Benthic diatoms as indicators of herbicide toxicity in rivers – A new SPEcies At Risk (SPEARherbicides) index

Ecological Indicators 99 (2019) 203–213

Contents lists available at ScienceDirect

Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind

Benthic diatoms as indicators of herbicide toxicity in rivers – A new SPEcies At Risk (SPEARherbicides) index

T

⁎

Rebecca J. Wooda, , Simon M. Mitrovica, Richard P. Lima, Michael St. J. Warneb,c,d, Jason Dunlope, Ben J. Keffordf a Freshwater and Estuarine Research Group, Ecosystem Security Team, School of Life Sciences, University of Technology Sydney, PO Box 123, Broadway, NSW 2000, Australia b School of Earth and Environmental Sciences, University of Queensland, St Lucia, Queensland 4072, Australia c Water Quality and Investigations, Environmental Monitoring and Assessment, Department of Environment and Science, Dutton Park, Queensland 4102, Australia d Centre for Agroecology, Water and Resilience, Coventry University, West Midlands, United Kingdom e Water Assessment and Systems, Environmental Monitoring and Assessment, Department of Environment and Science, Dutton Park, Queensland 4102, Australia f Institute for Applied Ecology, Faculty of Science and Technology, University of Canberra, Canberra ACT 2601, Australia

A R T I C LE I N FO

A B S T R A C T

Keywords: Traits Biomonitoring Herbicide tolerance Diatoms Field effects

Benthic diatom communities are used widely as indicators of river health due to their rapid response to changes in water quality. The ability for diatom-based indices to detect eutrophication has been well documented; however, an index designed specifically to detect herbicide impacts is yet to be established. This is required as herbicide contamination of rivers is common in agricultural regions and poses a potential threat to aquatic ecosystems. This study developed a new biomonitoring index (SPEARherbicides) using benthic diatom communities to detect the toxic impacts of herbicides in rivers, and tested it across 14 rivers in the Great Barrier Reef catchment area, Australia. The new index uses diatom species traits to classify diatoms as either sensitive or tolerant to herbicides and calculates the fraction of sensitive taxa within a sample. The SPEARherbicides index showed a decline in herbicide sensitive diatoms with increasing herbicide toxicity of the sites. The impacts of herbicide toxicity on the diatom community were only apparent after the wet season when aqueous herbicide concentrations typically peak and diatoms were able to recover during the dry season when herbicide concentrations were lower. SPEARherbicides values had a negative relationship with the percentage of grazing and cropping in catchments but had a positive relationship with the percentage of conservation in catchments. SPEARherbicides also had a negative relationship with co-occurring potential stressors such as nutrients and total suspended solids.

1. Introduction Benthic diatoms are important biological components of freshwater ecosystems and can be used to assess the ecological health of rivers (Van Dam et al., 1994; Kelly et al., 1998). The ability to use changes in benthic diatom communities as indicators of declining water quality and anthropogenic impacts has been well established for eutrophication (Bellinger et al., 2006), urbanisation (Newall and Walsh, 2005) and inorganic pollution (Dela-Cruz et al., 2006). These indices utilise the differing sensitivities of the diatom taxa (usually to nutrients) to indicate trophic impacts in rivers (Rimet, 2012). However, previous studies have found that diatom indices of nutrient pollution in rivers were not suitable for detecting the impacts of herbicide toxicity (Morin et al., 2009; Blanco and Bécares, 2010). Therefore, there is a need for a ⁎

biomonitoring index that is designed to detect the impacts of herbicides in rivers (Larras et al., 2017). Herbicide pollution of rivers in agricultural regions is an issue of concern worldwide (Graymore et al., 2001; Schäfer et al., 2007; Dorigo et al., 2010; Roubeix et al., 2010; Dalton et al., 2015). Herbicides can have toxic impacts on aquatic phototrophic organisms; with effects on photosynthesis and growth in benthic diatoms (Debenest et al., 2009; Tlili et al., 2011). Exposure of benthic diatom communities to herbicides can result in the loss of sensitive species, thus altering species composition (Schmitt-Jansen and Altenburger, 2005; Magnusson et al., 2012). Although most agricultural herbicides are not designed to affect invertebrates, fish and other aquatic animals, they may be indirectly influenced as a result of impacts on primary producers (Rohr and Crumrine, 2005). Benthic diatoms play a vital role as primary producers

Corresponding author. E-mail address: [email protected] (R.J. Wood).

https://doi.org/10.1016/j.ecolind.2018.12.035 Received 5 October 2018; Received in revised form 14 December 2018; Accepted 16 December 2018 1470-160X/ © 2018 Elsevier Ltd. All rights reserved.

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

hereafter referred to as the dry and wet season sampling, respectively.

and are ubiquitous in shallow streams and rivers (Rimet and Bouchez, 2011). Diatoms vary in their sensitivity to herbicides (Wood et al., 2014, 2016a,b, 2017) and are highly responsive to changes in environmental conditions (Lange et al., 2011). For these reasons benthic diatoms have the potential to be used as biomonitors of herbicide pollution in rivers. Traits-based indices can be used to assess the ecological status of rivers. The abundance of particular traits in a community can be expected to shift according to prevailing environmental conditions. A traits-based approach, similar to the SPEcies At Risk (SPEAR) index (Liess and Ohe, 2005), has the potential to be utilised to separate the effects of herbicides on diatom communities from other stressors. SPEAR has been successfully used with stream macroinvertebrates affected by stressors such as salinity (Schäfer et al., 2011a) and pesticides (Schäfer et al., 2011b; Beketov et al., 2013). SPEAR indicates the proportion of sensitive taxa in a community based on traits that make them at risk from the stressor of interest, such as their physiological sensitivity to herbicides as determined by toxicity tests. Changes in the proportion of sensitive taxa (SPEAR taxa) in the community can then be linked to the impacts of that stressor in the field. The SPEAR approach has the potential to be adapted to utilise diatoms to detect herbicide impacts. The first objective of this study was to develop a new diatom biomonitoring index (SPEARherbicides) that can assess herbicide impacts in rivers. The new traits-based index classifies the diatom species as either SPEcies At Risk (SPEAR) from herbicide toxicity or not at risk (notSPEAR). The second objective was to determine if the diatom based SPEARherbicides index responds similarly to the macroinvertebrate based SPEARpesticides. The final objective was to test the new SPEARherbicides index with field data collected from 14 river sites that flow into the Great Barrier Reef (GBR) in Australia. Diatom and macroinvertebrate communities were sampled from river reaches that are impacted by varying intensities of agricultural activity, from upstream catchments devoted to conservation with no agriculture or grazing to high intensity crop areas. The relationship of the new SPEARherbicides index to pollution gradients and catchment land uses was assessed during both the dry and wet seasons for two successive years 2011–2013.

