Article
To read the full-text of this research, you can request a copy directly from the authors.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... To perform a prediction competition as presented here, we aimed to build data sets with around 200 samples. This number is needed to arrive at stable and reliable estimation results as suggested by Feld et al. (2016). A high number of records is especially important if the number of stressors M increases (in our case varying between 7 and 10). ...
... Another approach, not applied here, has been followed by Feld et al. (2016), Gieswein et al. (2017) and Herrero et al. (2018) who propose to analyze multiple-stressor models as a two-stage procedure: ...
... And these differences are not all. Data sets will differ as for (i) the number of samples, (ii) the presence or absence of extreme EQRs, the width of gradients in stressors, transformations of stressor indicators, such as Box-Cox transformations (Feld et al., 2016). ...
Article
Full-text available
What policy is needed to ensure that good-quality water is available for both people's needs and the environment? The EU Water Framework Directive (WFD), which came into force in 2000, established a framework for the assessment, management, protection and improvement of the status of water bodies across the European Union. However, recent reviews show that the ecological status of the majority of surface waters in the EU does not meet the requirement of good status. Thus, it is an important question what measures water management authorities should take to improve the ecological status of their water bodies. To find concrete answers, several institutes in the Netherlands cooperated to develop a software tool, the WFD Explorer, to assist water managers in selecting efficient measures. This article deals with the development of prediction tools that allow one to calculate the effect of restoration and mitigation measures on the biological quality, expressed in terms of Ecological Quality Ratios (EQRs). To find the ideal modeling tool we give a review of 11 predictive models: 10 models from the field of Machine Learning and, additionally, the Multiple Regression model. We present our results in terms of a ‘prediction-interpretation competition’. All these models were tested in a multiple-stressor setting: the values of 15 stressors (or steering factors) are available to predict the EQR values of four biological quality elements (phytoplankton, other aquatic flora, benthic invertebrates and fish). Analyses are based on 29 data sets from various water clusters (streams, ditches, lakes, channels). All 11 models were ranked by their predictive performance and their level of model transparency. Our review shows a trade-off between these two aspects. Models that have the best EQR prediction performance show non-transparent model structures. These are Random Forest and Boosting. However, models with low prediction accuracies show transparent response relationships between EQRs on the one hand and individual steering factors on the other hand. These models are Multiple Regression, Regression Trees and Product Unit Neural Networks. To acknowledge both aspects of model quality – predictive power and transparency – we recommend that models from both groups are implemented in the WFD Explorer software.
... Three datasets were not assigned to regional groups due to insufficient replication and/or impact gradient length ( Figure S1; Table S1). We also analysed four individual datasets (CY, ES_NE1, ES_NE2 and ES_S), which represented all datasets with sufficient impact and drying gradients to warrant individual analysis (Feld et al., 2016; Figure S1). ...
... We used marginal and conditional goodness-of-fit statistics (R 2 m and R 2 c , respectively) to evaluate model performance (Mac Nally et al., 2018). Independent (single, additive) and interactive (antagonistic, opposing, synergistic) response types were classified using the sign and significance of responses to predictors and their interactions (Feld et al., 2016). We used significance levels of p < 0.01 and <0.001 for response variables violating one or both of the assumptions of normality and homoscedasticity, respectively. ...
... Characterization of distinctive regional river types and their associated communities is a priority to underpin improvements in temporary river biomonitoring and management (Clarke et al., 2003;Stubbington et al., 2018). In addition, our capacity to detect biological responses to impacts was hampered by a short impact gradient in the Atlantic region (Feld et al., 2016), highlighting the need to collect data representing the full range of impact levels experienced across the breadth of European temporary rivers. ...
Article
Full-text available
Rivers are dynamic ecosystems in which both human impacts and climate‐driven drying events are increasingly common. These anthropogenic and natural stressors interact to influence the biodiversity and functioning of river ecosystems. Disentangling ecological responses to these interacting stressors is necessary to guide management actions that support ecosystems adapting to global change. We analysed the independent and interactive effects of human impacts and natural drying on aquatic invertebrate communities—a key biotic group used to assess the health of European freshwaters. We calculated biological response metrics representing communities from 406 rivers in eight European countries: taxonomic richness, functional richness and redundancy, and two biomonitoring indices that indicate ecological status. We analysed metrics based on the whole community and a group of taxa with traits promoting resistance and/or resilience (‘high RR’) to drying. We also examined how responses vary across Europe in relation to climatic aridity. Most community metrics decreased independently in response to impacts and drying. A richness‐independent biomonitoring index (the average score per taxon; ASPT) showed particular potential for use in biomonitoring, and should be considered alongside new metrics representing high RR diversity, to promote accurate assessment of ecological status. High RR taxonomic richness responded only to impacts, not drying. However, these predictors explained little variance in richness and other high RR metrics, potentially due to low taxonomic richness. Metric responsiveness could thus be enhanced by developing region‐specific high RR groups comprising sufficient taxa with sufficiently variable impact sensitivities to indicate ecological status. Synthesis and applications. Metrics are needed to assess the ecological status of dynamic river ecosystems—including those that sometimes dry—and thus to identify priority sites requiring action to tackle the causes of environmental degradation. Our results inform recommendations guiding the development of such metrics. We propose concurrent use of richness‐independent ‘average score per taxon’ indices and metrics that characterize the richness of resistant and resilient taxa. We observed interactions between aridity, impacts and drying, highlighting that these new metrics should be region‐specific, type‐specific and adaptable, promoting their ability to inform management actions that protect biodiversity in river ecosystems responding to climate change.
... The dependence of combined effect types on scales suggests that the observed combined effect types are not solely dependent on the environmental setting, but also on the sampling strategy. An increase in scale can be associated with an increase in the size of datasets or the stressor gradient length (e.g. an increase in the temperature gradient length from 15 -22 • C to 15 -31 • C). Feld et al. (2016) showed that sample size and stressor gradient in survey-based multiple stressor studies needed to be sufficient to accurately detect the combined stressor effect type (sample size ≥ 150 and gradient length ≥ 75 % of the prevalent gradient). However, systematic analyses of the role of the stressor gradient length on multiple stressor effects are lacking. ...
... All analyses were conducted in R (version 4.0.3, R Core Team) based on the approach suggested by Feld et al. (2016) to assess the impacts of multiple stressors and the analytical procedure detailed in Birk et al. (2020). The following provides a short overview of the data processing, modelling, model evaluation and statistical synthesis. ...
... Further, stressor correlation was investigated using a correlation matrix chart (Peterson and Carl, 2020). Cases with a Spearman correlation of ≥ 0.7 were excluded to avoid collinearity problems (Feld et al., 2016). ...
Article
Multiple stressors are continuously deteriorating surface waters worldwide, posing many challenges for their conservation and restoration. Combined effect types of multiple stressors range from single-stressor dominance to complex interactions. Identifying prevalent combined effect types is critical for environmental management, as it helps to prioritise key stressors for mitigation. However, it remains unclear whether observed single and combined stressor effects reflect true ecological processes unbiased by sample size and length of stressor gradients. Therefore, we examined the role of sample size and stressor gradient lengths in 158 paired-stressor response cases with over 120,000 samples from rivers, lakes, transitional and marine ecosystems around the world. For each case, we split the overall stressor gradient into two partial gradients (lower and upper) and investigated associated changes in single and combined stressor effects. Sample size influenced the identified combined effect types, and stressor interactions were less likely for cases with fewer samples. After splitting gradients, 40 % of cases showed a change in combined effect type, 30 % no change, and 31 % showed a loss in stressor effects. These findings suggest that identified combined effect types may often be statistical artefacts rather than representing ecological processes. In 58 % of cases, we observed changes in stressor effect directions after the gradient split, suggesting unimodal stressor effects. In general, such non-linear responses were more pronounced for organisms at higher trophic levels. We conclude that observed multiple stressor effects are not solely determined by ecological processes, but also strongly depend on sampling design. Observed effects are likely to change when sample size and/or gradient length are modified. Our study highlights the need for improved monitoring programmes with sufficient sample size and stressor gradient coverage. Our findings emphasize the importance of adaptive management, as stress reduction measures or further ecosystem degradation may change multiple stressor-effect relationships, which will then require associated changes in management strategies.
... These groups of variables characterize a hierarchical arrangement (catchment, riparian, reach, and microhabitat) usually present in river systems (Allan, 2004). Also, we included a set of natural descriptors to account for natural variability and quantify their effects on the calculated metrics (Feld et al., 2016b). ...
... To select the most ecologically meaningful stressors, we employed different statistical approaches. First, we examined independence among explanatory variables following Feld et al. (2016aFeld et al. ( , 2016b. We used Shapiro-Wilk tests to assess normality and stressors were transformed (square-root, logarithm, logit) if necessary to meet normality. ...
... RF is a technique allowing continuous, categorical and proportional variables during modelling and it is capable of analysing complex interactions and non-linear relationships (Breiman, 2001). All RFs were adequately parametrised (Feld et al., 2016b) and run with the 'randomForestSRC' package (Ishwaran and Kogalur, 2017). ...
Article
However, knowledge about multiple-stressors effects on urbanised Andean streams is lacking. In southern Ecuador, we assessed how multiple stressors determine the structural (aquatic invertebrate metrics) and functional (organic matter breakdown and delta N of primary consumers) attributes of streams in a densely populated watershed without wastewater treatment and with contrasting land uses. We found that urbanised streams exhibited individual-stressor effects and that stressor interactions were rare. While structural and function attributes responded negatively to urbanisation, ecosystem functioning metrics were influenced most. Stream ecosystem functions were influenced by water-chemistry stressors, whereas aquatic invertebrate metrics were influenced by physical-habitat stressors. We suggest that managers of urbanised streams in the Andes immediately focus on the most important stressors by reducing inputs of inorganic N and P, re-establishing stream flow and substrate heterogeneity, and restoring riparian vegetation instead of attempting to elucidate intricate interactions among stressors. Our result also demonstrate that stream biomonitoring programs would benefit from a combination of structural and functional indicators to assess anthropogenic effects in a multiple-stressors scenario.
... Evidence of changing trophic states and the creation of these nutrient criteria are considered essential for controlling coastal eutrophication (Xie et al., 2021). Likewise, understanding the effects of multiple stressors is necessary for managing coastal waters and related ecosystems (Feld et al., 2016). ...
... The approach of multiple stressors is fundamental for the interpretation of anthropogenic effects on estuarine biotic communities. This is because the analysis combines adequate pressure and stressor variables (Feld et al., 2016). ...
... However, there is a lack of robust analytical frameworks and guidance to analyze these data in the context of multiple stressors (Feld et al., 2016). Multivariate trophic indices provide an efficient way to assess and classify the eutrophication level and ecological status of a given water body, but the use of only traditional methods that are based on a nutrient index has led to overestimated results and ambiguous interpretations (Xiao et al., 2007;Béjaoui et al., 2018). ...
Article
Full-text available
Modeling approaches are useful tools for assessing the general state of eutrophication and understanding the effects of multiple stressors on coastal ecosystems. Thus, we hypothesized that anthropogenic pressures and multiple stressors would increase the trophic state of macrotidal estuaries. Datasets from long-term (2012−2020) environmental and biological monitoring in the Itapecuru River estuary (IRE) were analyzed using the pressure-state-response (PSR) approach and nonlinear methods. Our results indicated a low dilution of nutrients and a moderate flushing potential of urban effluent (2150.94 ton year⁻¹) in the estuary. The estuary was consequently classified as being in a high trophic state, being susceptible to and suffering high estuarine pressure to develop eutrophic symptoms. The Assessment of estuarine trophic status (ASSETS) model indicates that eutrophication is seasonal and depends on climatic variation. La Niña events (2019–2020) contributed to chlorophyll-a (>40 μg L⁻¹) and orthophosphate (0.04 mg L⁻¹) concentrations, principally during periods of low river discharge. According to the GAM's model, brackish waters (salinity>10) with high temperatures (> 30 °C) and high dissolved oxygen (>4 mg L⁻¹) have more intense trophic conditions, especially in the mixing zone. The low dissolved oxygen (DO) levels (DO<3 mg L⁻¹) and high concentrations of chlorophyll-a in the seawater zone indicate that the lower portion of the estuary was the most susceptible. In addition, the random forest model selected salinity, DIP and Chl-a as the principal stressors that intensified eutrophication in the macrotidal systems. According to the ASSETS final ranking (worsen-high) for the next decade (2021−2031), the primary planned strategies should be to reduce anthropogenic contributions and improve trophic conditions in the IRE. From these results, the interactions and predictions of eco-hydrological effects could facilitate the characterization of future risks and the management of macrotidal estuarine systems.
... Further efforts in Iberian multiple-stressor research should be directed to (i) intensify the study of lentic systems, (ii) explore more observational data, (iii) autotrophic organisms and (vi) biodiversity-ecosystem functioning responses, and (v) cover a wider range of stressors and (vi) more complex interacting stressor scenarios. sidered simultaneously (Côté et al., 2016;Crainet al., 2008;Feld et al., 2016;Piggott et al., 2015). In these cases, combined effects can be additive (i.e., non-interactive), when the combined effect equals the sum of the individual stressor effects, or interactive, when the combined effect is lower (antagonism) or greater (synergism) than the expected additive effect. ...
