Automated discovery of food webs from ecological data using logic-based machine learning

Ecological network structure in response to community assembly processes over evolutionary time

The dynamic structure of ecological communities results from interactions among taxa that change with shifts in species composition in space and time. However, our ability to study the interplay of ecological and evolutionary processes on community assembly remains relatively unexplored due to the difficulty of measuring community structure over long temporal scales. Here, we made use of a geological chronosequence across the Hawaiian Islands, representing 50 years to 4.15 million years of ecosystem development, to sample 11 communities of arthropods and their associated plant taxa using semiquantitative DNA metabarcoding. We then examined how ecological communities changed with community age by calculating quantitative network statistics for bipartite networks of arthropod-plant associations. The average number of interactions per species (linkage density), ratio of plant to arthropod species (vulnerability) and uniformity of energy flow (interaction evenness) increased significantly in concert with community age. The index of specializationH 2 ' has a curvilinear relationship with community age. Our analyses suggest that younger communities are characterized by fewer but stronger interactions, while biotic associations become more even and diverse as communities mature. These shifts in structure became especially prominent on East Maui (~0.5 million years old) and older volcanos, after enough time had elapsed for adaptation and specialization to act on populations in situ. Such natural progression of specialization during community assembly is probably impeded by the rapid infiltration of non-native species, with special risk to younger or more recently disturbed communities that are composed of fewer specialized relationships.

Patterns and drivers of soil surface-dwelling Oribatida diversity along an altitudinal gradient on the Changbai Mountain, China

Distribution patterns of biodiversity and environmental interactions are dominant themes in ecology. In montane ecosystems, biodiversity is closely associated with altitudinal gradients. However, studies of biodiversity in montane ecosystems are focused on plants and vertebrates, with relatively less on invertebrates. Here, the present study used a Vortis arthropod suction sampler to explore the biodiversity patterns of soil surface-dwelling Oribatida and their drivers along an altitudinal gradient (600, 800, 1600, 2000, and 2300 m) from typical temperate forests, evergreen coniferous forests, subalpine birch forests to alpine tundra on the north slope of Changbai Mountain, Northeast China. Trichoribates berlesei, Platynothrus peltifer, and Oribatula tibialis were the dominant soil surface-dwelling species on Changbai Mountain. Generally, alpha diversity and beta diversity of soil surface-dwelling Oribatida decreased with the rising altitude, with a peaking density value at 2000 m. The result of beta diversity showed that the structures of community were more influenced by the species turnover component than the nestedness component. Nonmetric multidimensional scaling (NMDS) ordination showed that the community structure of soil surface-dwelling Oribatida varied significantly along the altitudinal gradient. The variance partitioning showed that the elevation and climatic conditions determined the soil surface-dwelling Oribatida community. Spatial filtering represented by geographic and elevation distances was particularly associated with soil surface-dwelling Oribatida community variation between altitudes on Changbai Mountain. However, the variation of the Oribatida community between adjacent altitudes was only associated with geographic distance. Our study provides supportive evidence for the biodiversity analyzing of soil surface-dwelling Oribatida in montane ecosystems along an altitudinal gradient.

Arthropod food webs predicted from body size ratios are improved by incorporating prey defensive properties

Trophic interactions are often deduced from body size differences, assuming that predators prefer prey smaller than themselves because larger prey are more difficult to subdue. This has mainly been confirmed in aquatic ecosystems, but rarely in terrestrial ecosystems, especially in arthropods. Our goal was to validate whether body size ratios can predict trophic interactions in a terrestrial, plant-associated arthropod community and whether predator hunting strategy and prey taxonomy could explain additional variation. We conducted feeding trials with arthropods from marram grass in coastal dunes to test whether two individuals, of the same or different species, would predate each other. From the trial results, we constructed one of the most complete, empirically derived food webs for terrestrial arthropods associated with a single plant species. We contrasted this empirical food web with a theoretical web based on body size ratios, activity period, microhabitat, and expert knowledge. In our feeding trials, predator-prey interactions were indeed largely size-based. Moreover, the theoretical and empirically based food webs converged well for both predator and prey species. However, predator hunting strategy, and especially prey taxonomy improved predictions of predation. Well-defended taxa, such as hard-bodied beetles, were less frequently consumed than expected based on their body size. For instance, a beetle of average size (measuring 4 mm) is 38% less vulnerable than another average arthropod with the same length. Body size ratios predict trophic interactions among plant-associated arthropods fairly well. However, traits such as hunting strategy and anti-predator defences can explain why certain trophic interactions do not adhere to size-based rules. Feeding trials can generate insights into multiple traits underlying real-life trophic interactions among arthropods.

