Fostering integration of freshwater ecology with ecotoxicology
2016; Wiley; Volume: 61; Issue: 12 Linguagem: Inglês
10.1111/fwb.12852
ISSN1365-2427
Autores Tópico(s)Aquatic Ecosystems and Phytoplankton Dynamics
ResumoFreshwater BiologyVolume 61, Issue 12 p. 1991-2001 Special IssueFree Access Fostering integration of freshwater ecology with ecotoxicology Mark O. Gessner, Corresponding Author Mark O. Gessner gessner@igb-berlin.de Department of Experimental Limnology, Leibniz Institute of Freshwater Ecology and Inland Fisheries (IGB), Stechlin, Germany Department of Ecology, Berlin Institute of Technology (TU Berlin), Berlin, GermanyCorrespondence: Mark O. Gessner, Department of Experimental Limnology, Leibniz Institute of Freshwater Ecology and Inland Fisheries (IGB), Alte Fischerhütte 2, 16775 Stechlin, Germany. E-mail: gessner@igb-berlin.deSearch for more papers by this authorAhmed Tlili, Ahmed Tlili Department of Environmental Toxicology, Swiss Federal Institute of Aquatic Science and Technology (Eawag), Dübendorf, SwitzerlandSearch for more papers by this author Mark O. Gessner, Corresponding Author Mark O. Gessner gessner@igb-berlin.de Department of Experimental Limnology, Leibniz Institute of Freshwater Ecology and Inland Fisheries (IGB), Stechlin, Germany Department of Ecology, Berlin Institute of Technology (TU Berlin), Berlin, GermanyCorrespondence: Mark O. Gessner, Department of Experimental Limnology, Leibniz Institute of Freshwater Ecology and Inland Fisheries (IGB), Alte Fischerhütte 2, 16775 Stechlin, Germany. E-mail: gessner@igb-berlin.deSearch for more papers by this authorAhmed Tlili, Ahmed Tlili Department of Environmental Toxicology, Swiss Federal Institute of Aquatic Science and Technology (Eawag), Dübendorf, SwitzerlandSearch for more papers by this author First published: 08 November 2016 https://doi.org/10.1111/fwb.12852Citations: 71AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Summary Ecology and ecotoxicology have different historical roots, despite their similar names, but are slowly converging to meet the challenge of addressing the massive global proliferation and release of chemicals in the environment. The conceptual, methodological, review and standard research papers in this special issue reflect this emerging trend of blending ecological and ecotoxicological perspectives to assess impacts in freshwater ecosystems. Assessing community and ecosystem impacts of chemical contaminants is complex, however, and will require approaches that explicitly consider biological and chemical diversity as well as the natural variability of environmental factors at multiple spatial and temporal scales. Central themes of the papers in this issue are (i) the importance of indirect effects of chemical contaminants on species interactions and food webs; (ii) effects of multiple stressors, especially interactions between contaminants and environmental factors; (iii) consequences of chemical exposure on ecosystem processes such as primary production and litter decomposition; (iv) the need to account for context dependency and (v) potentially harmful community and ecosystem effects of emerging contaminants, among which nanoparticles are prominently represented. Collectively, these papers show that integrating ecological principles into the design and implementation of ecotoxicological research is essential for assessing and predicting contaminant impacts on biological communities and ecosystems. Conversely, applied ecology and bioassessment would benefit from concepts and approaches developed in ecotoxicology and from fully embracing chemical contaminants as key drivers of community structure and ecosystem processes. Introduction The Chemical Abstracts Registry, an open global database maintained by the American Chemical Society (www.cas.org), currently lists over 100 million unique chemical substances, less than 0.36% of which are regulated (Fig. 1). Although only a small fraction of the registered chemicals is marketed in notable amounts, the total chemical production around the globe amounted to 400 million tonnes in the year 2000, 400 000 times more than 70 years earlier in 1930. This corresponds to a global revenue of the chemical industry in 2015 to 5.2 trillion US$ (Fig. 1). These numbers illustrate the vast scope of chemicals in use and the ultimate release into the environment of either the original substances or their degradation products. As a consequence, manufactured chemicals are now ubiquitous in fresh waters, along with other toxic substances, such as mercury, which can undergo long-distance atmospheric transport. Indeed, the diversity and industrial production of chemicals has reached an extent that the massive proliferation affects even the most remote areas on the earth's surface and has evolved as a major component of global environmental change, to the point that the rates of increase in chemical diversity, production and