2.2. Diatom and macroinvertebrate sampling, preservation and identification At each study site and sampling occasion natural diatom and macroinvertebrate communities were collected concurrently. Macroinvertebrates were sampled within a 20 m length edge habitat of the river using a 250um mesh size aquatic net. One composite sample was collected per site on each sampling occasion, the macroinvertebrate samples were then live-sorted in the field and preserved in 70% ethanol for later identification to family level. Full collection protocols are detailed in Queensland Department of Natural Resources and Mines (Queensland Department of Natural Resources and Mines, 2001). Various submerged substrates from the edge habitat within a 20 m length of river were sampled for diatoms. Submerged pebbles and cobbles were the preferred substrates. If rocky substrates were unavailable then leaves and other submerged objects such as branches were used. The attached benthic diatoms were removed by scrubbing and scraping the surfaces of collected substrates (e.g., rocks and plant material) with a toothbrush and a sharp knife, respectively. The detached benthic material were combined into one composite sample per site, per sampling occasion and preserved in 70% ethanol in a 20 mL vial (Chessman et al., 1999). The preserved diatom samples were stored for later identification to species level where possible by Dr. Jennie Fluin from the University of Adelaide. Diatoms were cleaned and mounted on permanent slides for identification using an Olympus BH-2 (Olympus) light microscope at 1000x magnification. Ten transects of the mounted slide were counted and up to 600 diatom cells were identified per sample. 2.3. Environmental data Water temperature, dissolved oxygen (DO), electrical conductivity (EC) and pH were recorded in situ during diatom and macroinvertebrate collection (Supplementary Table S2). Herbicide, nutrient, discharge and total suspended solids (TSS) data for the 10 monitored sites (Supplementary Table S2) were provided by the Great Barrier Reef Catchment Loads Monitoring Program (GBRCLMP) (Wallace et al., 2014; Wallace et al., 2015) as part of water quality monitoring conducted for the Reef Water Quality Protection Plan 2013 (Australian Government and Queensland Government, 2013). Discharge data were available for two of the reference sites (Bluewater Creek and Finch Hatton Creek), at locations less than 12 km downstream of the diatom collection sites. For the other two reference sites, discharge was estimated from the averages of the Bluewater and Finch Hatton creeks. Herbicide and nutrient data were not available for the four reference sites. Herbicide concentrations were assumed to be negligible at all reference sites based on the absence of agricultural land uses upstream of these sites and prior sampling had not detect herbicides at these sites; Bulgun Creek (Lewis et al., 2009), Finch Hatton Creek (Mitchell et al., 2005; Lewis et al., 2009). For the same reasons nutrient concentrations at sampling locations have been observed to be low and were considered negligible – including Bulgan Creek (Faithful et al., 2007); Bluewater Creek (Lewis et al., 2008); Russell River and Finch Hatton Creek (Mitchell et al., 2005). Therefore, missing herbicide and nutrient data for these reference sites were assumed to be half the minimum values reported at the monitored sites as an attempt to estimate background (natural) nutrient concentrations. Water quality samples for the GBRCLMP were collected using both manual grab sampling techniques and automated samplers (e.g., Wallace et al., 2014; Wallace et al., 2015). Sampling occurred every few hours to daily intervals during high flow events and at a reduced frequency (usually monthly) during low or base-flow conditions (including the dry season). Nutrient concentrations were analysed using Flow

2. Methods 2.1. Study sites and study design Benthic diatom communities were collected from 14 sites within the Great Barrier Reef (GBR) catchment area (Fig. 1); see Supplementary Table S1 for site locations. These rivers are located across coastal catchment areas in freshwater reaches (above tidal influence) that drain directly into the GBR Marine Park. Ten sites, marked by red circles in Fig. 1, are considered contaminated by herbicides to various degrees and were part of the Reef Water Quality Protection Plan 2013 (Reef Plan, Australian and Queensland Government, 2013) and are now part of the Reef 2050 Water Quality Improvement Plan (Australian Government and Queensland Government, 2018) which replaced the Reef Plan in 2018. The remaining four sites, marked by green triangles in Fig. 1, have no agricultural, industrial, mining or urban areas present upstream, but have nature conservation and recreational activities and are thus extremely unlikely to have any significant herbicide, pesticide or other types of contamination. They are nominated as reference sites. The climate of the study region is characterised by the summer (monsoonal) wet season, typically December – March where most of the annual rainfall occurs, which is often intense leading to elevated discharges in the region’s rivers (Waterhouse et al., 2012). The wet season coincides with peak concentrations of herbicides in most of these rivers (Lewis et al., 2009). Our sampling regime targeted two successive wet seasons (2011/2012 and 2012/2013). Sites were sampled at the end of the dry season (i.e., November 2011 and September 2012) and immediately after these wet seasons (i.e., May 2012 and April 2013), 204

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

Fig. 1. Map of the Great Barrier Reef catchment area and study sites. Green triangles are the reference sites and red circles are the monitored sites. Natural resource management (NRM) regions are marked in different colours. Source modified from Reef Water Quality Protection Plan (Reef Plan, 2011).

dates were used (this occurred in four of the 49 sampling occasions). Catchment land use data were also available for the monitored sites (Supplementary Table S3). The area of land used for bananas, conservation, cropping, forestry, horticulture, grazing, and sugar cane in each of the monitored catchments was obtained from the Queensland Government Queensland Spatial Catalogue (DSITI, 2016). This land use data were generated by the Queensland Land Use Monitoring Program, which is part of the Australian Collaborative Land Use and Management Program (http://www.agriculture.gov.au/abares/aclump/aboutaculmp). The land use for the reference sites was estimated as 100% conservation.

Injection Analysis (colourimetric techniques) at the Science Division Chemistry Centre (Dutton Park, Queensland), a National Association of Testing Authorities, Australia (NATA) accredited laboratory. Herbicide concentrations in the grab samples were analysed using solid phase extraction followed by liquid chromatography-mass spectrometry (LCMS) at Queensland Health Forensic and Scientific Services (Coopers Plains, Queensland), also a NATA accredited laboratory. Samples were stored and transported in accordance with the Environmental Protection (Water) Policy Monitoring and Sampling Manual (DERM, 2009). In order to summarize the environmental data collected at each site, the mean concentrations of nutrients (oxidised nitrogen (NOx), ammonia, filterable reactive phosphorus (FRP), discharge and TSS were calculated over the 60 days prior to diatom sampling using the GBRCLMP data. In cases where there were no herbicide or nutrient data available in the 60 days prior to sampling due to low river discharge, the site average 60 day mean values across the three other sampling

2.4. Classification of SPEAR taxa A total of 289 benthic diatom taxa was identified in the samples collected from the 14 rivers (Supplementary Table S4). Each species 205

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

have been shown to be additive (Magnusson et al., 2010). In order to estimate the effects of herbicide mixtures in a given sample the toxic equivalency quotient (TEQ) was calculated. The concentrations of PSII herbicides in a sample can be added together after each component has a toxic equivalency factor (TEF) applied, using the equation derived by Safe (1998):

was classified as either SPEAR or not SPEAR based on two specific traits; 1) sensitivity to herbicides and 2) motility. The reasons for including diatom motility into the index are that mobile taxa are able to adjust their position within the biofilm and this may allow them to avoid herbicide exposure by using the biofilm for protection, and also to seek out more favourable conditions during exposure events, making them more resilient to herbicide exposure (Rimet and Bouchez, 2011). The motility trait refers to any diatom belonging to the motile guild as defined by Rimet and Bouchez (2012). For any given taxon to be classified as SPEAR it would need to be both sensitive to herbicides and non-motile, otherwise it was classified as notSPEAR. There was only one taxon for which the sensitivity classification was changed based on its motility – Navicula gregaria. All other species sensitive to herbicides were also non-motile (see Supplementary Table S4), so the motility trait had minimal effect on the overall classification. For 57 taxa, herbicide sensitivity data were available for the taxon in question (Supplementary Table S5). The sensitivities of the remaining taxa were extrapolated from the sensitivity data of related taxa in the database (Supplementary Table S5). Extrapolation was initially conducted at the Genus level, and then at the Order level. If there were several species within a Genus with differing sensitivity classifications then the dominant classification was applied to other species within that Genus. Conflicting classifications at the Order level did not occur. Where there was no traits data available at the Order level, that taxon was excluded from the index (this applied to 2% of taxa, which collectively accounted for 0.4% of individuals observed). The complete taxon list with species traits and SPEAR classification is given in Supplementary Table S4.