Machine learning ecological networks

Deep-learning tools can help to construct historical, modern-day, and future food webs.

Inductive logic programming at 30

Inductive logic programming (ILP) is a form of logic-based machine learning. The goal is to induce a hypothesis (a logic program) that generalises given training examples and background knowledge. As ILP turns 30, we review the last decade of research. We focus on (i) new meta-level search methods, (ii) techniques for learning recursive programs, (iii) new approaches for predicate invention, and (iv) the use of different technologies. We conclude by discussing current limitations of ILP and directions for future research.

A roadmap towards predicting species interaction networks (across space and time)

Networks of species interactions underpin numerous ecosystem processes, but comprehensively sampling these interactions is difficult. Interactions intrinsically vary across space and time, and given the number of species that compose ecological communities, it can be tough to distinguish between a true negative (where two species never interact) from a false negative (where two species have not been observed interacting even though they actually do). Assessing the likelihood of interactions between species is an imperative for several fields of ecology. This means that to predict interactions between species-and to describe the structure, variation, and change of the ecological networks they form-we need to rely on modelling tools. Here, we provide a proof-of-concept, where we show how a simple neural network model makes accurate predictions about species interactions given limited data. We then assess the challenges and opportunities associated with improving interaction predictions, and provide a conceptual roadmap forward towards predictive models of ecological networks that is explicitly spatial and temporal. We conclude with a brief primer on the relevant methods and tools needed to start building these models, which we hope will guide this research programme forward. This article is part of the theme issue 'Infectious disease macroecology: parasite diversity and dynamics across the globe'.

Inferring species interactions from ecological survey data: A mechanistic approach to predict quantitative food webs of seed feeding by carabid beetles

Ecological networks are valuable for ecosystem analysis but their use is often limited by a lack of data because many types of ecological interaction, for example, predation, are short-lived and difficult to observe or detect. While there are different methods for inferring the presence of interactions, they have rarely been used to predict the interaction strengths that are required to construct weighted, or quantitative, ecological networks.Here, we develop a trait-based approach suitable for inferring weighted networks, that is, with varying interaction strengths. We developed the method for seed-feeding carabid ground beetles (Coleoptera: Carabidae) although the principles can be applied to other species and types of interaction.Using existing literature data from experimental seed-feeding trials, we predicted a per-individual interaction cost index based on carabid and seed size. This was scaled up to the population level to create inferred weighted networks using the abundance of carabids and seeds from empirical samples and energetic intake rates of carabids from the literature. From these weighted networks, we also derived a novel measure of expected predation pressure per seed type per network.This method was applied to existing ecological survey data from 255 arable fields with carabid data from pitfall traps and plant seeds from seed rain traps. Analysis of these inferred networks led to testable hypotheses about how network structure and predation pressure varied among fields.Inferred networks are valuable because (a) they provide null models for the structuring of food webs to test against empirical species interaction data, for example, DNA analysis of carabid gut regurgitates and (b) they allow weighted networks to be constructed whenever we can estimate interactions between species and have ecological census data available. This permits ecological network analysis even at times and in places when interactions were not directly assessed.

Ecosystems monitoring powered by environmental genomics: A review of current strategies with an implementation roadmap

A decade after environmental scientists integrated high-throughput sequencing technologies in their toolbox, the genomics-based monitoring of anthropogenic impacts on the biodiversity and functioning of ecosystems is yet to be implemented by regulatory frameworks. Despite the broadly acknowledged potential of environmental genomics to this end, technical limitations and conceptual issues still stand in the way of its broad application by end-users. In addition, the multiplicity of potential implementation strategies may contribute to a perception that the routine application of this methodology is premature or "in development", hence restraining regulators from binding these tools into legal frameworks. Here, we review recent implementations of environmental genomics-based methods, applied to the biomonitoring of ecosystems. By taking a general overview, without narrowing our perspective to particular habitats or groups of organisms, this paper aims to compare, review and discuss the strengths and limitations of four general implementation strategies of environmental genomics for monitoring: (a) Taxonomy-based analyses focused on identification of known bioindicators or described taxa; (b) De novo bioindicator analyses; (c) Structural community metrics including inferred ecological networks; and (d) Functional community metrics (metagenomics or metatranscriptomics). We emphasise the utility of the three latter strategies to integrate meiofauna and microorganisms that are not traditionally utilised in biomonitoring because of difficult taxonomic identification. Finally, we propose a roadmap for the implementation of environmental genomics into routine monitoring programmes that leverage recent analytical advancements, while pointing out current limitations and future research needs.

