Studying interactions among anthropogenic stressors in freshwater ecosystems: A systematic review of 2396 multiple-stressor experiments.

Understanding the interactions among anthropogenic stressors is critical for effective conservation and management of ecosystems. Freshwater scientists have invested considerable resources in conducting factorial experiments to disentangle stressor interactions by testing their individual and combined effects. However, the diversity of stressors and systems studied has hindered previous syntheses of this body of research. To overcome this challenge, we used a novel machine learning framework to identify relevant studies from over 235,000 publications. Our synthesis resulted in a new dataset of 2396 multiple-stressor experiments in freshwater systems. By summarizing the methods used in these studies, quantifying trends in the popularity of the investigated stressors, and performing co-occurrence analysis, we produce the most comprehensive overview of this diverse field of research to date. We provide both a taxonomy grouping the 909 investigated stressors into 31 classes and an open-source and interactive version of the dataset (https://jamesaorr.shinyapps.io/freshwater-multiple-stressors/). Inspired by our results, we provide a framework to help clarify whether statistical interactions detected by factorial experiments align with stressor interactions of interest, and we outline general guidelines for the design of multiple-stressor experiments relevant to any system. We conclude by highlighting the research directions required to better understand freshwater ecosystems facing multiple stressors.

Effects of herbivorous fish on shallow lake ecosystems increase at moderate nutrient conditions.

Submerged macrophytes are vital in shallow lakes, as they provide critical ecosystem functions and services and can stabilize the clear-water conditions by various mechanisms. Nutrient enrichment reduces the resilience of macrophyte dominance in shallow lakes, thereby making them susceptible to shifts towards phytoplankton dominance following perturbations. Here, we conducted a mesocosm experiment to examine the individual and combined effects of nutrient enrichment and the addition of grass carp (Ctenopharyngodon idella) on the abundance of submerged macrophytes, epiphyton, and phytoplankton. We hypothesized that moderate nutrient enrichment facilitates macrophyte abundance, but also phytoplankton abundance after macrophyte removal by herbivorous fish. Our data showed that herbivory by grass carp could trigger a shift from macrophytes to algal dominance in mesocosms with moderate nutrient concentrations, but not in those with low nutrient concentrations. Moderate nutrient enrichment alone promoted submerged macrophyte growth, whereas the introduction of grass carp induced a collapse of submerged macrophytes regardless of nutrient conditions. Moreover, the introduction of grass carp showed more negative effects on light conditions of the water column in mesocosms with moderate nutrient concentrations compared to those with low nutrient concentrations. A recovery of submerged macrophytes might thus be limited by low light availability in lakes with moderate nutrient conditions suffering grass carp perturbation. Our results suggest that submerged macrophyte-dominated shallow lakes with moderate nutrient conditions are vulnerable to perturbation by herbivorous fish such as grass carp. In turn, managing the abundance of herbivores in these lakes can support the dominance of macrophytes and associated clear water conditions.

Short-term effects of macrophyte removal on aquatic biodiversity in rivers and lakes.

Mass development of macrophytes is an increasing problem in many aquatic systems worldwide. Dense mats of macrophytes can negatively affect activities like boating, fishing or hydropower production and one of the management measures often applied is mechanical removal. In this study, we analyzed the effect of mechanical macrophyte removal on phytoplankton, zooplankton, and macroinvertebrate (pelagic and benthic samples) assemblages. Our study covered five sites in four countries in Europe and Africa with highly variable characteristics. In all sites, dense mats of different macrophyte species (Juncus bulbosus in a river in Norway; a mix of native macrophytes in a German river, Elodea nuttallii in a lake in Germany, Ludwigia spp. In a French lake and Pontederia crassipes in a South African lake) are problematic and mechanical removal was applied. In every country, we repeated the same BACI (Before-After-Control-Impact) design, including "before", "one week after", and "six weeks after" sampling in a control and an impact section. Repeating the same experimental design at all sites allowed us to disentangle common effects across all sites from site-specific effects. For each taxonomic group, we analyzed three structural and three functional parameters, which we combined in a scoring system. Overall, the removal of macrophytes negatively affected biodiversity, in particular of zooplankton and macroinvertebrate assemblages. In contrast, plant removal had positive effects on the phytoplankton assemblages. Effects were more pronounced one week after removal than six weeks after. Consequently, we suggest a stronger consideration of the effect of plant removal on biodiversity to arrive at more sustainable management practices in the future.

