Bayesian networks as a novel tool to enhance interpretability and predictive power of ecological models

In today's world, it is becoming increasingly important to have the tools to understand, and ultimately to predict, the response of ecosystems to disturbance. However, understanding such dynamics is not simple. Ecosystems are a complex network of species interactions, and therefore any change to a population of one species will have some degree of community level effect. In recent years, the use of Bayesian networks (BNs) has seen successful applications in molecular biology and ecology, where they were able to recover plausible links in the respective systems they were applied to. The recovered network also comes with a quantifiable metric of interaction strength between variables. While the latter is an invaluable piece of information in ecology, an unexplored application of BNs would be using them as a novel variable selection tool in the training of predictive models. To this end, we evaluate the potential usefulness of BNs in two aspects: (1) we apply BN inference on species abundance data from a rocky shore ecosystem, a system with well documented links, to test the ecological validity of the revealed network; and (2) we evaluate BNs as a novel variable selection method to guide the training of an artificial neural network (ANN). Here, we demonstrate that not only was this approach able to recover meaningful species interactions networks from ecological data, but it also served as a meaningful tool to inform the training of predictive models, where there was an improvement in predictive performance in models with BN variable selection. Combining these results, we demonstrate the potential of this novel application of BNs in enhancing the interpretability and predictive power of ecological models; this has general applicability beyond the studied system, to ecosystems where existing relationships between species and other functional components are unknown.     

Movement,impacts and management of plant distributions in response to climate change: insights from invasions

Synthesis Prediction and management of species responses to climate change is an urgent but relatively young research field. Therefore, climate change ecology must by necessity borrow from other fields. Invasion ecology is particularly well‐suited to informing climate change ecology because both invasion ecology and climate change ecology address the trajectories of rapidly changing novel systems. Here we outline the broad range of active research questions in climate change ecology where research from invasion ecology can stimulate advances. We present ideas for how concepts, case‐studies and methodology from invasion ecology can be adapted to improve prediction and management of species responses to climate change. A major challenge in this era of rapid climate change is to predict changes in species distributions and their impacts on ecosystems, and, if necessary, to recommend management strategies for maintenance of biodiversity or ecosystem services. Biological invasions, studied in most biomes of the world, can provide useful analogs for some of the ecological consequences of species distribution shifts in response to climate change. Invasions illustrate the adaptive and interactive responses that can occur when species are confronted with new environmental conditions. Invasion ecology complements climate change research and provides insights into the following questions: 1) how will species distributions respond to climate change? 2) how will species movement affect recipient ecosystems? And 3) should we, and if so how can we, manage species and ecosystems in the face of climate change? Invasion ecology demonstrates that a trait‐based approach can help to predict spread speeds and impacts on ecosystems, and has the potential to predict climate change impacts on species ranges and recipient ecosystems. However, there is a need to analyse traits in the context of life‐history and demography, the stage in the colonisation process (e.g. spread, establishment or impact), the distribution of suitable habitats in the landscape, and the novel abiotic and biotic conditions under which those traits are expressed. As is the case with climate change, invasion ecology is embedded within complex societal goals. Both disciplines converge on similar questions of ‘when to intervene?‘ and ‘what to do?‘ which call for a better understanding of the ecological processes and social values associated with changing ecosystems.     

Climate‐change impacts on sandy‐beach biota: crossing a line in the sand

Sandy ocean beaches are iconic assets that provide irreplaceable ecosystem services to society. Despite their great socioeconomic importance, beaches as ecosystems are severely under‐represented in the literature on climate‐change ecology. Here, we redress this imbalance by examining whether beach biota have been observed to respond to recent climate change in ways that are consistent with expectations under climate change. We base our assessments on evidence coming from case studies on beach invertebrates in South America and on sea turtles globally. Surprisingly, we find that observational evidence for climate‐change responses in beach biota is more convincing for invertebrates than for highly charismatic turtles. This asymmetry is paradoxical given the better theoretical understanding of the mechanisms by which turtles are likely to respond to changes in climate. Regardless of this disparity, knowledge of the unique attributes of beach systems can complement our detection of climate‐change impacts on sandy‐shore invertebrates to add rigor to studies of climate‐change ecology for sandy beaches. To this end, we combine theory from beach ecology and climate‐change ecology to put forward a suite of predictive hypotheses regarding climate impacts on beaches and to suggest ways that these can be tested. Addressing these hypotheses could significantly advance both beach and climate‐change ecology, thereby progressing understanding of how future climate change will impact coastal ecosystems more generally.     

