Conservation implications of a mismatch between data availability and demographic impact.

Cost-effective use of limited conservation resources requires understanding which data most contribute to alleviating biodiversity declines. Interventions might reasonably prioritise life-cycle transitions with the greatest influence on population dynamics, yet some contributing vital rates are particularly challenging to document. This risks managers making decisions without sufficient empirical coverage of the spatiotemporal variation experienced by the species. Here, we aimed to explore whether the number of studies contributing estimates for a given life-stage transition aligns with that transition's demographic impact on population growth rate, λ. We parameterised a matrix population model using meta-analysis of vital rates for the common eider (Somateria mollissima), an increasingly threatened yet comparatively data-rich species of seaduck, for which some life stages are particularly problematic to study. Female common eiders exhibit intermittent breeding, with some established breeders skipping one or more years between breeding attempts. Our meta-analysis yielded a breeding propensity of 0.72, which we incorporated into our model with a discrete and reversible 'nonbreeder' stage (to which surviving adults transition with a probability of 0.28). The transitions between breeding and nonbreeding states had twice the influence on λ than fertility (summed matrix-element elasticities of 24% and 11%, respectively), whereas almost 15 times as many studies document components of fertility than breeding propensity (n = 103 and n = 7, respectively). The implications of such mismatches are complex because the motivations for feasible on-the-ground conservation actions may be different from what is needed to reduce uncertainty in population projections. Our workflow could form an early part of the toolkit informing future investment of finite resources, to avoid repeated disconnects between data needs and availability thwarting evidence-led conservation.

Understanding 'it depends' in ecology: a guide to hypothesising, visualising and interpreting statistical interactions.

Ecologists routinely use statistical models to detect and explain interactions among ecological drivers, with a goal to evaluate whether an effect of interest changes in sign or magnitude in different contexts. Two fundamental properties of interactions are often overlooked during the process of hypothesising, visualising and interpreting interactions between drivers: the measurement scale - whether a response is analysed on an additive or multiplicative scale, such as a ratio or logarithmic scale; and the symmetry - whether dependencies are considered in both directions. Overlooking these properties can lead to one or more of three inferential errors: misinterpretation of (i) the detection and magnitude (Type-D error), and (ii) the sign of effect modification (Type-S error); and (iii) misidentification of the underlying processes (Type-A error). We illustrate each of these errors with a broad range of ecological questions applied to empirical and simulated data sets. We demonstrate how meta-analysis, a widely used approach that seeks explicitly to characterise context dependence, is especially prone to all three errors. Based on these insights, we propose guidelines to improve hypothesis generation, testing, visualisation and interpretation of interactions in ecology.

Ecology of large felids and their prey in small reserves of the Yucatán Peninsula of Mexico.

Jaguars and pumas are top-predator species in the Neotropics that are threatened by habitat destruction, illegal poaching of their body parts and their favored prey, and by the human-wildlife conflicts that arise when predators attack livestock. Much of the remaining felid habitat in the Americas is in protected nature reserves that are too small and isolated to support local populations. Surrounding forests therefore play a vital role in felid conservation. Successful long-term conservation of these two felids requires evidence-based knowledge of their biological and ecological requirements. We studied population distributions of jaguars and pumas and their prey in and between two small, private reserves of the Northern Yucatán Peninsula, Mexico, with areas of 25 and 43 km2. During 2 years of camera trapping (2015 and 2016), we detected 21 jaguars, from which we estimated an average space requirement of 28-45 km2/individual. Dietary niche overlap exceeded random expectation. The most frequently occurring prey items in jaguar and puma diets were collared peccary and deer. Jaguar also favored nine-banded armadillos and white-nosed coati, while puma favored canids. Both felids avoided ocellated turkey. Overall, diet of jaguars was less species-rich, but similar in niche breadth, to that of pumas. A fluid use of space by both species, in 2015 tending toward mutual attraction and in 2016 toward partial exclusion of pumas by jaguars, combined with the high dietary overlap, is consistent with a dominance hierarchy facilitating coexistence. Jaguars and pumas favor the same prey as the people in local communities who hunt, which likely will intensify human-wildlife impacts when prey become scarce. We conclude that even small reserves play an important role in increasing the continuity of habitat for prey and large felids, whose generalist habits suppress interspecific competition for increasingly limiting prey that are largely shared between them and humans.

