The problem of scale in the prediction and management of pathogen spillover

Bats and ectoparasites: exploring a hidden link in zoonotic disease transmission

Bats are increasingly in the focus of disease surveillance studies as they harbor pathogens that can cause severe human disease. In other host groups, ectoparasitic arthropods play an important role in transmitting pathogens to humans. Nevertheless, we currently know little about the role of bat-associated ectoparasites in pathogen transmission, not only between bats but also to humans and other species, even though some of these parasites occasionally feed on humans and harbor potentially zoonotic organisms. In this work, we summarize current knowledge on the zoonotic risks linked to bat-associated ectoparasites and provide novel risk assessment guidelines to improve targeted surveillance efforts. Additionally, we suggest research directions to help adjust surveillance strategies and to better understand the eco-epidemiological role of these parasites.

A protocol and a data-based prediction to investigate virus spillover at the wildlife interface in human-dominated and protected habitats in Thailand: The Spillover Interface project

The Spillover Interface Project aims at assessing the encounter of wildlife, domestic animals, and humans along a landscape gradient from a protected area to a residential community, through areas of reforestation and agricultural land. Here, we present the protocols of the project that combine virus screening in humans, bats, rodents and dogs with camera trapping, land-use characterization, and network analyses. The project is taking place in the sub-district of Saen Thong (Nan Province, Thailand) in collaboration with local communities, the District Public Health Office, and Nanthaburi National Park. To formulate a predictive hypothesis for the Spillover Interface Project, we assess the wildlife diversity and their viral diversity that could be observed in Saen Thong through a data science analysis approach. Potential mammalian species are estimated using data from the International Union for Conservation of Nature (IUCN) and their associated viral diversity from a published open database. A network analysis approach is used to represent and quantify the transmission of the potential viruses hosted by the mammals present in Saen Thong, according to the IUCN. A total of 57 viruses are expected to be found and shared between 43 host species, including the domestic dog and the human species. By following the protocols presented here, the Spillover Interface Project will collect the data and samples needed to test this data-driven prediction.

Predicting the fine-scale spatial distribution of zoonotic reservoirs using computer vision

Zoonotic diseases threaten human health worldwide and are often associated with anthropogenic disturbance. Predicting how disturbance influences spillover risk is critical for effective disease intervention but difficult to achieve at fine spatial scales. Here, we develop a method that learns the spatial distribution of a reservoir species from aerial imagery. Our approach uses neural networks to extract features of known or hypothesized importance from images. The spatial distribution of these features is then summarized and linked to spatially explicit reservoir presence/absence data using boosted regression trees. We demonstrate the utility of our method by applying it to the reservoir of Lassa virus, Mastomys natalensis, within the West African nations of Sierra Leone and Guinea. We show that, when trained using reservoir trapping data and publicly available aerial imagery, our framework learns relationships between environmental features and reservoir occurrence and accurately ranks areas according to the likelihood of reservoir presence.

Using drivers and transmission pathways to identify SARS-like coronavirus spillover risk hotspots

The emergence of SARS-like coronaviruses is a multi-stage process from wildlife reservoirs to people. Here we characterize multiple drivers-landscape change, host distribution, and human exposure-associated with the risk of spillover of zoonotic SARS-like coronaviruses to help inform surveillance and mitigation activities. We consider direct and indirect transmission pathways by modeling four scenarios with livestock and mammalian wildlife as potential and known reservoirs before examining how access to healthcare varies within clusters and scenarios. We found 19 clusters with differing risk factor contributions within a single country (N = 9) or transboundary (N = 10). High-risk areas were mainly closer (11-20%) rather than far ( < 1%) from healthcare. Areas far from healthcare reveal healthcare access inequalities, especially Scenario 3, which includes wild mammals and not livestock as secondary hosts. China (N = 2) and Indonesia (N = 1) had clusters with the highest risk. Our findings can help stakeholders in land use planning, integrating healthcare implementation and One Health actions.