TEQ =

where Ci = the concentration of the individual herbicides and TEFi = the toxic equivalency factor of the individual herbicides. The TEF for each herbicide was derived from Ma et al. (2006) for the freshwater microalgal species, Scenedesmus obliquus (TEQSO). The TEQso values (toxic equivalent quotient for S. obliquus) were calculated for each site following the method in Smith et al. (2012), using the GBRCLMP herbicide concentration data for ametryn, atrazine, diuron, simazine and prometryn. The 95th percentile was then calculated for TEQSO, over a 60 day period preceding each of the diatom sampling dates (TEQSO 60 days). The 60 day period was chosen as an estimate of the herbicide toxicity of each site immediately prior to sampling. This duration was intended to be sufficiently long to capture the potential effects of both acute and chronic herbicide toxicities to the benthic diatom community (Wood et al., 2017). 2.8. Statistical analysis The data set was divided into two groups - dry season (i.e., the November 2011 and September 2012 samples) and wet season (i.e., the May 2012 and April 2013 samples). For each data set (Dry and Wet) linear regression analysis was conducted to model the relationship between SPEARherbicides and TEQSO (60 days). Linear regression was also performed to assess the relationship of SPEARherbicides to environmental variables and catchment land use categories. Analysis of covariance (ANCOVA) was performed on all regression equations to test the significance of the regression slope, as well as to determine whether this was consistent between the two sampling years (alpha = 0.05). Principal Component Analysis (PCA) was used to explore the relationship of sites to the measured environmental gradients. PCA was performed on the combined data sets (Dry and Wet data) using all available predictors; DO (in situ), pH (in situ), EC (in situ), Temperature (in situ), TEQSO (60 days), Discharge (60 days), TSS (60 days), Ammonia (60 days), NOx (60 days) and FRP (60 days). Canonical correspondence analysis (CCA) was performed to investigate the distribution of the benthic diatom taxa along gradients of selected environmental variables. All environmental variables were individually checked for normality and homoscedasticity and were log transformed where appropriate prior to analysis. Variables with high variance inflation factors (VIF > 10) were excluded from the CCA to reduce collinearity. The diatom community data included taxa with > 5% relative abundance and occurred in at least two samples. Diatom relative abundance data were square root transformed prior to analysis to down-weight the common taxa. PCA, CCA, Regression analysis and ANCOVA were performed using the statistical software R, using packages - stats (R Core and Team, 2017) and vegan (Oksanen et al., 2017).

2.5. Calculation of SPEARherbicides index The SPEARherbicides index is a measure of the relative abundance of sensitive taxa in the benthic diatom community. The SPEARherbicides index is calculated as per the invertebrate SPEAR index described in Schäfer et al. (2011a): n

SPEARherbicides =

∑i = 1 log(xi+1) y n

∑i = 1 log(xi+1)

where n is the number of diatom taxa in a sample, xi is the abundance of taxon i and y is 1 if the taxon is classified as a SPEcies At Risk (SPEAR), otherwise y is 0. The species traits used for SPEAR classification in the SPEARherbicides index are ‘motility’ shown in Supplementary Table S4 and ‘sensitivity classification’ shown in Supplementary Table S5. 2.6. Calculation of SPEARpesticides index The SPEARpesticides index is a measure of the relative abundance of sensitive taxa in the macroinvertebrate community. The SPEARpesticides index is calculated as per the invertebrate SPEAR index described in Schäfer et al. (2011a): n

SPEARpesticides =

∑ Ci × TEFi

∑i = 1 log(xi+1) y n

∑i = 1 log(xi+1)

where n is the number of taxa observed in a sample, xi is the abundance of taxon i and y is 1 if taxon i is classified as SPEAR, otherwise 0. The traits used in the SPEARpesticides index are physiological sensitivity called ‘Sorg’ (Von der Ohe and Liess, 2004), reproduction capacity; which is based on both ‘number of generations per year’ and ‘time until reproduction’, and the traits ‘dispersal capacity’ and ‘out of water assessment’ as per Liess and Ohe (2005) (Supplementary Table S6).

3. Results 3.1. Spearherbicides index A new trait-based SPEARherbicides index was developed utilising diatom communities collected from rivers of the GBR catchment area (Supplementary Table S4). The SPEARherbicides values calculated at the 14 rivers ranged from 2 to 68%, with a mean value of 23% (Supplementary Table S7). The SPEARherbicides values were higher at the reference sites (ANOVA p = 0.014) with mean values of 33% sensitive taxa, compared to a mean of 20% at the other sites. The lowest

2.7. Calculating the toxicity of mixtures of Photosystem II inhibiting (PSII) herbicides The effects of herbicides with similar modes of action in a mixture 206

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

Fig. 2. Proportion of sensitive diatoms in the benthic community as characterised by SPEARherbicides (%) against calculated mixture toxicity expressed as log TEQSO 95th percentiles calculated over 60 days prior to sampling the dry and wet season rainfall events. Circles = samples relevant for the 2011/12 wet season, triangles = 2012/13 wet season. Solid line is the regression line, the dashed lines are the 95% confidence intervals.

that Barratta Creek is located in an irrigated region. Irrigation during the dry season can lead to off-site transport of pesticides to this creek without much dilution as might occur with rainfall over the entire or large parts of the catchment (O’Brien et al., 2016). The highest herbicide toxicity during the wet season occurred at Sandy Creek (43 μg atrazine equivalent L−1). Despite being an outlier, we have included Barratta Creek in all analyses; however, we note that the results were altered slightly by its inclusion. With the inclusion of this site there was no significant correlation between SPEARherbicides and TEQSO when both wet and dry data sets were combined, while with its exclusion there was a significant correlation (p = 0.011, Table S8). Regardless, there was still a significant relationship between herbicide toxicity (TEQSO 95th percentile) and SPEARherbicides during the wet season with Barratta Creek excluded (p = 0.049, Table S8), although this relationship was clearly stronger with its inclusion (p = 0.019, Table 1). There was no significant relationship during the dry season either with the inclusion (p = 0.633) or exclusion (p = 0.118, Table S8 and Supplementary Fig. S1) of Barratta Creek. The SPEARherbicides index showed a significant correlation with conservation land use (Table S9), as the proportion of conservation land use in the catchment area increased so did the SPEARherbicides value (Fig. S2). Sites with higher percentages of grazing and cropping areas in their catchments showed a decline in the SPEARherbicides values (Fig. S2) and these relationships were statistically significant (Table S9). The land use categories; bananas, horticulture, forestry and sugarcane did not show a significant relationship with the SPEARherbicides index (Table S9, Fig. S2).

SPEARherbicides values occurred in the Burdekin River during the wet season with just 2% in 2012 and 5% in 2013. The highest SPEARherbicides values were found in the Herbert River samples during the dry season of 2012 (68%) and at Bluewater Creek during the 2012/ 13 wet season (54%). The SPEARherbicides value declined with increasing herbicide toxicity of the sites in the samples collected during the wet season rainfall (Fig. 2). SPEARherbicides was significantly negatively correlated with TEQ toxicity (60 day 95th percentile TEQSO) in the wet season (p = 0.019, Table 1). However, there was no significant relationship between SPEARherbicides and the TEQ of the sites in the dry season (p = 0.633, Table 1, Fig. 2) or when both dry and wet season data were combined (p = 0.058, Table 1). These trends were consistent over two successive sampling years (p > 0.05, Table 1). There was no correlation between the new diatom based SPEARherbicides and the macroinvertebrate based SPEARpesticides for the wet season samples (p = 0.103, Table 1), dry season samples (p = 0.978, Table 1) or with both data sets combined (p = 0.199, Table 1). Likewise there was no correlation between SPEARpesticides and the TEQ 95th percentile for the wet or dry season data sets or the combined data set (p > 0.05, Table 1). There was also no difference in SPEARpesticides values between the reference and monitored sites (ANOVA p = 0.297). The highest herbicide toxicity occurred at Barratta Creek in the dry season (209 μg atrazine equivalent L−1). However, this was not representative of the majority of our study sites, which are exposed to peak herbicide concentrations during the wet season (Comet Weir and Tully River were other exceptions). This is most likely due to the fact

Table 1 Results of ANCOVA for relationship of SPEARherbicides and SPEARpesticides to calculated mixture toxicity, expressed as log TEQSO 95th percentile, during the dry and wet seasons and with both data sets combined. The model also compares the sampling occasions (each visit is compared to the first visit). Bold type indicates statistical significance (p < 0.05). Season