Soil Biodiversity: State-of-the-Art and Possible Implementation in Chemical Risk Assessment

Protecting the structure and functioning of soil ecosystems is one of the central aims of current regulations of chemicals. This is, for instance, shown by the emphasis on the protection of key drivers and ecosystem services as proposed in the protection goal options for soil organisms by the European Food Safety Authority (EFSA). Such targets require insight into soil biodiversity, its role in the functioning of ecosystems, and the way it responds to stress. Also required are tools and methodologies for properly assessing biodiversity. To address these issues, the Society of Environmental Toxicology and Chemistry (SETAC) Europe 14th Special Science Symposium (SESSS14) was held 19 to 20 November 2019 in Brussels, Belgium. The central aim of the SESSS14 was to provide information on how to include soil biodiversity and soil functions as protection goal options in the risk assessment and quantification of the effects of chemicals and other stressors (including their respective regulations). This paper is based on the presentations and discussions at the SESSS14 and will give a brief update on the scientific state-of-the art on soil biodiversity, novel scientific developments, experimental and modeling approaches, as well as case studies. It will also discuss how these approaches could inform future risk assessment of chemicals and other stressors in the regulatory context of protecting soil ecosystems. Integr Environ Assess Manag 2021;17:541-551. © 2020 The Authors. Integrated Environmental Assessment and Management published by Wiley Periodicals LLC on behalf of Society of Environmental Toxicology & Chemistry (SETAC).

Ecological plasticity governs ecosystem services in multilayer networks

Agriculture is under pressure to achieve sustainable development goals for biodiversity and ecosystem services. Services in agro-ecosystems are typically driven by key species, and changes in the community composition and species abundance can have multifaceted effects. Assessment of individual services overlooks co-variance between different, but related, services coupled by a common group of species. This partial view ignores how effects propagate through an ecosystem. We conduct an analysis of 374 agricultural multilayer networks of two related services of weed seed regulation and gastropod mollusc predation delivered by carabid beetles. We found that weed seed regulation increased with the herbivore predation interaction frequency, computed from the network of trophic links between carabids and weed seeds in the herbivore layer. Weed seed regulation and herbivore interaction frequencies declined as the interaction frequencies between carabids and molluscs in the carnivore layer increased. This suggests that carabids can switch to gastropod predation with community change, and that link turnover rewires the herbivore and carnivore network layers affecting seed regulation. Our study reveals that ecosystem services are governed by ecological plasticity in structurally complex, multi-layer networks. Sustainable management therefore needs to go beyond the autecological approaches to ecosystem services that predominate, particularly in agriculture.

Microbial networks inferred from environmental DNA data for biomonitoring ecosystem change: Strengths and pitfalls

Environmental DNA contains information on the species interaction networks that support ecosystem functions and services. Next-generation biomonitoring proposes the use of this data to reconstruct ecological networks in real time and then compute network-level properties to assess ecosystem change. We investigated the relevance of this proposal by assessing: (i) the replicability of DNA-based networks in the absence of ecosystem change, and (ii) the benefits and shortcomings of community- and network-level properties for monitoring change. We selected crop-associated microbial networks as a case study because they support disease regulation services in agroecosystems and analysed their response to change in agricultural practice between organic and conventional systems. Using two statistical methods of network inference, we showed that network-level properties, especially β-properties, could detect change. Moreover, consensus networks revealed robust signals of interactions between the most abundant species, which differed between agricultural systems. These findings complemented those obtained with community-level data that showed, in particular, a greater microbial diversity in the organic system. The limitations of network-level data included (i) the very high variability of network replicates within each system; (ii) the low number of network replicates per system, due to the large number of samples needed to build each network; and (iii) the difficulty in interpreting links of inferred networks. Tools and frameworks developed over the last decade to infer and compare microbial networks are therefore relevant to biomonitoring, provided that the DNA metabarcoding data sets are large enough to build many network replicates and progress is made to increase network replicability and interpretation.