Applications and advances in acoustic monitoring for infectious disease epidemiology

Emerging infectious diseases continue to pose a significant burden on global public health, and there is a critical need to better understand transmission dynamics arising at the interface of human activity and wildlife habitats. Passive acoustic monitoring (PAM), more typically applied to questions of biodiversity and conservation, provides an opportunity to collect and analyse audio data in relative real time and at low cost. Acoustic methods are increasingly accessible, with the expansion of cloud-based computing, low-cost hardware, and machine learning approaches. Paired with purposeful experimental design, acoustic data can complement existing surveillance methods and provide a novel toolkit to investigate the key biological parameters and ecological interactions that underpin infectious disease epidemiology.

Ecological conditions predict the intensity of Hendra virus excretion over space and time from bat reservoir hosts

The ecological conditions experienced by wildlife reservoirs affect infection dynamics and thus the distribution of pathogen excreted into the environment. This spatial and temporal distribution of shed pathogen has been hypothesised to shape risks of zoonotic spillover. However, few systems have data on both long-term ecological conditions and pathogen excretion to advance mechanistic understanding and test environmental drivers of spillover risk. We here analyse three years of Hendra virus data from nine Australian flying fox roosts with covariates derived from long-term studies of bat ecology. We show that the magnitude of winter pulses of viral excretion, previously considered idiosyncratic, are most pronounced after recent food shortages and in bat populations displaced to novel habitats. We further show that cumulative pathogen excretion over time is shaped by bat ecology and positively predicts spillover frequency. Our work emphasises the role of reservoir host ecology in shaping pathogen excretion and provides a new approach to estimate spillover risk.

Reservoir population ecology, viral evolution and the risk of emerging infectious disease

The ecology and life history of wild animals influences their potential to harbour infectious disease. This observation has motivated studies identifying empirical relationships between traits of wild animals and historical patterns of spillover and emergence into humans. Although these studies have identified compelling broad-scale patterns, they are generally agnostic with respect to underlying mechanisms. Here, we develop mathematical models that couple reservoir population ecology with viral epidemiology and evolution to clarify existing verbal arguments and pinpoint the conditions that favour spillover and emergence. Our results support the idea that average lifespan influences the likelihood of an animal serving as a reservoir for human infectious disease. At the same time, however, our results show that the magnitude of this effect is sensitive to the rate of viral mutation. Our results also demonstrate that viral pathogens causing persistent infections or a transient immune response within the reservoir are more likely to fuel emergence. Genetically explicit stochastic simulations enrich these mathematical results by identifying relationships between the genetic basis of transmission and the risk of spillover and emergence. Together, our results clarify the scope of applicability for existing hypotheses and refine our understanding of emergence risk.

Optimising predictive models to prioritise viral discovery in zoonotic reservoirs

Despite the global investment in One Health disease surveillance, it remains difficult and costly to identify and monitor the wildlife reservoirs of novel zoonotic viruses. Statistical models can guide sampling target prioritisation, but the predictions from any given model might be highly uncertain; moreover, systematic model validation is rare, and the drivers of model performance are consequently under-documented. Here, we use the bat hosts of betacoronaviruses as a case study for the data-driven process of comparing and validating predictive models of probable reservoir hosts. In early 2020, we generated an ensemble of eight statistical models that predicted host-virus associations and developed priority sampling recommendations for potential bat reservoirs of betacoronaviruses and bridge hosts for SARS-CoV-2. During a time frame of more than a year, we tracked the discovery of 47 new bat hosts of betacoronaviruses, validated the initial predictions, and dynamically updated our analytical pipeline. We found that ecological trait-based models performed well at predicting these novel hosts, whereas network methods consistently performed approximately as well or worse than expected at random. These findings illustrate the importance of ensemble modelling as a buffer against mixed-model quality and highlight the value of including host ecology in predictive models. Our revised models showed an improved performance compared with the initial ensemble, and predicted more than 400 bat species globally that could be undetected betacoronavirus hosts. We show, through systematic validation, that machine learning models can help to optimise wildlife sampling for undiscovered viruses and illustrates how such approaches are best implemented through a dynamic process of prediction, data collection, validation, and updating.