Model R2

Model p value

F

Regression coefficient p values

SPEARherbicides vs. TEQSO Dry Wet Combined

−0.091 0.181 0.042

0.887 0.035 0.210

0.121 3.878 1.53

TEQSO p value 0.633 0.019 0.058

Visit p value 0.913 0.157 (0.148, 0.838, 0.733)

SPEARpesticides vs. TEQSO Dry Wet Combined

0.071 −0.041 −0.026

0.192 0.619 0.601

1.804 0.489 0.692

TEQSO p value 0.366 0.333 0.959

Visit p value 0.102 0.864 (0.719, 0.135, 0.668)

SPEARherbicides vs. SPEARpesticides Dry Wet Combined

−0.105 0.075 −0.005

0.997 0.151 0.453

0.003 2.05 0.935

SPEARpesticides p value 0.978 0.103 0.199

Visit p value 0.936 0.289 (0.211, 0.713, 0.759)

207

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

4. Discussion 4.1. Spearherbicides as an indicator of herbicide toxicity The new SPEARherbicides index was able to distinguish the difference between reference and impacted rivers within the GBR catchment, showing a decline in the abundance of sensitive diatom species at sites known to be affected by herbicide contamination and other agricultural activities. SPEARherbicides was significantly negatively correlated with herbicide toxicity (TEQSO) across the study sites across the two wet seasons studied. These results show that the relative abundance of herbicide sensitive diatom species declined with increasing herbicide toxicity of the sites. The decline in abundance of herbicide sensitive diatom species was only significant during the wet season rains and not during the dry seasons, when herbicide concentrations are generally lower (Davis et al., 2012). Other studies have also found that the diatom community has the ability to recover after exposure to herbicides (Dorigo et al., 2010; Proia et al., 2011) and it is well established that the effects of short-term exposure of algae to PSII herbicides are reversible (Vallotton et al., 2008a; Vallotton et al., 2009; Brain et al., 2012). However, exposure to other herbicides or to PSIIs for longer periods and in sunlight can lead to sustained or irreversible effects in photosynthetic organisms (Jones, 2005; Vallotton et al., 2008b). Taking SPEARherbicides measurements immediately after peak herbicide concentrations will therefore maximise the likelihood of detecting impacts. The SPEARherbicides index did not show a significant relationship to SPEARpesticides, thus both indices were responding to different environmental factors (i.e., to herbicides and pesticides, respectively). Therefore additional information can be gained from SPEARherbicides that is not obtained from SPEARpesticides. The fact that there was not a significant relationship between SPEARpesticides and TEQSO is not surprising as the TEQSO only considers the toxicity of PSII herbicides that exert their toxicity by inhibiting the photosystem II component of photosynthesis which the macroinvertebrates assessed in SPEARpesticides do not have. This does not necessarily indicate that changes in the diatom community were not affecting the macroinvertebrate community, only that changes in the diatom community were not affecting the macroinvertebrate index, SPEARpesticides. The SPEARherbicides index responded to changes in land use categories, showing that sites with the greatest percentages of grazing and cropping activities in the catchment area also had the lowest SPEARherbicides values. It was unexpected that there was no evidence of a relationship between the SPEARherbicides index and the sugarcane land use category, given that this industry is known to use herbicides in GBR catchments (Davis et al., 2013) and that waterways in sugarcane growing regions are the most toxic waterways, in terms of pesticides, that discharge to the GBR (Smith et al., 2012; Lewis et al., 2013; Waterhouse et al., 2017). In addition, statistically significant negative relationships have been found between higher percentage of land used for growing sugarcane and the severity of endocrine biomarkers in barramundi (Kroon et al., 2015) and a range of histopathological and transcriptomic effects in barramundi has been attributed to exposure to pesticides associated with sugarcane growing activities (Hook et al., 2017a,b; Hook et al., 2018).

Fig. 3. Principal Component Analysis of the physico-chemical variables of the sites during sampling at the monitored (circles) and reference (triangles) sites, during the dry (closed) and wet (open) seasons. Blue arrows represent the selected environmental vectors.

3.2. Environmental conditions and diatom communities at the study sites The reference sites were typified by higher DO and lower EC, TEQ, TSS and nutrients compared to that for the monitored sites Fig. 3). The monitored sites were spread across a gradient of discharge with wet season sites somewhat concentrated in the bottom left area of the plot and the dry season samples being more dispersed. PC axis 1 and 2 were negatively correlated with discharge, TSS and NOx; whereas, TEQ, EC, Ammonia and FRP were collinear and independent from discharge. The most significant environmental vectors distinguishing the sites were EC, TEQ, Ammonia, FRP, TSS, NOx and Discharge (p < 0.05). The CCA analysis (Fig. 4) shows the distribution of diatom species and the strength of selected environmental variables influencing diatom communities at the sites. Selected environmental variables are those identified as significant in the PCA (Fig. 3). FRP was excluded from the analysis due to multi-collinearity (VIF > 10). TEQ, TSS, EC and nutrients had a strong influence on diatom community composition (p < 0.05). The measured environmental variables explained 21% of the variance of species distributions (total inertia = 75). The CCA axes represent 54% and 27% of the constrained variance, respectively. The diatom species related to the strongest pollution gradient and sites with the greatest herbicide contamination were – Bacillaria paxillifer (BPAX), Cyclotella meneghiniana (CMEN), Diadesmis confervacea (DCOF), Gomphonema angustum var. subminutum (GASU), Luticola goeppertiana (LGOE), Melosira spp. (MELO), Navicula schroeterii (NSHR), Navicula viridula (NVIR), Nitzschia perminuta (NIPM), Stauroneis anceps (STAN) and Tabularia fasciculata (TFAS) (Fig. 4). Reference and monitored sites associated with the lowest gradient of pollution were typified by the following species – Achnanthidium minutissimum (AMIN), Encyonema gracilis (EGRS), Epithemia adnata (EADN), Fragilaria capucina var. capucina (FCAP), Fragilaria capucina var. rumpens (FCRU), Fragilaria tenera (FTEN), Gomphonema clevei (GCLE), Navicula radiosa (NRAD) and Rossithidium pusillum (RPUS). The SPEARherbicides index was negatively correlated with Ammonia (Fig. 5A), NOx (Fig. 5B), TSS (Fig. 5C), EC (Fig. 5D) and FRP (Fig. 5E) during the wet season; however, they were not correlated during the dry season (Table 2). There was no relationship (p > 0.05) between the SPEARherbicides index and discharge, pH or temperature in either the dry or wet seasons (Table 2). SPEARherbicides showed a significant positive correlation with DO in the wet season (Table 2, Fig. 5F). These results were consistent in both sampling years for all variables (p < 0.05).

4.2. Agricultural impacts and multiple stressor effects In our study region of the GBR catchment, agricultural land development has led to increased loads of suspended sediments, nutrients and herbicides in rivers (Kroon et al., 2012; Waterhouse et al., 2012). At our study sites the main factors emanating from agricultural activities that impacted the diatom community were herbicide toxicity (TEQSO), EC, TSS and nutrients (FRP, NOx, Ammonia) (Fig. 3). These findings are consistent with other studies that have found diatom community composition to be influenced by agricultural impacts including changes in nutrient concentrations (Sonneman et al., 2001; Stevenson et al., 208

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

Fig. 4. Constrained correspondence analysis (CCA) plots of a) diatom species and b) environmental vectors overlayed on sites. Diatom species names corresponding to the unique four letter diatom codes are listed in Supplementary Table S4.

also negatively correlated with EC, TSS and nutrients (FRP, Ammonia, NOx). These other environmental factors most likely contributed to some of the variability in the SPEARherbicides – TEQSO relationship seen in Fig. 2. It was therefore difficult to distinguish the effects of herbicide toxicity from other agricultural impacts at the sites. The effect of herbicides on the diatom community has the potential to be influenced by environmental factors such as nutrients, light and temperature (Bonnineau et al., 2012; Larras et al., 2013; Dalton et al., 2015), although the relative sensitivities of diatom taxa appear to be unchanged at different light levels (Wood et al., 2016b). Herbicide pollution in agricultural regions often co-occurs with other stressors such as nutrients, which also have a strong influence on diatom community composition (Guasch et al., 1998; Roubeix et al., 2010; Ponsatí et al., 2016; Larras et al., 2017). These pollutants are mobilised generally at the same time as herbicides i.e., after rain events, making diagnosis of the potentially toxic impacts of herbicide pollution alone on aquatic