... However, the context-dependency and high variability of some multiple-stressor effects call for further investigation to provide more robust foundations for freshwater management and restoration (Feld et al., 2018). In addition, multiple-stressor effects also depend on the spatial variation of stressor gradient lengths as well as on the biogeographic, socio-economic and environmental settings of each region (Feld et al., 2016;Newbold et al., 2020). Thus, there is an urgent need to provide regional-scale assessments of multiple-stressor effects to reduce such context-dependency and provide more solid guidance to water managers Côté et al., 2016). ...
... The category "other environmental factors" included variables typically unaffected by anthropogenic activity (e.g., altitude, lithology, slope), which are used in some studies as co-variates to better predict biological or ecosystem responses. For simplicity, we used the term stressor in a wide context including also overarching anthropogenic pressures (e.g., land-use), although we acknowledge that their impacts act at different temporal and spatial scales (Feld et al., 2016;Jackson et al., 2021). In all cases, we defined stressors as abiotic or biotic environmental conditions exceeding the normal range experienced by organisms, which cause potential injurious changes to biological systems (Bijlsma & Loeschcke, 2005). ...
Article
Full-text available
Freshwater ecosystems are exposed to an increasing number of stressors, challenging their biomonitoring and management. Despite recent advances in multiple-stressor research, regional-scale assessments in areas with high freshwater biodiversity and increasing anthropogenic pressure are urgently needed. We reviewed 61 studies focused on freshwater individuals, populations and communities from the Iberian Peninsula to (i) quantify the frequency of experimental approaches used (manipulative, observational), aquatic systems, biological organization levels, and types of organisms and modelled responses, (ii) identify key individual stressors and the frequency of significant positive (increase in response magnitude) and negative (decrease) effects and (iii) determine types of interacting stressors and the frequency of their combined effects. Our dataset comprised 409 unique responses to 13 types of individual stressors, 34 stressor pairs and 12 high-order interactions (3- and 4-way). We found a higher prevalence of manipulative (85 %) respect to observational studies, and a greater focus on lotic systems (59 %) and heterotrophic organisms (58 %). The most studied stressors were nutrient (Nutr), thermal (Therm), hydrologic (Hydr), ultraviolet radiation (UVR), toxic (Toxic) and salinity (Sal) stress and land-use pressure. Individual stressors showed a higher proportion of negative (34 %) than positive effects (26 %). Nutr × UVR, Toxic × Toxic, Therm × Toxic, Hydr × Toxic, Sal × Therm, and Nutr × Therm were the most studied stressor pairs. Non-interactive (50 %) and interactive responses (50 %) were balanced. Antagonistic effects (18 %) were slightly more common than synergisms (15 %), reversal or opposing (13 %) and high-order interactions (4 %). Such proportions varied within experimental approaches, biological organization levels and organism types. Our findings are helpful to manage certain stressor combinations in Iberian freshwaters. Further efforts in Iberian multiple-stressor research should be directed to (i) intensify the study of lentic systems, (ii) explore more observational data, (iii) autotrophic organisms and (vi) biodiversity-ecosystem functioning responses, and (v) cover a wider range of stressors and (vi) more complex interacting stressor scenarios.
... To address more specifically the extent to which diversity and abiotic factors alone determine E. canadensis abundance, as a second complementary analytical step, we used a combination of boosted regression tree analysis (BRT; Elith, Leathwick, & Hastie, 2008) and random forest analysis (RFA; Breiman, 2001). BRT was used to partition the variation in E. canadensis abundance explained by diversity and environmental descriptors alone, and how they might together reflect habitat heterogeneity at the landscape scale Feld, Segurado, & Gutiérrez-Cánovas, 2016). BRT constitutes a machine-learning method that combines classical regression tree analysis with boosting (Elith et al., 2008). ...
... BRT was ideal for our study as it can accommodate collinear data (e.g., latitude and longitude) and han- To reduce any spatial autocorrelation in the data arising due to the underlying hydrological network and to evaluate whether the importance of diversity and abiotic predictors in explaining E. canadensis abundance shifted with degree of lake connectivity and eutrophication, we ran independent pBRTs for each lake group using the "dismo" (Hijmans, Phillips, Leathwick, Elith, & Hijmans, 2017), and RFAs were then used to assess the extent to which diversity predictors explain E. canadensis abundances through time. Like BRTs, RFA is suited to analysing non-linear relationships by fitting several models (regression trees) to bootstrapped data subsets with the advantage of handling datasets with a low number of observations and predictors, that is, our palaeo-data (Feld, Segurado, & Gutiérrez-Cánovas, 2016). RFAs were run using the function rfsrc of the package "randomForestSRC" (Ishwaran & Kogalur, 2016). ...
... dle linear and non-linear descriptors with missing values. BRT partitioning (pBRT) was assessed through an additive partial regression scheme followingFeld, Birk, et al. (2016) andFeld, Segurado, & Gutiérrez-Cánovas (2016). This analysis decomposed each BRTexplained variation into four fractions: (a) pure diversity, (b) pure abiotic, (c) shared diversity/abiotic, and (d) unexplained variation. ...
Article
Full-text available
Macrophyte invasive alien species (IAS) fitness is often hypothesised to be associated with beneficial environmental conditions (environmental matching) or species-poor communi- ties. However, positive correlations between macrophyte IAS abundance and native plant richness can also arise, due to habitat heterogeneity (defined here as variation in abiotic and native biotic conditions over space and time). We analysed survey and pal- aeoecological data for macrophytes in satellite lakes along the Upper Lough Erne (ULE) system (Northern Ireland, UK), covering a gradient of eutrophication and connectivity to partition how environmental conditions, macrophyte diversity and habitat heterogeneity explained the abundance of Elodea canadensis, a widely distributed non-native macrophyte in Europe. E. canadensis abundance positively correlated with macrophyte richness at both the within- and between-lake scales indicating coexistence of native and invasive species over time. E. canadensis was also more prolific in highly connected and macrophyte-rich lakes, but sparser in the more eutrophic-isolated ones. Partial boosted regression trees rev- ealed that in eutrophic-isolated lakes, E. canadensis abundances correlated with water clar- ity (negatively), plant diversity (positively), and plant cover (negatively) whereas in diverse- connected lakes, beta diversity (both positively and negatively) related to most greatly E. canadensis abundance. Dense macrophyte cover and unfavourable environmental condi- tions thus appear to confer invasibility resistance and sufficient habitat heterogeneity to mask any single effect of native biodiversity or environmental matching in controlling E. canadensis abundance. Therefore, in shallow lake landscapes, habitat heterogeneity vari- ously enables the coexistence of native macrophytes and E. canadensis, reducing the often-described homogenisation effects of invasive macrophytes.
... Interactions may lead to synergistic responses, when joint stressor effects are stronger than the sum of their isolated effects, or antagonistic responses when joint stressor effects are weaker than the sum of their isolated effects (Coors and De Meester, 2008;Crain et al., 2008;Belarde and Railsback, 2016;Feld et al., 2016). Alternatively, in the absence of interactions, stressor joint effects are equal to the sum of their isolated effects, leading to an additive response. ...
... Here, we claim that context-dependent responses to stressors may also arise from multiple stressor gradients being covered inconsistently in different regions or time frames. In this sense, Feld et al. (2016) highlighted the need to cover at least 75% of the stressors gradient to identify the most important interactions. Here, using simulations, we analyze more thoroughly how encompassing different stressor gradients may affect the identification and parameterization of interactions among stressors. ...
... Joint effects of pairs of variables are often represented by 'heat-map' contour plots, where the response is expressed as a surface of varying color intensities against two explanatory variables (Feld et al., 2016). The patterns that arise from these plots are dependent on the relative coefficient magnitude and sign of each stressor and their interaction terms in the model. ...
Article
This study aims at understanding how observed inconsistencies in the response of biotic indicators to multiple stressors may result from different stressor gradient lengths being represented at different areas or temporal windows, either as the result of intrinsic natural causes or as the result of sampling bias. We simulated a pool of sites showing five types of interactive responses of indicators to two co-occurring virtual stressors, as well as several sampling constraints, resulting in different portions of each stressor's gradient being covered. The sampled gradient length showed a strong influence on the detection of single stressor effects, both in terms of statistical significance and goodness-of-fit. Increasing constraints on gradient coverage also led to an increasingly deficient identification of stressor interactions. The fail in detecting significant interactions largely dominated over switches between interaction types. The simulations indicated that datasets not fully capturing stressor gradients may hinder the ability to unveil underlying multiple stressor effects. As distinct portions of stressor gradients may be present at different contexts and may change over time, our simulations stress the importance of adaptive management strategies based on robust sampling designs to minimize potential statistical artefacts and uncertainties.
... While resources for environmental conservation are usually restricted, the co-occurrence of multiple stressors is a challenge for ecosystem management where priorities are defined through mitigation actions. In such situations, disentangling the effects of multiple stressors is a preliminary requirement in order to focus restoration efforts on the dominant anthropogenic stressors and their impacts (Feld et al. 2016, Teichert et al. 2016. The complex relationships between the ecological response and stressors can lead to unexpected or ineffective management outcomes (Paine et al. 1998), especially when non-linear and interactive settings are involved (Hewitt et al. 2016, Samhouri et al. 2017). ...
... hydro-morphological alterations and pollution commonly occur in urban areas), the collinearity between stressor descriptors prevents the determination of their relative contribution to the response variation (Dormann et al. 2013, Colin et al. 2018. In this situation, the number of explanatory variables can be reduced by selecting the most influential ones based on preliminary analysis or a priori knowledge (Feld et al. 2016), or by developing a synthetic disturbance index (Teichert et al. 2018b). Nevertheless, fish responses to human pressures can be evaluated at various levels of organisation, from molecular to community levels, so that suitable methods should be implemented depending on the study context and the response variables. ...
... additive, synergistic or antagonistic). Under these circumstances, Feld et al. (2016) proposed an analytical framework based on GLMs to disentangle the effects of multiple stressors based on monitoring surveys, such as those conducted in estuaries. ...
... While resources for environmental conservation are usually restricted, the co-occurrence of multiple stressors is a challenge for ecosystem management where priorities are defined through mitigation actions. In such situations, disentangling the effects of multiple stressors is a preliminary requirement in order to focus restoration efforts on the dominant anthropogenic stressors and their impacts (Feld et al. 2016, Teichert et al. 2016. The complex relationships between the ecological response and stressors can lead to unexpected or ineffective management outcomes (Paine et al. 1998), especially when non-linear and interactive settings are involved (Hewitt et al. 2016, Samhouri et al. 2017). ...
... hydro-morphological alterations and pollution commonly occur in urban areas), the collinearity between stressor descriptors prevents the determination of their relative contribution to the response variation (Dormann et al. 2013, Colin et al. 2018. In this situation, the number of explanatory variables can be reduced by selecting the most influential ones based on preliminary analysis or a priori knowledge (Feld et al. 2016), or by developing a synthetic disturbance index (Teichert et al. 2018b). Nevertheless, fish responses to human pressures can be evaluated at various levels of organisation, from molecular to community levels, so that suitable methods should be implemented depending on the study context and the response variables. ...
... additive, synergistic or antagonistic). Under these circumstances, Feld et al. (2016) proposed an analytical framework based on GLMs to disentangle the effects of multiple stressors based on monitoring surveys, such as those conducted in estuaries. ...
Chapter
This chapter addresses the use of fish as indicators of environmental health. The main anthropogenic pressures impacting estuarine fishes are reviewed, as well as the main types of responses by fishes at different levels of biological organisation. Fishes have been widely used to assess estuarine health through different methodological approaches, namely comparisons with historical data or reference conditions, experimental approaches, environmental impact or risk assessment methods, as well as qualitative or quantitative indicators and models. A large number of multi‐metric indices based on fish have been proposed and are routinely used in environmental assessments, although to disentangle natural variability from anthropogenic pressures in a multi‐stress context of global change is still a major challenge.
... Multiple linear regression on the other hand can identify stressor interaction types from the direction and magnitude of estimated model coefficients (e.g. Thrush et al., 2008b;Feld et al., 2016;Ellis et al., 2017a;Ellis et al., 2019;Birk et al., 2020). Although this provides a way to identify additive, antagonistic, or synergistic effects, the form of stressor interactions along increasing gradients of stress has only recently been explored. ...
... Although this provides a way to identify additive, antagonistic, or synergistic effects, the form of stressor interactions along increasing gradients of stress has only recently been explored. Visualising these effects across gradients of stress may aid better interpretation and communication of multiplicative effects (Feld et al., 2016). ...
... This is problematic for management and consistency is needed. Previous studies using regression models distinguish additive, antagonistic, and synergistic effects from model coefficient estimates (but see Feld et al., 2016). To aid transparency and consistency in categorising stressor interactions, visualisation of stressor effects through 3D-PDP reduces the risk of mis-interpretation by showing the direction, magnitude, and gradients of change (Feld et al., 2016). ...