Los jaguares y pumas son las principales especies depredadoras del Neotrópico. Se encuentran amenazados por la destrucción de su hábitat, la caza furtiva de sus partes corporales, así como de sus presas favoritas, y por los impactos entre humanos y vida silvestre que surgen cuando estas especies atacan al ganado. Gran parte del hábitat protegido de los felinos restante en las Américas lo constituyen reservas naturales que son demasiado pequeñas y aisladas para por sí mismas sustentar las poblaciones locales de estas especies. Por lo tanto, los bosques circundantes juegan un papel vital para la conservación de estos felinos. La conservación exitosa a largo plazo de estas dos especies de felinos necesita conocimiento basado en evidencia de sus requerimientos biológicos y ecológicos. Estudiamos la distribución de poblaciones de jaguares y pumas, y sus presas, en dos pequeñas áreas protegidas privadas del norte de la península de Yucatán, México, con áreas de 25 y 43 km2, y en el área no protegida de 250 km2 que se encuentra entre ellas. Durante un estudio de foto-trampeo de dos años (2015 y 2016), detectamos 21 jaguares, a partir de los cuales estimamos requerimientos espaciales de 28­45 km2/individuo en promedio. La superposición entre nichos alimentarios superó las expectativas aleatorias. Las presas más frecuentes en las dietas del jaguar y el puma fueron el pecarí de collar y los venados. El jaguar también favoreció al armadillo de nueve bandas y coatí de nariz blanca, mientras que el puma favoreció a los cánidos. Ambos felinos evitaron al pavo ocelado. En general, la dieta de los jaguares presentó menor riqueza específica, pero similar amplitud de nicho a la de los pumas. Un uso fluido del espacio por parte de ambas especies hizo que en un año tendieran a tener atracción mutua y en otro a una exclusión parcial por parte de los jaguares a los pumas, lo cual, en combinación con la alta superposición alimentaria, es consistente con una jerarquía de dominancia que facilita la convivencia. Los jaguares y los pumas favorecieron las mismas presas que la gente que caza en las comunidades locales, lo que probablemente intensificará los impactos entre humanos y vida silvestre cuando las presas escaseen. Concluimos que incluso las reservas pequeñas desempeñan un papel importante en el aumento de la continuidad del hábitat para presas y grandes felinos, cuyos hábitos generalistas suprimen la competencia inter-específica por presas cada vez más limitadas que en gran parte comparten con los humanos.

Improving quantitative synthesis to achieve generality in ecology.

Synthesis of primary ecological data is often assumed to achieve a notion of 'generality', through the quantification of overall effect sizes and consistency among studies, and has become a dominant research approach in ecology. Unfortunately, ecologists rarely define either the generality of their findings, their estimand (the target of estimation) or the population of interest. Given that generality is fundamental to science, and the urgent need for scientific understanding to curb global scale ecological breakdown, loose usage of the term 'generality' is problematic. In other disciplines, generality is defined as comprising both generalizability-extending an inference about an estimand from the sample to the population-and transferability-the validity of estimand predictions in a different sampling unit or population. We review current practice in ecological synthesis and demonstrate that, when researchers fail to define the assumptions underpinning generalizations and transfers of effect sizes, generality often misses its target. We provide guidance for communicating nuanced inferences and maximizing the impact of syntheses both within and beyond academia. We propose pathways to generality applicable to ecological syntheses, including the development of quantitative and qualitative criteria with which to license the transfer of estimands from both primary and synthetic studies.

Implications of scale dependence for cross-study syntheses of biodiversity differences.