2008). In contrast, a recent study of rivers in South East Queensland, Australia (Tan et al., 2017) found that diatom indices developed in Europe responded more strongly to pH than to nutrients. The range of pH values in the current study and Tan et al. (2017) is very similar being 6.2–8.7 and 5.4–8.1, respectively, and the ranges for phosphorus were also similar — FRP ranged from 0.001 to 0.13 mg L−1 (current study) and SRP ranged from 0.00 to 0.16 mg L−1 (Tan et al., 2017). One might therefore expect pH and nutrients to have similar effects in both studies of diatoms in Queensland rivers. Tan et al. (2017) suggested that pH had such a major influence due to the lack of a strong pollution gradient in the rivers they studied. Their hypothesis is supported by our findings where pH was not a strong influence on diatoms as the rivers monitored in the current study had a number of strong pollution gradients. Our field data showed that the new SPEARherbicides index was negatively correlated with herbicide toxicity (TEQSO); however, it was 209

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

Fig. 5. Proportion of sensitive diatoms in the benthic community as characterised by SPEARherbicides (%) against the logarithms of the environmental variables; a) Ammonia, b) NOx, c) TSS, d) EC, e) FRP and f) DO. Only wet season data shown. Circles = 2011/12 wet season, triangles = 2012/13 wet season. Solid line is the regression line, dashed lines are 95% confidence intervals.

optimal EC values for individual diatom species ranged from 40 to 902 μS cm−1 (Wilson et al., 1994). The measured EC values at our study sites ranged from 47 to 1400 μS cm−1, with the median value of 362 μS cm−1. A number of other studies has found EC to be a dominant covariate influencing diatom community structure (Blinn and Bailey, 2001; Sonneman et al., 2001; Chessman and Townsend, 2010). A study on the effects of salinity and the herbicide prometryn as co-stressors did not suggest co-tolerance (Rotter et al., 2013). As a single stressor on diatom community composition the effects of EC occur at levels higher than the ranges observed in the current study (2000–5000 μS cm−1) (Rotter et al., 2013; Cañedo-Argüelles et al., 2017). This would indicate that EC has a limited role in the observed differences in benthic diatom communities.

ecosystems more difficult (Morin et al., 2009; Debenest et al., 2010; Andrus et al., 2015). The SPEARherbicides index was most strongly negatively correlated with TEQ, TSS, EC and nutrients during the wet season, showing that the effects of these factors are dependent on the seasonal discharge regime. Concentrations of nutrients and EC in rivers tend to co-increase with increased herbicide loads due to agricultural land practices (Schäfer et al., 2011a). Salinity has an influence on the diatom species present in the diatom community, with individual diatom taxa showing affinities for particular ions in freshwater (Wilson et al., 1994; Potapova and Charles, 2003). Previous studies have shown that relative abundances of diatoms shift along conductivity gradients; for example, a study of diatom communities at 1109 sites across the US found that 210

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

was sensitive to changes in water quality over two successive wet seasons as it indicated changes in diatom composition following exposure to contaminants in the wet season and recovery in species composition during the dry season when contaminant exposure was minimal. The new SPEARherbicides index provided additional ecological information that was not indicated by the macroinvertebrate based SPEARpesticides. The percentage of land assigned to conservation had a positive effect on SPEARherbicides values, whilst grazing and cropping land uses were associated with lower SPEARherbicides values. Our results demonstrated that SPEARherbicides is capable of detecting agricultural impacts on benthic diatom communities, including the effects of herbicide toxicity and nutrient pollution. SPEARherbicides could be used to identify communities at risk of agricultural impacts and to assess changes in management practices on sensitive diatom taxa over time. Further research is required to elucidate the specific impact of herbicides on benthic diatom communities in a multi-stressor context. Investigations into the herbicide sensitivity traits of benthic diatom species and their interaction with nutrient exposure may provide useful data with which to incorporate into the SPEARherbicides index in future.

Table 2 Results of ANCOVA analysis to determine the relationship of SPEAR to environmental variables for dry and wet season data. – indicates that the parameter was incalculable. Variable Ammonia (mg L NOx (mg L TSS (mg L

−1

−1

−1

)

)

)

EC (μS cm−1 @ 25˚C) FRP (mg L−1) Discharge (m3 s−1) pH Temp (°C) DO (mg L−1)

Season

R2

F

p

Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet

−0.044 0.290 −0.083 0.152 0.089 0.215 0.097 0.367 0.095 0.424 −0.098 0.106 −0.103 0.057 −0.077 −0.004 −0.090 0.329

0.554 6.332 0.200 3.323 2.025 4.551 2.13 8.526 2.108 10.57 0.064 2.533 0.019 1.792 0.250 0.943 0.133 7.371

0.307 0.003 0.538 0.030 0.058 0.011 0.053 0.001 0.054 < 0.001 0.731 0.063 0.861 0.135 0.491 0.391 0.617 0.001

Acknowledgements This study was funded by the Australian and Queensland Governments through the Caring for Our Country Reef Rescue Water Quality Research & Development Program (project no. RRD058) and RJW received a scholarship from these funds. The authors would like to thank all members of the Great Barrier Reef Catchment Loads Monitoring Program for supplying the herbicide, total suspended solids, nutrient and land-use information for the monitored sites and associated catchments. Special thanks to Dr. Rachael Smith for her assistance with the herbicide mixture toxicity calculations and to Dr. Rajesh Prasad and Perceval Depresle for their efforts in the field, and Dr. Glenn McGregor and Dr. Satish Choy for their support of this project. Many thanks also to Dr. Mike Holmes, Dr. Suzanne Vardy and two anonymous referees for their helpful comments on the draft manuscript. We would like to acknowledge Dr. John Tibby and Dr. Jennie Fluin from the University of Adelaide for their contribution to diatom identifications.

4.3. Occurrence of diatom taxa and traits Diatoms strongly associated with the herbicide polluted sites were D. confervacea (DCON), L. goeppertiana (LGOE), N. perminuta (NIPM) and Stauroneis anceps (STAN), and are also considered to be tolerant to organic pollution and indicators of eutrophic conditions (Van Dam et al., 1994; Kelly and Whitton, 1995). B. paxillifer (BPAX) was another herbicide tolerant taxon strongly associated with the high EC sites that is known to tolerate brackish conditions and occurs exclusively at pH > 7 (Van Dam et al., 1994). T. fasciculata (TFAS) and G. angustum var. subminutum (GASU) were classified as sensitive based on the sensitivities of related taxa, yet were found at herbicide polluted sites, indicating these species are more tolerant than suggested by their phylogenetic relatives. C. meneghiniana was classified as sensitive based on its response to herbicides in laboratory based toxicity tests (Larras et al., 2012; Larras et al., 2014). However, in our field data this species was abundant at sites highly contaminated with herbicides, and has been noted as pollution tolerant (including nutrients) in other field studies (Morin et al., 2009; Duong et al., 2012). This suggests that this species is highly adaptable and that prior exposure to pollutants in the field may be inducing herbicide tolerance in a multi-stressor environment (Schmitt-Jansen and Altenburger, 2005; Wood et al., 2017). Temperature is an important environmental variable that affects both diatom growth and herbicide sensitivity (Larras et al., 2013). C. meneghiniana has a fast growth rate and is known to form blooms in temperate rivers in Australia (Mitrovic et al., 2008). Measured spot temperatures ranged from 16 °C at the densely forested reference site of Finch Hatton Gorge, to 31 °C at the highly impacted and open (cleared of vegetation) site on the Burdekin River. The optimal growth temperature for C. meneghiniana is 25 °C (Mitrovic et al., 2010), whereas a sub-optimal temperature (i.e., 21 °C) was used in the laboratory-based tests to assess herbicide sensitivity (Larras et al., 2012). If the herbicide sensitivity of this taxon is dependant on temperature, its sensitivity to herbicides may be reduced at temperatures closer to 25 °C due to its faster growth rate.

Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ecolind.2018.12.035. References Andrus, J.M., Winter, D., Scanlan, M., Sullivan, S., Bollman, W., Waggoner, J.B., Hosmer, A.J., Brain, R.A., 2015. Spatial and temporal variation of algal assemblages in six Midwest agricultural streams having varying levels of atrazine and other physicochemical attributes. Sci. Total Environ. 505, 65–89. Australian Government and Queensland Government, 2013. Reef Water Quality Protection Plan 2013. Securing the health and resilience of Great Barrier Reef World Heritage Area and adjacent catchments. Reef Water Quality Protection Plan Secretariat, the State of Queensland., Brisbane (QLD), Australia. Australian Government and Queensland Government, 2018. Reef 2050 Water Quality Improvement Plan: 2017–2022, Available from: https://www.reefplan.qld.gov.au/ about/assets/reef-2050-water-quality-improvement-plan-2017-22.pdf. Beketov, M.A., Kefford, B.J., Schäfer, R.B., Liess, M., 2013. Pesticides reduce regional biodiversity of stream invertebrates. In: Proceedings of the National Academy of Sciences. Bellinger, B.J., Cocquyt, C., O’Reilly, C.M., 2006. Benthic diatoms as indicators of eutrophication in tropical streams. Hydrobiologia 573, 75–87. Blanco, S., Bécares, E., 2010. Are biotic indices sensitive to river toxicants? A comparison of metrics based on diatoms and macro-invertebrates. Chemosphere 79, 18–25. Blinn, D.W., Bailey, P.C.E., 2001. Land-use influence on stream water quality and diatom communities in Victoria, Australia: a response to secondary salinization. Hydrobiologia 466, 231–244. Bonnineau, C., Sague, I., Urrea, G., Guasch, H., 2012. Light history modulates antioxidant and photosynthetic responses of biofilms to both natural (light) and chemical (herbicides) stressors. Ecotoxicology 21, 1208–1224. Brain, R.A., Arnie, J.R., Porch, J.R., Hosmer, A.J., 2012. Recovery of photosynthesis and growth rate in green, blue–green, and diatom algae after exposure to atrazine. Environ. Toxicol. Chem. 31, 2572–2581.

5. Conclusions SPEARherbicides is a new freshwater diatom based bioassessment tool for the management, protection and restoration of impacted rivers. There was a significant decline in the SPEARherbicides index at river sites contaminated with herbicides, TSS and nutrients compared with that at reference sites, within the GBR catchment. The SPEARherbicides index 211

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

Larras, F., Lambert, A.-S., Pesce, S., Rimet, F., Bouchez, A., Montuelle, B., 2013. The effect of temperature and a herbicide mixture on freshwater periphytic algae. Ecotoxicol. Environ. Saf. 98, 162–170. Larras, F., Keck, F., Montuelle, B., Rimet, F., Bouchez, A., 2014. Linking diatom sensitivity to herbicides to phylogeny: a step forward for biomonitoring? Environ. Sci. Technol. 48, 1921–1930. Larras, F., Coulaud, R., Gautreau, E., Billoir, E., Rosebery, J., Usseglio-Polatera, P., 2017. Assessing anthropogenic pressures on streams: a random forest approach based on benthic diatom communities. Sci. Total Environ. 586, 1101–1112. Lewis, S., Bainbridge, Z., Brodie, J., Butler, B., Maughan, M., 2008. Water Quality Monitoring of the Black Ross Basins: 2007/08 Wet Season, Report No. 08/04. Australian Centre for Tropical Freshwater Research, Queensland, Australia. Lewis, S.E., Brodie, J.E., Bainbridge, Z.T., Rohde, K.W., Davis, A.M., Masters, B.L., Maughan, M., Devlin, M.J., Mueller, J.F., Schaffelke, B., 2009. Herbicides: a new threat to the great barrier reef. Environ. Pollut. 157, 2470–2484. Lewis, S., Smith, R., O’Brien, D., Warne, M.St.J., Negri, A., Petus, C., da Silva, E., Zeh, D., Turner, R.D.R., Davis, A., Mueller, J., Brodie, J., 2013. 2013 Scientific Consensus Statement. Chapter 4: Assessing the risk of additive pesticide exposure in Great Barrier Reef ecosystems. In: Assessment of the relative risk of water quality to ecosystems of the Great Barrier Reef: Supporting Studies. A report to the Department of the Environment and Heritage Protection, Queensland Government, Brisbane. TropWATER Report 13/30, Townsville, Australia. 36p. Avalilable from: https:// research.jcu.edu.au/tropwater/publications/copy2_of_ 1330AssessmentofdegradedwatertoGBRecosystemsSupportingStudies.pdf/view. Liess, M., Ohe, P.C.V.D., 2005. Analyzing effects of pesticides on invertebrate communities in streams. Environ. Toxicol. Chem. 24, 954–965. Ma, J., Wang, S., Wang, P., Ma, L., Chen, X., Xu, R., 2006. Toxicity assessment of 40 herbicides to the green alga Raphidocelis subcapitata. Ecotoxicol. Environ. Saf. 63, 456–462. Magnusson, M., Heimann, K., Quayle, P., Negri, A.P., 2010. Additive toxicity of herbicide mixtures and comparative sensitivity of tropical benthic microalgae. Mar. Pollut. Bull. 60, 1978–1987. Magnusson, M., Heimann, K., Ridd, M., Negri, A.P., 2012. Chronic herbicide exposures affect the sensitivity and community structure of tropical benthic microalgae. Mar. Pollut. Bull. 65, 363–372. Mitchell, C., Brodie, J., White, I., 2005. Sediments, nutrients and pesticide residues in event flow conditions in streams of the Mackay Whitsunday Region, Australia. Mar. Pollut. Bull. 51, 23–36. Mitrovic, S.M., Chessman, B.C., Davie, A.W., Avery, E.L., Ryan, N., 2008. Development of blooms of Cyclotella meneghiniana and Nitzschia spp. (Bacillariophyceae) in a shallow river and estimation of effective suppression flows. Hydrobiologia 596, 173–185. Mitrovic, S.M., Hitchcock, J.N., Davie, A.W., Ryan, D.A., 2010. Growth responses of Cyclotella meneghiniana (Bacillariophyceae) to various temperatures. J. Plankton Res. 32, 1217–1221. Morin, S., Bottin, M., Mazzella, N., Macary, F., Delmas, F., Winterton, P., Coste, M., 2009. Linking diatom community structure to pesticide input as evaluated through a spatial contamination potential (Phytopixal): a case study in the Neste river system (SouthWest France). Aquat. Toxicol. 94, 28–39. Newall, P., Walsh, C.J., 2005. Response of epilithic diatom assemblages to urbanization influences. Hydrobiologia 532, 53–67. O’Brien, D., Lewis, S., Davis, A., Gallen, C., Smith, R., Turner, R., Warne, M., Turner, S., Caswell, S., Mueller, J.F., Brodie, J., 2016. Spatial and temporal variability in pesticide exposure downstream of a heavily irrigated cropping area: application of different monitoring techniques. J. Agric. Food. Chem. 64, 3975–3989. Oksanen, J., Guillaume, B.F., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P.R., O'Hara, R.B., Simpson, G.L., Solymos, P., Henry, M., Stevens, H., Szoecs, E., Wagner, H., 2017. vegan: Community Ecology Package. R package version 2.4-2. Ponsatí, L., Corcoll, N., Petrović, M., Picó, Y., Ginebreda, A., Tornés, E., Guasch, H., Barceló, D., Sabater, S., 2016. Multiple-stressor effects on river biofilms under different hydrological conditions. Freshw. Biol. 61, 2102–2115. Potapova, M., Charles, D.F., 2003. Distribution of benthic diatoms in U.S. rivers in relation to conductivity and ioic composition. Freshw. Biol. 48, 1311–1328. Proia, L., Morin, S., Peipoch, M., Romaní, A.M., Sabater, S., 2011. Resistance and recovery of river biofilms receiving short pulses of Triclosan and Diuron. Sci. Total Environ. 409, 3129–3137. Queensland Department of Natural Resources and Mines, 2001. Queensland Australian River Assessment System (AusRivAS) SAMPLING AND PROCESSING MANUAL, in: Queensland, T.S.o. (Ed.), Queensland, Australia. R Core Team, 2017. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Reef Plan, 2011. Reef Water Quality Protection Plan. Great Barrier Reef First Report Card 2009 Baseline. The State of Queensland, Reef Water Quality Protection Plan Secretariat. Rimet, F., 2012. Recent views on river pollution and diatoms. Hydrobiologia 683, 1–24. Rimet, F., Bouchez, A., 2011. Use of diatom life-forms and ecological guilds to assess pesticide contamination in rivers: lotic mesocosm approaches. Ecol. Ind. 11, 489–499. Rimet, F., Bouchez, A., 2012. Life-forms, cell-sizes and ecological guilds of diatoms in European rivers. Knowl. Managt. Aquatic Ecosyst. 1. Rohr, J.R., Crumrine, P.W., 2005. Effects of an herbicide and an insecticide on pond community structure and processes. Ecol. Appl. 15, 1135–1147. Rotter, S., Heilmeier, H., Altenburger, R., Schmitt-Jansen, M., 2013. Multiple stressors in periphyton - comparison of observed and predicted tolerance responses to high ionic loads and herbicide exposure. J. Appl. Ecol. 50, 1459–1468. Roubeix, V., Mazzella, N., Delmas, F., Coste, M., 2010. In situ evaluation of herbicide