Drivers of Perceived Nuisance Growth by Aquatic Plants.

Mass developments of macrophytes occur frequently worldwide and are often considered a nuisance when interfering with human activities. It is crucial to understand the drivers of this perception if we are to develop effective management strategies for ecosystems with macrophyte mass developments. Using a comprehensive survey spanning five sites with different macrophyte species in four countries (Norway, France, Germany and South Africa), we quantified the perception of macrophyte growth as a nuisance among residents and visitors, and for different recreational activities (swimming, boating, angling, appreciation of biodiversity, appreciation of landscape and birdwatching). We then used a Bayesian network approach to integrate the perception of nuisance with the consequences of plant removal. From the 1234 responses collected from the five sites, a range of 73-93% of the respondents across the sites considered macrophyte growth a nuisance at each site. Residents perceived macrophytes up to 23% more problematic than visitors. Environmental mindedness of respondents did not influence the perception of nuisance. Perceived nuisance of macrophytes was relatively similar for different recreational activities that were possible in each case study site, although we found some site-specific variation. Finally, we illustrate how Bayesian networks can be used to choose the best management option by balancing people's perception of macrophyte growth with the potential consequences of macrophyte removal.

Predator Group Composition Indirectly Influences Food Web Dynamics through Predator Growth Rates.

AbstractConsiderable theoretical work predicts that intraspecific trait variation can have profound ecological consequences by altering species interactions. Because of their high flexibility, behavioral traits may be especially relevant in mediating how species respond to one another, thus affecting food web dynamics and ecosystem functioning. However, empirical evidence supporting this idea is limited. Here, we generated predator groups where we manipulated the composition of behavioral types within the groups to assess effects on predator growth rates, prey communities, basal resources, and ecosystem functioning in replicated outdoor ponds. Using European perch (Perca fluviatilis), we created three types of predator populations: two where all individuals expressed either bold or shy phenotypes and one that contained a mix of individuals of the two behavioral types. Bold perch grew faster in mixed populations, indicating that predator growth depends on each individual's behavioral type and that of its group members. However, there was no evidence that the behavioral composition of the perch population directly altered the dynamics of lower trophic levels. Instead, final perch biomass, not behavioral composition, had the strongest influence on lower trophic levels. Thus, the central question may not be whether predator behavior matters at all for trophic dynamics but rather when behavioral effects will predominate over effects of other influences, such as predator biomass variation.

Blue Waters, Green Bottoms: Benthic Filamentous Algal Blooms Are an Emerging Threat to Clear Lakes Worldwide.

Nearshore (littoral) habitats of clear lakes with high water quality are increasingly experiencing unexplained proliferations of filamentous algae that grow on submerged surfaces. These filamentous algal blooms (FABs) are sometimes associated with nutrient pollution in groundwater, but complex changes in climate, nutrient transport, lake hydrodynamics, and food web structure may also facilitate this emerging threat to clear lakes. A coordinated effort among members of the public, managers, and scientists is needed to document the occurrence of FABs, to standardize methods for measuring their severity, to adapt existing data collection networks to include nearshore habitats, and to mitigate and reverse this profound structural change in lake ecosystems. Current models of lake eutrophication do not explain this littoral greening. However, a cohesive response to it is essential for protecting some of the world's most valued lakes and the flora, fauna, and ecosystem services they sustain.

Empirical correspondence between trophic transfer efficiency in freshwater food webs and the slope of their size spectra.