Biodiversity studies are sensitive to well-recognised temporal and spatial scale dependencies. Cross-study syntheses may inflate these influences by collating studies that vary widely in the numbers and sizes of sampling plots. Here we evaluate sources of inaccuracy and imprecision in study-level and cross-study estimates of biodiversity differences, caused by within-study grain and sample sizes, biodiversity measure, and choice of effect-size metric. Samples from simulated communities of old-growth and secondary forests demonstrated influences of all these parameters on the accuracy and precision of cross-study effect sizes. In cross-study synthesis by formal meta-analysis, the metric of log response ratio applied to measures of species richness yielded better accuracy than the commonly used Hedges' g metric on species density, which dangerously combined higher precision with persistent bias. Full-data analyses of the raw plot-scale data using multilevel models were also susceptible to scale-dependent bias. We demonstrate the challenge of detecting scale dependence in cross-study synthesis, due to ubiquitous covariation between replication, variance and plot size. We propose solutions for diagnosing and minimising bias. We urge that empirical studies publish raw data to allow evaluation of covariation in cross-study syntheses, and we recommend against using Hedges' g in biodiversity meta-analyses.

Trophic resource partitioning drives fine-scale coexistence in cryptic bat species.

Understanding the processes that enable species coexistence has important implications for assessing how ecological systems will respond to global change. Morphology and functional similarity increase the potential for competition, and therefore, co-occurring morphologically similar but genetically unique species are a good model system for testing coexistence mechanisms. We used DNA metabarcoding and high-throughput sequencing to characterize for the first time the trophic ecology of two recently described cryptic bat species with parapatric ranges, Myotis escalerai and Myotis crypticus. We collected fecal samples from allopatric and sympatric regions and from syntopic and allotopic locations within the sympatric region to describe the diets both taxonomically and functionally and compare prey consumption with prey availability. The two bat species had highly similar diets characterized by high arthropod diversity, particularly Lepidoptera, Diptera and Araneae, and a high proportion of prey that is not volant at night, which points to extensive use of gleaning. Diet overlap at the prey item level was lower in syntopic populations, supporting trophic shift under fine-scale co-occurrence. Furthermore, the diet of M. escalerai had a marginally lower proportion of not nocturnally volant prey in syntopic populations, suggesting that the shift in diet may be driven by a change in foraging mode. Our findings suggest that fine-scale coexistence mechanisms can have implications for maintaining broad-scale diversity patterns. This study highlights the importance of including both allopatric and sympatric populations and choosing meaningful spatial scales for detecting ecological patterns. We conclude that a combination of high taxonomic resolution with a functional approach helps identify patterns of niche shift.

Borrowing from Peter to pay Paul: managing threatened predators of endangered and declining prey species.

Conservation policy and practice can sometimes run counter to their mutual aims of ensuring species survival. In Kenya, where threatened predators such as lion deplete endangered prey such as Grevy's zebra, conservation practitioners seek to ensure species success through exclusive strategies of protection, population increase and preservation. We found strong selection for the endangered Grevy's zebra by both lion and hyena on two small fenced conservancies in Kenya. Despite abundant diversity of available prey, Grevy's zebra were selected disproportionately more than their availability, while other highly available species such as buffalo were avoided. Lions were therefore not alone in presenting a credible threat to Grevy's zebra survival. Conservation practitioners must consider interlinked characteristics of prey selection, resource availability and quality, the interplay between carnivore guild members and landscape scale population trends performance in wildlife management decisions.

Microbial epidemiology and carriage studies for the evaluation of vaccines.