release now exceed trends observed for other drivers of global change such as atmospheric CO2 concentrations (Bernhardt, Rosi-Marshall & Gessner, 2017). Until recently, this global dimension of chemical proliferation was not well appreciated (e.g. Rockström et al., 2009; Steffen et al., 2015), although large-scale assessments of chemicals in surface waters are emerging (Vörösmarty et al., 2010; Malaj et al., 2014; Ippolito et al., 2015; Stehle & Schulz, 2015). Importantly, many of the manufactured chemicals, particularly pharmaceuticals and pesticides, are designed to have specific biological effects even at low concentrations, and so are likely to be harmful virtually by definition. Figure 1Open in figure viewerPowerPoint Growth of chemical substances and revenue of the global chemical industry over the last 60 years. The number of chemicals refers to unique substances registered in the Chemical Abstract Service of the American Chemical Society (https://www.cas.org). Revenue estimates of the chemical industry were taken from OECD (2001) for the period between 1970 and 1998, and from ACC (2015) for the period of 2000–2015. The revenue data were adjusted for inflation (base year 2015) based on the Consumer Price Index provided by the Bureau of Labor Statistics of the US Department of Labor (http://data.bls.gov/cgi-bin/cpicalc.pl). Assessing impact Assessing the environmental impacts of chemicals is the domain of ecotoxicology. A host of concepts, approaches and methods has been developed since the inception of the discipline in the early 1960s (Truhaut, 1977). The result is a rich field with its own body of theories and methodologies, a range of established journals, an active community of professionals organised in an influential association, the Society of Environmental Toxicology and Chemistry (SETAC; https://www.setac.org), and well-attended scientific meetings. Traditionally, ecotoxicology has focused on the environmental fate and effects of chemicals on individual organisms. This focus is a straightforward extension of concepts developed to assess toxic effects of chemicals on humans. It reflects ecotoxicology's roots in medical toxicology as well as environmental chemistry, but not in ecology. As a result, the standard ecotoxicological approach is to assess impacts on individuals in highly standardised laboratory settings. This perspective has yielded valuable insights and largely underpins current legislation regulating the use of chemical products with potential environmental risks. However, despite broadening concepts and large progress over the past 50 years (e.g. Cairns, 1988; Chapman, 2002; Schmitt-Jansen et al., 2008; Clements & Rohr, 2009; Halstead et al., 2014; Liess et al., 2016), a fundamental challenge remains as to how the discrepancy can be resolved between ecotoxicology's traditional focus on single organisms and the societal requirement to assess consequences of contaminant exposure at levels of biological organisation most relevant to risk assessment: populations, communities and whole ecosystems. Analysing patterns, processes and relationships at these levels is the domain of ecology, suggesting that a systematic consideration of ecological concepts and approaches could greatly strengthen ecotoxicology. It has indeed become increasingly recognised that the traditional focus of ecotoxicology on individuals (or lower levels of biological organisation) is insufficient to guide environmental assessment (e.g. Schmitt-Jansen et al., 2008). In response, a number of specific community-level and ecosystem-level metrics have been developed. Examples are the concepts of pollution-induced community tolerance (PICT; Blanck, 2002), species sensitivity distribution (De Zwart & Posthuma, 2005), SPEcies At Risk (SPEAR: Liess & von der Ohe, 2005) or Stress Addition Model (SAM: Liess et al., 2016), which are increasingly applied as part of a radically expanded methodology and conceptual framework that aims both to detect ecotoxicological effects and to unravel the underlying mechanisms operating at the community and ecosystem level (Newman & Clements, 2008; Peters, Bundschuh & Schäfer, 2013). An important insight in this context is that ecotoxicological impacts can have multiple causes, including both direct toxic effects on key organisms and indirect effects emerging when species interaction networks are disrupted, which can ramify into community structure and other ecosystem properties (Relyea & Hoverman, 2006; Halstead et al., 2014). Such indirect mechanisms may weaken contaminant effects because of compensatory processes or they may exacerbate impacts (Relyea & Hoverman, 2006; Halstead et al., 2014), which can also be amplified over time (Liess et al., 2013). Species interactions, context dependency and multiple stressors Advocacy of the importance of indirect effects arising from species interactions appears to have led to an intellectual divide