The density of organisms declines with size, because larger organisms need more energy than smaller ones and energetic losses occur when larger organisms feed on smaller ones. A potential expression of density-size distributions are Normalized Biomass Size Spectra (NBSS), which plot the logarithm of biomass independent of taxonomy within bins of logarithmic organismal size, divided by the bin width. Theoretically, the NBSS slope of multi-trophic communities is exactly -1.0 if the trophic transfer efficiency (TTE, ratio of production rates between adjacent trophic levels) is 10% and the predator-prey mass ratio (PPMR) is fixed at 104 . Here we provide evidence from four multi-trophic lake food webs that empirically estimated TTEs correspond to empirically estimated slopes of the respective community NBSS. Each of the NBSS considered pelagic and benthic organisms spanning size ranges from bacteria to fish, all sampled over three seasons in 1 yr. The four NBSS slopes were significantly steeper than -1.0 (range -1.14 to -1.19, with 95% CIs excluding -1). The corresponding average TTEs were substantially lower than 10% in each of the four food webs (range 1.0% to 3.6%, mean 1.85%). The overall slope merging all biomass-size data pairs from the four systems (-1.17) was almost identical to the slope predicted from the arithmetic mean TTE of the four food webs (-1.18) assuming a constant PPMR of 104 . Accordingly, our empirical data confirm the theoretically predicted quantitative relationship between TTE and the slope of the biomass-size distribution. Furthermore, we show that benthic and pelagic organisms can be merged into a community NBSS, but future studies have yet to explore potential differences in habitat-specific TTEs and PPMRs. We suggest that community NBSS may provide valuable information on the structure of food webs and their energetic pathways, and can result in improved accuracy of TTE-estimates.

Warming enhances sedimentation and decomposition of organic carbon in shallow macrophyte-dominated systems with zero net effect on carbon burial.

Temperatures have been rising throughout recent decades and are predicted to rise further in the coming century. Global warming affects carbon cycling in freshwater ecosystems, which both emit and bury substantial amounts of carbon on a global scale. Currently, most studies focus on the effect of warming on overall carbon emissions from freshwater ecosystems, while net effects on carbon budgets may strongly depend on burial in sediments. Here, we tested whether year-round warming increases the production, sedimentation, or decomposition of particulate organic carbon and eventually alters the carbon burial in a typical shallow freshwater system. We performed an indoor experiment in eight mesocosms dominated by the common submerged aquatic plant Myriophyllum spicatum testing two temperature treatments: a temperate seasonal temperature control and a warmed (+4°C) treatment (n = 4). During a full experimental year, the carbon stock in plant biomass, dissolved organic carbon in the water column, sedimented organic matter, and decomposition of plant detritus were measured. Our results showed that year-round warming nearly doubled the final carbon stock in plant biomass from 6.9 ± 1.1 g C in the control treatment to 12.8 ± 0.6 g C (mean ± SE), mainly due to a prolonged growing season in autumn. DOC concentrations did not differ between the treatments, but organic carbon sedimentation increased by 60% from 96 ± 9.6 to 152 ± 16 g C m-2  yaer-1 (mean ± SE) from control to warm treatments. Enhanced decomposition of plant detritus in the warm treatment, however, compensated for the increased sedimentation. As a result, net carbon burial was 40 ± 5.7 g C m-2  year-1 in both temperature treatments when fluxes were combined into a carbon budget model. These results indicate that warming can increase the turnover of organic carbon in shallow macrophyte-dominated systems, while not necessarily affecting net carbon burial on a system scale.

Impact of trematode infections on periphyton grazing rates of freshwater snails.

In freshwater ecosystems, snails can significantly influence the competition between primary producers through grazing of periphyton. This activity can potentially be modified by trematodes, a large group of parasites which mostly use molluscs as the first intermediate host. Available studies, however, show contradictory effects of trematodes on snail periphyton grazing. Here, we used four different freshwater snail-trematode systems to test whether a general pattern can be detected for the impact of trematode infections on snail periphyton grazing. In our experimental systems, mass-specific periphyton grazing rates of infected snails were higher, lower, or similar to rates of non-infected conspecifics, suggesting that no general pattern exists. The variation across studied snail-trematode systems may result from differences on how the parasite uses the resources of the snail and thus affects their energy budget. Trematode infections can significantly alter the grazing rate of snails, where, depending on the system, the mass-specific grazing rate can double or halve. This underlines both, the high ecological relevance of trematodes and the need for comprehensive studies at the species level to allow an integration of these parasite-host interactions into aquatic food web concepts.