Virus isolation data improve host predictions for New World rodent orthohantaviruses

Identifying reservoir host species is crucial for understanding the ecology of multi-host pathogens and predicting risks of pathogen spillover from wildlife to people. Predictive models are increasingly used for identifying ecological traits and prioritizing surveillance of likely zoonotic reservoirs, but these often employ different types of evidence for establishing host associations. Comparisons between models with different infection evidence are necessary to guide inferences about the trait profiles of likely hosts and identify which hosts and geographical regions are likely sources of spillover. Here, we use New World rodent-orthohantavirus associations to explore differences in the performance and predictions of models trained on two types of evidence for infection and onward transmission: RT-PCR and live virus isolation data, representing active infections versus host competence, respectively. Orthohantaviruses are primarily carried by muroid rodents and cause the diseases haemorrhagic fever with renal syndrome (HFRS) and hantavirus cardiopulmonary syndrome (HCPS) in humans. We show that although boosted regression tree (BRT) models trained on RT-PCR and live virus isolation data both performed well and capture generally similar trait profiles, rodent phylogeny influenced previously collected RT-PCR data, and BRTs using virus isolation data displayed a narrower list of predicted reservoirs than those using RT-PCR data. BRT models trained on RT-PCR data identified 138 undiscovered hosts and virus isolation models identified 92 undiscovered hosts, with 27 undiscovered hosts identified by both models. Distributions of predicted hosts were concentrated in several different regions for each model, with large discrepancies between evidence types. As a form of validation, virus isolation models independently predicted several orthohantavirus-rodent host associations that had been previously identified through empirical research using RT-PCR. Our model predictions provide a priority list of species and locations for future orthohantavirus sampling. More broadly, these results demonstrate the value of multiple data types for predicting zoonotic pathogen hosts. These methods can be applied across a range of systems to improve our understanding of pathogen maintenance and increase efficiency of pathogen surveillance.

A Data-driven Horizon Scan of Bacterial Pathogens at the Wildlife-livestock Interface

Many livestock diseases rely on wildlife for the transmission or maintenance of the pathogen, and the wildlife-livestock interface represents a potential site of disease emergence for novel pathogens in livestock. Predicting which pathogen species are most likely to emerge in the future is an important challenge for infectious disease surveillance and intelligence. We used a machine learning approach to conduct a data-driven horizon scan of bacterial associations at the wildlife-livestock interface for cows, sheep, and pigs. Our model identified and ranked from 76 to 189 potential novel bacterial species that might associate with each livestock species. Wildlife reservoirs of known and novel bacteria were shared among all three species, suggesting that targeting surveillance and/or control efforts towards these reservoirs could contribute disproportionately to reducing spillover risk to livestock. By predicting pathogen-host associations at the wildlife-livestock interface, we demonstrate one way to plan for and prevent disease emergence in livestock.

Zoonotic malaria transmission and land use change in Southeast Asia: what is known about the vectors

Zoonotic Plasmodium infections in humans in many Southeast Asian countries have been increasing, including in countries approaching elimination of human-only malaria transmission. Most simian malarias in humans are caused by Plasmodium knowlesi, but recent research shows that humans are at risk of many different simian Plasmodium species. In Southeast Asia, simian Plasmodium species are mainly transmitted by mosquitoes in the Anopheles leucosphyrus and Anopheles dirus complexes. Although there is some evidence of species outside the Leucosphyrus Group transmitting simian Plasmodium species, these await confirmation of transmission to humans. The vectors of monkey malarias are mostly found in forests and forest fringes, where they readily bite long-tailed and pig-tailed macaques (the natural reservoir hosts) and humans. How changing land-uses influence zoonotic malaria vectors is still poorly understood. Fragmentation of forests from logging, agriculture and other human activities is associated with increased zoonotic Plasmodium vector exposure. This is thought to occur through altered macaque and mosquito distributions and behaviours, and importantly, increased proximity of humans, macaques, and mosquito vectors. Underlying the increase in vector densities is the issue that the land-use change and human activities create more oviposition sites and, in correlation, increases availably of human blood hosts. The current understanding of zoonotic malaria vector species is largely based on a small number of studies in geographically restricted areas. What is known about the vectors is limited: the data is strongest for distribution and density with only weak evidence for a limited number of species in the Leucosphyrus Group for resting habits, insecticide resistance, blood feeding habits and larval habitats. More data are needed on vector diversity and bionomics in additional geographic areas to understand both the impacts on transmission of anthropogenic land-use change and how this significant disease in humans might be controlled.