Article
To enable environmental management actions to be more effectively prioritized, cumulative effects between multiple stressors need to be accounted for in risk-assessment frameworks. Ecological risk and uncertainty are generally high when multiple stressors occur. In the face of high uncertainty, transparent communication is essential to inform decision-making. The impact of stressor interactions on risk and uncertainty was assessed using generalized linear models for additive and multiplicative effect of six anthropogenic stressors on the abundance of estuarine macrofauna across New Zealand. Models that accounted for multiplicative stressor interactions demonstrated that non-additive effects dominated, had increased explanatory power (6 to 73 % relative increase between models), and thereby reduced the risk of unexpected ecological responses to stress. Secondly, 3D-plots provide important insights in the direction, magnitude and gradients of change, and aid transparency and communication of complex stressor effects. Notably, small changes in a stressor can cause a disproportionally steep gradient of change for a synergistic effect where the tolerance to stressors are lost, and would invoke precautionary management. 3D-plots were able to clearly identify directional shifts where the nature of the interaction changed from antagonistic to synergistic along increasing stressor gradients. For example, increased nitrogen load and exposure caused a shift from positive to negative effect on the abundance of a deposit-feeding polychaete (Magelona). Assessments relying on model coefficient estimates, which provide one effect term, could not capture the complexities observed in 3D-plots and are at risk of mis-identifying interaction types. Finally, visualising model uncertainty demonstrated that although error terms were higher for multiplicative models, they better captured the uncertainty caused by data availability. Together, the steep gradients of change identified in 3D-plots and the higher uncertainty in model predictions in multiplicative models urges more conservative limits to be set for management that account for risk and uncertainty in multiple stressor effects.
... BRT is a combination of statistical and machine learning technique, that optimize predictive performance by using boosting to combine large numbers of relatively simple tree models adaptively (Elith et al. 2008). BRT can be used in water resources studies to model and analyze the linkage between stressors and indicators, either biological [e.g., (Feld et al. 2016;Segurado et al. 2018)], or hydrological [e.g., (Mazor et al. 2018;Radinger et al. 2018)]. ...
... Empirical modeling and linkages between pressures and discharge were performed based on the guidance for the analytical process and the interpretation of results proposed by Feld et al. (Feld et al. 2016), although these were modified to meet the needs of the present study. BRT modeling was performed with the packages gbm (Ridgeway 2010; Greenwell et al. 2020) and dismo (Elith et al. 2008;Hijmans et al. 2020) in R v.4.1.0 ...
... Outliers and variable skewness were checked. In almost all predictors, a left skewness was observed, and therefore, a data transformation was applied to allow the data to approximate a normal distribution (Feld et al. 2016). After the transformation procedure, all variables were centered and standardized to allow them to be in a comparable numerical range. ...
Article
Full-text available
The quantitative and qualitative status of a riverine ecosystem is mainly controlled by the characteristics of the catchment (topography, land use, and geological structure) and climatic factors (precipitation and temperature), both of which can be affected by anthropogenic activities. To quantify the impact of humanly imposed pressures on river discharge and to determine the dominant stressors that affect the hydrological regime of Greek rivers, two discharge datasets consisting of discharge/ichthyofauna and discharge/benthic macroinvertebrate measurements were employed, and 36 predictor variables were statistically elaborated. Impacted samplings were defined based on the classification of the corresponding biotic indices sensitive to hydrological alterations. The assessment indicated that the median discharge of impacted rivers was lower in relation to unimpacted conditions in all river types by up to 85% (R-M3), except for the case of R-M5 (temporary rivers), where discharge was higher by up to 46%. The most important variables affecting discharge values were the proximity of the dam upstream of the sampling site, the distance to source, the catchment area upstream, the presence of siliciclastic rocks upstream, annual precipitation, and the presence of artificial surfaces. Surprisingly, irrigated land area and water abstractions volume were not indicated as major driving variables affecting the hydrology of Greek rivers, possibly due to limitations of the current methodological approach. The development of a hydrological regime alteration index, specifically for Greek rivers, based on the deviation of the current state from the unimpacted conditions can be a valuable tool for the implementation of Water Framework Directive 2000/60 objectives concerning the hydromorphological quality of riverine ecosystems.
... Whilst process-based models realistically represent the sensitivities of the system to key drivers (Hrachowitz et al. 2014) and capture the short term dynamics (Jankowski et al. 2021), empirical models have long-standing pedigree in providing insight into ecosystems' response to multiple stressors (Izagirre et al. 2008;Beaulieu et al. 2013). Specifically, with the development of machine learning techniques, empirical models can utilize data to learn and increasingly improve model performance (Elith et al. 2008;Feld et al. 2016). Therefore, in addition to the process-based model, we also use an empirical approach to assess the sensitivity of modeled metabolism rates to multiple stressors. ...
... GLS was used to account for the residual autocorrelation using the nlme package (Pinheiro et al. 2017). Selection of relevant predictors for GLS models was carried out following Feld et al. (2016). Specifically, we performed an exploratory analysis using the random forest (Breiman 2001) machine learning technique (randomForestSRC package, Ishwaran and Kogalur 2017) to derive the hierarchy of the most influential stressors and interactions that explain GPP and ER dynamics. ...
... Empirical approaches also provide information about important environmental stressors and their interactions for GPP and ER variation. These established relationships between metabolism rates and environmental stressors can be useful to infer the degradation or recovery of river health following management actions (Jankowski et al. 2021), although the variable importance and effect sizes of environmental stressors should be considered (Feld et al. 2016). A process-based approach, on the other hand, presents a readily available tool to study river ecosystem functioning in response to changing multiple environmental stressors (Heathwaite 2010). ...
Article
Full-text available
High‐resolution monitoring of water quality and ecosystem functioning over large spatial scales in expansive lowland river catchments is challenging. Therefore, we need modeling tools to predict these processes at locations where observations are absent. Here, we present a new approach to estimate ecosystem metabolism underpinned by a high‐resolution, process‐based model of in‐stream flows and water quality. The model overcomes the current challenges in metabolism modeling by accounting for oxygen transport under varying flows and oxygen transformations due to biogeochemical processes. We implement the model in a 62‐km‐long stretch of the River Thames, England, using observations spanning 2 yr. Model outputs suggest that the river is primarily autotrophic from mid‐spring to mid‐summer due to high biomass during low‐flow periods, and is heterotrophic during the rest of the year. Ecosystem respiration in upstream reaches is driven mainly by biochemical oxygen demand, autotrophic respiration, and nitrification processes, whereas downstream sites also show a control of benthic oxygen demand in addition to the aforementioned processes. Using empirical modeling, we analyze the sensitivity of our estimated metabolism rates to multiple environmental stressors. Results demonstrate that empirical models could be useful for rapid river health assessments, but need improvements to reproduce peak metabolism rates. The process‐based model, although more complex than existing in situ approaches to metabolism quantification, allows inference when gaps in continuous observations are present. The model offers additional benefits for predicting metabolism rates under future scenarios of environmental change incorporating multiple stressor effects.
... We identified the most appropriate distributions for all variables using the fitdistrplus package (Delignette-Muller and Dutang, 2015; Table 1). We used an AICc model selection approach following Feld et al. (2016). All predictor variables were transformed to best approximate normal distributions and scaled to have a mean of zero and standard deviation of 1 (zscale function) and assessed for collinearity (library usdm, vifstep function), ensuring variance inflation factors were ≤ 5 (Feld et al., 2016). ...
... We used an AICc model selection approach following Feld et al. (2016). All predictor variables were transformed to best approximate normal distributions and scaled to have a mean of zero and standard deviation of 1 (zscale function) and assessed for collinearity (library usdm, vifstep function), ensuring variance inflation factors were ≤ 5 (Feld et al., 2016). Using fitdistrplus (Delignette-Muller and Dutang, 2015) after transformation showed that transformed variables best fit a normal distribution. ...
Article
Full-text available
Urbanization in stream catchments can have strong effects on stream channel hydrogeomorphic features including channel dimensions, channel-floodplain connectivity, and flood regime. However, the consequences of hydrogeomorphic alterations on aquatic-terrestrial subsidy dynamics are largely unexplored. We examined the associations among hydrogeomorphic characteristics, emergent aquatic insect assemblages, and the density and trophic dynamics of riparian spiders of the family Tetragnathidae at 23 small urban stream reaches in the Columbus, OH (United States) Metropolitan Area. Naturally abundant stable isotopes of ¹³ C and ¹⁵ N were used to quantify the relative contribution of aquatically derived energy (i.e., nutritional pathways deriving from algae) to tetragnathid spiders and their trophic position. Bankfull discharge was negatively related to both emergence rate and family richness. On average, tetragnathid spiders relied on aquatically derived energy for 36% of their nutrition, with the greatest reliance found for spiders next to channels with wider flood-prone widths and proportionally fewer emergent insects of the family Chironomidae. Mean emergent aquatic insect reliance on aquatically derived energy was 32% and explained 44% of the variation in tetragnathid aquatically derived energy. A positive relationship between δ ¹³ C of tetragnathid spiders and emergent insects provides additional evidence of tetragnathid reliance on emergent insects. Mean tetragnathid trophic position was 2.85 and was positively associated with channel sinuosity and negatively associated with aquatic insect emergence rate. Sinuosity was also positively related to aquatically derived energy of emergent aquatic insects; as well as emergent insect family richness; % Ephemeroptera, Plecoptera, and Trichoptera (EPT); and aquatic insect emergence rate; implicating instream habitat-mediated shifts in emergent aquatic insect communities as an indirect mechanistic link between hydrogeomorphic processes and spiders. Our findings underscore that the impacts of stream hydrogeomorphic alterations can cascade into terrestrial food webs. These results suggest that monitoring and restoration of fluvial geomorphic form and function (e.g., sinuosity, slope, and hydrology) confer benefits to both aquatic and terrestrial riparian ecosystems in urban catchments.
... The master thesis of Alex Piro (2020) served as an example for the statistical analysis Erwin Lautsch assisted the analysis as advisor and supporter. Furthermore, a cookbook for multi-stressor analysis in aquatic biomonitoring (Feld et al. 2016) and peer reviewed paper with an approach similar to this thesis (Hayes et al. 2021) was consulted. ...
... Target variables have no normal distribution, so that parametric statistical test were not possible, and non-parametric tests had to be used. Further for this reason, a statistical solution was chosen, transforming the target variables (di-and trichotomizing the data) to gain a ordinally-scaled normal distribution of continuous data, downweighing extreme values and outliers (Feld et al. 2016). ...
Thesis
The brown trout is a near threatened species in Austrian riverine ecosystems. This calls for timely protection measures. Yet, as the effects of climate change become more alarming, and subsequently hydropower development more interesting, brown trout populations are increasingly threatened. In particular, hydropower induced sub-daily flow fluctuations (hydropeaking) are known to entail adverse effects for fish. This thesis aims to investigate the impact of hydropeaking on the brown trout’s population level, also including multiple anthropogenic stressors as well as natural parameters. Electrofishing data collected in 108 different streams and rivers of the rhithral section of Austria, during 125 fishing occasions, were evaluated. To test different hypotheses, statistical tests such as correlations, visual methods, cross-tabulations and classification and regression tress were applied. Results showed clear trends, indicating a negative impact of hydropower operations on the size and structure of local brown trout populations, when comparing impacted to hydrological control sites. In detail, the combination of hydropeaking event frequency and riverbed width could explain up to two thirds of the variation in the dataset. The results of this thesis underline the need to mitigate hydropeaking effects to protect brown trout populations.
... Variance Inflation Factor (VIF) test was used to detect collinearity among the variables in R package 'usdm' with the threshold set at 8, and variables which exceeded the threshold value were excluded from further analyses (Feld et al., 2016;Cremona et al., 2018). Consequently, average air temperature was removed because of collinearity. ...
... Consequently, average air temperature was removed because of collinearity. To pre-select the most suitable predictive variables for each fish species, the machine learning method Random Forests (RF) was performed with 1000 trees and 5 nodes by using the package 'RandomForestSRC' (Feld et al., 2016;Ishwaran and Kogalur, 2017). ...
Article
Climate change shows itself in many different ways on marine life. The fishery is also a part of marine life and affected by climate change-driven weather conditions directly or indirectly. In the present study, relationships between commercial species (grey mullet -≈90% of total capture- and gilthead seabream) that were captured from lagoon traps in Köyceğiz lagoon (Turkey) and local weather conditions were analysed. The machine learning method Random Forests (RF) was used to pre-select the model predictors. RF results showed that while temperature-related parameters, cloudy days, and wind speed were the most effective parameters, precipitation-parameters were the least important parameters for these two species catch. Generalized linear models (GLMs) were applied to each fish species with the best pre-selected parameters, with the resulting equation being used for future prediction of the two fish species. Future prediction of predictors was calculated by monthly autoregressive integrated moving average (ARIMA) and 20th/80th percentile intervals were used as the scenarios. Simulations showed that an increase in some weather parameters (wind speed, seawater temperature, maximum air temperature, cloudy days) lead to an increase in grey mullet and (wind speed) gilthead seabream catch. Models proved that the impact of the weather parameters differs for those two fish species although they live in the same environment. We recommend that individual fish species (and/or catch) should be used in the models, not the whole fish yield. Moreover, the model can also be used for non-commercial species in ecosystem-based studies. Changes in weather parameters due to climate change should be monitored to make proper decision on fishery management.