Boom-bust dynamics in biological invasions: towards an improved application of the concept.

Boom-bust dynamics - the rise of a population to outbreak levels, followed by a dramatic decline - have been associated with biological invasions and offered as a reason not to manage troublesome invaders. However, boom-bust dynamics rarely have been critically defined, analyzed, or interpreted. Here, we define boom-bust dynamics and provide specific suggestions for improving the application of the boom-bust concept. Boom-bust dynamics can arise from many causes, some closely associated with invasions, but others occurring across a wide range of ecological settings, especially when environmental conditions are changing rapidly. As a result, it is difficult to infer cause or predict future trajectories merely by observing the dynamic. We use tests with simulated data to show that a common metric for detecting and describing boom-bust dynamics, decline from an observed peak to a subsequent trough, tends to severely overestimate the frequency and severity of busts, and should be used cautiously if at all. We review and test other metrics that are better suited to describe boom-bust dynamics. Understanding the frequency and importance of boom-bust dynamics requires empirical studies of large, representative, long-term data sets that use clear definitions of boom-bust, appropriate analytical methods, and careful interpretations.

Effects of light and autochthonous carbon additions on microbial turnover of allochthonous organic carbon and community composition.

The fate of allochthonous dissolved organic carbon (DOC) in aquatic systems is primarily controlled by the turnover of heterotrophic bacteria. However, the roles that abiotic and biotic factors such as light and DOC release by aquatic primary producers play in the microbial decomposition of allochthonous DOC is not well understood. We therefore tested if light and autochthonous DOC additions would increase allochthonous DOC decomposition rates and change bacterial growth efficiencies and community composition (BCC). We established continuous growth cultures with different inocula of natural bacterial communities and alder leaf leachates (DOCleaf) with and without light exposure before amendment. Furthermore, we incubated DOCleaf together with autochthonous DOC from lysed phytoplankton cultures (DOCphyto). Our results revealed that pretreatments of DOCleaf with light resulted in a doubling of bacterial growth efficiency (BGE), whereas additions of DOCphyto or combined additions of DOCphyto and light had no effect on BGE. The change in BGE was not accompanied by shifts in the phylogenetic structure of the BCC, but BCC was influenced by the DOC source. Our results highlight that a doubling of BGE is not necessarily accompanied by a shift in BCC and that BCC is more strongly affected by resource properties.

Causes of macrophyte mass development and management recommendations.

Aquatic plants (macrophytes) are important for ecosystem structure and function. Macrophyte mass developments are, however, often perceived as a nuisance and are commonly managed by mechanical removal. This is costly and often ineffective due to macrophyte regrowth. There is insufficient understanding about what causes macrophyte mass development, what people who use water bodies consider to be a nuisance, or the potential negative effects of macrophyte removal on the structure and function of ecosystems. To address these gaps, we performed a standardized set of in situ experiments and questionnaires at six sites (lakes, reservoirs, and rivers) on three continents where macrophyte mass developments occur. We then derived monetary values of ecosystem services for different scenarios of macrophyte management ("do nothing", "current practice", "maximum removal"), and developed a decision support system for the management of water courses experiencing macrophyte mass developments. We found that (a) macrophyte mass developments often occur in ecosystems which (unintentionally) became perfect habitats for aquatic plants, that (b) reduced ecosystem disturbance can cause macrophyte mass developments even if nutrient concentrations are low, that (c) macrophyte mass developments are indeed perceived negatively, but visitors tend to regard them as less of a nuisance than residents do, that (d) macrophyte removal lowers the water level of streams and adjacent groundwater, but this may have positive or negative overall societal effects, and that (e) the effects of macrophyte removal on water quality, greenhouse gas emissions, and biodiversity vary, and likely depend on ecosystem characteristics and macrophyte life form. Overall, we found that aquatic plant management often does not greatly affect the overall societal value of the ecosystem, and we suggest that the "do nothing" option should not be easily discarded in the management of perceived nuisance mass developments of aquatic plants.

Ecosystem-level studies of terrestrial carbon reveal contrasting bacterial metabolism in different aquatic habitats.