A protocol for a longitudinal, observational cohort study of infection and exposure to zoonotic and vector-borne diseases across a land-use gradient in Sabah, Malaysian Borneo: a socio-ecological systems approach

Introduction. Landscape changes disrupt environmental, social and biological systems, altering pathogen spillover and transmission risks. This study aims to quantify the impact of specific land management practices on spillover and transmission rates of zoonotic and vector-borne diseases within Malaysian Borneo. This protocol describes a cohort study with integrated ecological sampling to assess how deforestation and agricultural practices impact pathogen flow from wildlife and vector populations to human infection and detection by health facilities. This will focus on malaria, dengue and emerging arboviruses (Chikungunya and Zika), vector-borne diseases with varying contributions of simian reservoirs within this setting. Methods. A prospective longitudinal observational cohort study will be established in communities residing or working within the vicinity of the Stability of Altered Forest Ecosystems (SAFE) Project, a landscape gradient within Malaysian Borneo encompassing different plantation and forest types. The primary outcome of this study will be transmission intensity of selected zoonotic and vector-borne diseases, as quantified by changes in pathogen-specific antibody levels. Exposure will be measured using paired population-based serological surveys conducted at the beginning and end of the two-year cohort study. Secondary outcomes will include the distribution and infection rates of Aedes and Anopheles mosquito vectors, human risk behaviours and clinical cases reported to health facilities. Longitudinal data on human behaviour, contact with wildlife and GPS tracking of mobility patterns will be collected throughout the study period. This will be integrated with entomological surveillance to monitor densities and pathogen infection rates of Aedes and Anopheles mosquitoes relative to land cover. Within surrounding health clinics, continuous health facility surveillance will be used to monitor reported infections and febrile illnesses. Models will be developed to assess spillover and transmission rates relative to specific land management practices and evaluate abilities of surveillance systems to capture these risks.

Seasonality of coronavirus shedding in tropical bats

Anticipating cross-species transmission of zoonotic diseases requires an understanding of pathogen infection dynamics within natural reservoir hosts. Although bats might be a source of coronaviruses (CoVs) for humans, the drivers of infection dynamics in bat populations have received limited attention. We conducted a fine-scale 2-year longitudinal study of CoV infection dynamics in the largest colony of Reunion free-tailed bats (Mormopterus francoismoutoui), a tropical insectivorous species. Real-time PCR screening of 1080 fresh individual faeces samples collected during the two consecutive years revealed an extreme variation of the detection rate of bats shedding viruses over the birthing season (from 0% to 80%). Shedding pulses were repeatedly observed and occurred both during late pregnancy and within two months after parturition. An additional shedding pulse at the end of the second year suggests some inter-annual variations. We also detected viral RNA in bat guano up to three months after bats had left the cave. Our results highlight the importance of fine-scale longitudinal studies to capture the rapid change of bat CoV infection over months, and that CoV shedding pulses in bats may increase spillover risk.

Challenges in modelling the dynamics of infectious diseases at the wildlife-human interface

The Covid-19 pandemic is of zoonotic origin, and many other emerging infections of humans have their origin in an animal host population. We review the challenges involved in modelling the dynamics of wildlife–human interfaces governing infectious disease emergence and spread. We argue that we need a better understanding of the dynamic nature of such interfaces, the underpinning diversity of pathogens and host–pathogen association networks, and the scales and frequencies at which environmental conditions enable spillover and host shifting from animals to humans to occur. The major drivers of the emergence of zoonoses are anthropogenic, including the global change in climate and land use. These, and other ecological processes pose challenges that must be overcome to counterbalance pandemic risk. The development of more detailed and nuanced models will provide better tools for analysing and understanding infectious disease emergence and spread.