Respiratory tract infections are responsible for over 2.8 million deaths per year worldwide. Colonization is the first step in the process of microbes occupying the respiratory tract, which may lead to subsequent infection. Carriage, in contrast, is defined as the occupation of microbial species in the respiratory tract. The duration of carriage may be affected by host immunity, the composition and interactions between members of the microbial community, and the characteristics of colonizing bacteria, including physiology associated with being present in a bacterial biofilm. Numerous vaccines have been implemented to control infections caused by bacteria that can colonize and be subsequently carried. Such vaccines are often species-specific and may target a limited number of strains thereby creating a vacant niche in the upper respiratory tract. Epidemiological changes of bacteria found in both carriage and disease have therefore been widely reported, since the vacant niche is filled by other strains or species. In this review, we discuss the use of carriage-prevalence studies in vaccine evaluation and argue that such studies are essential for (1) examining the epidemiology of carriage before and after the introduction of new vaccines, (2) understanding the dynamics of the respiratory tract flora and (3) identifying the disease potential of emerging strains. In an era of increasing antibiotic resistance, bacterial carriage-prevalence studies are essential for monitoring the impact of vaccination programmes.

Network parameters quantify loss of assemblage structure in human-impacted lake ecosystems.

Lake biodiversity is an incomplete indicator of exogenous forcing insofar as it ignores underlying deformations of community structure. Here, we seek a proxy for deformation in a network of diatom assemblages comprising 452 species in 273 lakes across China. We test predictions from network theory that nodes of similar type will tend to self-organize in an unstressed system to a positively skewed frequency distribution of nodal degree. The empirical data reveal shifts in the frequency distributions of species associations across regions, from positive skew in lakes in west China with a history of low human impacts, to predominantly negative skew amongst lakes in highly disturbed regions in east China. Skew values relate strongly to nutrient loading from agricultural activity and urbanization, as measured by total phosphorus in lake water. Reconstructions through time show that positive skew reduces with temporal intensification of human impacts in the lake and surrounding catchments, and rises as lakes recover from disturbance. Our study illustrates how network parameters can track the loss of aquatic assemblage structure in lakes associated with human pressures.

Meta-analysis of management effects on biodiversity in plantation and secondary forests of Japan.

Conservation of temperate forest biodiversity has historically focused on natural old-growth. Less than 3% of the world's temperate forests remain unmodified by humans, however, and much of temperate-forest biodiversity is held in the predominating planted and secondary forests. Japan provides a widely applicable model for examining how to maximize biodiversity in managed temperate forests, because of its richness of forestry research generated from its vast forest area, albeit largely in Japanese, and the wide practice of its dominant management interventions across the northern temperate zone. Management for plantations includes thinning, extended rotation cycles and clear-cutting. For secondary forests regenerating from past clearance, traditional management varies in its intensities, from clear-cutting as coppices to small-scale understory clearance. Here we provide a first synthesis of published research on biodiversity in planted and secondary forests of Japan, relevant to management of these types of forest in northern temperate regions. Systematic review and meta-analyses of papers published in English and Japanese quantified management impacts on species richness and abundance of several taxa, in relation to moderator variables including stand age and management intensity. Plantation thinning substantially increases the richness and abundance of several taxa. Effect sizes decline with time since thinning for the abundance of regenerating saplings and seedlings, necessitating repeated thinning treatments every 6 years to sustain this positive effect. Taxonomic groups exhibit variable relationships with stand age in both planted and secondary forests, indicating a need to include both young and old forest stands in managed forest mosaics. We find an insufficient evidence base is available to allow for a meaningful synthesis of low-intensity management effects in historically managed secondary forests, with studies varying widely in scale and reported outcomes. We outline an agenda for the research community to achieve a systematic evaluation of scale-dependent effects of traditional forest management on biodiversity.

Correction for bias in meta-analysis of little-replicated studies.