... However, the achieved reduction of complexity might have several downsides. First, the detectability of a relevant predictor variable in a statistical model is directly dependant on the gradient length within a dataset (Feld et al., 2016;Gieswein et al., 2017). So, given the spatial clustering and potential reduction of predictor variables gradients, the impact of an environmental variable might be underestimted or even not detected (Feld et al., 2016;Gieswein et al., 2017). ...
... First, the detectability of a relevant predictor variable in a statistical model is directly dependant on the gradient length within a dataset (Feld et al., 2016;Gieswein et al., 2017). So, given the spatial clustering and potential reduction of predictor variables gradients, the impact of an environmental variable might be underestimted or even not detected (Feld et al., 2016;Gieswein et al., 2017). Further, the developed relationships might be valid only for smaller predictor ranges, with limited upscaling possibilities (Yates et al., 2018). ...
Article
Full-text available
Freshwater bioassessment programmes yield valuable information for assessing the diversity and distribution of freshwater organisms, and can be related to environmental variables through the use of multivariate methods. Regionalization and subsetting strategies are widely used in this regard, but the effects on the discerned relationships are largely unexplored. In this paper, we used a partial redundancy analysis (pRDA) to investigate the influence of different environmental variables on (i) fish community structure, (ii) the explanatory power of spatial and environmental variables, and (iii) ranking of the most relevant variables for the fish community structure in Poland. We performed the analysis at the national level and for different regionalization/subsetting strategies based on hydrography (coarse resolution: river basins, fine resolution: water regions), topography, biogeography, and fish‐based river typology. Depth, slope and sediment type were three most relevant predictors at the national level and for the majority of subsets. However, compared to the national level, a significant misalignment in predictors rankings was found for a large fraction of the identified subsets. Overall, river basin subsetting provided limited gain in information compared to the national level, while water region subsetting yielded higher variability in the predictive capacity of the pRDA models and increased the share of variance explained by spatial pattern which might obscure environmental effects. Thus, it is recommended to use biotic relevant subsetting methods based on elevation or fish indicators to better capture the variability of the dataset and provide simple and informative relationships between the fish community structure and the environmental variables.
... In many instances, BRT and other ML algorithms have been shown to perform better than classical statistical models to explain and predict data patterns (Marcos Rodrigues and De la Riva 2014; Oliveira et al. 2012). Due to the aforementioned possibilities, they are well-suited for exploratory analyses with many variables (Feld, Segurado, and Gutiérrez-Cánovas 2016). ...
Thesis
Full-text available
Forest policy increasingly mobilizes the forest sector to address environmental concerns. Owing to the forest sector’s complexity and time scales involved, simulation models are often used as research methods to explore the future. This thesis investigates the contributions of Forest Sector Models (FSM), bio-economic simulation models commonly used for prospective analysis, to this transition.We first adopt a conceptual perspective and, through a parallel exploration of the early literature in forest economics and the epistemology of model use, we show that forest policy has been, and still is, a strong driver of FSM research, influencing representation processes in models as well as narratives used to drive research. We also highlight that the nature of facts within the forest sector, the local context, data availability and past practices are other important determinants of model-based research. We subsequently review more recent literature to assess the extent to which environmental issues have been addressed. While originally focused on timber production and trade, a majority of the research now focuses on goals such as renewable energy production or the conservation of biodiversity. The treatment of such objectives has however been unequal, and those closer to the models’ original target are treated more often and more deeply. On the contrary, modelling is hindered when economic values are hard to estimate or when models cannot handle spatialized data, hence objectives related to cultural and some regulation services are less commonly studied.The remainder of the thesis addresses two aspects of climate change, namely mitigation and adaptation, and brings methodological contributions by leveraging two ways of overcoming obstacles to the investigation of environmental objectives with large-scale bio-economic models: model couplings and the consideration of local environmental conditions. Both chapters focus on France, where the diversity of local contexts makes analyses focused on the upstream forest sector relevant, and use the French Forest Sector Model (FFSM).First, using the FFSM and Hartman’s model for optimal rotations with non-timber amenities, we investigate consequences for forestry and landscapes of management practices aiming at both producing timber and sequestering carbon. We show that, while postponing harvests can increase carbon stocks in the short-term, changes in management regimes and species choice yield additional benefits in the long-term. Over time, these changes lead to more diverse forest landscapes in terms of composition and structure, with potential implications for policy and environmental co-benefits. However, trends show a high level of spatial variability across and within regions, highlighting the importance of considering the local context.In-situ carbon stocks are however exposed to risks of non-permanence. We assess implications for the forest sector of climate-induced changes in wildfire regimes, as well as implications for model projections of uncertainties related to these changes. To do so, we use a probabilistic model of wildfire activity, which we couple to the FFSM, and we carry out multiple simulations using various radiative forcing levels and different climate models. Although locally significant, wildfires’ impacts remain limited at the sectoral scale. Fires affect a limited amount of the resource every year but in a cumulative manner, and the influence of climate change is mostly witnessed in the latter half of the century. Inter-annual fluctuations in fire activity only marginally propagate to the forest sector, and most uncertainty comes from the choice of climate models and scenarios. Stochasticity in the fire process, although never predominant, accounts for a significant share of uncertainty. These results stress the importance of considering multiple possible outcomes and the inherent variability in environmental processes in large-scale model projections.
... Using the 'vifstep' function, watershed variables (see Section 2.2) were stepwise excluded using a conservative VIF threshold of 5 (Zuur et al., 2007), the remaining variables were then carried over to further analyses. To examine the relationship between the parameters selected from the VIF analysis we conducted LMM with a Gaussian link function (Feld et al., 2016) (Bates et al., 2015) and built models involving all possible combinations, using the 'dredge' function of the package MuMIn. The best models were identified using the Akaike Information Criterion corrected for small sample size (AICc). ...
Article
Climate change is causing drastic landscape changes in the Arctic, but how these changes modify stream biogeochemistry is not clear yet. We examined how catchment properties influence stream nitrogen (N) and dissolved organic carbon concentrations (DOC) in a high-Arctic environment. We sampled two contrasting headwater streams (10-15 stations over 4.8 and 6.8 km, respectively) in Northeast Greenland (74°N). We characterized the geomorphology (i.e. bedrock, solifluction and alluvial types) and the vegetation (i.e. barren, fell field, grassland and tundra types) cover of each subcatchment area draining into each sampling station and collected water samples for hydrochemistry characterization. The two sampled streams differed in geomorphology and vegetation cover in the catchment. Aucellaelv catchment was mostly covered by a `bedrock´ geomorphology (71%) and `fellfield´ vegetation (51%), whereas Kæerelv was mostly covered by `alluvial´ geomorphology (65%) and `grassland´ and `tundra` vegetation (42 and 41% respectively). Hydrochemistry also differed between the two study streams, with higher concentrations of inorganic N forms in Aucellaelv and lower DOC concentrations, compared to Kærelv. The results from the linear mixed model selection showed that vegetation and geomorphology had contrasting effects on stream hydrochemistry. Subcatchments with higher solifluction sheets and limited vegetation had higher nitrate concentrations but lower dissolved organic carbon concentrations. Interestingly, we also found high variability on the production and removal of nitrate across subcatchments. These results indicate landscape controls to nutrient and organic matter exports via flow paths, soil organic matter stocks and nutrient retention via terrestrial vegetation. Moreover, the results suggest that climate change induced alterations to vegetation cover and soil physical disturbance in high-Arctic catchments will affect stream hydrochemistry, with potential effects in stream productivity, trophic relations as well as change of solute export to downstream coastal areas. This article is protected by copyright. All rights reserved.
... There is an ongoing debate on how to set nutrient targets when other stressors are present and definitive guidance cannot yet be offered. In the meantime, Feld et al. (2016) provide a toolkit for investigating the role of multiple stressors whilst Phillips et al. (2019) use synthetic datasets to examine the extent to which interactions among stressors might affect relationships. The complexity of multiple stressor interactions has also raised interest in the use of more sophisticated Fig. 1. ...
Article
One key component of any eutrophication management strategy is establishment of realistic thresholds above which negative impacts become significant and provision of ecosystem services is threatened. This paper introduces a toolkit of statistical approaches with which such thresholds can be set, explaining their rationale and situations under which each is effective. All methods assume a causal relationship between nutrients and biota, but we also recognise that nutrients rarely act in isolation. Many of the simpler methods have limited applicability when other stressors are present. Where relationships between nutrients and biota are strong, regression is recommended. Regression relationships can be extended to include additional stressors or variables responsible for variation between water bodies. However, when the relationship between nutrients and biota is weaker, categorical approaches are recommended. Of these, binomial regression and an approach based on classification mismatch are most effective although both will underestimate threshold concentrations if a second stressor is present. Whilst approaches such as changepoint analysis are not particularly useful for meeting the specific needs of EU legislation, other multivariate approaches (e.g. decision trees) may have a role to play. When other stressors are present quantile regression allows thresholds to be established which set limits above which nutrients are likely to influence the biota, irrespective of other pressures. The statistical methods in the toolkit may be useful as part of a management strategy, but more sophisticated approaches, often generating thresholds appropriate to individual water bodies rather than to broadly defined “types”, are likely to be necessary too. The importance of understanding underlying ecological processes as well as correct selection and application of methods is emphasised, along with the need to consider local regulatory and decision-making systems, and the ease with which outcomes can be communicated to non-technical audiences.
... Relations between multiple stressors and biotic condition are most detectable where the full range of land-use conditions is encountered (large disturbance gradient) (Waite, 2014;Feld et al., 2016;Herlihy et al., 2020). The relations between number of exceedances and MMI generally are strongest for the three regions with sites in settings that range from undistrubed (forested) reference to highly urban: Southeast, Northwest, and Northeast. ...
Article
Biological assemblages in streams are affected by a wide variety of physical and chemical stressors associated with land-use development, yet the importance of combinations of different types of stressors is not well known. From 2013 to 2017, the U.S. Geological Survey completed multi-stressor/multi-assemblage stream ecological assessments in five regions of the United States (434 streams total). Diatom, invertebrate, and fish communities were enumerated, and five types of potential stressors were quantified: habitat disturbance, excess nutrients, high flows, basic water quality, and contaminants in water and sediment. Boosted regression tree (BRT) models for each biological assemblage and region generally included variables from all five stressor types and multiple stressors types in each model was the norm. Classification and regression tree (CART) models then were used to determine thresholds for each BRT model variable above which there appeared to be adverse effects (multi-metric index (MMI) models only). In every region and assemblage there was a significant inverse relation between the MMI and the number of stressors exerting potentially adverse effects. The number of elevated instream stressors often varied substantially for a given level of land-use development and the number of elevated stressors was a better predictor of biological condition than was development. Using the adverse effects-levels that were developed based on the BRT model results, 68% of the streams had two or more stressors with potentially adverse effects and 35% had four or more. Our results indicate that relatively small increases in the number of stressors of different types can have a large effect on a stream ecosystem.
... Fortunately, advances in the use of statistical models and machine learning techniques offer the capability of tackling with many of the aforementioned issues, such as interactions between multiple stressors, stochasticity of flows, and data inadequacies (Arthington, Kennen, et al., 2018;Feld et al., 2016). Particularly machine-learning techniques, such as boosted regression trees and random forest and Bayesian networks (BNs) have many advantages compared with more conventional modeling approaches as they are more flexible in terms of statistical requirements, something which is very useful for catchments that lack extensive monitoring and field data and is less likely to meet these requirements. ...
Chapter
In all available methodologies for the assessment of the environmental flow requirements, a sufficient knowledge of the natural hydrological regime is essential. In this chapter the hydrological data that are required in environmental flow assessment studies, their main characteristics, and their importance as well as the specific challenges in the case of mountainous areas are analyzed. The various available data sources, the measurement and processing of hydrological data, and the utilization of modeling techniques for the estimation of streamflow data in the case of ungauged or poorly gauged watersheds and for the naturalization of streamflow data are also presented. A short description of hydrological data series analysis for the determination of environmental water requirements is provided as well. Finally, sources for further reading are provided in each section.
... The Generalized Linear Models (GLM) were implemented for two response variables: (1) density and (2) mean size, with Tweedie and Gamma distributions, respectively, and log-link functions (Tweedie, 1984). The AIC was the measure derived from the AIC (Akaike information criterion) that was used as the model selection criterion and significant interactions were assessed considering the signs of the coefficients for the interaction and individual variables and with surface plots (Feld et al., 2016). A thorough description on the used models and the analytical procedures that supported the interpretation of the results are detailed in Supplementary Material 2. For relevant predictors in the models that showed a variations in time, a causality test, based on the ccf() cross-correlation function, was used to determine relationships between two time series: the relevant predictor variable and the response variable, across the sampling period. ...