Connected Carabids: Network Interactions and Their Impact on Biocontrol by Carabid Beetles

Carabid beetles can greatly contribute to biocontrol in agroecosystems, reducing both insect pests and weed seeds. However, insect foraging and feeding behavior can be highly dependent on the interaction network and spatial structure of the environment, which can make their biocontrol contributions variable. In the present article, we explore how the interaction network of carabids can affect their behavior and how spatial vegetation structure and specific agronomy practices can, in turn, affect the strength of interactions in their network. We suggest that research on carabid biocontrol should move toward an approach in which the network of interactions among pests, carabids, and other organisms within its spatial structure is evaluated, with equal focus on direct and indirect interactions, and provide examples of tools to do so. Overall, we believe this approach will improve our knowledge of carabid networks, help to elucidate the underlying mechanisms of biocontrol, and lay the foundation for future biocontrol strategies.

Characterizing ecological interaction networks to support risk assessment in classical biological control of weeds

A key element in weed biological control is the selection of a biological control agent that minimizes the risks of non-target attack and indirect effects on the recipient community. Network ecology is a promising approach that could help decipher tritrophic interactions in both the native and the invaded ranges, to complement quarantine-based host-specificity tests and gain insights on potential interactions of biological control agents. This review highlights practical questions addressed by networks, including 1) biological control agent selection, based on specialization indices, 2) risk assessment of biological control agent release into a novel environment, via particular patterns of association such as apparent competition between agent(s) and native herbivore(s), 3) network comparisons through structural metrics, 4) potential of network modelling and 5) limits of network construction methods.

Ecological networks reveal resilience of agro-ecosystems to changes in farming management

Sustainable management of ecosystems and growth in agricultural productivity is at the heart of the United Nations' Sustainable Development Goals for 2030. New management regimes could revolutionize agricultural production, but require an evaluation of the risks and opportunities. Replacing existing conventional weed management with genetically modified, herbicide-tolerant (GMHT) crops, for example, might reduce herbicide applications and increase crop yields, but remains controversial owing to concerns about potential impacts on biodiversity. Until now, such new regimes have been assessed at the species or assemblage level, whereas higher-level ecological network effects remain largely unconsidered. Here, we conduct a large-scale network analysis of invertebrate communities across 502 UK farm sites to GMHT management in different crop types. We find that network-level properties were overwhelmingly shaped by crop type, whereas network structure and robustness were apparently unaltered by GMHT management. This suggests that taxon-specific effects reported previously did not escalate into higher-level systemic structural change in the wider agricultural ecosystem. Our study highlights current limitations of autecological assessments of effect in agriculture in which species interactions and potential compensatory effects are overlooked. We advocate adopting the more holistic system-level evaluations that we explore here, which complement existing assessments for meeting our future agricultural needs.

Mapping Portuguese Natura 2000 sites in risk of biodiversity change caused by atmospheric nitrogen pollution

In this paper, we assess and map the risk that atmospheric nitrogen (atN) pollution poses to biodiversity in Natura 2000 sites in mainland Portugal. We first review the ecological impacts of atN pollution on terrestrial ecosystems, focusing on the biodiversity of Natura 2000 sites. These nature protection sites, especially those located within the Mediterranean Basin, are under-characterized regarding the risk posed by atN pollution. We focus on ammonia (NH3) because this N form is mostly associated with agriculture, which co-occurs at or in the immediate vicinity of most areas of conservation interest in Portugal. We produce a risk map integrating NH3 emissions and the susceptibility of Natura 2000 sites to atN pollution, ranking habitat sensitivity to atN pollution using expert knowledge from a panel of Portuguese ecological and habitat experts. Peats, mires, bogs, and similar acidic and oligotrophic habitats within Natura 2000 sites (most located in the northern mountains) were assessed to have the highest relative risk of biodiversity change due to atN pollution, whereas Natura 2000 sites in the Atlantic and Mediterranean climate zone (coastal, tidal, and scrubland habitats) were deemed the least sensitive. Overall, results allowed us to rank all Natura 2000 sites in mainland Portugal in order of evaluated risk posed by atN pollution. The approach is of great relevance for stakeholders in different countries to help prioritize site protection and to define research priorities. This is especially relevant in countries with a lack of expertise to assess the impacts of nitrogen on biodiversity and can represent an important step up from current knowledge in such countries.