Meta-analyses conventionally weight study estimates on the inverse of their error variance, in order to maximize precision. Unbiased variability in the estimates of these study-level error variances increases with the inverse of study-level replication. Here, we demonstrate how this variability accumulates asymmetrically across studies in precision-weighted meta-analysis, to cause undervaluation of the meta-level effect size or its error variance (the meta-effect and meta-variance).Small samples, typical of the ecological literature, induce big sampling errors in variance estimation, which substantially bias precision-weighted meta-analysis. Simulations revealed that biases differed little between random- and fixed-effects tests. Meta-estimation of a one-sample mean from 20 studies, with sample sizes of 3-20 observations, undervalued the meta-variance by c. 20%. Meta-analysis of two-sample designs from 20 studies, with sample sizes of 3-10 observations, undervalued the meta-variance by 15%-20% for the log response ratio (lnR); it undervalued the meta-effect by c. 10% for the standardized mean difference (SMD).For all estimators, biases were eliminated or reduced by a simple adjustment to the weighting on study precision. The study-specific component of error variance prone to sampling error and not parametrically attributable to study-specific replication was replaced by its cross-study mean, on the assumptions of random sampling from the same population variance for all studies, and sufficient studies for averaging. Weighting each study by the inverse of this mean-adjusted error variance universally improved accuracy in estimation of both the meta-effect and its significance, regardless of number of studies. For comparison, weighting only on sample size gave the same improvement in accuracy, but could not sensibly estimate significance.For the one-sample mean and two-sample lnR, adjusted weighting also improved estimation of between-study variance by DerSimonian-Laird and REML methods. For random-effects meta-analysis of SMD from little-replicated studies, the most accurate meta-estimates obtained from adjusted weights following conventionally weighted estimation of between-study variance.We recommend adoption of weighting by inverse adjusted-variance for meta-analyses of well- and little-replicated studies, because it improves accuracy and significance of meta-estimates, and it can extend the scope of the meta-analysis to include some studies without variance estimates.

Ecology and diversity in upper respiratory tract microbial population structures from a cross-sectional community swabbing study.

PURPOSE: Respiratory tract infections (RTIs) are responsible for over 2.8 million deaths per year worldwide with pathobiont carriage a required precursor to infection. We sought to determine carriage epidemiology for both bacterial and viral respiratory pathogens as part of a large population-based cross-sectional carriage study. METHODOLOGY: Nose self-swab samples were collected in two separate time-points, May to August 2012 (late spring/summer) and February to April 2013 (winter/early spring). The presence of six bacterial species: S. pneumoniae, H. influenzae, M. catarrhalis, S. aureus, P. aeruginosa and N. meningitidis in addition to respiratory syncytial virus, influenza viruses A and B, rhinovirus/enterovirus, coronavirus, parainfluenza viruses 1-3 and adenovirus was determined using culture and PCR methods.Results/Key findings. Carriage was shown to vary with age, recent RTI and the presence of other species. Spatial structures of microbial communities were more disordered in the 0-4 age group and those with recent RTI. Species frequency distributions were flatter than random expectation in young individuals (X2=20.42, P=0.002), indicating spatial clumping of species consistent with facilitative relationships. Deviations from a neutral model of ecological niches were observed in summer samples and from older individuals but not in the winter or younger individuals (0-4 years), suggesting the presence of seasonal and age-dependent niche processes in respiratory community assembly. CONCLUSION: The application of epidemiological methods and ecological theory to respiratory tract samples has yielded novel insights into the factors that drive microbial community composition.

Pneumococcal conjugate vaccine implementation in middle-income countries.

BACKGROUND: Since 2000, the widespread adoption of pneumococcal conjugate vaccines (PCVs) has had a major impact in the prevention of pneumonia. Limited access to international financial support means some middle-income countries (MICs) are trailing in the widespread use of PCVs. We review the status of PCV implementation, and discuss any needs and gaps related to low levels of PCV implementation in MICs, with analysis of possible solutions to strengthen the PCV implementation process in MICs. MAIN BODY: We searched PubMed, PubMed Central, Ovid MEDLINE, and SCOPUS databases using search terms related to pneumococcal immunization, governmental health policy or programmes, and MICs. Two authors independently reviewed the full text of the references, which were assessed for eligibility using pre-defined inclusion and exclusion criteria. The search terms identified 1,165 articles and the full texts of 21 were assessed for suitability, with eight articles included in the systematic review. MICs are implementing PCVs at a slower rate than donor-funded low-income countries and wealthier developed countries. A significant difference in the uptake of PCV in lower middle-income countries (LMICs) (71%) and upper middle-income countries (UMICs) (48%) is largely due to an unsuccessful process of "graduation" of MICs from GAVI assistance, an issue that arises as countries cross the income eligibility threshold and are no longer eligible to receive the same levels of financial assistance. A lack of country-specific data on disease burden, a lack of local expertise in economic evaluation, and the cost of PCV were identified as the leading causes of the slow uptake of PCVs in MICs. Potential solutions mentioned in the reviewed papers include the use of vaccine cost-effectiveness analysis and the provision of economic evidence to strengthen decision-making, the evaluation of the burden of disease, and post-introduction surveillance to monitor vaccine impact. CONCLUSION: The global community needs to recognise the impediments to vaccine introduction into MICs. Improving PCV access could help decrease the incidence of pneumonia and reduce the selection pressure for pneumococcal antimicrobial resistance.

