RRRD058: Biomonitoring effects of pesticides in rivers draining on to the Great Barrier Reef.

Ben J Kefford1, Rebecca J. Wood1, Simon Mitrovic and Peter von der Ohe

1 Centre for Environmental Sustainability, University of Technology Sydney, Sydney

2 UFZ, Helmholtz Centre for Environmental Research, Leipzig, German

# Current address: Institue for Applied Ecology, University of Canberra, Bruce, ACT


pdfDownload the RRRD058 Research Outcomes Report here3.99 MB

Executive Summary

Coral cover across large parts of the Great Barrier Reef (GBR) in north Queensland, Australia has declined since the 1960's, with the exception of the northern section, where little intensive agriculture takes place in the respective river catchments draining in the Great Barrier Reef. Agricultural runoff - containing nutrients, sediments and pesticides - that reaches the marine environment via rivers is believed to be a significant factor in this decline. Agricultural runoff is also believed to be an important stressor for other estuarine and marine ecosystems within the GBR world heritage area (GBRWHA), including seagrass and mangrove systems. Of the pesticides in agricultural runoff, persistent herbicides, particularly the photosystem II inhibitors (hereafter PSII), are believed to pose the greatest risk to ecosystems and organisms in the GBRWHA.
While it will not replace chemical monitoring, biological monitoring (biomonitoring) of effects of pesticides on biotic communities can complement existing chemical monitoring programs and together both methods can provide a more comprehensive picture of the state of environmental contamination. Aquatic organisms live continuously in the water and are therefore exposed to all chemicals in the water over their lifetime in contrast to grab samples that quantify selected chemicals in a sample at an instant in time. So biomonitoring studies can assure that the effects of peak concentrations of pesticides are captured and that no ecologically relevant chemicals in the waterare overlooked due to low concentrations or lack of available testing method. There is uncertainty as to the ecological relevance of concentrations of many pesticides and much uncertainty as to the effects of mixtures of multiple pesticides. Biomonitoring by measuring the ecological effects, and not concentrations of a potentially incomplete set of chemicals, side-steps these areas of uncertainty. Biomonitoring detects the effect of the pesticides and if an effect is detected it indicates ecologically relevant exposure (although not which particular chemical or mixture caused the effect). Biomonitoring also allows, at relatively modest cost, greater spatial coverage than current chemical monitoring programs. Finally, biomonitoring allows for the re-analysis of historical biotic data, which in the case of rivers draining into the GBR, allows the historical trends to be considered, assome biotic data precede chemical monitoring.
This document reports on the findings of a study that developed biomonitoring methods to detect the impacts of pesticides on freshwater organisms in rivers draining into the GBRWHA. In particular, we modified an existing stream macroinvertebrate-based SPEcies At Risk from pesticides (SPEARpesticides) index, that has up until now been used in regions with temperate climates, so that it is suitable for use in tropical north Queensland. This index has been successfully used across Europe and in Victoria, Australia and responds specifically to impacts on macroinvertebrate communities as a result of pesticides, but appears to be largely unaffected by other environmental factors. As macroinvertebrates are not photosynthetic organisms they are generally more tolerant of direct effects of herbicides compared to algae and plants, and thus SPEARpesticides largely detects impacts from insecticides and fungicides.
To detect impacts of herbicides, we devised a new biomonitoring index based on freshwater benthic diatoms (a common group of micro photosynthetic algae with a cell wall made of silica) to detect impacts of herbicides. This new index, called SPEARherbicides, is based on similar principles as the macroinvertebrate-based SPEARpesticides. (Note that we retain the name SPEARpesticides for the macroinvertebrate-based index, despite the fact that herbicides are pesticides, to be consistent with the usage of the term SPEARpesticides elsewhere).
The SPEARherbicides index was developed by conducting a series of rapid toxicity tests designed to test the sensitivity of large number of diatom taxa collected from rivers draining into the GBR. These rapid tests were specifically developed as part of the current study. The results of these toxicity tests were used to classify diatom taxa as either SPEcies At Risk (SPEAR) or SPEcies not At Risk (SPEnotAR) from herbicides. For species which we did not test, their classification was based on extrapolation from related species of the same genus. Genera that were untested were unclassified and not used in the calculation of SPEARherbicides that appears in this report. From this classification, the final SPEARherbicides index was derived. The rapid tests included, in addition to herbicide concentration, experimentally manipulated up to 8 types of herbicide, light intensity, nutrient levels and source of diatoms. Results showed that the relative sensitivity of diatom genera was independent of herbicide type, light intensity, nutrient level and collection sites.
We sampled the stream macroinvertebrate and benthic diatom communities at the 11 Department of Science, Information Technology, Innovation and the Arts (DSITIA) pesticide monitoring sites that are part of the Great Barrier Reef Catchment Loads Monitoring Program and at 3 reference sites in rivers before and after the 2011/12 and 2012/13 wet-seasons. At present we have received pesticide concentration data from the 2011/12 wet-season but not the 2012/13 wet-season. When the data from the 2012/13 wet-season are available, analysis of the biotic data from the season will occur.
After the 2011/12 wet-season the SPEARherbicides at the pesticide monitoring sites was inversely correlated with the estimates of the peak concentration of photosynthesis II inhibiting herbicides as expressed as atrazine Toxic Equivalent concentrations for Scenedesmus obliquus (TEQso) during that wet-season. This indicated that the relative abundance of herbicide sensitive diatom species declined with increasing herbicide concentration. This result suggest that the chemical monitoring program is not missing ecologically significant peak concentrations, un-measured chemicals nor unexpected effects of mixtures of chemicals. If the relationship is confirmed in the 2011/13 wet-season, it would strengthen the case that the developed SPEARherbicides index can detect the effects of herbicides on diatom communities.
The relative abundance of pesticide sensitive diatom species appeared to recover during the dry-season, so that SPEARherbicides index prior to the 2011/12 and 2012/13 wet-seasons was not related to TEQso concentrations.
The values of the macroinvertebrate-based SPEARpesticides index suggested that these chemicals are generally no major problem to invertebrates at the sites sampled in 2011/12. Furthermore the values of the SPEARpesticides index suggest that there are no ecologically relevant (for invertebrates) effects that have been missed by the pesticide monitoring, mixtures of multiple chemicals or un-measured chemicals.
The project has successfully modified the SPEARpesticides index and developed the new SPEARherbicides index. They appear to be promising tools to complement chemical monitoring of pesticides in the catchments of the GBR. We recommend that they are used in conjunction with the chemical monitoring program before and after subsequent wet-seasons. Some further research is also recommended involving mesocosm experiments to be able to prove causality between the indexes and pesticides exposure.
The cost of biomonitoring in conjunction with chemical monitoring of pesticides is extremely cost effective. The collection and processing of diatom and macroinvertebrates samples is approximately $6,000 per site per year in comparison to some $85,000 per site per year for chemical monitoring of pesticides (note neither of these estimates include data analysis and reporting costs). So at relatively modest costs additional costs, biomonitoring can add considerable new information as the ecological effects of the pesticides and increased confidence that all ecologically relevant compounds and their peak concentrations are being captured by the chemical monitoring program. Additionally biomonitoring could be used to screen a large number of sites for indications of pesticide toxicity, where chemical monitoring does not occur.