Article
Full-text available
There is an increasing demand for sea cucumbers, for human consumption, mainly from Asian markets and, as a consequence, NE-Atlantic species are now new targets for exploitation and exportation. Holothuria mammata is one of the most valuable species in Europe. However, the lack of historical economic interest in this species in most European countries has also led to a lack of studies concerning biological and ecological aspects on wild populations and this is a major issue for stock management. This study aims to determine the temporal and spatial patterns of distribution of H. mammata, considering its abundance and demographic structure in a NE-Atlantic area, SW Portugal, as a function of environmental conditions. For that, a population from a marine protected area was followed for 1 year at 1.5-month intervals. Throughout the coastal area, six sites were selected and at each sampling campaign three random transects per site and substrate (rock and sand) in which all H. mammata individuals were counted and measured. For each site and survey several environmental parameters of interest, from the water column, the sediment and substrate cover, were also measured. Generalized Linear Models were used to model the spatial and temporal distribution of the species according to environmental conditions, to determine the species' habitat preferences. The distribution models indicate that abiotic and biotic parameters of the water column are not the main drivers shaping the distribution of H. mammata. The species has a patchy distribution, and its habitat preferences depend on environmental stability, the presence of shelter and habitat complexity, which is more important for smaller, more vulnerable, individuals, while bigger size classes tend to venture more into less stable environments in an opportunistic fashion. The knowledge of these population traits is determinant to develop stock management measures, which are Frontiers in Marine Science | www.frontiersin.org 1 August 2021 | Volume 8 | Article 675330 Félix et al. Habitat Preferences for Holothuria mammata now urgent to prevent the depletion of commercial sea cucumber populations in the NE-Atlantic. Sustainable fisheries policies should be developed and start by considering to delimit fishing areas and periods, considering the species spatial and temporal distribution patterns.
... What is clear is that all lakes will likely be impacted by multiple and potentially interacting stressors in the future, and more adaptive management approaches must be considered to address the stressors (Spears et al. 2015). To support this approach, we draw attention to advances in conceptualising (Piggott et al. 2015), detecting (Birk et al. 2020), and predicting (Feld et al. 2016, Spears et al. 2021) the ecological responses associated with the management of interacting stressors. Julius et al. (2008) conducted an analysis of resilience to climate change for various ecosystem types in the United States. ...
Article
Globally, anthropogenic actions of land use change and intensification and deliberate or unintentional species invasions have adversely affected lakes, resulting in widespread loss of benefits to society. In recognition of these impacts, restoration efforts have increased in recent years. Restoration is a challenging and expensive process, however, and success rates are variable and often unpredictable. Here, we demonstrate that early actions to prevent degradation of lakes currently in good ecological condition are preferable to attempting to restore lakes that have been allowed to degrade, to allow continuity of ecosystem services. We compare case studies for 3 lakes that use preventative approaches to mitigate the effects of anthropogenic pressures. These initiatives aim to protect or enhance long-term societal benefits through building resilient ecosystems and maintaining ecological integrity. They differ from restoration projects, where lakes are often in an advanced state of degradation resilient to modest restoration efforts. We identify the need to mainstream preventative lake management including building a robust evidence base to support initiatives aimed at reversing the early stages of changes in ecological state.
... Fortunately, advances in the use of statistical models and machine learning techniques offer the capability of tackling with many of the aforementioned issues, such as interactions between multiple stressors, stochasticity of flows, and data inadequacies (Arthington, Kennen, et al., 2018;Feld et al., 2016). Particularly machine-learning techniques, such as boosted regression trees and random forest and Bayesian networks (BNs) have many advantages compared with more conventional modeling approaches as they are more flexible in terms of statistical requirements, something which is very useful for catchments that lack extensive monitoring and field data and is less likely to meet these requirements. ...
Chapter
The modification of sediment and flow regimes caused by damming and river regulation has deleterious effects on the ecological and morphological river processes. This alteration of river systems triggered the implementation of safeguarding environmental flows (e-flows) defined as “the quantity, timing, and quality of water flows required to sustain freshwater and estuarine ecosystems and the human livelihoods and wellbeing that depend on these ecosystems”. In the last decades, physical habitat simulation approaches emerged as fundamental stand-alone or supplementary methods for e-flow assessment. These approaches combine three main components: (1) hydraulic simulation, (2) habitat suitability modeling, to determine the quality of the available habitat, and (3) hydrological analyses (under current and climate change scenarios). E-flow regimes are finally defined, by assessing the spatial and temporal habitat variability for the target taxa or community, after combining these three components. During the process of physical habitat simulation some river processes, such as sediment transport and morphological changes, are often neglected while uncertainties arise from every component. We reviewed the elements that should be considered in every component of the physical habitat simulation to reduce uncertainties with emphasis on the actual trends on the topic and how sediment transport and river morphodynamics can be included within this methodological framework.
... Fortunately, advances in the use of statistical models and machine learning techniques offer the capability of tackling with many of the aforementioned issues, such as interactions between multiple stressors, stochasticity of flows, and data inadequacies (Arthington, Kennen, et al., 2018;Feld et al., 2016). Particularly machine-learning techniques, such as boosted regression trees and random forest and Bayesian networks (BNs) have many advantages compared with more conventional modeling approaches as they are more flexible in terms of statistical requirements, something which is very useful for catchments that lack extensive monitoring and field data and is less likely to meet these requirements. ...
Chapter
Mountains and mountain rivers provide a multitude of invaluable goods and services to a profound portion of the planet’s population. As “water towers” of the Earth mountains are sources of the mightiest world rivers and play a pivotal role for global biodiversity, freshwater, and sediment supply. Distinct morphological, climatic, hydrological, hydrochemical, and biological features of mountainous river ecosystems, compared to lowland ones, make them particularly fragile and vulnerable to human interference. Despite a number of remote mountain areas and rivers still remaining intact from direct human pressures, the majority of mountain ecosystems, are being increasingly threatened by adverse local and global changes driven by market economy. To efficiently conserve and sustainably use mountain ecosystems and contribute to the survival of the planet, it is critical to change our standards and life attitudes by realizing and appreciating our immediate connection to the global ecosystem, change attitudes and current consumption patterns, and stimulate the ways our global society functions and interacts with the natural environment.
... Fortunately, advances in the use of statistical models and machine learning techniques offer the capability of tackling with many of the aforementioned issues, such as interactions between multiple stressors, stochasticity of flows, and data inadequacies (Arthington, Kennen, et al., 2018;Feld et al., 2016). Particularly machine-learning techniques, such as boosted regression trees and random forest and Bayesian networks (BNs) have many advantages compared with more conventional modeling approaches as they are more flexible in terms of statistical requirements, something which is very useful for catchments that lack extensive monitoring and field data and is less likely to meet these requirements. ...
... Fortunately, advances in the use of statistical models and machine learning techniques offer the capability of tackling with many of the aforementioned issues, such as interactions between multiple stressors, stochasticity of flows, and data inadequacies (Arthington, Kennen, et al., 2018;Feld et al., 2016). Particularly machine-learning techniques, such as boosted regression trees and random forest and Bayesian networks (BNs) have many advantages compared with more conventional modeling approaches as they are more flexible in terms of statistical requirements, something which is very useful for catchments that lack extensive monitoring and field data and is less likely to meet these requirements. ...
... To evaluate and rank the most important environmental or microbial community variables predicting biotransformation rate constants we used a Random Forest model as it seemed appropriate for the type and amount of data generated in our study. Random Forest is a robust machine learning algorithm, able to afford non-parametric regression analysis of ecological data (Feld et al., 2016;Grizzetti et al., 2017). It can handle a large number of correlated variables and low number of observations, and is suited to analyse non-linear relationships and complex interactions, even with numerous missing values (Breiman, 2001). ...
Article
Full-text available
For most micropollutants (MPs) present in surface waters, such as pesticides and pharmaceuticals, the contribution of biotransformation to their overall removal from lake ecosystems is largely unknown. This study aims at empirically determining the biotransformation rate constants for 35 MPs at different periods of the year and depths of a meso-eutrophic lake. We then tested statistically the association of environmental parameters and microbial community composition with the biotransformation rate constants obtained. Biotransformation was observed for 14 out of 35 studied MPs for at least one sampling time. Large variations in biotransformation rate constants were observed over the seasons and between compounds. Overall, the transformation of MPs was mostly influenced by the lake's temperature, phytoplankton density and bacterial diversity. However, some individual MPs were not following the general trend or association with microorganism biomass. The antidepressant mianserin, for instance, was transformed in all experiments and depths, but did not show any relationship with measured environmental parameters, suggesting the importance of specific microorganisms in its transformation. The results presented here contribute to our understanding of the fate of MPs in surface waters and thus support improved risk assessment of contaminants in the environment.
... We analyzed the global differences in the survival of Nectopsyche sp. and leaf decomposition between treatments using ANOVA and Tukey tests, after checking that the assumptions of normality and homoscedasticity were met. To assess changes in survival and leaf decomposition over time in the different experiments, we followed the statistical framework developed by Feld, Segurado and Gutiérrez-Cánovas [32]. First, we built Random Forests, using survival as the response variable, and treatment intensity as well as all of the measured environmental variables as predictors. ...
Article
Full-text available
Andean streams are becoming increasingly impacted by agricultural activities. However, the potential effects of pesticides on their aquatic biodiversity remain unassessed. In order to address this knowledge gap, we conducted an experiment over 37 days in microcosms to assess the effect of two pesticides commonly used in Ecuador (Engeo and Chlorpyrifos) on the aquatic insect Nectopsyche sp. (Trichoptera: Leptoceridae) at 0, 0.10, 5 and 10 μg L−1 concentra-tions. The highest concentration corresponds to the maximum concentration allowed by the Equatorian legislation. We assessed insect mortality every 24 h, with leaf litter decomposition rates of organic matter determined by deploying Andean alder (Alnus acuminata) dry leaf packs in the microcosms. We found significant mortality of Nectopsyche sp. at high concentrations of Chlorpyrifos, whereas leaf litter was not significantly affected by any of the treatments. We con-clude that the environmental legislation of Ecuador might not be fully protecting aquatic biodi-versity from pesticide pollution. Further studies are needed, especially when considering that the maximum permitted concentration is very likely exceeded in many areas of the country. We also suggest that the maximum permissible values should be reviewed, considering each pesticide individually.
... All models were checked for model fit adequacy statistics (Burnham et al., 2011;Harrison et al., 2018), including overdispersion and homoscedasticity (Feld et al., 2016;. Using the 'mediation' package (Tingley et al., 2014), we ran 5000 simulations for each model (Hayes, 2009), using the bias-corrected and accelerated bootstrapping method for estimating mediation effects to correct for non-normality and address power limitations (Preacher and Hayes, 2008). ...
Article
Accelerating rates of urbanisation are contributing to biodiversity declines worldwide. However, urban green (e.g. parks) and blue spaces (e.g. coast) provide important habitat for species. Emerging evidence also shows that green and blue spaces can benefit human psychological wellbeing, although few studies originate from the Global South and it is unclear whether more biodiverse spaces offer greater wellbeing gains. We examine how bird diversity (abundance, species richness, Shannon diversity, and community composition) in green and coastal blue space in Georgetown, Guyana, is associated with people's wellbeing (positive and negative affect, anxiety) in situ, using point counts and questionnaires. Bird community composition differed between green and coastal sites, and diversity was significantly higher in green sites. Positive affect and anxiety did not differ between green and coastal sites, but negative affect was higher in coastal sites. Mixed-effect models showed no associations between biodiversity and wellbeing, implying other features are contributing to people's positive wellbeing. Despite no association between biodiversity and wellbeing, both green and coastal blue sites are important for wellbeing and supporting different bird communities. City planning authorities and public health professionals should ensure these social and environmental needs are met in developing cities in the Global South.
... Then, the random forest model (Feld et al., 2016) was employed to examine the most important predictors of the LCBD and its two components from the water and sediment variables above. Notably, we applied cross-validation (Elith et al., 2008) to get the optimal number of 2,000 trees. ...
Article
Full-text available
Microbial beta diversity has been recently studied along the water depth in aquatic ecosystems, however its turnover and nestedness components remain elusive especially for multiple taxonomic groups. Based on the beta diversity partitioning developed by Baselga and Local Contributions to Beta Diversity (LCBD) partitioning by Legendre, we examined the water- depth variations in beta diversity components of bacteria, archaea and fungi in surface sediments of Hulun Lake, a semi-arid lake in northern China, and further explored the relative importance of environmental drivers underlying their patterns. We found that the relative abundances of Proteobacteria, Chloroflexi, Euryarchaeota, and Rozellomycota increased toward deep water, while Acidobacteria, Parvarchaeota, and Chytridiomycota decreased. For bacteria and archaea, there were significant (p < 0.05) decreasing water-depth patterns for LCBD and LCBD Repl (i.e., species replacement), while increasing patterns for total beta diversity and turnover, implying that total beta diversity and LCBD were dominated by species turnover or LCBD Repl . Further, bacteria showed a strong correlation with archaea regarding LCBD, total beta diversity and turnover. Such parallel patterns among bacteria and archaea were underpinned by similar ecological processes like environmental selection. Total beta diversity and turnover were largely affected by sediment total nitrogen, while LCBD and LCBD Repl were mainly constrained by water NO2− -N and NO3− -N. For fungal community variation, no significant patterns were observed, which may be due to different drivers like water nitrogen or phosphorus. Taken together, our findings provide compelling evidences for disentangling the underlying mechanisms of community variation in multiple aquatic microbial taxonomic groups.
