves that close when touched, and has invaded vast areas of northern Australian floodplains.
Strategic control of mimosa has concentrated on integration of mechanical/physical, biological and chemical control methods. Chemical control has been the most widely used approach to date and involves the application of herbicides to mimosa stands, although in some situations all three methods of control are employed. Several herbicides have been used to control mimosa but probably the most widely used has been Tebuthiuron-the active ingredient of the commercial formulation Graslan®. Tebuthiuron is a thiadiazole urea herbicide that acts to kill plants by uncoupling electron transport and thereby inhibiting photosynthesis. Graslan® contains Tebuthiuron at concentrations of either 10, 20, or 30%, and is applied to soils in clay pellet form, with primary uptake by plants being through root absorption.
In 1991 approximately 62 000 kg of Graslan® (~ 12 000 kg Tebuthiuron) were applied to a mimosa infestation of approximately 5800 ha at Oenpelli in western Arnhem Land, highlighting the extensive use of the herbicide in one area of northern Australia. Such large-scale application of herbicides in northern Australian wetlands is of particular environmental concern, particularly considering that there were no toxicological data available on the effects of Tebuthiuron to non-target tropical freshwater species. Adding to concerns about northern Australian wetland environments, regulatory authorities and the product's manufacturer recommend that Tebuthiuron not be applied near established watercourses or where surface water is present.
While no data exist on the aquatic toxicity of Tebuthiuron to Australian tropical freshwater species, its effects on northern hemisphere temperate species have been extensively studied. The studies indicated that Tebuthiuron toxicity to aquatic animals was very low compared to aquatic plants. Nevertheless, considering the large amounts of Tebuthiuron used in northern Australia for mimosa control, it was imperative that an assessment of the sensitivity of local aquatic organisms to the herbicide be performed.
The aims of the present study were to:
1 Assess the toxicity of Tebuthiuron to the following freshwater animals:
- the purple spotted gudgeon, Mogurnda mogurnda
- the green hydra, Hydra viridissima
- the cladoceran, Moinodaphnia macleayi
2 Use the above toxicity results, and all other relevant information, to undertake a preliminary ecological risk assessment on the use of Tebuthiuron for the chemical control of mimosa in wetland habitats of northern Australia.
Toxicity test results showed that the toxicity of Tebuthiuron to the three organisms tested decreased in the following order:
cladoceran (M. macleayi)>hydra (H. viridissima)>gudgeon (M. mogurnda).
The 10% bounded effect concentration (BEC10) and EC50 for M. macleayi, H. viridissima and M. mogurnda were 17.4 and 134, 40.6 and 153, and 108 and 214 (LC50) mg L-1, respectively. Overall there was little difference in the toxicity of Tebuthiuron to Australian tropical species compared to northern hemisphere temperate species, although M. mogurnda was approximately 1.3-1.9 times more sensitive than northern hemisphere fish species. However, it was recommended that further data on local species, including plants, be obtained in order to perform a more substantial comparison.
As a means of evaluating statistical endpoints for use in deriving water quality guidelines, a comparison of various statistical endpoint values for the three test species was undertaken. In all cases, the BEC10 was lower than the corresponding no-observed-effect concentration (NOEC). However, more confident estimates of the NOEC (ie when a large number of concentrations were tested and sample size (n) was high, eg in the hydra and cladoceran experiments) closely approximated the BEC10, indicating that the BEC10 was an appropriate estimate of a no adverse biological effect concentration. The EC10 was not considered an appropriate indicator of such a 'no-effect' concentration for Tebuthiuron.
A literature review on the fate and behaviour of Tebuthiuron in the aquatic environment, particularly northern Australian floodplain environments, was carried out in order to estimate a likely level of exposure to local aquatic organisms. As this study represented only a preliminary phase of an overall ecological risk assessment of Tebuthiuron, and some data were still lacking, a precautionary principle approach was adopted for determining the likely exposure level. Thus the likely maximum level of Tebuthiuron aquatic organisms would be exposed to was estimated to be 4.9 mg L-1, being the highest concentration of Tebuthiuron measured in the water column following a major application to a mimosa infestation in northern Australia.
While this level significantly exceeded the recommended Australian water quality guideline value for Tebuthiuron of 1 µg L-1 (ANZECC 1992), a comparison with the toxicity data from the three species assessed in the present study indicated that the risk of adverse effects to the non-target aquatic organisms was minimal. However, to date, insufficient local data are available to quantitatively assess the risks of Tebuthiuron to the tropical wetland environments of northern Australia. In particular, the response of local aquatic plants needs to be assessed, as does the ability of organisms and plants to recover from short-term exposure to potentially toxic Tebuthiuron concentrations. Experiments assessing the toxicity of Tebuthiuron to an aquatic macrophyte and green alga were already underway at the time of completion of the present study. It is anticipated that the results will be combined with those of the present study, in order to undertake a quantitative ecological risk assessment of the use of Tebuthiuron to control Mimosa pigra.