Fig 5 - uploaded by Tano Gutiérrez-Cánovas
Content may be subject to copyright.
Interaction plot showing the response of native fish richness in the Sorraia basin, Portugal, against combinations of annual mean flow and % agriculture upstream. Percent agriculture was fixed at three intensities: 10 th , 50 th and 90 th percentiles. Original flow values (m 3 /s) are shown in a logarithmic scale under the log-transformed and standardised scale, to allow for a direct interpretation of flow.
Contexts in source publication
Context 1
... increase in % agriculture in the upstream catchment attenuates the strong positive effect of annual mean flow. Until finally, native fish species richness decreases, when a high annual mean flow exacerbates the negative effect of % agriculture (Fig. 4). ...
Context 2
... models is overdispersion, frequently caused by an excessive number of zeros in the response variable. As theta was equal to 1 in the final model (see model summary above), the response variable NFR was not overdispersed in our data. The diagnostic plots do not suggest any serious violation of normality and homoscedasticity of the residuals (Fig. S5.4). ...
Similar publications
During the last decades, a great number of indices have been developed throughout Europe to fulfil the requirements of the Water Framework Directive (WFD) and to assess the ecological impacts of anthropogenic pressures. The development of a methodology for assessing the ecological quality status of Greek rivers and streams was urgently needed becau...
p>Temporary rivers comprise a significant proportion of river networks globally and their prevalence is expected to increase as a result of future climate change and anthropogenic water resource pressures. Despite this, the influence of drying events on freshwater biomonitoring tools within temporary rivers has received limited research attention w...
Citations
... Accurately modelling multiple, linear and nonlinear stressor-response relationships benefits from the increase in spatial and temporal scale. As scale increases so does the ability to comprehensively characterise individual effects over longer gradients and thus to more accurately capture the nature of the interactions and their effects, thereby improving the prediction of ecosystem responses (Feld et al., 2016;Mack et al., 2022). Field surveys and experiments allow for the inclusion of larger spatial and temporal scales and of environmental contexts and thus tend to be characterised by a higher degree of realism. ...
Stressor-response models are used to detect and predict changes within ecosystems in response to anthropogenic and naturally occurring stressors. While nonlinear stressor-response relationships and interactions between stressors are common in nature, predictive models often do not account for them due to perceived difficulties in the interpretation of results. We used Irish river monitoring data from 177 river sites to investigate if multiple stressor-response models can be improved by accounting for nonlinearity, interactions in stressor-response relationships and environmental context dependencies. Out of the six models of distinct biological responses, five models benefited from the inclusion of nonlinearity while all six benefited from the inclusion of interactions. The addition of nonlinearity means that we can better see the exponential increase in Trophic Diatom Index (TDI3) as phosphorus increases, inferring ecological conditions deteriorating at a faster rate with increasing phosphorus. Furthermore, our results show that the relationship between stressor and response has the potential to be dependent on other variables, as seen in the interaction of elevation with both siltation and nutrients in relation to Ephemeroptera, Plecoptera and Trichoptera (EPT) richness. Both relationships weakened at higher elevations, perhaps demonstrating that there is a decreased capacity for resilience to stressors at lower elevations due to greater cumulative effects. Understanding interactions such as this is vital to managing ecosystems. Our findings provide empirical support for the need to further develop and employ more complex modelling techniques in environmental assessment and management.
... Variability in these pressures over different temporal and spatial scales, and for multiple systems, can create challenges for resource managers, planners, and policy makers to effectively achieve or maintain key environmental outcomes and thus ecosystem integrity (Naiman and Dudgeon, 2011;Scholes et al., 2013). Assessing ecosystem integrity is often based on how well-functioning and complete an ecosystem is, relative to its natural state, and how ecosystem services are impacted by environmental pressure(s) (Dorren et al., 2004;Feld et al., 2016;Kandziora et al., 2013). Environmental pressure may result from an individual activity such as a transportation corridor, or the aggregate of multiple pressures that may adversely impact the maintenance of ideal environmental outcomes at various spatial scales and through time. ...