even within ecotoxicology. Some ecotoxicologists have embraced the need to devise assessments beyond the level of individuals, arguing that this broadened perspective is already well established. Others maintain that tests targeting single organisms (or lower levels of biological organisation) are sufficient, at least for initial assessments (EFSA PPR Panel, 2013). The argument goes that the common practice of assessing risk based on effect levels determined for particular (model) organisms and applying safety margins accounts for variation in the sensitivity among species. However, this traditional reliance on single-species tests in controlled settings fails to recognise that qualitatively distinctive impacts can emerge when species interact. Consider a simple system consisting of an insect grazer consuming algae (Fig. 2a). Exposure of the species pair to an insecticide will adversely affect the consumer and release the algae from herbivory. Subsequent proliferation of the algae results in an apparent stimulation of biological production in response to insecticide exposure. The mechanism causing this effect is readily understood in this simple scenario with only two organisms involved and the selective effect of the insecticide being known. However, the outcome could be more complicated to predict and sometimes counterintuitive if the effect of the contaminant on particular organisms is less clear, as is often the case. Outcomes can be reversed when even a single additional species is present, for instance, an insect predator that is more sensitive to the insecticide than its prey (Fig. 2b); or effects may differ between competing species or functional groups at the base of a food web when one is vulnerable to grazing and the other is not, even when neither one is sensitive to the chemical contaminant (Fig. 2c); or there might be no net effect of a contaminant at the base of the food web, for instance, when a second consumer is more sensitive than its competitor and a predator but not susceptible to predation (Fig. 2d). In much more diverse real communities, interacting species obviously possess many more degrees of freedom for interactions that further complicate the prediction of outcomes of contaminant exposure. Clearly, much ingenuity is needed to develop assessment approaches accounting for effects arising from such ecological interactions. Figure 2Open in figure viewerPowerPoint Four scenarios depicting cascading effects of a contaminant, such as an insecticide on species interactions, highlighting that even simple food webs can show complex responses to chemical exposure. Red and black arrows denote negative and positive effects, respectively, with the thickness of lines indicating the effect strength. Solid lines denote direct chemical effects and stippled lines denote indirect effects mediated by species interactions. In the first scenario (a), an insecticide has a direct negative effect on the consumer and thus releases the base of the food web (resource) from consumption. In the second scenario (b), the effect on the base of the food web is reversed because an assumed stronger impact of the insecticide on a predator than on the primary consumer results in increased consumption of the basal species. In the third scenario (c), outcomes on different components of the food-web base differ depending on whether competing species are vulnerable to grazing (R1) or not (R2). Finally, in the fourth scenario (d), no net effect of the insecticide on the base of the food web is expected, because a strong toxic effect on a consumer not vulnerable to predation (C1) is balanced by a combination of effects on a second consumer (C2) caused by the insecticide (negative), release from predation (positive) and release from competition (strongly positive), resulting in reduced consumption by C1 that is fully compensated by an enhanced consumption of C2. Including species interactions is only one important aspect of attempts to strengthen ecotoxicology by integrating ecological considerations. Another thrust to improve predictions of contaminant effects in real-world ecosystems is the recognition that impacts of chemicals depend on a large number of environmental factors, such as temperature, pH, flow velocity or nutrient and organic matter supply. Moreover, some of these environmental factors are beneficial at low levels, where they could buffer negative contaminant effects, but become detrimental themselves as exposure levels increase, resulting in a unimodal pattern. This has been conceptualised in the subsidy-stress concept (Odum, Finn & Franz, 1979; Wagenhoff et al., 2011) and is related to the hormetic dose–response relationship known in ecotoxicology (Calabrese, 2004). Further, the influence of particular environmental factors varies in space and time, which adds another layer of complexity to assessing contaminant effects in real-world communities and