Climate change, biotic interactions and ecosystem services

Climate change is real. The wrangling debates are over, and we now need to move onto a predictive ecology that will allow managers of landscapes and policy makers to adapt to the likely changes in biodiversity over the coming decades. There is ample evidence that ecological responses are already occurring at the individual species (population) level. The challenge is how to synthesize the growing list of such observations with a coherent body of theory that will enable us to predict where and when changes will occur, what the consequences might be for the conservation and sustainable use of biodiversity and what we might do practically in order to maintain those systems in as good condition as possible. It is thus necessary to investigate the effects of climate change at the ecosystem level and to consider novel emergent ecosystems composed of new species assemblages arising from differential rates of range shifts of species. Here, we present current knowledge on the effects of climate change on biotic interactions and ecosystem services supply, and summarize the papers included in this volume. We discuss how resilient ecosystems are in the face of the multiple components that characterize climate change, and suggest which current ecological theories may be used as a starting point to predict ecosystem-level effects of climate change.     

Temporal ecology in the Anthropocene

Two fundamental axes – space and time – shape ecological systems. Over the last 30 years spatial ecology has developed as an integrative, multidisciplinary science that has improved our understanding of the ecological consequences of habitat fragmentation and loss. We argue that accelerating climate change – the effective manipulation of time by humans – has generated a current need to build an equivalent framework for temporal ecology. Climate change has at once pressed ecologists to understand and predict ecological dynamics in non‐stationary environments, while also challenged fundamental assumptions of many concepts, models and approaches. However, similarities between space and time, especially related issues of scaling, provide an outline for improving ecological models and forecasting of temporal dynamics, while the unique attributes of time, particularly its emphasis on events and its singular direction, highlight where new approaches are needed. We emphasise how a renewed, interdisciplinary focus on time would coalesce related concepts, help develop new theories and methods and guide further data collection. The next challenge will be to unite predictive frameworks from spatial and temporal ecology to build robust forecasts of when and where environmental change will pose the largest threats to species and ecosystems, as well as identifying the best opportunities for conservation.     

Spatial perspectives enhance modeling of nanomaterial risks

Novel engineered nanomaterials (ENMs) are increasingly being manufactured and integrated into renewable energy generation and storage technologies. Past research estimated the potential impact of this increased demand on environmental systems, due to both the life cycle impact of ENM production and the potential for their direct release into ecosystems. However, many models treat ENM production and use as spatially implicit, without considering the specific geographic location of potential emissions. By not considering geographical context, ENM accumulation or impact may be underestimated. Here, we introduce an integrated predictive model that forecasts likely ENM manufacturing locations and potential emissions to the environment, with a focus on critical environmental areas and freshwater ecosystems. Spatially explicit ENM concentrations are estimated for four case study ENMs that have promising application in lithium‐ion battery production. Results demonstrate that potential ENM exposure from manufacturing locations within buffer zones of sensitive ecosystems would accumulate to levels associated with measured ecotoxicity risk under high release scenarios, underscoring the importance of adding a spatial and temporal perspective to life cycle toxicity impact assessment. This predictive integrated modeling approach is novel to the nanomaterial literature and can be adapted to other regions and material case studies to proactively inform life cycle tradeoffs and decision‐making.     

Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers

Invasions by exotic species represent both threats to ecosystems as well as opportunities to learn more about them. Among the invaders that will have the largest impacts are those that directly modify ecosystems and thus have cascading effects for resident biota. Exotics can affect ecosystems by altering system-level flows, availability, or quality of nutrients, food, and physical resources (e.g. living space, water, heat or light). The invader-mediated control of physical resources, typically achieved through the modification of habitats, has received limited attention in invasion biology. This reflects a general trend in ecology, and only recently has the concept of ecosystem engineering been developed to account for the role of species that shape habitats. Plants and animals in terrestrial and aquatic systems can both create and destroy structure. When introduced into ecosystems, these exotic engineers cause physical state changes with effects that ramify throughout the system. Although the consequences of these modifications are varied and complex, insight gained from general ecological principles offers an opportunity to predict what invaders will do upon their integration into systems. Examples from the literature suggest that introduced ecosystem engineers that increase habitat complexity or heterogeneity tend to cause abundances and/or species richness to rise, while those that decrease complexity tend to have the reverse effect. In assessing such patterns, however, it is critical to also consider spatial scales and the life habits of resident organisms. In addition to providing predictive power, recognition of engineering as a major means by which invasive species affect ecosystems provides a unifying theme for invasion biology and offers a chance to consider more fully the general role of species in ecosystems.     

The emergence of the social‐ecological restoration concept

Many ecosystems in the world are the result of a close interaction between local people and their environment, which are currently recognized as social‐ecological systems (SoES). Natural catastrophes or long‐standing social and political turmoil can degrade these SoES to a point where human societies are no longer autonomous and their supporting ecosystems are highly degraded. Here, we focus on the special case of the restoration of SoES that we call social‐ecological restoration (SoER), which is characterized as a restoration process that cannot avoid simultaneously dealing with ecological and social issues. In practice, SoER is analogous in many ways to the general principles of ecological restoration, but it differs in three key aspects: (1) the first actions may be initially intended for human groups that need to recover minimum living standards; (2) the SoER process would often be part of a healing process for local people where cultural values of ecosystems play an essential role; and (3) there is a strong dependency on external economic inputs, as the people belonging to the SoES may be incapable of reorganizing themselves on their own and supporting ecosystems can no longer self‐recover. Although it might not be desirable or necessary to call all restoration projects with a social component an SoER, the use of this concept may help in defining early restoration targets that may prevent conflicts among users in the long term. From the perspective of other disciplines, SoER would be more appropriately perceived as programs of “social‐ecological recovery” in the long term.     

微生物生态学理论框架

微生物是生态系统的重要组成部分,直接或间接地参与所有的生态过程。微生物生态学是基于微生物群体的科学,利用微生物群体DNA/RNA等标志物,重点研究微生物群落构建、组成演变、多样性及其与环境的关系,在生态学理论的指导和反复模型拟合下由统计分析得出具有普遍意义的结论。其研究范围从基因尺度到全球尺度。分子生物学技术的发展,使人们可以直接从基因水平上考查其多样性,从而使得对微生物空间分布格局及其成因的深入研究成为可能。进而可以从方法学探讨微生物生物多样性、分布格局、影响机制及其对全球变化的响应等。在微生物生态学研究中,群落构建与演化、分布特征(含植物-微生物相互关系)、执行群体功能的机理(生物地球化学循环等)、对环境变化的响应与反馈机理是今后需要关注的重点领域。概述了微生物生态学的概念,并初步提出其理论框架,在对比宏观生态学基础理论和模型的基础上,分析微生物多样性的研究内容、研究方法和群落构建的理论机制,展望了今后研究的重点领域。     

Fundamental unpredictability in multispecies competition

One of the central goals of ecology is to predict the distribution and abundance of organisms. Here, we show that, in ecosystems of high biodiversity, the outcome of multispecies competition can be fundamentally unpredictable. We consider a competition model widely applied in phytoplankton ecology and plant ecology in which multiple species compete for three resources. We show that this competition model may have several alternative outcomes, that the dynamics leading to these alternative outcomes may exhibit transient chaos, and that the basins of attraction of these alternative outcomes may have an intermingled fractal geometry. As a consequence of this fractal geometry, it is impossible to predict the winners of multispecies competition in advance.     