In aquatic systems, terrestrial dissolved organic matter (t-DOM) is known to stimulate bacterial activities in the water column, but simultaneous effects of autumnal leaf input on water column and sediment microbial dynamics in littoral zones of lakes remain largely unknown. The study's objective was to determine the effects of leaf litter on bacterial metabolism in the littoral water and sediment, and subsequently, the consequences for carbon cycling and food web dynamics. Therefore, in late fall, we simultaneously measured water and sediment bacterial metabolism in the littoral zone of a temperate shallow lake after adding terrestrial particulate organic matter (t-POM), namely, maize leaves. To better evaluate bacterial production (BP) and community respiration (CR) in sediments, we incubated sediment cores with maize leaves of different quality (nonleached and leached) under controlled laboratory conditions. Additionally, to quantify the incorporated leaf carbon into microbial biomass, we determined carbon isotopic ratios of fatty acids from sediment and leaf-associated microbes from a laboratory experiment using 13C-enriched beech leaves. The concentrations of dissolved organic carbon (DOC) increased significantly in the lake after the addition of maize leaves, accompanied by a significant increase in water BP. In contrast, sediment BP declined after an initial peak, showing no positive response to t-POM addition. Sediment BP and CR were also not stimulated by t-POM in the laboratory experiment, either in short-term or in long-term incubations, except for a short increase in CR after 18 hours. However, this increase might have reflected the metabolism of leaf-associated microorganisms. We conclude that the leached t-DOM is actively incorporated into microbial biomass in the water column but that the settling leached t-POM (t-POML) does not enter the food web via sediment bacteria. Consequently, t-POML is either buried in the sediment or introduced into the aquatic food web via microorganisms (bacteria and fungi) directly associated with t-POM(L) and via benthic macroinvertebrates by shredding of t-POM(L). The latter pathway represents a "benthic shortcut" which efficiently transfers t-POM(L) to higher trophic levels.

Microcosm experiment combined with process-based modeling reveals differential response and adaptation of aquatic primary producers to warming and agricultural run-off.

Fertilizers, pesticides and global warming are threatening freshwater aquatic ecosystems. Most of these are shallow ponds or slow-flowing streams or ditches dominated by submerged macrophytes, periphyton or phytoplankton. Regime shifts between the dominance of these primary producers can occur along a gradient of nutrient loading, possibly triggered by specific disturbances influencing their competitive interactions. However, phytoplankton dominance is less desirable due to lower biodiversity and poorer ecosystem function and services. In this study, we combined a microcosm experiment with a process-based model to test three hypotheses: 1) agricultural run-off (ARO), consisting of nitrate and a mixture of organic pesticides and copper, differentially affects primary producers and enhances the risk of regime shifts, 2) warming increases the risk of an ARO-induced regime shift to phytoplankton dominance and 3) custom-tailored process-based models support mechanistic understanding of experimental results through scenario comparison. Experimentally exposing primary producers to a gradient of nitrate and pesticides at 22°C and 26°C supported the first two hypotheses. ARO had direct negative effects on macrophytes, while phytoplankton gained from warming and indirect effects of ARO like a reduction in the competitive pressure exerted by other groups. We used the process-based model to test eight different scenarios. The best qualitative fit between modeled and observed responses was reached only when taking community adaptation and organism acclimation into account. Our results highlight the importance of considering such processes when attempting to predict the effects of multiple stressors on natural ecosystems.

Disentangling the mechanisms sustaining a stable state of submerged macrophyte dominance against free-floating competitors.