... Visualization along both dimensions of a gradient (PC1 and PC2) was achieved through two-dimensional surface plots displaying fitted response values from the GLM against a surface defined by the two PCs (PC1 on the X-axis and PC2 on the Y-axis; (Feld et al., 2016)). In addition to results P < 0.05, we also highlighted results P < 0.1, i.e. that explain variation in the data but where we lacked statistical power to detect any effects at the 5% level. ...
Article
Full-text available
Ecosystem functioning and community structure are recognized as key components of ecosystem integrity, but comprehensive, standardized studies of the responses of both structural and functional indicators to different types of anthropogenic pressures remain rare. Consequently, we lack an empirical basis for (i) identifying when monitoring ecosystem structure alone misses important changes in ecosystem functioning, (ii) recommending sets of structural and functional metrics best suited for detecting ecological change driven by different anthropogenic pressures, and (iii) understanding the cumulative effects of multiple, co-occurring stressors on structure and function. We investigated variation in community structure and ecosystem functioning of stream ecosystems along three gradients (10–16 independent stream sites each) of increasing impact arising from agriculture, forestry and river regulation for hydropower, respectively. For each stream, we quantified variation in (i) the abiotic environment, (ii) community composition of four organism groups and (iii) three basal ecosystem processes underpinning carbon and nutrient cycling in streams. We assessed the responsiveness of multiple biodiversity, community structure and ecosystem functioning indicators based on variance explained and effect size metrics. Along a gradient of increasing agricultural impact, diatoms and fish were the most responsive groups overall, but significant variation was detected in at least one aspect of community composition, abundance and/or biodiversity of every organism group . In contrast, most of our functional metrics did not vary significantly along the agricultural gradient, possibly due to contrasting, antagonistic effects of increasing nutrient concentrations and turbidity on ecosystem process rates. The exception was detritivore-mediated litter decomposition which increased up to moderate levels of nutrient. Impacts of river regulation were most marked for diatoms, which were responsive to both increasingly frequent hydropeaking and to increasing seasonal river regulation. Among functional indicators, both litter decomposition and algal biomass accrual declined significantly with increasing hydropeaking. Few structural or functional metrics varied with forest management, with macroinvertebrate diversity increasing along the forestry gradient, as did algal and fungal biomass accrual. Together, these findings highlight the challenges of making inferences about the impacts of anthropogenic disturbances at the ecosystem level based on community data alone, and pinpoint the need to identify optimal sets of functional and structural indicators best suited for detecting ecological changes associated with different human activities.
... 53 Furthermore, served as a crucial tool, the traditional 54 statistical analysis can help us to elucidate the knowledge 55 from the monitoring of complex environments ( Zhong 56 et al, 2018 ;Zhang et al., 2021 ). For instance, Principal 57 Components Analysis (PCA) and other multivariate tech-58 niques provide a means to identify stressor gradients 59 in the data, including their correlation with each other 60 ( Ewaid et al, 2018 ;Feld et al, 2016 ). These methods are re-61 liable on account of their capacity to efficiently and practi-62 cally quantify the situation ( Béjaoui et al., 2016 ). ...
Article
Full-text available
The Paciência river estuary (PRE) is an Amazonian estuary with relevant socio-economic importance; however, it is undergoing a continuous anthropogenic pressure such as the input of untreated domestic effluents. Hence, the phytoplankton community and trophic indices were analysed during three seasons in 2017 to assess the water quality and understand the degree of human influence in a macrotidal estuary. The trophic state of the PRE was determined by applying the trophic state index (TSI), trophic index, and Karydis trophic index, and ecological status by diversity metrics (diversity, evenness, richness, and dominance). Multivariate analysis indicated that TSI and dominance index were the most sensitive indices to changes in environmental conditions. The precipitation regime characterised the estuary in two different trophic scenarios. PRE was hypereutrophic in the dry season and mesotrophic in the rainy/transitional seasons based on the TSI. A high dominance index was recorded in the dry season with blooms of Thalassiosira rotula, reflecting an ecological disturbance. A higher phytoplankton diversity was recorded in the rainy/transitional seasons, indicating a tendency toward a lower stressed ecosystem. Therefore, understanding the trophic state variability in tropical estuaries and its response to changes in phytoplankton ecology is essential to support the management of coastal ecosystems.
... 53 Furthermore, served as a crucial tool, the traditional 54 statistical analysis can help us to elucidate the knowledge 55 from the monitoring of complex environments ( Zhong 56 et al, 2018 ;Zhang et al., 2021 ). For instance, Principal 57 Components Analysis (PCA) and other multivariate tech-58 niques provide a means to identify stressor gradients 59 in the data, including their correlation with each other 60 ( Ewaid et al, 2018 ;Feld et al, 2016 ). These methods are re-61 liable on account of their capacity to efficiently and practi-62 cally quantify the situation ( Béjaoui et al., 2016 ). ...
Article
Biodiversity maintenance is a main goal in ecology. Hence, phytoplankton diversity and biomass were analyzed in a coastal bay (Cumã Bay) of the Amazon Macrotidal Mangrove Coast, which has been designated as an international hotspot for conservation (Ramsar site) with high biological productivity and diversity that provides crucial ecosystem services and elevated fish production. An ecohydrology-based approach was applied to identify the main factors that drive the patterns of phytoplankton diversity and biomass, considering spatio-temporal analyses of physical, chemical, and biological variables from May 2019 to June 2020. Phytoplankton dynamics were investigated using multivariate analyses, correlations, and generalized additive models. Seven indices were tested to select the most efficient biodiversity metric. The hydrological conditions of Cumã Bay were governed primarily by elevated precipitation and macrotidal dynamics, resulting in two different functional zones based on environmental variability: the freshwater influence zone and marine influence zone. Seasonally, the maximum freshwater discharge, low salinity and light availability promoted cell abundance and biomass increase, with blooms of Skeletonema costatum, which reduced the taxonomic diversity of the community in the rainy season. During the dry season, turbid waters resulting from macrotidal dynamics and wind speed limited light penetration and phytoplankton photosynthesis, leading to a higher uniformity in the species distribution. Shannon index was the most sensitive biodiversity metric to environmental changes. This study found that deterministic processes governed the community, which rainfall on the Amazon coast, along with wind speed, salinity, light availability and nutrients were the main controlling factors for phytoplankton diversity and richness.
... Finally, the interaction coefficient c 11 of the regression model informs about whether the presence of one stressor affects the response of the dependent variable to the other stressor. Stressors may interact in different ways, they can be synergistic (larger combined affect than the sum of the individual effects), antagonistic (smaller combined affect than the sum of the individual effects) or additive (effects linearly add to each other) (Feld et al., 2016). In line with Ellis et al. (2017) and Mahon et al. (2019), we can interpret the different parameters of our regression model to quantify the additive and interactive effects of our two different stressors. ...
Article
Using sodium chloride (NaCl) for de-icing roads is known to have severe consequences on freshwater organisms when washed into water bodies. N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylenediamine, also known as 6PPD, is an antiozonant mainly found in automobile tire rubber to prevent ozone mediated cracking or wear-out. Especially the ozonated derivate, 6PPD-quinone, which is washed into streams after storm events, has been found to be toxic for coho salmon. Studies on other freshwater organisms could not confirm those findings, pointing towards distinct species-specific differences. Storm events result in greater run-offs from all water-soluble contaminants into freshwater bodies, potentially enhancing the concentrations of both chloride and 6PPD during winter. Here we show that these two contaminants have synergistic negative effects on the population growth of the rotifer Brachionus calyciflorus, a common freshwater herbivore. Hence, while only high concentrations of 6PPD and even higher concentrations of 6PPD-quinone, beyond environmentally relevant concentrations, had lethal effects on rotifers, the addition of NaCl enhanced the sensitivity of the rotifers towards the application of 6PPD so that their negative effects were more pronounced at lower concentrations. Similarly, 6PPD increased the lethal effect of NaCl. Our results support the species-specific toxicity of 6PPD and demonstrate a synergistic effect of the antiozonant on the toxicity of other environmentally relevant stressors, such as road salt contamination.
... In addition, a non-metric multidimensional scaling (NMDS) analysis was used to compare the differences during seawater-freshwater transition and freshwater gut bacterial communities based on Bray-Curtis distances. We applied random forest model (Feld et al., 2016) and redundancy analysis (RDA) to identify the important factors affecting each OTU. Pearson's correlation coefficients were used to evaluate the statistical correlation between the explanatory variables, and the variables with high correlation coefficients (Pearson r > 0.7) were excluded from the models. ...
Article
Full-text available
Biological migration is usually associated with disturbances and environmental changes that are key drivers in determining the diversity, community compositions, and function of gut microbiome. However, little is known about how gut microbiome is affected by disturbance such as salinity changes during migration from seawater to freshwater. Here, we tracked the gut microbiome succession of Chinese mitten crabs ( Eriocheir sinensis ) during their migrations from seawater to freshwater and afterward using 16S rDNA sequencing for 127 days, and explored the temporal patterns in microbial diversity and the underlying environmental factors. The species richness of gut microbiome showed a hump-shaped trend over time during seawater–freshwater migration. The community dissimilarities of gut microbiome increased significantly with day change. The turnover rate of gut microbiome community was higher during seawater–freshwater transition (1–5 days) than that in later freshwater conditions. Salinity was the major factor leading to the alpha diversity and community dissimilarity of gut microbiome during seawater–freshwater transition, while the host selection showed dominant effects during freshwater stage. The transitivity, connectivity, and average clustering coefficient of gut microbial co-occurrence networks showed decreased trends, while modularity increased during seawater–freshwater migration. For metabolic pathways, “Amino Acid Metabolism” and “Lipid Metabolism” were higher during seawater–freshwater transition than in freshwater. This study advances our mechanistic understanding of the assembly and succession of gut microbiota, which provides new insights into the gut ecology of other aquatic animals.
... Finally, the interaction coefficient c 11 of the regression model informs about whether the presence of one stressor affects the response of the dependent variable to the other stressor. Stressors may interact in different ways, they can be synergistic (larger combined affect than the sum of the individual effects), antagonistic (smaller combined affect than the sum of the individual effects) or additive (effects linearly add to each other) (Feld et al., 2016). In line with Ellis et al. (2017) and Mahon et al. (2019), we can interpret the different parameters of our regression model to quantify the additive and interactive effects of our two different stressors. ...
Article
The human-caused proliferation of cyanobacteria severely impacts consumers in freshwater ecosystems. Toxicity is often singled out as the sole trait to which consumers can adapt, even though cyanobacteria are not necessarily toxic and the lack of nutritionally critical sterols in cyanobacteria is known to impair consumers. We studied the relative significance of toxicity and dietary sterol deficiency in driving the evolution of grazer resistance to cyanobacteria in a large lake with a well-documented history of eutrophication and oligo-trophication. Resurrecting decades-old Daphnia genotypes from the sediment allowed us to show that the evolution and subsequent loss of grazer resistance to cyanobacteria involved an adaptation to changes in both toxicity and dietary sterol availability. The adaptation of Daphnia to changes in cyanobacteria abundance revealed a sterol-mediated gleaner-opportunist trade-off. Genotypes from peak-eutrophic periods showed a higher affinity for dietary sterols at the cost of a lower maximum growth rate, whereas genotypes from more oligotrophic periods showed a lower affinity for dietary sterols in favour of a higher maximum growth rate. Our data corroborate the significance of sterols as limiting nutrients in aquatic food webs and highlight the applicability of the gleaner-opportunist trade-off for reconstructing eco-evolutionary processes.
... To analyse how flow intermittence and environmental characteristics affected the decomposition rates, automated model selection for the linear regression model was performed using the R package MuMIn (Barton, 2009) to select the best combination of all predictor variables based on AICc (Akaike Information Criterion corrected version for small samples). To assess collinearity between explanatory variables, pairwise Pearson correlations and the variance inflator factor (vifstep of usdm R Package) (Feld et al., 2016) were previously estimated. Therefore, agricultural land uses (highly correlated with DIN 3 concentration), conductivity (correlated with Ca and DOC) and Mg (correlated with Ca) were discarded (Table S2). ...