ecosystems. Thus, ecotoxicological assessments must acknowledge context dependency. Add to this the presence of multiple stressors, whether contaminants (i.e. chemical mixtures), environmental factors or both (Fig. 3), which can interact with one other (Woodward, Perkins & Brown, 2010). In such complex situations, one of the key challenges is to predict the combined effects of chemicals with other stressors occurring in the environment (e.g. Liess et al., 2016; Magbanua et al., 2016) and to pinpoint the specific impacts of the chemicals or mixtures of chemicals (Fischer, Pomati & Eggen, 2013). As for the problem of impacts arising from species interactions, the standard application of generic safety factors falls short of capturing impacts of multiple interacting chemicals when their effects are not simply additive but synergistic or antagonistic. The analytical challenges to assess such combined effects of chemical mixtures or multiple environmental stress factors are similar (Jonker et al., 2005; Piggott, Townsend & Matthaei, 2015) and also resemble the problem of assessing species diversity effects on ecosystem processes (e.g. Loreau & Hector, 2001), a field that has received much attention in ecology. Therefore, sound concepts and appropriate methodologies are in place to address the problem of interaction between multiple stressors in experiments, whether the stressors are anthropogenic chemicals or environmental factors. Nevertheless, thorough assessments of impacts arising from interacting factors remain a tedious task. Figure 3Open in figure viewerPowerPoint Schematic representation on the potential effects of chemicals and environmental factors acting in concert on different trophic levels of food webs. Red arrows indicate that besides having a direct influence, environmental factors can modify the bioavailability and toxicity of contaminants, and thus have indirect effects on food-web components that are mediated by chemicals. Assessing and predicting impacts of contaminants on species, species interactions and diverse communities are not enough. As has been argued in the context of river bioassessment (Gessner & Chauvet, 2002), comprehensive assessments require the consideration of processes, including ecosystem processes, in addition to variables describing the structure of populations, communities and ecosystems (Rosi-Marshall & Royer, 2012). For some processes, such as primary production or nitrification, effects of chemical substances can be directly derived from the impairments of the activities of distinct groups of organisms that share particular physiological capabilities. For others, however, such as the decomposition of organic matter, multiple functionally distinct groups of organisms may be involved, which offers many opportunities for species interactions and compensatory effects to influence the net outcome of contaminant exposure on process rates (Gessner et al., 2010). Directly determining the effects of contaminants on ecosystem process rates in combination with assessments of impacts on community structure and activities is a promising avenue to quantify and understand the effects of chemicals on these processes (e.g. Bundschuh et al., 2009; Rosi-Marshall et al., 2013). Overview of contributions The 18 papers in this special issue offer a number of important insights that can be gained by exploring the interface between ecotoxicology and ecology. Most of the papers elucidate the effects of chemical contaminants in fresh waters at the community and ecosystem level. Specifically, they address the following interconnected themes: (i) the importance of indirect effects of contaminants through species interactions (Fig. 2); (ii) trophic transfer and biomagnification of chemicals; (iii) effects of multiple stressors, including interactions between contaminants and environmental factors (Fig. 3); (iv) consequences of chemical exposure on ecosystem processes; (v) the complementarity between field observations and experimental approaches at different scales; (vi) the development of community tolerance to contaminants; (vii) context dependency; and (viii) potentially harmful effects of emerging contaminants such as nanoparticles. These themes are not distinctly separate and additional topics relating to these themes also appear repeatedly throughout this special issue, although the diversity of relationships between freshwater ecology and ecotoxicology cannot be completely covered. Clearly, however, this collection of papers illustrates the richness of linkages between these disciplines and the power of ecological concepts, relationships and methodologies to enhance ecotoxicological impact and risk assessment. Contaminant effects on species interactions Concepts of community ecology, one of the recurrent themes of this special issue, have been proposed as an integrative framework to predict impacts of contaminants in natural ecosystems (Rohr, Kerby & Sih, 2006). At the heart of this approach lies the premise that improved understanding of how chemicals affect trophic interactions and food-web structure can provide insights into ecotoxicological effects beyond those provided by standard assays targeting individual species or lower levels of biological organisation (Relyea & Hoverman, 2006; Clements & Rohr, 2009). A corollary of this perspective is that framing ecotoxicological assessments in a community-ecology context can help reveal indirect effects of contaminants that superimpose and potentially overwhelm direct toxic effects. Guasch et al. (2016) report in this issue the results of a microcosm study assessing the single and combined effects of grazing and exposure to a widespread biocide, triclosan, on structural and functional properties of stream periphyton. Similarly, Pacioglu et al. (2016) investigated how phosphorus immobilisation by polyaluminium chloride applied at large scale to restore eutrophied lakes affects interactions between leaf-litter, fungi and detritivorous consumers, and the consequences of these interactions on the process of leaf-litter decomposition. Despite the fundamental differences between the two study systems, both show synergistic as well as antagonistic interactions between the contaminants and consumers. Such results underscore that detailed mechanistic investigations into the effects of contaminants on species interactions are essential to assess the risks of chemical substances released into natural ecosystems. Besides direct negative effects of a fungicide, tebuconazole, on fungal communities associated with decomposing leaves in streams, Donnadieu et al. (2016) found that bacterial communities (non-target organisms) were also affected indirectly. A likely reason is that resource competition experienced by the bacteria was reduced when the competing fungi were stressed by tebuconazole exposure, but it is also possible that nutrients released from killed fungi were supplied to the bacteria as an additional resource. Thompson et al. (2016) analysed an insecticide spill in a field investigation involving citizen scientists and found that a wide range of metrics reflecting the structure and functioning of a river ecosystem were altered by the spill. Both direct and indirect effects of the insecticide were evident, and these occurred across multiple taxa and levels of biological organisation, from genes to ecosystems. While elucidating indirect effects of contaminants arising from trophic interactions is clearly important, an additional argument for considering trophic interactions in ecotoxicology is to understand the transfer of toxic chemicals through food webs and to assess the consequences for biomagnification. This prompted Laws et al. (2016) to investigate the transfer of tributyltin (TBT), a widespread organotin compound, not only within aquatic food webs, but also from freshwater to terrestrial consumers feeding on aquatic prey. Although there was no evidence for biomagnification in these consumers, the study discovered a disturbing environmental legacy of butyltins, showing that TBT persisted in the examined food webs 25 years after its last known use in the country, the U.K., and that it continued to be transferred to terrestrial consumers even after this long period. Effects of multiple-stressor interactions Another central theme covered by several contributions to this special issue is the potential for chemicals to interact with a range of environmental factors that affect freshwater communities and ecosystem processes (Fig. 3). There is clear evidence that considering ecosystem responses to multiple stressors increases understanding of the consequences of environmental change, which is critical for setting guidelines to protect fresh waters (Ormerod et al., 2010). On the one hand, interactions between stressors may exacerbate effects of individual stressors and result in unexpectedly strong ecological effects. On the other hand, ecological effects of chemicals can be mitigated by environmental factors such as organic matter availability. Bundschuh & McKie (2016) review fine-particulate organic matter (FPOM) dynamics in streams and how FPOM can modulate the impacts of chemicals on communities and ecosystem processes. This literature analysis reveals that little attention has been given to the influence of FPOM on the transfer, bioavailability and toxicity of chemicals, although such information is important for realistic assessments of contaminant effects and FPOM dynamics per se have been extensively studied in streams. Nutrient content of organic matter is assumed to be an important factor determining food quality for detritivorous consumers (Danger, Gessner & Bärlocher, 2016). This led Arce-Funck et al. (2016) to hypothesise that the consumption of leaf litter enriched with phosphorus (P) can mitigate metal stress on litter-consuming detritivores. Consumption