The influence of increasing anthropogenic pressure on ecosystem integrity, such as land use change, is resulting in many ecosystems experiencing a decline in their ability to maintain balanced functions and services. Identifying and quantifying these pressures over different scales is challenging and thus impacting the achievement or maintenance of key environmental outcomes. In this study, a GIS-based and scalable tool was developed, the Relative Environmental Pressure (REP) Tool, to address these challenges. The REP tool combines an ecosystem
integrity conceptual framework with a weighted linear combination analysis to quantify and rank relative environmental pressure across the scale of interest. The REP Tool was developed as an automated Python-based model in a PyCharm working environment using ArcGIS Pro Arcpy scripting. The REP Tool was applied to spatially contiguous geospatial data for the Province of Alberta, Canada and dynamically scaled relative to
Hydrologic Unit Codes at level 8 (HUC8) along with regional and sub-regional scale sub-watersheds. Both cumulative and individual relative pressure levels were calculated and mapped for specific ecosystem integrity framework-derived Environmental Pressure Groups (EPGs) including Atmospheric Alteration, Sedimentation, Habitat Alteration, Hydrologic Alteration, and Social Pressure. Data driven Jenks natural breaks were then
applied to classify the relative environmental pressures into a nine-level ranking system. The resulting visualization and data outputs from the REP Tool clearly show that the highest cumulative relative environmental pressure values align with the distribution of major population centres, zones of intense agriculture and major industrial activity. These regions reflect the physiography of Alberta with the Rocky Mountain and Boreal natural regions dominated by low relative environmental pressure. As scales become smaller and more refined, the location of the higher relative environmental pressure levels typically become more subdivided with greater spatial precision where higher pressured areas are located. These patterns are repeated when looking at individual EPGs but with enhanced differentiation of pressure as scales are refined. The framework and geospatial
science driven approach behind the REP Tool can be universally applied to support enhanced understanding of relative environmental pressures in, or between regions, as well as informing adaptive environmental resource management and monitoring activities.
... Thus, full model-averaged coefficients are more conservative (Burnham and Anderson, 2002). Significant interactions were assessed considering the signs of the coefficients for the interaction and for individual terms, to assess the type, and surface plots for interpretation (Feld et al., 2016). For relevant time varying predictors, a causality test, based on the cross-correlation function, was used to determine relationships between two time series: the response variable and the relevant predictor variable, across the sampling period. ...
... To test the first hypothesis (i.e., photoreactivity is positively correlated to the abundance of aromatic terrestrially derived DOM and negatively correlated to WRT), the relationships between AQY, the initial DOC concentration, DOM quality at the start of the incubation and WRT were tested with linear regressions. In order to assess the effect of DOC and DOM quality (SUVA 254, α 420, %C 3, %C 5 , Sr) and the effect of other environmental variables (WRT, pH, TN, TP, POM, Chla) on AQY, we performed a stepwise regression (function step; R core Team, 2020) after transforming, scaling and removing collinear variables according to Feld et al. (2016). Collinear variables were either removed according to their correlation coefficients (Pearson |r| > 0.7) or automatically with the function vifstep (package usdm; Naimi et al., 2013). ...
Photochemical degradation of dissolved organic matter (DOM) has been the subject of numerous studies; however, its regulation along the inland water continuum is still unclear. We aimed to unravel the DOM photoreactivity and concurrent DOM compositional changes across 30 boreal aquatic ecosystems including peat waters, streams, rivers, and lakes distributed along a water residence time (WRT) gradient. Samples were subjected to a standardized exposure of simulated sunlight. We measured the apparent quantum yield (AQY), which corresponds to DOM photomineralization per photon absorbed, and the compositional change in DOM at bulk and individual compound levels in the original samples and after irradiation. AQY increased with the abundance of terrestrially derived DOM and decreased at higher WRT. Additionally, the photochemical changes in both DOM optical properties and molecular composition resembled changes along the natural boreal WRT gradient at low WRT (<3 years). Accordingly, mass spectrometry revealed that the abundance of photolabile and photoproduced molecules decreased with WRT along the boreal aquatic continuum. Our study highlights the tight link between DOM composition and DOM photodegradation. We suggest that photodegradation is an important driver of DOM composition change in waters with low WRT, where DOM is highly photoreactive.