Early warning of critical transitions in biodiversity from compositional disorder.

Global environmental change presents a clear need for improved leading indicators of critical transitions, especially those that can be generated from compositional data and that work in empirical cases. Ecological theory of community dynamics under environmental forcing predicts an early replacement of slowly replicating and weakly competitive "canary" species by slowly replicating but strongly competitive "keystone" species. Further forcing leads to the eventual collapse of the keystone species as they are replaced by weakly competitive but fast-replicating "weedy" species in a critical transition to a significantly different state. We identify a diagnostic signal of these changes in the coefficients of a correlation between compositional disorder and biodiversity. Compositional disorder measures unpredictability in the composition of a community, while biodiversity measures the amount of species in the community. In a stochastic simulation, sequential correlations over time switch from positive to negative as keystones prevail over canaries, and back to positive with domination of weedy species. The model finds support in empirical tests on multi-decadal time series of fossil diatom and chironomid communities from lakes in China. The characteristic switch from positive to negative correlation coefficients occurs for both communities up to three decades preceding a critical transition to a sustained alternate state. This signal is robust to unequal time increments that beset the identification of early-warning signals from other metrics.

Similar biodiversity of ectomycorrhizal fungi in set-aside plantations and ancient old-growth broadleaved forests.

Setting aside overmature planted forests is currently seen as an option for preserving species associated with old-growth forests, such as those with dispersal limitation. Few data exist, however, on the utility of set-aside plantations for this purpose, or the value of this habitat type for biodiversity relative to old-growth semi-natural ecosystems. Here, we evaluate the contribution of forest type relative to habitat characteristics in determining species richness and composition in seven forest blocks, each containing an ancient old-growth stand (> 1000 yrs) paired with a set-aside even-aged planted stand (ca. 180 yrs). We investigated the functionally important yet relatively neglected ectomycorrhizal fungi (EMF), a group for which the importance of forest age has not been assessed in broadleaved forests. We found that forest type was not an important determinant of EMF species richness or composition, demonstrating that set-aside can be an effective option for conserving ancient EMF communities. Species richness of above-ground EMF fruiting bodies was principally related to the basal area of the stand (a correlate of canopy cover) and tree species diversity, whilst richness of below-ground ectomycorrhizae was driven only by tree diversity. Our results suggest that overmature planted forest stands, particularly those that are mixed-woods with high basal area, are an effective means to connect and expand ecological networks of ancient old-growth forests in historically deforested and fragmented landscapes for ectomycorrhizal fungi.

The influence of simulated exploitation on Patella vulgata populations: protandric sex change is size-dependent.