Next-Generation Global Biomonitoring: Large-scale, Automated Reconstruction of Ecological Networks

We foresee a new global-scale, ecological approach to biomonitoring emerging within the next decade that can detect ecosystem change accurately, cheaply, and generically. Next-generation sequencing of DNA sampled from the Earth's environments would provide data for the relative abundance of operational taxonomic units or ecological functions. Machine-learning methods would then be used to reconstruct the ecological networks of interactions implicit in the raw NGS data. Ultimately, we envision the development of autonomous samplers that would sample nucleic acids and upload NGS sequence data to the cloud for network reconstruction. Large numbers of these samplers, in a global array, would allow sensitive automated biomonitoring of the Earth's major ecosystems at high spatial and temporal resolution, revolutionising our understanding of ecosystem change.

Ecoinformatics (Big Data) for Agricultural Entomology: Pitfalls, Progress, and Promise

Ecoinformatics, as defined in this review, is the use of preexisting data sets to address questions in ecology. We provide the first review of ecoinformatics methods in agricultural entomology. Ecoinformatics methods have been used to address the full range of questions studied by agricultural entomologists, enabled by the special opportunities associated with data sets, nearly all of which have been observational, that are larger and more diverse and that embrace larger spatial and temporal scales than most experimental studies do. We argue that ecoinformatics research methods and traditional, experimental research methods have strengths and weaknesses that are largely complementary. We address the important interpretational challenges associated with observational data sets, highlight common pitfalls, and propose some best practices for researchers using these methods. Ecoinformatics methods hold great promise as a vehicle for capitalizing on the explosion of data emanating from farmers, researchers, and the public, as novel sampling and sensing techniques are developed and digital data sharing becomes more widespread.

Trophic and Non-Trophic Interactions in a Biodiversity Experiment Assessed by Next-Generation Sequencing

Plant diversity affects species richness and abundance of taxa at higher trophic levels. However, plant diversity effects on omnivores (feeding on multiple trophic levels) and their trophic and non-trophic interactions are not yet studied because appropriate methods were lacking. A promising approach is the DNA-based analysis of gut contents using next generation sequencing (NGS) technologies. Here, we integrate NGS-based analysis into the framework of a biodiversity experiment where plant taxonomic and functional diversity were manipulated to directly assess environmental interactions involving the omnivorous ground beetle Pterostichus melanarius. Beetle regurgitates were used for NGS-based analysis with universal 18S rDNA primers for eukaryotes. We detected a wide range of taxa with the NGS approach in regurgitates, including organisms representing trophic, phoretic, parasitic, and neutral interactions with P. melanarius. Our findings suggest that the frequency of (i) trophic interactions increased with plant diversity and vegetation cover; (ii) intraguild predation increased with vegetation cover, and (iii) neutral interactions with organisms such as fungi and protists increased with vegetation cover. Experimentally manipulated plant diversity likely affects multitrophic interactions involving omnivorous consumers. Our study therefore shows that trophic and non-trophic interactions can be assessed via NGS to address fundamental questions in biodiversity research.

Assessing the impact of arthropod natural enemies on crop pests at the field scale

There are many reasons why it is important that we find ways to conserve, and better utilize natural enemies of invertebrate crop pests. Currently, measures of natural enemy impact are rarely incorporated into studies that purport to examine pest control. Most studies examine pest and natural enemy presence and/or abundance and then qualitatively infer impact. While this provides useful data to address a range of ecological questions, a measure of impact is critical for guiding pest management decision-making. Often some very simple techniques can be used to obtain an estimate of natural enemy impact. We present examples of field-based studies that have used cages, barriers to restrict natural enemy or prey movement, direct observation of natural enemy attack, and sentinel prey items to estimate mortality. The measure of natural enemy impact used in each study needs to be tailored to the needs of farmers and the specific pest problems they face. For example, the magnitude of mortality attributed to natural enemies may be less important than the timing and consistency of that mortality between seasons. Tailoring impact assessments will lead to research outcomes that do not simply provide general information about how to conserve natural enemies, but how to use these natural enemies as an integral part of decision-making.