... The use of species distribution models (SDMs) to identify the areas in which biodiversity is potentially concentrated, based on predicted species distributions (from species occurrence and environmental factors), represents a useful approach to overcome the limitation of incom-plete geographical databases, and has been widely used to inform conservation planning considering both current and future environmental conditions (Bellard et al. 2012;Titeux et al. 2016;Mori et al. 2018;Barnes and Delborne 2019;Lyon et al. 2019;Della Rocca et al. 2019, 2021Milanesi 2020, 2022;Howard et al. 2023;Belhaj et al. 2023b). Therefore, if we aim to anticipate biodiversity loss, area prioritization for conservation should not focus only on biodiversity values but also on the potential threats driven by these changing scenarios (Bruno et al. 2014;Newbold et al. 2015;Feld et al. 2016;Milanesi et al. 2017). Here, we present a novel approach to identify priority basins for the preservation of freshwater biodiversity using both SDMs, which enable to account for future climate change scenarios into their predictions, and estimates of future LC dynamics. ...
Freshwater ecosystems are among the most threatened worldwide. A great part of this threat comes from climate and land cover changes. This situation is specially worrying in areas and ecosystems that are highly relevant in terms of biodiversity but severely impacted by these two factors, such as water bodies in North Africa. Using water beetles as surrogates of freshwater biodiversity, we present a novel approach to identify priority basins for the preservation of freshwater biodiversity in Morocco, using Species Distribution Models to identify potential biodiversity hotspots under future climate change scenarios, and estimates of future Land Cover dynamics. The mountainous areas of the Rif and Prerif, Middle Atlas and northern Central Plateau areas, as well as some Atlantic coastal basins were identified as priority areas for water beetles conservation and will play a crucial role for the conservation of freshwater biodiversity. Indeed, these areas can act as a dead end for a number of African species in the future. However, most of these areas are poorly covered by the national protected areas network (i.e., protected area extent < 10%). In addition, our analyses revealed a general negative trend of the natural vegetation coverage in the study area, and an increase of non-irrigated cropland. Finally, we identified specific areas that need special attention because are priority for conservation but are expected to suffer intense land cover changes. These results, which consider not only biodiversity values, but also the potential threats driven by climate and land cover changes, can provide a solid scientific basis to policy-makers to define specific conservation measures to anticipate freshwater biodiversity loss in North Africa.
... occurrence as a proxy of primary success) and respectively form abundant populations (i.e. relative abundance as a proxy of secondary success), we examined changes in the occurrence and relative abundances of the species following the methodological framework proposed by Feld et al. (2016). To identify the most important variables for predicting changes in the occurrence and relative abundances of D. polymorpha and to remove redundant predictors, we used a random forest (i.e. ...
The zebra mussel Dreissena polymorpha is one of the most successful, notorious, and detrimental aquatic invasive non-native species worldwide, having invaded Europe and North America while causing substantial ecological and socioeconomic impacts. Here, we investigated the spatiotemporal trends in this species' invasion success using 178 macroinvertebrate abundance time series, containing 1451 records of D. polymorpha collected across nine European countries between 1972-2019. Using these raw (absolute) abundance data, we examined trends and drivers of occurrences and relative abundances of D. poly-morpha within invaded communities. Meta-regression models revealed non-significant trends both at the European level and for the majority of the invaded countries, except for France (significant decreasing trend) and Hungary (marginally positive trend). At the European level, the number of D. polymorpha occurrences over time followed a flat-top bell-shaped distribution, with a steep increase between 1973-1989 followed by a plateau phase prior to significantly declining post-1998. Using a series of climatic and hydromorphological site-specific characteristics of invaded and uninvaded sites from two periods (1998-2002; 2011-2015), we found that native richness, non-native abundance, distance to the next barrier, and elevation were associated with the occurrence of D. polymorpha. We also found that higher native richness and lower latitude were related to lower relative abundances. Using Cohen's D as a measure of D. polymorpha impact, we found that biodiversity within the invaded sites was initially higher than in uninvaded ones, but then declined, suggesting differences in biodiversity trends across invaded and uninvaded sites. While our results emphasise the high invasion success of D. polymorpha, increasing stressors within the context of global change-particularly ongoing climate change-are likely to enhance invasion rates and the impact of D. polymorpha in the near future, exacerbated by the lack of timely and effective management actions.