Article
Flow interruption in intermittent rivers (IRs) generates a mosaic of terrestrial and aquatic habitats across the river network affecting ecosystem processes, as organic matter (OM) decomposition. Water use for farming in arid and semi-arid climates intensifies the dry conditions and affects local river characteristics. In that way, flow intermittence and the distribution of land uses may affect the OM processing along the river. To understand the role of IRs in global OM dynamics and how global change affecting water flow regimes determines these dynamics, it is important to estimate OM-processing rates at a basin scale. The aim of this study was to evaluate the effect of the intensity of flow intermittence on OM processing, and how this effect was modulated by local environmental factors related to land uses across a Mediterranean river basin. To do this, wood decomposition (mass loss and fungal biomass) was selected as a functional indicator. Drying duration and frequency were measured to characterize flow intermittence in different reaches along the river, as well as local environmental factors. Linear models stablished the role of factors on decomposition. The results showed that differences in decomposition rates across the river network were negatively related to the duration of flow interruption. Dissolved inorganic nitrogen associated with agriculture counteracted the negative effect of intermittence on mass loss (increasing by up to three times); but with a higher duration of dry conditions, its effect was insignificant. An increase of 20% of canopy (higher in natural areas) resulted in increases of up to 5% of mass loss. Overall, our study is relevant to understanding the interaction between flow intermittence and land uses on OM processing, especially considering the intensification of flow intermittence and its increased distribution to other regions, which is expected to be a consequence of climate warming and human activities.
... The dredge analysis for model fitting was computed in R using Mumin library with lm function (Barton and Barton 2015) by lowest Akaike Information Criteria (AIC c ) is the most parsimonious model. The analysis from the dredge provided basic insights to nature of interaction (Feld et al. 2016) and impact towards response variable. ...
Article
Full-text available
Globally the inland urban freshwater wetland biodiversity is facing threat due to anthropogenic activities. Scientists worldwide understand the complex linkages between ecological impacts due to human induced pollution. Wetland systems are susceptible to changes in quality and quantity of water. The paper will focus on water quality based stressors and indicative changes in biodiversity including resident and wintering avifauna density. The avifauna is an indicator towards wetland and health of its ecosystem services. The present study provides an overview of a tropical inland wetland (Najafgarh Jheel) in Delhi, India which is exposed to multiple anthropogenic stressors. The wetland is a proposed bird sanctuary and is facing acute conservation issues due to expanding urbanization and extensive agriculture. We measure the data for number of birds at nine census transect points. The census was conducted once a month for a period of one year (2017), early morning 6 am-9 am. To each census transect we associated eleven points for water quality parameters. The wetland is organically loaded with agricultural runoff and unregulated sewage drains deteriorating quality of water. Therefore, water quality is an immediate local stressor for the resident and wintering birds. The water quality parameters were partially correlated and further modelled with yearly bird density. The generalized linear mixed models depicted that DO, BOD and TDS have higher effect on conglomeration of birds. Further, CCA plot between wintering birds and water quality from two consecutive year data was plotted. The analysis depicts the inter-linkages of wintering birds with specific stressor. The combination of these multiple stressors has synergistic and antagonistic effects that are decisive for management and restoration of a lake. The current hypothesis based on understanding short term freshwater research, implication towards local management policy and baseline for long term comprehensive studies with holistic approaches. The concept of understanding short-term linkages using multivariate data analysis between water quality and bird density is relatively untested in urban tropical wetlands of India.
... Prior to the analyses, we checked for collinearity between all predictors using the function VIF (Feld et al., 2016). Since all predictors showed VIF values ≤ 3, they were all included in the full model, as recommended by Zuur et al. (2010). ...
Article
Full-text available
The frequency and intensity of extreme events in coastal wetlands is increasing due to climate change. These, in combination with other threats such as habitat loss, can have strong effects on the biodiversity and ecosystem services of coastal wetland ecosystems. Here we examined how traditional (community composition and taxonomic diversity) and alternative indices based on body size (size diversity and the slope of size spectra) respond to single and combined effects of anthropogenic disturbances, extreme event disturbances, and key environmental drivers, for both aquatic and terrestrial invertebrates in Mediterranean coastal wetlands in Chile. We studied 18 coastal wetlands over more than 500 km along the Chilean coast. We applied both univariate (GLMMs and GAMMs) and multivariate statistics (RDA and variation partitioning) to identify additive and non-additive effects on invertebrates. We also examined latitudinal trends, and tested the potential of alternative metrics to detect interactions between disturbances. We found latitude-specific vulnerability to disturbances in Chilean coastal wetlands. Non-additive effects were more important for aquatic invertebrates, while additive effects were more important for terrestrial invertebrates. In many cases, disturbance effects depended on the environmental conditions of study sites, especially salinity. This study suggests that size-based metrics may be better than taxonomic metrics at detecting interactions among different disturbances, especially for aquatic invertebrates.
Article
Full-text available
We assessed long-term impacts of multiple stressors and their interaction on the zooplankton community of the large, eutrophic, cyanobacteria-dominated Lake Peipsi (Estonia, Russia). Stressor dataset consisted in time series (1997–2018) of temperature, nutrients, pH, water transparency, phytoplankton biomass and taxonomic richness. The best predictors were selected with random forests machine-learning algorithms and the subsequent models were constructed with generalized linear modeling. We also aimed to identify graphical thresholds representing non-linear, marked responses of abundance or biomass to stressors. Temperature was the dominant stressor for explaining zooplankton abundance and biomass, followed by cyanobacteria biomass, total nitrogen concentration and water transparency. The effect of water temperature was positive, whereas the effect of cyanobacteria became negative after their biomass exceeded a threshold of ~ 2 mg l−1. However, the two stressors together had antagonistic effects on zooplankton, causing a decrease in biomass and abundance. For zooplankton, critical thresholds of total nitrogen (~ 700 μg l−1), total phosphorus (~ 70 μg l−1), and water transparency (~ 1.4 m) after which zooplankton metrics changed drastically, were determined. These findings show that although lake warming alone could be positive for zooplankton, the necessity of reducing interacting stressors that influence harmful cyanobacteria growth and biomass, especially nitrogen loads, must be considered.
Article
Natural and anthropogenic factors form stream macroinvertebrate communities depending on their combination, intensity, and spatial pattern. The study aimed to identify macroinvertebrate indicators that respond to land cover, hydromorphology, and wastewater releases individually and to their multiple-pressure pattern. Environmental and macroinvertebrate data from 36 sites were used in the study. Pressure parameters representing hierarchy of their complexity and spatial scale were included in analyses. Correlation analyses were used for evaluation of relationships among pressure characteristics and also pressure–macroinvertebrate relationships. The pressure-based and biological classification of sites was compared and indicator taxa were identified. The arable land in the sub-corridor extending 2–10 km upstream of an investigated site was the main pressure factor influencing the structure of macroinvertebrate communities in the studied streams. The biological effects of small-scale land cover were followed by catchment-scale land cover and hydromorphology. Almost no association of macroinvertebrates with the risk of point source pollution were detected. Classifications based on pressures and community composition corresponded only by the separation of most degraded sites from others. Among the macroinvertebrate indicators characterizing the severe impairment threshold, chironomids and oligochaetes dominated. Different responses of macroinvertebrates to hydromorphological degradation were observed under conditions of high small- and large-scale agricultural pressures (decrease in macroinvertebrate evenness and increase in oligochaete taxa richness, respectively). Linking biological indicators to pressure components and their combinations improves the efficiency of conservation and restoration strategies applied in fluvial ecosystems.
Article
Freshwater macroinvertebrates provide valuable indicators for biomonitoring ecosystem change in relation to natural and anthropogenic drivers. DNA metabarcoding is an efficient approach for estimating such indicators, but its results may differ from morphotaxonomic approaches traditionally used in biomonitoring. Here we test the hypothesis that despite differences in the number and identity of taxa recorded, both approaches may retrieve comparable patterns of community change, and detect similar ecological gradients influencing such changes. We compared results obtained with morphological identification at family level of macroinvertebrates collected at 80 streams under a Water Framework Directive biomonitoring program in Portugal, with results obtained with metabarcoding from the ethanol preserving the bulk samples, using either single (COI-M19BR2, 16S-Inse01, 18S-Euka02) or multiple markers. Metabarcoding recorded less families and different communities compared to morphotaxonomy, but community sensitivities to disturbance estimated with the IASPT index were more similar across approaches. Spatial variation in local community metrics and the factors influencing such variation were significantly correlated between morphotaxonomy and metabarcoding. After reducing random noise in the dissimilarity matrices, the spatial variation in community composition was also significantly correlated across methods. A dominant gradient of community change was consistently retrieved, and all methods identified a largely similar set of anthropogenic stressors strongly influencing such gradient. Overall, results confirm our initial hypothesis, suggesting that morphotaxonomy and metabarcoding can estimate consistent spatial patterns of community variation and their main drivers. These results are encouraging for macroinvertebrate biomonitoring using metabarcoding approaches, suggesting that they can be intercalibrated with morphotaxonomic approaches to recover equivalent spatial and temporal gradients of ecological change.
Article
The majority of the carbon worldwide is in soil. In a river catchment, the tight relationship between soil, water and climate makes carbon likely to be eroded and transported from the soil to the rivers. There are multiple variables which can trigger and accelerate the process. In order to assess the importance of the factors involved, and their interactions resulting in the changes in the carbon cycle within catchments, we have studied the catchments of 26 Finnish rivers from 2000 to 2019. These catchments are distributed all over Finland, but we have grouped them into three categories: southern, peatland and northern. We have run a boosted regression tree (BRT) analysis on chemical, physical, climatic and anthropogenic factors to determine their influence on the variations of total organic carbon (TOC) concentration. TOC concentration has decreased in Finland between 2000 and 2019 by 0.91 mg/l, driven principally by forest ditching and % old forest in the catchment. Old forest is especially dominant in the northern catchments with an influence on TOC of 40.5%. In southern and peatland catchments, average precipitation is an important factor to explain the changes in TOC whilst in northern catchments, organic fields have more influence.
Article
Full-text available
Freshwater fish biodiversity is experiencing an alarming decline worldwide. Understanding the main factors behind its deterioration is a key step for ecosystem restoration. In this work, large‐scale and long‐term data were used to identify the causes of the decline of native species richness in Castilla‐La Mancha. This region in central Spain covers part of six river basins belonging to four of the 11 biogeographical provinces for freshwater fish in the Iberian Peninsula. Firstly, we built a dataset that associates the presence of several fish species and a wide range of environmental variables (e.g. hydrological and hydromorphological indicators, land use classes, presence of alien fish species) at selected river sites for two different time periods (1980–2000 and 2001–2020). Secondly, we conducted an exploratory data analysis to identify possible temporal trends in the dataset. Finally, we applied the random forest algorithm to predict the response of different ecological guild‐based metrics of fish richness to the selected variables. The exploratory data analysis revealed a decrease in native fish species richness in 74% of the area studied. There was no sustained temporal trend for stressor variables, except for the number of alien species, which increased in most river sites (63%). The models of the richness of native rheophilic, native intolerant, alien rheophilic, and alien limnophilic species performed satisfactorily. Magnitude of maximum discharge, presence of alien species, land use in the catchment area and altitude were the most important predictors of richness of native intolerant and rheophilic species. Alien limnophilic species proved to be sensitive to variables related to flow regime alteration, such as the presence of dams and the number of river flow reversals, while a less degraded habitat was found to be favourable to alien rheophilic species. The results suggest that the cumulative effect of persistent altered flow regimes and water pollution, coupled with a strong increase in the number of alien species, have led to the decline of native species in the area studied. The restoration of near‐natural magnitudes of high flows when implementing environmental flows emerged as a key measure to restore ecosystem integrity. Starting from a long‐term and large‐scale dataset, this study provides new, quantitative insights into stressor–ecosystem relationships in rivers and could inform future environmental policy initiatives because it has identified the main factors leading to native fish decline and alien fish proliferation. Our findings emphasise the importance of considering metrics based on fish assemblage composition and ecological functional groups in order to disentangle the effects of stressors on fish communities.
Technical Report
Full-text available
Project objectives and research questions The overall aim of this project was to compile and assess the key evidence required to improve our understanding of climate change impacts on the water quality of Scottish standing waters at national, regional and local scales. The project focussed on the interactions between climate change, the drivers of eutrophication problems and their impacts. We synthesised information from the literature, expert opinion and monitoring data, using statistical analyses and visualisation (mainly mapping) combined with climate change scenario modelling to meet two project objectives and address the six strategic water research questions outlined below: Objective 1 aimed to establish and deliver a preliminary evidence-base to enable the evaluation of the extent to which: 1.1. Climate change impacts are driving current and future risk of water quality issues arising in Scottish standing waters. 1.2. Climate change impacts on water quality are mediated through catchment management practices, in-loch processes and other interacting factors (e.g., prevailing weather; hydrological extremes) under current and projected climate change scenarios. Objective 2 aimed to use expert opinion and best available evidence (from outputs from Objective 1 such as literature review, data exploration and modelling), to address the following key strategic water research questions: Drivers and Impacts 1. Is there evidence of a causal link between climate change impacts and water quality issues in Scottish standing waters at national, regional and local scales? 2. What are the main types of climate-driven water quality impacts under current and projected climate change scenarios? Risk 3. Which areas, locations and types of Scottish standing waters are currently most to least at risk of developing water quality issues due to climate change impacts at national, regional and local scales? 4. Which areas, locations and types of Scottish standing waters are likely to experience exacerbated water quality risks under projected climate change scenarios? Factors 5. What factors contribute to the risk of water quality issues from climate change impacts in Scottish standing waters at national, regional and local scales? 6. What factors need to be considered for mitigating climate-driven risks to water quality under current and projected climate change scenarios?