of leaves rich in P did indeed increase the energy reserves and activity of the tested detritivore, the amphipod Gammarus fossarum. However, stress induced by exposure to elevated silver concentrations reduced the energetic reserves and activity of the detritivore irrespective of the previously consumed food. This suggests that a positive fitness effect of high food quality was insufficient to produce a measurable increase in metal tolerance. In a similar vein, Alexander et al. (2016) examined the effects of subsidy–stress responses on predator–prey relationships in a benthic macroinvertebrate community exposed to insecticides, nutrients and predation pressure. In their experiments conducted in 72 small streamside flow-through mesocosms, they found that the outcomes of scenarios involving multiple factors do not necessarily conform to conventional exposure–response models of decreased macroinvertebrate density with increased stress. Rather, both studies (i.e. Alexander et al., 2016; Arce-Funck et al., 2016) indicate that in addition to indirect nutrient effects, characteristics of populations and communities can cause unexpected responses to contaminant exposure (see Fig. 2), and thus require consideration in ecotoxicological assessments and environmental risk management of chemicals. Multiple-stressor analyses at large scale A completely different approach to experiments in mesocosms, microcosms or simple laboratory systems is to use statistical analyses on large data sets collected in the field to identify associations between contaminants, environmental factors and variables describing populations and communities of freshwater organisms as well as ecosystem properties. This approach is routinely used in applied freshwater ecology but has not been commonly adopted in ecotoxicology. However, Ponsatí et al. (2016) conducted field surveys at multiple river sites to investigate the relative importance of environmental factors and a very wide range of synthetic chemicals (i.e. herbicides, insecticides, antibiotics, pharmaceuticals, personal care products and industrial organic compounds) to the structure and function of periphyton (Fig. 3) under various hydrological conditions and land uses. Results of their multivariate statistics suggest that well-developed periphyton during base flow were more strongly associated with various synthetic chemicals than with environmental factors, whereas thinner periphyton developing during periods of high river flow appeared to be less affected by chemicals than by environmental factors. Schäfer et al. (2016) also assessed the relative importance of synthetic chemicals and other stressors, based on large data sets collected as part of river monitoring programmes in Germany. Estimated levels of stress caused by chemicals were only weakly correlated with those arising from nutrient pollution, habitat degradation and invasive species, but exceeded risk thresholds at about 50% of the investigated sites, as many as noted for invasive species. Although this percentage was lower than for habitat degradation and nutrient pollution (c. 85%), its magnitude suggests that chemical contaminants contribute notably to risks of ecological impacts on river ecosystems. Attributing effect to cause in field settings Many freshwater ecosystems are affected by multiple stressors (Ormerod et al., 2010; Vörösmarty et al., 2010), which limits the power of field studies to attribute ecological effects to specific causes such as exposure to particular chemicals. A partial remedy is a novel versatile method that has potential to assess impacts of multiple chemical stressors on biofilms in fresh waters (Costello et al., 2016). This method is based on an adaptation of the widely used nutrient-diffusing substrata that can be deployed in situ (e.g. Capps et al., 2011). Similarly, contaminant exposure substrata contain chemical mixtures or single chemicals in agar poured in plastic cups and covered by a permeable surface (e.g. fritted glass) for biofilm growth to assess effects of the diffusing chemicals on biofilms developing in these conditions. The main strength of the method is that many combinations of chemicals can be tested directly in the field. Another approach to establish community-level causalities is based on the concept of PICT suggested by Blanck, Wängberg & Molander (1988). It rests on the fact that differences in the tolerance of species to contaminants lead to changes in community structure in response to chronic chemical stress (Millward & Klerks, 2002). As a consequence, increased tolerance of a community to stress can be used as an indicator of previous contaminant exposure. In a perspective discussing the PICT concept, Tlili et al. (2016) argue that major advantages of the approach are that unlike in single-species tests, the observed community responses account for differences in species sensitivity and intersp
Referência(s)