Effects of priority herbicides and their breakdown products on tropical, estuarine microalgae of the Great Barrier Reef Lagoon

Herbicide contamination is ubiquitous in rivers and coastal waters in Queensland, Australia. Globally, most ecotoxicological studies to date investigating the toxicity of herbicides have focused on temperate test-organisms and test-conditions, and there is a paucity of information on herbicide toxicity in tropical ecosystems. Despite the established ecological importance of benthic microflora in estuarine habitats, research investigating the effects of herbicides on these communities is particularly under-represented in the tropics. Due to a documented variability in organismal and chemical responses to environmental parameters, the validity of extrapolating from effect-concentrations derived using temperate organisms to tropical environments is debatable. Recent detections of herbicides in the Great Barrier Reef (GBR) lagoon underscore the need to advance the understanding of their toxicity to local non-target species. The research presented in this thesis therefore explores the extent and effects of herbicide pollution on locally isolated tropical estuarine microphytobenthos (MPB), using the photosystem II (PSII)-inhibitors diuron and tebuthiuron (urea-derivatives), atrazine and simazine (s-triazines), and hexazinone (triazinone), and the amino acid synthesis-inhibitor imazapic (imidazolinone) as model compounds.\ud \ud The composition and structure of previously un-described diatom (Ochrophyta) communities in six creeks and estuaries in the Wet and Dry Tropics of the GBR lagoon is presented, along with a review of the ecological importance of the MPB. Nearly 300 diatom taxa were observed in these coastal samples, with average abundances of 1.01 x 106 and 1.27 x 106 cells mL-1 sediment in the Wet and Dry Tropics, respectively. Through analysis of similarity (ANOSIM) based on cell counts of digested material, it was established that estuarine benthic diatom-communities in the GBR are comparable across climate zones (coastal Wet and Dry Tropics).\ud \ud Pulse amplitude modulation (PAM) fluorometry is ideally suited to measure the sublethal impacts of PSII-inhibitors on microalgae, but key relationships between effective quantum yield [Y(II)] measured by PAM fluorometry and the traditional endpoints growth rate (μ) and biomass increase has not previously been established. The effects of the herbicides diuron, hexazinone, and atrazine on Y(II), μ, and biomass increase were therefore examined during standardized three-day growth inhibition tests coupled with PAM fluorometry using the locally isolated benthic microalgae Navicula sp. (Ochrophyta) and Nephroselmis pyriformis (Chlorophyta). The relationships between Y(II), μ and biomass increase were consistent (r2 ≥ 0.90) and linear (1:1), validating the utility of PAM fluorometry as a rapid and reliable technique to measure sub-lethal toxicity of PSII-inhibitors in these microalgae. The order of toxicity (EC50-range) was: diuron (16 – 33 nM) > hexazinone (25 – 110 nM) > atrazine (130 – 620 nm) for both algal species.\ud \ud To provide significant information on the sensitivity of local tropical microalgae to herbicides under environmentally relevant conditions, the acute effects of diuron, tebuthiuron, atrazine, simazine, and hexazinone, along with binary mixtures and herbicide breakdown products (desethyl-atrazine [DEA] and 3,4-dichloroaniline [3,4-DCA]) were investigated using a Maxi-Imaging PAM bioassay (I-PAM) measuring the previously validated endpoint Y(II). The sensitivities of the tropical N. pyriformis, Navicula sp. and Cylindrotheca closterium (Ochrophyta), were compared with the temperate standard testorganism Phaeodactylum tricornutum (Ochrophyta). The order of toxicity (IC50-range) was diuron (8.8 – 18.3 nM) > hexazinone (9.5 – 27.2 nM) > tebuthiuron (52 – 412 nM) > atrazine (66 – 356 nM) > simazine (121 – 1200 nM) > DEA (1150 – 5640 nM) > 3,4-DCA (no effect) for all species. Nephroselmis pyriformis was always the most sensitive species, whereas the responses of the tropical and temperate diatoms tested here were similar to each other. All binary mixtures exhibited additive toxicity.\ud \ud To estimate local MPB exposure to herbicides, sediment and porewaters from the Herbert, Johnstone, Tully, and Daintree Rivers were analysed for these contaminants. Diuron was detected in samples from all sites at concentrations ranging between 0.01 – 0.23 nM in porewater and 1.3 – 42.9 nmol kg-1 dry weight (DW) in sediments, whereas atrazine, simazine, hexazinone, and tebuthiuron were detected less frequently. Porewater extracts were tested for acute phytotoxicity using the I-PAM bioassay and the results were compared with chemical analyses of the same water samples. Porewaters elicited up to 3.6% inhibition of Y(II) in N. pyriformis. Agreement between photoinhibition measured with the bioassay and analytically determined concentrations of PSII-inhibitors in the porewaters was excellent; however, measured inhibition was slightly higher than expected based on the analytical results, indicating the presence of unidentified phytotoxins.\ud \ud Acute, single-species tests may not adequately assess the effects of contaminants on intact, multi-species communities. MPB biofilms established in laboratory microcosms were therefore exposed to a concentration-series of diuron for four weeks plus two weeks recovery. Effects on Y(II) was measured using multiwavelength-fluorescence (Phyto-PAM), while community structure was measured by cell counts for single-celled organisms, percent area coverage for filamentous organisms, and whole-community pigment profiles using high performance liquid chromatography (HPLC) during weekly sampling. This combination of techniques enabled the first identification of pollution-induced community tolerance (PICT) in tropical estuarine biofilms: an increased tolerance to diuron coincided with changes in community structure. PICT was quantified in acute exposures to diuron following the 4- week exposures as an increase in IC50 (inhibition of Y(II), I-PAM bioassay) from 39 ± 5 nM in controls, to 82 ± 9 nM and 155 ± 20 nM in the 28 nM and 57.6 nM treatments, respectively. A threshold concentration for development of PICT in these biofilms could thus be determined to be below 28 nM after four weeks. Although the photosynthetic activity of a herbicide-exposed community eventually recovers as herbicide concentrations decrease, structural changes were shown to require in excess of two weeks for full recovery.\ud \ud Further research to determine whether PICT is already a characteristic of microphytobenthos in the GBR lagoon through exposure to environmental concentrations of herbicides is recommended. Mechanistic explanations for development of PICT should be researched through measurements of gene-regulation and heterotrophic nutrient-utilization. Additional chronic-exposure studies using herbicide mixtures and pulsed dosing are needed for tropical species. Overall, the tropical organisms tested here were more sensitive to atrazine compared to their temperate counterparts as reported in the available literature, whereas diuron was similarly toxic regardless of climatic origin of the test-organism. It is therefore recommended that extrapolations from temperate effect-concentrations to tropical species are implemented with caution