Grazing mollusks are used as a food resource worldwide, and limpets are harvested commercially for both local consumption and export in several countries. This study describes a field experiment to assess the effects of simulated human exploitation of limpets Patella vulgata on their population ecology in terms of protandry (age-related sex change from male to female), growth, recruitment, migration, and density regulation. Limpet populations at two locations in southwest England were artificially exploited by systematic removal of the largest individuals for 18 months in plots assigned to three treatments at each site: no (control), low, and high exploitation. The shell size at sex change (L 50: the size at which there is a 50:50 sex ratio) decreased in response to the exploitation treatments, as did the mean shell size of sexual stages. Size-dependent sex change was indicated by L 50 occurring at smaller sizes in treatments than controls, suggesting an earlier switch to females. Mean shell size of P. vulgata neuters changed little under different levels of exploitation, while males and females both decreased markedly in size with exploitation. No differences were detected in the relative abundances of sexual stages, indicating some compensation for the removal of the bigger individuals via recruitment and sex change as no migratory patterns were detected between treatments. At the end of the experiment, 0-15 mm recruits were more abundant at one of the locations but no differences were detected between treatments. We conclude that sex change in P. vulgata can be induced at smaller sizes by reductions in density of the largest individuals reducing interage class competition. Knowledge of sex-change adaptation in exploited limpet populations should underpin strategies to counteract population decline and improve rocky shore conservation and resource management.

Drivers of the composition and diversity of carabid functional traits in UK coniferous plantations.

Functional diversity (FD) is increasingly used as a metric to evaluate the impact of forest management strategies on ecosystem functioning. Management interventions that aim to maximise FD require knowledge of multiple environmental drivers of FD, which have not been studied to date in temperate coniferous production forests. We quantified the relative importance of abiotic (forest management) and biotic (ground vegetation community) drivers of carabid FD and trait distribution in 44 coniferous plantation forest stands across the UK. Carabid FD declined with canopy cover and carabid body length correlated negatively with the percentage of open semi-natural area surrounding a plot. We conclude that forest management could enhance carabid FD through initiatives that emulate natural disturbance regimes through gap creation. We found that neither functional nor taxonomic metrics of vegetation diversity correlated with carabid FD, suggesting that restoration of plant communities, a major goal of forest restoration efforts, will not necessarily enhance carabid FD in coniferous plantations.

What can ecosystems learn? Expanding evolutionary ecology with learning theory.

BACKGROUND: The structure and organisation of ecological interactions within an ecosystem is modified by the evolution and coevolution of the individual species it contains. Understanding how historical conditions have shaped this architecture is vital for understanding system responses to change at scales from the microbial upwards. However, in the absence of a group selection process, the collective behaviours and ecosystem functions exhibited by the whole community cannot be organised or adapted in a Darwinian sense. A long-standing open question thus persists: Are there alternative organising principles that enable us to understand and predict how the coevolution of the component species creates and maintains complex collective behaviours exhibited by the ecosystem as a whole? RESULTS: Here we answer this question by incorporating principles from connectionist learning, a previously unrelated discipline already using well-developed theories on how emergent behaviours arise in simple networks. Specifically, we show conditions where natural selection on ecological interactions is functionally equivalent to a simple type of connectionist learning, 'unsupervised learning', well-known in neural-network models of cognitive systems to produce many non-trivial collective behaviours. Accordingly, we find that a community can self-organise in a well-defined and non-trivial sense without selection at the community level; its organisation can be conditioned by past experience in the same sense as connectionist learning models habituate to stimuli. This conditioning drives the community to form a distributed ecological memory of multiple past states, causing the community to: a) converge to these states from any random initial composition; b) accurately restore historical compositions from small fragments; c) recover a state composition following disturbance; and d) to correctly classify ambiguous initial compositions according to their similarity to learned compositions. We examine how the formation of alternative stable states alters the community's response to changing environmental forcing, and we identify conditions under which the ecosystem exhibits hysteresis with potential for catastrophic regime shifts. CONCLUSIONS: This work highlights the potential of connectionist theory to expand our understanding of evo-eco dynamics and collective ecological behaviours. Within this framework we find that, despite not being a Darwinian unit, ecological communities can behave like connectionist learning systems, creating internal conditions that habituate to past environmental conditions and actively recalling those conditions.

Ecosystems: the rocky road to regime-shift indicators.

Ecosystems can undergo dramatic shifts from one stable state to another. While indicators of such shifts are well known, experimental tests are few and far between. A new study on rocky shore ecosystems now offers a test of warning indicators.