Cañedo-Argüelles, M., Brucet, S., Carrasco, S., Flor-Arnau, N., Ordeix, M., Ponsá, S., Coring, E., 2017. Effects of potash mining on river ecosystems: an experimental study. Environ. Pollut. 224, 759–770. Chessman, B., Gell, P., Newall, P., Sonneman, J., 1999. Draft protocol for sampling and laboratory processing of diatoms for the monitoring and assessment of streams. An Illustrated Key to the Common Diatom Genera from Southern Australia. Murray Darling Freshwater Research Centre Identification Guide Series 26, An Illustrated Key to the Common Diatom Genera from Southern. Cooperative Research Centre for Freshwater Ecology, Thurgoona, Australia. Chessman, B.C., Townsend, S.A., 2010. Differing effects of catchment land use on water chemistry explain contrasting behaviour of a diatom index in tropical northern and temperate southern Australia. Ecol. Ind. 10, 620–626. Dalton, R.L., Boutin, C., Pick, F.R., 2015. Determining in situ periphyton community responses to nutrient and atrazine gradients via pigment analysis. Sci. Total Environ. 515, 70–82. Davis, A.M., Lewis, S.E., Bainbridge, Z.T., Glendenning, L., Turner, R.D.R., Brodie, J.E., 2012. Dynamics of herbicide transport and partitioning under event flow conditions in the lower Burdekin region, Australia. Mar. Pollut. Bull. 65, 182–193. Davis, A.M., Thorburn, P.J., Lewis, S.E., Bainbridge, Z.T., Attard, S.J., Milla, R., Brodie, J.E., 2013. Environmental impacts of irrigated sugarcane production: herbicide runoff dynamics from farms and associated drainage systems. Agric. Ecosyst. Environ. 180, 123–135. Debenest, T., Pinelli, E., Coste, M., Silvestre, J., Mazzella, N., Madigou, C., Delmas, F., 2009. Sensitivity of freshwater periphytic diatoms to agricultural herbicides. Aquat. Toxicol. 93, 11–17. Debenest, T., Silvestre, J., Coste, M., Pinelli, E., 2010. Effects of pesticides on freshwater diatoms. Rev. Environ. Contam. Toxicol. 203, 87–103. Dela-Cruz, J., Pritchard, T.I.M., Gordon, G., Ajani, P., 2006. The use of periphytic diatoms as a means of assessing impacts of point source inorganic nutrient pollution in southeastern Australia. Freshw. Biol. 51, 951–972. DERM, 2009. (Department of Environment and Resource Management). Monitoring and Sampling Manual 2009, Version 2, Brisbane, Australia. Dorigo, U., Berard, A., Rimet, F., Bouchez, A., Montuelle, B., 2010. In situ assessment of periphyton recovery in a river contaminated by pesticides. Aquat. Toxicol. 98, 396–406. DSITI (Department of Science, Information Technology and Innovation). 2016. Queensland Land Use and Management Program (QLUMP). Available from: http:// qldspatial.information.qld.gov.au/catalogue/custom/index.page. Downloaded: March 2016. Duong, T.T., Coste, M., Feurtet-Mazel, A., Dang, D.K., Ho, C.T., Le, T.P.Q., 2012. Responses and structural recovery of periphytic diatom communities after short-term disturbance in some rivers (Hanoi, Vietnam). J. Appl. Phycol. 1–13. Faithful, J., Liessmann, L., Brodie, J., Ledee, E., Sydes, D., Maughan, M., 2007. Water quality characteristics of water draining different land uses in the Tully/Murray Rivers Region (DECEMBER 2005 TO JUNE 2006), Report No. 06/25. Australian Centre for Tropical Freshwater Research, Queensland, Australia. Graymore, M., Stagnitti, F., Allinson, G., 2001. Impacts of atrazine in aquatic ecosystems. Environ. Int. 26, 483–495. Guasch, H., Ivorra, N., Lehmann, V., Paulsson, M., Real, M., Sabater, S., 1998. Community composition and sensitivity of periphyton to atrazine in flowing waters: the role of environmental factors. J. Appl. Phycol. 10, 203–213. Hook, S.E., Kroon, F.J., Greenfield, P.A., Warne, M., St, J., Smith, R.A., Turner, R.D.R., 2017b. Hepatic transcriptomic profiles from barramundi, Lates calcarifer, as a means of assessing organism health and identifying stressors in rivers in northern Queensland. J. Mar. Environ. Res. 129, 166–179. Hook, S.E., Kroon, F.J., Metcalfe, S., Greenfield, P.A., Moncuquet, P., McGrath, A., Smith, R.A., Warne, M., St, J., Turner, R.D., McKeown, A., Westcott, D.A., 2017a. Global transcriptomic profiling in barramundi Lates calcarcifer from rivers impacted by differing agricultural land-uses. Environ. Toxicol. Chem. 36, 103–112. Hook, S.E., Mondon, J., Revill, A.T., Greenfield, P.A., Smith, R.A., Turner, R.D.R., Corbett, P., Warne, M., St, J., 2018. Transcriptomic, lipid, and histological profiles suggest changes in health in fish from a pesticide hot spot. Mar. Environ. Res. 140, 299–321. https://doi.org/10.1016/j.marenvres.2018.06.020. Jones, R., 2005. The ecotoxicological effects of Photosystem II herbicides on corals. Mar. Pollut. Bull. 51, 495–506. Kelly, M., Cazaubon, A., Coring, E., Dell'Uomo, A., Ector, L., Goldsmith, B., Guasch, H., Hürlimann, J., Jarlman, A., Kawecka, B., Kwandrans, J., Laugaste, R., Lindstrøm, E.A., Leitao, M., Marvan, P., Padisák, J., Pipp, E., Prygiel, J., Rott, E., Sabater, S., van Dam, H., Vizinet, J., 1998. Recommendations for the routine sampling of diatoms for water quality assessments in Europe. J. Appl. Phycol. 10, 215–224. Kelly, M., Whitton, B., 1995. The Trophic Diatom Index: a new index for monitoring eutrophication in rivers. J. Appl. Phycol. 7, 433–444. Kroon, F.J., Kuhnert, P.M., Henderson, B.L., Wilkinson, S.N., Kinsey-Henderson, A., Abbott, B., Brodie, J.E., Turner, R.D., 2012. River loads of suspended solids, nitrogen, phosphorus and herbicides delivered to the Great Barrier Reef lagoon. Mar. Pollut. Bull. 65, 167–181. Kroon, F.J., Hook, S.E., Jones, D., Metcalfe, S., Henderson, B., Smith, R., Warne, M., St, J., Turner, R.D., McKeown, A., Westcott, D.A., 2015. Altered transcription levels of endocrine associated genes in two fisheries species collected from the Great Barrier Reef catchment and lagoon. Ocean Coast. Manag. 104, 51–61. Lange, K., Liess, A., Piggott, J.J., Townsend, C.R., Matthaei, C.D., 2011. Light, nutrients and grazing interact to determine stream diatom community composition and functional group structure. Freshw. Biol. 56, 264–278. Larras, F., Bouchez, A., Rimet, F., Montuelle, B., 2012. Using Bioassays and species sensitivity distributions to assess herbicide toxicity towards benthic diatoms. PLoS One 7, 1–9.