... Prior to statistical analysis, lake and species-specific adult fish body sizes (L ∞ ) and lake (biogeochemical, climate, and angling pressure) data were visualized using histograms to identify and remove outliers. Collinearity between variables was evaluated using the variance inflation factor and a threshold set at ≤8 (Feld et al. 2016). Continuous covariates were standardized (subtracting the mean from individual values and dividing by the standard deviation) prior to statistical analyses. ...
Recent studies have linked ecosystem warming to decreased mean body size in aquatic ectotherms but have often relied on experimental approaches. This study explores how climate change modifies adult body sizes in natural communities for five culturally, commercially, and recreationally important freshwater fish species that vary in thermal preferences (lake trout (Salvelinus namaycush), lake whitefish (Coregonus clupeaformis), walleye (Sander vitreus), yellow perch (Perca flavescens), and smallmouth bass (Micropterus dolomieu)). Using data for >170 000 sampling records from >600 lakes spanning the entire province of Ontario, Canada, we employ a spatial analysis technique based on Integrated Nested Laplace Approximation to detect body size changes due to warming lake conditions. With warming conditions, cold-adapted species (lake whitefish) decreased in adult body size, while cool- and warm-adapted species (walleye and smallmouth bass) increased in adult body size. Our study, like others recently conducted in marine and freshwater systems, demonstrates that the responses of fish body sizes to warming are not a unidirectional shift and may be linked to the habitat and thermal preferences of individual fish species.
... Empirical data are essential for validating predictions from mechanistic models, or to calibrate (parameterize) them to regional conditions. However, confounding factors should be controlled using statistical methods (e.g., covariates in regression; Yuan et al., 2010;Feld et al., 2016) and inference of causation from empirical SR functions should be based on strong mechanistic underpinnings of existing ecological knowledge (Poff and Zimmerman, 2010). Further, extrapolating empirical relationships beyond their data range or to other regions with different stressor contexts may be problematic. ...
Stressor-response (SR) functions quantify ecological responses to natural environmental variation or anthropogenic stressors. They are also core drivers of cumulative effects (CE) models, which are increasingly recognized as essential management tools to grapple with the diffuse footprint of human impacts. Here, we provide a process framework for the identification, development, and integration of SR functions into CE models, and highlight their consequential properties, behaviour, criteria for selecting appropriate stressors and responses, and general approaches for deriving them. Management objectives (and causal effect pathways) will determine the ultimate stressor and target response variables of interest (i.e., individual growth/survival, population size, community structure, ecosystem processes), but data availability will constrain whether proxies need to be used for the target stressor or response variables. Available data and confidence in underlying mechanisms will determine whether empirical or mechanistic (theoretical) SR functions are optimal. Uncertainty in underlying SR functions is often the primary source of error in CE modelling, and monitoring outcomes through adaptive management to iteratively refine parameterization of SR functions is a key element of model application. Dealing with stressor interactions is an additional challenge, and in the absence of known or suspected interaction mechanisms, controlling main effects should remain the primary focus. Indicators of suspected interaction presence (i.e., much larger or smaller responses to stressor reduction than expected during monitoring) should be confirmed through adaptive management cycles or targeted stressor manipulations. Where possible, management decisions should selectively take advantage of interactions to strategically mitigate stressor impacts (i.e., by using antagonisms to suppress stressor impacts, and by using synergisms to efficiently reduce them).
... To quantify and hierarchically order the impact of multiple stressors on community metrics of riverine assemblages we calculated Random Forests (RF) regression models (RandomForestSRC V.3.1.1, Ishwaran et al., 2022) for each response variable (metric) according to the cookbook provided by Feld et al. (2016b). RF are a flexible, non-parametric Machine Learning tool and are able to handle large numbers of explanatory variables even in small datasets and have shown high performances in ecological models (Knudby et al., 2010;Visser et al., 2022;see Breiman (2001) for methodological details). ...
... Extensive monitoring activities are routinely reported, however, when information is condensed into proxies for levels of ecological status, it can no longer be used to identify relationships with typical environmental conditions (Voulvoulis et al., 2017;Feld et al., 2020). There are few robust frameworks that integrate the biological and environmental data to characterise and help manage coastal ecosystems for a range of outcomes (Feld et al., 2016). Management of aquatic environments requires an understanding of the links between environmental conditions and biological communities, including predictions of how changing climatic conditions may affect these communities (Philippart et al., 2011). ...