Free-floating and rootless submerged macrophytes are typical, mutually exclusive vegetation types that can alternatively dominate in stagnant and slow flowing inland water bodies. A dominance of free-floating plants has been associated with a lower number of aquatic ecosystem services and can be explained by shading of rootless submerged macrophytes. Vice versa, high pH and competition for several nutrients have been proposed to explain the dominance of rootless submerged macrophytes. Here, we performed co-culture experiments to disentangle the influence of limitation by different nutrients, by pH effects and by allelopathy in sustaining the dominance of rootless submerged macrophytes. Specifically, we compared the effects of nitrogen (N), phosphorus (P), iron (Fe) and manganese (Mn) deficiencies and an increased pH from 7 to 10 in reducing the growth of free-floating Lemna gibba by the rootless Ceratophyllum demersum. These macrophyte species are among the most common in highly eutrophic, temperate water bodies and known to mutually exclude each other. After co-culture experiments, additions of nutrients and pH neutralisation removed the growth inhibition of free-floating plants. Among the experimentally tested factors significantly inhibiting the growth of L. gibba, an increase in pH had the strongest effect, followed by depletion of P, N and Fe. Additional field monitoring data revealed that in water bodies dominated by C. demersum, orthophosphate concentrations were usually sufficient for optimal growth of free-floating plants. However, pH was high and dissolved inorganic N concentrations far below levels required for optimal growth. Low N concentrations and alkaline pH generated by dense C. demersum stands are thus key factors sustaining the stable dominance of rootless submerged vegetation against free-floating plants. Consequently, N loading from e.g. agricultural runoff, groundwater or stormwater is assumed to trigger regime shifts to a dominance of free-floating plants and associated losses in ecosystem services.

Warming lowers critical thresholds for multiple stressor-induced shifts between aquatic primary producers.

In aquatic ecosystems, excessive nutrient loading is a global problem that can induce regime shifts from macrophyte- to phytoplankton-dominated states with severe consequences for ecosystem functions. Most agricultural landscapes are sites of nutrient and pesticide loading, which can interact with other stressors (e.g., warming) in additive, antagonistic, synergistic or reversed forms. The effects of multiple stressors on the resilience of macrophyte-dominated states and on critical thresholds for regime shifts are, however, unknown. We test the effects of individual and combined stressors of warming, nitrate, and various pesticides typically found in agricultural run-off (ARO) on the growth of macrophytes, periphyton, and phytoplankton in microcosms. We applied a one-level replicated design to test whether ARO induces a regime shift and a multifactorial dose-response design to model stressor thresholds and disentangle stressor interactions along a gradient. The individual stressors did not induce a regime shift, but the full ARO did. Nitrate and pesticides acted synergistically, inducing a shift with increasing phytoplankton biomass and decreasing macrophyte biomass. Warming amplified this effect and lowered critical thresholds for regime shifts. Shallow aquatic ecosystems in agricultural landscapes affected by global warming thus increasingly risk shifting to a turbid, phytoplankton-dominated state, and negatively impacting ecosystem service provisioning. Multiple stressor interactions must be considered when defining safe operating spaces for aquatic systems.

Multiple-stressor exposure of aquatic food webs: Nitrate and warming modulate the effect of pesticides.

Shallow lakes provide essential ecological and environmental services but are exposed to multiple stressors, including agricultural runoff (ARO) and climate warming, which may act on different target receptors disrupting their normal functioning. We performed a microcosm experiment to determine the individual and combined effects of three stressors-pesticides, nitrate and climate warming-on two trophic levels representative of communities found in shallow lakes. We used three submerged macrophyte species (Myriophyllum spicatum, Potamogeton perfoliatus, Elodea nuttallii), eight benthic or pelagic microalgal species and three primary consumer species (Daphnia magna, Lymnaea stagnalis, Dreissena polymorpha) with different feeding preferences for benthic and pelagic primary producers. Eight different treatments consisted of a control, only nitrate, a pesticide cocktail, and a combination of nitrate and pesticides representing ARO, each replicated at ambient temperature and +3.5°C, mimicking climate warming. Pesticides negatively affected all functional groups except phytoplankton, which increased. Warming and nitrate modified these effects. Strong but opposite pesticide and warming effects on Myriophyllum drove the response of the total macrophyte biomass. Nitrate significantly suppressed Myriophyllum final biomass, but not overall macrophyte and microalgal biomass. Nitrate and pesticides in combination caused a macrophyte decline, and the system tipped towards phytoplankton dominance. Strong synergistic or even reversed stressor interaction effects were observed for macrophytes or periphyton. We emphasize the need for more complex community- and ecosystem-level studies incorporating multiple stressor scenarios to define safe operating spaces.