Article
We have implemented a specific data mining process to explore the relationship between biological indices and physico-chemical pressures in rivers. Data were collected in the framework of the French National monitoring network set up to assess the ecological status of rivers under the European Water Framework Directive (WFD). Chemical parameters and biological indices were collected regularly from 1.781 locations in metropolitan France from 2007 to 2013. The sequential pattern mining process generates closed partially ordered patterns representing a succession of physico-chemical events that precede a given biological index in a given status, validated using a subset of data. This paper focuses on the patterns and their occurrence. We showed that biological statuses depend on these temporal successions of alterations and not only on the last alterations. The physico-chemical statuses of water bodies usually appeared to be higher than their biological statuses, suggesting synergism between toxicants and/or an additive impact of other stressors related to hydromorphology or hydrology. Patterns found in the highest biological status for the biological indices based on macroinvertebrates, diatoms, macrophytes or fish, were characterised by the constancy of a high physico-chemical status over time. By contrast, before indices based on macroinvertebrates and macrophytes, two types of patterns were observed for bad biological status: (1) a chronic multi-pressure pattern, in which pressure categories such as nitrates, pesticides and other organic hydrocarbons, in moderate, poor or bad status, repeated themselves several times over time, or (2) a single occurrence of a degraded pressure category, such as one moderate nitrogen, excluding nitrate, or one poor oxidizable organic matter, among other pressure categories in good status. Extracting such patterns is a promising solution both to disentangle the effects of the different stressors on water quality, and to identify the key temporal sequences among them in a context of multi-stress conditions, which is a challenge currently facing the WFD.
Preprint
Full-text available
Microbial beta diversity has been recently studied along the water depth in aquatic ecosystems, however its turnover and nestedness components remain elusive especially for multiple taxonomic groups. Based on the beta diversity partitioning developed by Baselga and Local Contributions to Beta Diversity (LCBD) partitioning by Legendre, we examined the water-depth variations in beta diversity components of bacteria, archaea and fungi in surface sediments of Hulun Lake, a semi-arid lake in northern China, and further explored the relative importance of environmental drivers underlying their patterns. We found that the relative abundances of Proteobacteria , Chloroflexi , Euryarchaeota and Rozellomycota increased towards deep water, while Acidobacteria , Parvarchaeota and Chytridiomycota decreased. For bacteria and archaea, there were significant ( P < 0.05) decreasing water-depth patterns for LCBD and LCBD Repl (i.e., species replacement), while increasing patterns for total beta diversity and turnover, implying that total beta diversity and LCBD were dominated by species turnover or LCBD Repl . Further, bacteria showed a strong correlation with archaea regarding LCBD, total beta diversity and turnover. Such parallel patterns among bacteria and archaea were underpinned by similar ecological processes like environmental selection. Total beta diversity and turnover were largely affected by sediment total nitrogen, while LCBD and LCBD Repl were mainly constrained by water NO 2 ⁻ -N and NO 3 ⁻ -N. For fungal community variation, no significant patterns were observed, which may be due to different drivers like water nitrogen or phosphorus. Taken together, our findings provide compelling evidences for disentangling the underlying mechanisms of community variation in multiple aquatic microbial taxonomic groups.
Article
Full-text available
Although environmental impacts on the biodiversity and species composition of lakes have been studied in great detail at local and regional scales, unraveling the big picture of how lake communities respond to environmental variation across large spatial scales has received less attention. We performed a comprehensive analysis to assess how the phytoplankton community composition varies among >1,000 lakes across the conterminous United States of America. Our results show that lake-to-lake similarity in species composition was low even at the local scale, and slightly decreased with geographical distance. Analysis of the compositional data by Dirichlet regression revealed that the geographical variation in phytoplankton community composition was best explained by total phosphorus, water temperature, pH, and lake size. High total phosphorus concentrations were associated with high relative abundances of cyanobacteria and euglenophytes at the expense of other phytoplankton groups. High lake temperatures stimulated cyanobacteria, dinoflagellates, desmids and euglenophytes, whereas cryptophytes, golden algae and diatoms were relatively more abundant in colder lakes. Low lake pH correlated with high dissolved CO 2 concentrations, which may explain why it benefitted phytoplankton groups with inefficient carbon concentrating mechanisms such as golden algae and euglenophytes. Conversely, the relative abundance of cyanobacteria showed a pronounced increase with lake pH. Large lakes showed higher relative abundances of cyanobacteria and diatoms, whereas small lakes showed higher relative abundances of chlorophytes, desmids, dinoflagellates and euglenophytes. Biodiversity increased with lake temperature, but decreased at high total phosphorus concentrations and pH. The key environmental variables identified by our study (high phosphorus loads, warm temperature, low pH) are associated with anthropogenic pressures such as eutrophication, global warming and rising atmospheric CO 2 concentration. Hence, our results provide a comprehensive illustration of the major impact of these anthropogenic pressures on the biodiversity and taxonomic composition of lake phytoplankton communities.
Article
Full-text available
Interactions between multiple ecosystem stressors are expected to jeopardize biological processes, functions and biodiversity. The scientific community has declared stressor interactions—notably synergies—a key issue for conservation and management. Here, we review ecological literature over the past four decades to evaluate trends in the reporting of ecological interactions (synergies, antagonisms and additive effects) and highlight the implications and importance to conservation. Despite increasing popularity, and ever-finer terminologies, we find that synergies are (still) not the most prevalent type of interaction, and that conservation practitioners need to appreciate and manage for all interaction outcomes, including antagonistic and additive effects. However, it will not be possible to identify the effect of every interaction on every organism’s physiology and every ecosystem function because the number of stressors, and their potential interactions, are growing rapidly. Predicting the type of interactions may be possible in the near-future, using meta-analyses, conservation-oriented experiments and adaptive monitoring. Pending a general framework for predicting interactions, conservation management should enact interventions that are robust to uncertainty in interaction type and that continue to bolster biological resilience in a stressful world. © 2016 The Author(s) Published by the Royal Society. All rights reserved.
Article
Full-text available
Understanding and predicting ecosystem responses to multiple environmental pressures is a long-standing interest in ecology and environmental management. However, few studies have examined how the functional features of freshwater biological communities vary along multiple gradients of environmental stress. Furthermore, modelling these functional features for a whole river network constitutes a strong potential basis to improve ecosystem management. 2.We explored how functional redundancy of biological communities (FR, a functional feature related to the stability, resistance and resilience of ecosystems) responds to single and multiple environmental filters. We compared these responses with those of functional richness, evenness and divergence. We used riparian vegetation of a Mediterranean basin, and three of the main environmental filters affecting freshwater communities in such regions, i.e. drought, flow regulation and agricultural intensity, thus considering the potential effect of natural environmental variability. We also assessed the predictability of FR and estimated it for the entire river network. 3.We found that all functional measures decreased with increasing environmental filter intensity. However, FR was more sensitive to single and multiple environmental filters compared to other functional measures. The best-fitting model explained 59% of the FR variability and included agriculture, drought and flow regulation and the pairwise interactions of agriculture with drought and flow regulation. The parameters of the FR models differed from null model expectations reflecting a non-random decline along stress gradients. 4.Synthesis and applications. We found non-random detrimental effects along environmental filters for riparian functional redundancy (FR, the most sensitive functional index), meaning that increased stress could jeopardize stability, resistance and resilience of these systems. In general, agriculture caused the greatest impact on FR and functional diversity measures, being the most important stressor for riparian functionality in the study area. Temporary streams flowing through an agricultural, regulated basin had reduced values of FR, whereas the free-flowing medium-sized, perennial water courses flowing through unaltered sub-basins displayed higher values of FR and potentially greater stability against human impacts. All these findings along with the predicted basin-wide variation of FR can assist environmental managers in improving monitoring and ecosystem management. This article is protected by copyright. All rights reserved.
Article
Full-text available
The potential for complex synergistic or antagonistic interactions between multiple stressors presents one of the largest uncertainties when predicting ecological change but, despite common use of the terms in the scientific literature, a consensus on their operational definition is still lacking. The identification of synergism or antagonism is generally straightforward when stressors operate in the same direction, but if individual stressor effects oppose each other, the definition of synergism is paradoxical because what is synergistic to one stressor's effect direction is antagonistic to the others. In their highly cited meta-analysis, Crain et al. (Ecology Letters, 11, 2008: 1304) assumed in situations with opposing individual effects that synergy only occurs when the cumulative effect is more negative than the additive sum of the opposing individual effects. We argue against this and propose a new systematic classification based on an additive effects model that combines the magnitude and response direction of the cumulative effect and the interaction effect. A new class of “mitigating synergism” is identified, where cumulative effects are reversed and enhanced. We applied our directional classification to the dataset compiled by Crain et al. (Ecology Letters, 11, 2008: 1304) to determine the prevalence of synergistic, antagonistic, and additive interactions. Compared to their original analysis, we report differences in the representation of interaction classes by interaction type and we document examples of mitigating synergism, highlighting the importance of incorporating individual stressor effect directions in the determination of synergisms and antagonisms. This is particularly pertinent given a general bias in ecology toward investigating and reporting adverse multiple stressor effects (double negative). We emphasize the need for reconsideration by the ecological community of the interpretation of synergism and antagonism in situations where individual stressor effects oppose each other or where cumulative effects are reversed and enhanced.
Article
Full-text available
Human population growth, economic development, climate change, and the need to close the electricity access gap have stimulated the search for new sources of renewable energy. In response to this need, major new initiatives in hydropower development are now under way. At least 3,700 major dams, each with a capacity of more than 1 MW, are either planned or under construction, primarily in countries with emerging economies. These dams are predicted to increase the present global hydroelectricity capacity by 73 % to about 1,700 GW. Even such a dramatic expansion in hydropower capacity will be insufficient to compensate for the increasing electricity demand. Furthermore, it will only partially close the electricity gap, may not substantially reduce greenhouse gas emission (carbon dioxide and methane), and may not erase interdependencies and social conflicts. At the same time, it is certain to reduce the number of our planet’s remaining free-flowing large rivers by about 21 %. Clearly, there is an urgent need to evaluate and to mitigate the social, economic, and ecological ramifications of the current boom in global dam construction.
Article
Full-text available
Drastic biodiversity declines have raised concerns about the deterioration of ecosystem functions and have motivated much recent research on the relationship between species diversity and ecosystem functioning. A func-tional trait framework has been proposed to improve the mechanistic understanding of this relationship, but this has rarely been tested for organ-isms other than plants. We analysed eight datasets, including five animal groups, to examine how well a trait-based approach, compared with a more traditional taxonomic approach, predicts seven ecosystem functions below-and above-ground. Trait-based indices consistently provided greater explanatory power than species richness or abundance. The frequency distri-butions of single or multiple traits in the community were the best predictors of ecosystem functioning. This implies that the ecosystem functions we investigated were underpinned by the combination of trait identities (i.e. single-trait indices) and trait complementarity (i.e. multi-trait indices) in the communities. Our study provides new insights into the general mechan-isms that link biodiversity to ecosystem functioning in natural animal communities and suggests that the observed responses were due to the iden-tity and dominance patterns of the trait composition rather than the number or abundance of species per se.
Article
Full-text available
Aquatic and terrestrial ecosystems are linked through lateral interactions that support and maintain biodiversity in both regions. However, in many places, river regulation and channelization have isolated rivers from surrounding riparian areas. We evaluated the effects of channelization on the linkages between aquatic and terrestrial invertebrate assemblages in the Rio Grande, New Mexico via comparison of quantitative macroinvertebrate survey data and analyses of carbon and nitrogen isotopes to test for changes in nutrient flow between channelized and non-channelized reaches of this highly regulated system. Aquatic and terrestrial macroinvertebrates were surveyed in summer 2008 at channelized and non-channelized reaches. Average densities of aquatic macroinvertebrates were 50% lower in the channelized reaches. Taxonomic richness and densities of macroinvertebrates in the transition zone between the river and forested floodplain were also lower in channelized reaches and this effect was especially pronounced for predatory macroinvertebrate species. Carbon isotope ratios in consumers indicated that instream (i.e., benthic algae) production served as the major source of carbon for predaceous arthropods in the transition zone. Our results indicate that river regulation that leads to channelization can reduce diversity and macroinvertebrate densities at the landscape scale by severing linkages between the aquatic and riparian communities. These effects appear especially acute for predaceous macroinvertebrates, perhaps because preferred prey density is lowered. Restoration of natural flow regimes is vitally important to reduce channelization and maintain connectivity between the aquatic and terrestrial environments to conserve the unique assemblage of macroinvertebrates in the transition zone.