Predictive models in ecology: Comparison of performances and assessment of applicability

Ecological systems are governed by complex interactions which are mainly nonlinear. In order to capture the inherent complexity and nonlinearity of ecological, and in general biological systems, empirical models recently gained popularity. However, although these models, particularly connectionist approaches such as multilayered backpropagation networks, are commonly applied as predictive models in ecology to a wide variety of ecosystems and questions, there are no studies to date aiming to assess the performance, both in terms of data fitting and generalizability, and applicability of empirical models in ecology. Our aim is hence to provide an overview for nature of the wide range of the data sets and predictive variables, from both aquatic and terrestrial ecosystems with different scales of time-dependent dynamics, and the applicability and robustness of predictive modeling methods on such data sets by comparing different empirical modeling approaches. The models used in this study range from predicting the occurrence of submerged plants in shallow lakes to predicting nest occurrence of bird species from environmental variables and satellite images. The methods considered include k-nearest neighbor (k-NN), linear and quadratic discriminant analysis (LDA and QDA), generalized linear models (GLM) feedforward multilayer backpropagation networks and pseudo-supervised network ARTMAP.Our results show that the predictive performances of the models on training data could be misleading, and one should consider the predictive performance of a given model on an independent test set for assessing its predictive power. Moreover, our results suggest that for ecosystems involving time-dependent dynamics and periodicities whose frequency are possibly less than the time scale of the data considered, GLM and connectionist neural network models appear to be most suitable and robust, provided that a predictive variable reflecting these time-dependent dynamics included in the model either implicitly or explicitly. For spatial data, which does not include any time-dependence comparable to the time scale covered by the data, on the other hand, neighborhood based methods such as k-NN and ARTMAP proved to be more robust than other methods considered in this study. In addition, for predictive modeling purposes, first a suitable, computationally inexpensive method should be applied to the problem at hand a good predictive performance of which would render the computational cost and efforts associated with complex variants unnecessary.     

Biodiversity matters in a changing world

It is now widely accepted that the climate of our planet is changing, but it is still hard to predict the consequences of these changes on ecosystems. The impact is worst at the poles, with scientists concerned that impacts at lower latitudes will follow suit. Canada has a great responsibility and potential for studying the effects of climate changes on the ecological dynamics, given its geographical location and its scientific leadership in this field. The 5th annual meeting of the Canadian Society for Ecology and Evolution was held in the International Year of Biodiversity, to share recent advances in a wide variety of disciplines ranging from molecular biology to behavioural ecology, and to integrate them into a general view that will help us preserve biodiversity and limit the impact of climate change on ecosystems.     

On the inconsistency of pollinator species traits for predicting either response to land‐use change or functional contribution

The response and effect trait framework, if supported empirically, would provide for powerful and general predictions about how biodiversity loss leads to loss in ecosystem function. This framework proposes that species traits will explain how different species respond to disturbance (i.e. response traits) as well as their contribution to ecosystem function (i.e. effect traits). However, predictive response and effect traits remain elusive for most systems. Here, we use data on crop pollination services provided by native, wild bees to explore the role of six commonly used species traits in determining both species’ response to land‐use change and the subsequent effect on crop pollination. Analyses were conducted in parallel for three crop systems (watermelon, cranberry, and blueberry) located within the same geographical region (mid‐Atlantic USA). Bee species traits did not strongly predict species’ response to land‐use change, and the few traits that were weakly predictive were not consistent across crops. Similarly, no trait predicted species’ overall functional contribution in any of the three crop systems, although body size was a good predictor of per capita efficiency in two systems. Overall we were unable to make generalizable predictions regarding species responses to land‐use change and its effect on the delivery of crop pollination services. Pollinator traits may be useful for understanding ecological processes in some systems, but thus far the promise of traits‐based ecology has yet to be fulfilled for pollination ecology.     