Temperature response of aquatic greenhouse gas emissions differs between dominant plant types.

Greenhouse gas (GHG) emissions from small inland waters are disproportionately large. Climate warming is expected to favor dominance of algae and free-floating plants at the expense of submerged plants. Through different routes these functional plant types may have far-reaching impacts on freshwater GHG emissions in future warmer waters, which are yet unknown. We conducted a 1,000 L mesocosm experiment testing the effects of plant type and warming on GHG emissions from temperate inland waters dominated by either algae, free-floating or submerged plants in controls and warmed (+4 °C) treatments for one year each. Our results show that the effect of experimental warming on GHG fluxes differs between dominance of different functional plant types, mainly by modulating methane ebullition, an often-dominant GHG emission pathway. Specifically, we demonstrate that the response to experimental warming was strongest for free-floating and lowest for submerged plant-dominated systems. Importantly, our results suggest that anticipated shifts in plant type from submerged plants to a dominance of algae or free-floating plants with warming may increase total GHG emissions from shallow waters. This, together with a warming-induced emission response, represents a so far overlooked positive climate feedback. Management strategies aimed at favouring submerged plant dominance may thus substantially mitigate GHG emissions.

Heat waves rather than continuous warming exacerbate impacts of nutrient loading and herbicides on aquatic ecosystems.

Submerged macrophytes are vital components in shallow aquatic ecosystems, but their abundances have declined globally. Shading by periphyton and phytoplankton/turbidity plays a major role in this decline, and the competing aquatic primary producers are subject to the complex influence of multiple stressors such as increasing temperatures, nutrient loading and herbicides. Their joint impact has rarely been tested and is difficult to predict due to potentially opposing effects on the different primary producers, their interactions and their grazers. Here, we used 48 mesocosms (2500 L) to simulate shallow lakes dominated by two typical submerged macrophytes, bottom-dwelling Vallisneria denseserrulata and canopy-forming Hydrilla verticillata, and associated food web components. We applied a combination of nutrient loading, continuous warming, heat waves and glyphosate-based herbicides to test how these stressors interactively impact the growth of submerged macrophytes, phytoplankton and periphyton as competing primary producers. Warming or heat waves alone did not affect phytoplankton and periphyton abundance, but negatively influenced the biomass of V. denseserrulata. Nutrient loading alone increased phytoplankton biomass and water turbidity and thus negatively affected submerged macrophyte biomass, particularly for V. denseserrulata, by shading. Glyphosate alone did not affect biomass of each primary producer under ambient temperatures. However, heat waves facilitated phytoplankton growth under combined nutrient loading and glyphosate treatments more than continuous warming. As a consequence, H. verticillata biomass was lowest under these conditions indicating the potential of multiple stressors for macrophyte decline. Our study demonstrated that multiple stressors interactively alter the biomass of primary producers and their interactions and can eventually lead to a loss of macrophyte communities and shift to phytoplankton dominance. These results show the risks in shallow lakes and ponds in agricultural landscapes and underline the need for multiple stressor studies as a base for their future management.

Eye fluke infection changes diet composition in juvenile European perch (Perca fluviatilis).

Intraspecific diet specialization, usually driven by resource availability, competition and predation, is common in natural populations. However, the role of parasites on diet specialization of their hosts has rarely been studied. Eye flukes can impair vision ability of their hosts and have been associated with alterations of fish feeding behavior. Here it was assessed whether European perch (Perca fluviatilis) alter their diet composition as a consequence of infection with eye flukes. Young-of-the-year (YOY) perch from temperate Lake Müggelsee (Berlin, Germany) were sampled in two years, eye flukes counted and fish diet was evaluated using both stomach content and stable isotope analyses. Perch diet was dominated by zooplankton and benthic macroinvertebrates. Both methods indicated that with increasing eye fluke infection intensity fish had a more selective diet, feeding mainly on the benthic macroinvertebrate Dikerogammarus villosus, while less intensively infected fish appeared to be generalist feeders showing no preference for any particular prey type. Our results show that infection with eye flukes can indirectly affect interaction of the host with lower trophic levels by altering the diet composition and highlight the underestimated role of parasites in food web studies.