212

Ecological Indicators 99 (2019) 203–213

R.J. Wood et al.

exposure on Scenedesmus vacuolatus. Arch. Environ. Contam. Toxicol. 56, 442–449. Van Dam, H., Mertens, A., Sinkeldam, J., 1994. A coded checklist and ecological indicator values of freshwater diatoms from the Netherlands. Netherland J. Aquat. Ecology 28, 117–133. Von der Ohe, P.C., Liess, M., 2004. Relative sensitivity distribution of aquatic invertebrates to organic and metal compounds. Environ. Toxicol. Chem. 23, 150–156. Wallace, R., Huggins, R., Smith, R.A., Turner, R.D.R., Vardy, S., Warne, M.St.J., 2014. Total suspended solids, nutrient and pesticide loads (2011–2012) for rivers that discharge to the Great Barrier Reef – Great Barrier Reef Catchment Loads Monitoring Program 2011–2012. Department of Science, Information Technology and Innovation. Brisbane. Wallace, R., Huggins, R., Smith, R., Turner, R., Garzon-Garcia, A., Warne, M.St.J., 2015. Total suspended solids, nutrients and pesticide loads (2012–2013) for rivers that discharge to the Great Barrier Reef – Great Barrier Reef Catchment Loads Monitoring Program 2012–2013. In: Department of Science, I.T., Innovation and the Arts (Eds.), Brisbane, Queensland. Available from: http://www.reefplan.qld.gov.au/measuringsuccess/paddock-to-reef/assets/2012-2013-gbr-catchment-loads-report.pdf. Waterhouse, J., Brodie, J., Lewis, S., Mitchell, A., 2012. Quantifying the sources of pollutants in the Great Barrier Reef catchments and the relative risk to reef ecosystems. Mar. Pollut. Bull. 65, 394–406. Waterhouse J, Brodie J, Tracey D, Smith R, VanderGragt M, Collier C, Petus C, Baird M, Kroon F, Mann R, Sutcliffe T, Waters D and Adame F. 2017. Scientific Consensus Statement 2017: A synthesis of the science of land-based water quality impacts on the Great Barrier Reef, Chapter 3: The risk from anthropogenic pollutants to Great Barrier Reef coastal and marine ecosystems. State of Queensland, 2017. Wilson, S.E., Cumming, B.F., Smol, J.P., 1994. Diatom-salinity relationships in 11 lakes from the interior plateau of Brish Columbia, Canada: the development of diatombased models for paleosalinity reconstructions. J. Paleolimnol. 13, 197–221. Wood, R.J., Mitrovic, S.M., Kefford, B.J., 2014. Determining the relative sensitivity of benthic diatoms to atrazine using rapid toxicity testing: a novel method. Sci. Total Environ. 485–486, 421–427. Wood, R.J., Mitrovic, S.M., Lim, R.P., Kefford, B.J., 2016b. The influence of reduced light intensity on the response of benthic diatoms to herbicide exposure. Environ. Toxicol. Chem. 35, 2252–2260. Wood, R.J., Mitrovic, S.M., Lim, R.P., Kefford, B.J., 2016a. How benthic diatoms within natural communities respond to eight common herbicides with different modes of action. Sci. Total Environ. 557–558, 636–643. Wood, R.J., Mitrovic, S.M., Lim, R.P., Kefford, B.J., 2017. Chronic effects of atrazine exposure and recovery in freshwater benthic diatoms from two communities with different pollution histories. Aquat. Toxicol. 109, 200–208.

effects on the composition of river periphytic diatom communities in a region of intensive agriculture. Vie et milieu 60, 233–241. Safe, S.H., 1998. Hazard and risk assessment of chemical mixtures using the toxic equivalency factor approach. Environ. Health Perspect. 106, 1051. Schäfer, R.B., Caquet, T., Siimes, K., Mueller, R., Lagadic, L., Liess, M., 2007. Effects of pesticides on community structure and ecosystem functions in agricultural streams of three biogeographical regions in Europe. Sci. Total Environ. 382, 272–285. Schäfer, R.B., Kefford, B.J., Metzeling, L., Liess, M., Burgert, S., Marchant, R., Pettigrove, V., Goonan, P., Nugegoda, D., 2011a. A trait database of stream invertebrates for the ecological risk assessment of single and combined effects of salinity and pesticides in South-East Australia. Sci. Total Environ. 409, 2055–2063. Schäfer, R.B., Pettigrove, V., Rose, G., Allinson, G., Wightwick, A., von der Ohe, P.C., Shimeta, J., Kühne, R., Kefford, B.J., 2011b. Effects of pesticides monitored with three sampling methods in 24 sites on macroinvertebrates and microorganisms. Environ. Sci. Technol. 45, 1665–1672. Schmitt-Jansen, M., Altenburger, R., 2005. Predicting and observing responses of algal communities to photosystem II herbicide exposure using pollution-induced community tolerance and species sensitivity distributions. Environ. Toxicol. Chem. 24, 304–312. Smith, R., Middlebrook, R., Turner, R., Huggins, R., Vardy, S., Warne, M., St, J., 2012. Large-scale pesticide monitoring across Great Barrier Reef catchments – Paddock to Reef Integrated Monitoring, Modelling and Reporting Program. Mar. Pollut. Bull. 65, 117–127. Sonneman, J.A., Walsh, C.J., Breen, P.F., Sharpe, A.K., 2001. Effects of urbanization on streams of the Melbourne region, victoria, Australia II Benthic diatom communities. Freshw. Biol. 46, 553–565. Stevenson, R.J., Pan, Y., Manoylov, K.M., Parker, C.A., Larsen, D.P., Herlihy, A.T., 2008. Development of diatom indicators of ecological conditions for streams of the western US. J. North Am. Benthol. Soc. 27, 1000–1016. Tan, X., Zhang, Q., Burford, M.A., Sheldon, F., Bunn, S.E., 2017. Benthic diatom based indices for water quality assessment in two subtropical streams. Front. Microbiol. 8, 601. Tlili, A., Montuelle, B., Bérard, A., Bouchez, A., 2011. Impact of chronic and acute pesticide exposures on periphyton communities. Sci. Total Environ. 409, 2102–2113. Vallotton, N., Eggen, R.I.L., Escher, B.I., Krayenbühl, J., Chèvre, N., 2008a. Effect of pulse herbicidal exposure on Scenedesmus vacuolatus: a comparison of two photosystem II inhibitors. Environ. Toxicol. Chem. 27, 1399–1407. Vallotton, N., Moser, D., Eggen, R.I., Junghans, M., Chèvre, N., 2008b. S-metolachlor pulse exposure on the alga Scenedesmus vacuolatus: effects during exposure and the subsequent recovery. Chemosphere 73, 395–400. Vallotton, N., Eggen, R.I.L., Chèvre, N., 2009. Effect of sequential isoproturon pulse

213