Predicting phenology by integrating ecology,evolution and climate science

Forecasting how species and ecosystems will respond to climate change has been a major aim of ecology in recent years. Much of this research has focused on phenology – the timing of life‐history events. Phenology has well‐demonstrated links to climate, from genetic to landscape scales; yet our ability to explain and predict variation in phenology across species, habitats and time remains poor. Here, we outline how merging approaches from ecology, climate science and evolutionary biology can advance research on phenological responses to climate variability. Using insight into seasonal and interannual climate variability combined with niche theory and community phylogenetics, we develop a predictive approach for species’ reponses to changing climate. Our approach predicts that species occupying higher latitudes or the early growing season should be most sensitive to climate and have the most phylogenetically conserved phenologies. We further predict that temperate species will respond to climate change by shifting in time, while tropical species will respond by shifting space, or by evolving. Although we focus here on plant phenology, our approach is broadly applicable to ecological research of plant responses to climate variability.     

Functional group/guild modelling of inter-specific pathogen interactions: a potential tool for predicting the consequences of co-infection

Although co-infection is the norm in most human and animal populations, clinicians currently have no practical tool to assist them in choosing the best treatment strategy for such patients. Given the vast range of potential pathogens which may co-infect the host, obtaining such a practical tool may seem an intractable problem. In ecology the joint concepts of functional groups and guilds have been used to conceptually simplify complex ecosystems, in order to understand how their component parts interact and may be manipulated. Here we propose a mechanism by which to apply these concepts to pathogen co-infection systems. Further, we describe how these groups could be incorporated into a mathematical modelling framework which, after validation, could be used as a clinical tool to predict the outcome of any particular combination of pathogens co-infecting a host.     

How common are invasion-induced ecosystem impacts?

It would be an exaggeration to argue that most invasions produce ecosystem impacts, and the term should be reserved for cases in which many species in an ecosystem are affected. However, certain facts suggest that true ecosystem impacts are more common than is normally assumed. First, the term “ecosystem impact” has often been reserved for cases in which the nutrient regime or nutrient cycling is affected, whereas physical structural and other changes in ecosystems ought to be included. Second, as with all ecology, very few systems and species have been studied relative to all those that exist, so it is likely that many ecosystem impacts remain to be detected. Third, there are many types of impacts, many are idiosyncratic, many are subtle, and many are indirect, so it is likely that many impacts have simply not been recognized even in studied systems. Finally, the frequency of the lag phenomenon in invasions implies that at least some existing non-native species that are currently having little or no impact will eventually have much greater ones. These facts suggest that, even if it would be an overstatement to say that most invasions cause ecosystem impacts, it would not be more of an overstatement than the common assertion that very few introduced species have any significant impact.     

From molecules to dynamic biological communities

Microbial ecology is flourishing, and in the process, is making contributions to how the ecology and biology of large organisms is understood. Ongoing advances in sequencing technology and computational methods have enabled the collection and analysis of vast amounts of molecular data from diverse biological communities. While early studies focused on cataloguing microbial biodiversity in environments ranging from simple marine ecosystems to complex soil ecologies, more recent research is concerned with community functions and their dynamics over time. Models and concepts from traditional ecology have been used to generate new insight into microbial communities, and novel system-level models developed to explain and predict microbial interactions. The process of moving from molecular inventories to functional understanding is complex and challenging, and never more so than when many thousands of dynamic interactions are the phenomena of interest. We outline the process of how epistemic transitions are made from producing catalogues of molecules to achieving functional and predictive insight, and show how those insights not only revolutionize what is known about biological systems but also about how to do biology itself. Examples will be drawn primarily from analyses of different human microbiota, which are the microbial consortia found in and on areas of the human body, and their associated microbiomes (the genes of those communities). Molecular knowledge of these microbiomes is transforming microbiological knowledge, as well as broader aspects of human biology, health and disease.     

Assessing the ecology in ecotoxicology: a review and synthesis in freshwater systems

The field of ecotoxicology is experiencing a surge in attention among ecologists as we gain a deeper appreciation for how contaminants can impact natural ecosystems. This interest is particularly strong in aquatic systems where many non-target organisms experience pesticides. In this article, we assess how pesticides affect freshwater systems by applying the conceptual framework of density- and trait-mediated indirect effects from the field of basic ecology. We demonstrate the utility of this framework for understanding the conditions under which pesticides affect species interactions, communities and ecosystems. Through the integration of laboratory toxicity tests and this ecological framework, ecotoxicologists should be better able to identify the mechanisms through which pesticides affect communities and ecosystems. We also identify several areas of research that are in critical need of empirical attention including synergistic effects between pesticides and natural stressors, the importance of pesticides on community assembly via habitat preferences and oviposition effects, the timing and frequency of pesticide applications, pesticide effects on population dynamics, the evolution of pesticide resistance in non-target organisms and ecosystem recovery. With this knowledge, one can improve upon management decisions and help protect non-target species that are of conservation concern.     

Altered ignition catchments threaten a hyperdiverse fire‐dependent ecosystem

Human activities affect fire in many ways, often unintentionally or with considerable time‐lags before they manifest themselves. Anticipating these changes is critical, so that insidious impacts on ecosystems, their biodiversity and associated goods and services can be avoided, mitigated or managed. Here we explore the impact of anthropogenic land cover change on fire and biodiversity in adjacent ecosystems on the hyperdiverse Cape Peninsula, South Africa. We develop a conceptual framework based on the notion of an ignition catchment, or the spatial extent and temporal range where an ignition is likely to result in a site burning. We apply this concept using fire models to estimate spatial changes in burn probability between historical and current land cover. This change layer was used to predict the observed record of fires and forest encroachment into fire‐dependent Fynbos ecosystems in Table Mountain National Park. Urban expansion has created anthropogenic fire shadows that are modifying fire return intervals, facilitating a state shift to low‐diversity, non‐flammable forest at the expense of hyperdiverse, flammable Fynbos ecosystems. Despite occurring in a conservation area, these ecosystems are undergoing a hidden collapse and desperately require management intervention. Anthropogenic fire shadows can be caused by many human activities and are likely to be a universal phenomenon, not only contributing to the observed global decline in fire activity but also causing extreme fires in ecosystems where there is no shift to a less flammable state and flammable fuels accumulate. The ignition catchment framework is highly flexible and allows detection or prediction of changes in the fire regime, the threat this poses for ecosystems or fire risk and areas where management interventions and/or monitoring are required. Identifying anthropogenic impacts on ignition catchments is key for both understanding global impacts of humans on fire and guiding management of human‐altered landscapes for desirable outcomes.     

Riverine landscapes: taking landscape ecology into the water

1. Landscape ecology deals with the influence of spatial pattern on ecological processes. It considers the ecological consequences of where things are located in space, where they are relative to other things, and how these relationships and their consequences are contingent on the characteristics of the surrounding landscape mosaic at multiple scales in time and space. Traditionally, landscape ecologists have focused their attention on terrestrial ecosystems, and rivers and streams have been considered either as elements of landscape mosaics or as units that are linked to the terrestrial landscape by flows across boundaries or ecotones. Less often, the heterogeneity that exists within a river or stream has been viewed as a `riverscape' in its own right.
2. Landscape ecology can be unified about six central themes: (1) patches differ in quality (2) patch boundaries affect flows, (3) patch context matters, (4) connectivity is critical, (5) organisms are important, and (6) the importance of scale. Although riverine systems differ from terrestrial systems by virtue of the strong physical force of hydrology and the inherent connectivity provided by water flow, all of these themes apply equally to aquatic and terrestrial ecosystems, and to the linkages between the two.
3. Landscape ecology therefore has important insights to offer to the study of riverine ecosystems, but these systems may also provide excellent opportunities for developing and testing landscape ecological theory. The principles and approaches of landscape ecology should be extended to include freshwater systems; it is time to take the `land' out of landscape ecology.