Progress With Livestock Welfare in Extensive Production Systems: Lessons From Australia

The extensive livestock production industries are vital to the national economy of Australia. Continuing improvements to extensively-raised livestock welfare is desirable, necessary and in some situations mandatory, if the social license for animal sourced food and fiber production is to continue sustainably. However, meeting increasingly high welfare standards is challenging. The changing climate in this millennium, has seen the occurrence of two of the most severe drought periods on record in Australia, resulting in complex welfare issues arising from unforeseen disease, trade and environmental catastrophes. The onset of the first drought coincided with an uncontrolled epidemic of ovine paratuberculosis. It ended just prior to a temporary ban on live export of tropical cattle to Indonesia that induced a major market failure and led to severe morbidity and mortality on some beef properties. The second drought period progressed in severity and culminated in the most extreme bushfires recorded, causing unprecedented levels of mortality, morbidity and suffering in farmed animals and wildlife. Temperature extremes have also caused periodic heat-associated or cold-induced hyopthermia losses, requiring increased vigilance and careful management to reduce both temperature-induced stress during transport and the high ovine peri-parturient losses traditionally observed in extensive sheep farming. Several issues remain controversial, including surgical mulesing of wool sheep to manage flystrike, and the continuing live export trade of sheep and cattle. However, in reviewing the increasingly complex welfare challenges for the extensive livestock population industries that are export trade dependent and remain vulnerable to welfare activism, it appears progress has been made. These include development of prescribed livestock welfare Standards and Guidelines and the introduction of the Exporter Supply Chain Assurance System (ESCAS) to address export concerns. Further, the sheep mulesing crisis led to improved producer welfare attitudes and practices, including pain management during aversive husbandry procedures that is now occurring globally. Finally, innovations in animal welfare surveillance and assessment, are additional encouraging signs that suggest improving change management of extensive farm animal welfare is occurring that provides lessons well-beyond Australian shores.

From Protein to Pandemic: The Transdisciplinary Approach Needed to Prevent Spillover and the Next Pandemic

Pandemics are a consequence of a series of processes that span scales from viral biology at 10-9 m to global transmission at 106 m. The pathogen passes from one host species to another through a sequence of events that starts with an infected reservoir host and entails interspecific contact, innate immune responses, receptor protein structure within the potential host, and the global spread of the novel pathogen through the naive host population. Each event presents a potential barrier to the onward passage of the virus and should be characterized with an integrated transdisciplinary approach. Epidemic control is based on the prevention of exposure, infection, and disease. However, the ultimate pandemic prevention is prevention of the spillover event itself. Here, we focus on the potential for preventing the spillover of henipaviruses, a group of viruses derived from bats that frequently cross species barriers, incur high human mortality, and are transmitted among humans via stuttering chains. We outline the transdisciplinary approach needed to prevent the spillover process and, therefore, future pandemics.

Population structure and diet generalism define a preliminary ecological profile of zoonotic virus hosts in the Western Ghats, India

The rainforests of the Western Ghats exhibit some of the highest biodiversity on the planet, and yet are undergoing rapid land use change due to the expansion of agriculture and other industries. As the landscape of the region is transformed, more people are coming into conflict with wildlife and becoming exposed to pathogens that previously circulated beyond the boundaries of human incursion. Despite an ecological knowledge imperative, this emerging landscape is ill-defined with respect to the ecology of zoonotic viruses and their mammalian wildlife hosts. Without a better understanding of the underlying infection ecology, the epidemiology of viral spillover will remain elusive and unsuited to the task of predicting and preventing outbreaks. The current investigation explored the association between mammalian zoonotic virus richness and species-level landscape, life-history, and dietary traits to describe an initial ecological profile of zoonotic virus hosts in the Western Ghats. Social group composition and dietary forage were both non-linearly associated with greater zoonotic viral richness among these species, whereby species active in smaller social groups, albeit in higher population densities, and exhibiting a tendency toward a generalist diet hosted more zoonotic viruses. While these findings provide no definitive ecological demarcation of zoonotic virus hosts or their contribution to viral maintenance or amplification, it is expected that this preliminary profile can help to develop targeted wildlife pathogen surveillance programs and to expand the current approach to epidemiological modelling of emerging zoonoses in the region, which typically do not account for the macroecological parameters of infection transmission.

COVID-19 Effects on Livestock Production: A One Welfare Issue

The COVID-19 pandemic highlights that we exist in a global community. From a single city, it spread to 188 countries across the world and infected 30 million people by September 18, 2020. Decades of modeling pandemics predicted potential consequences, but COVID-19's impact on the food supply chain, and specifically livestock production was unexpected. Clusters of cases among workers in meat processing plants evolved quickly to affect human, animal, and environmental welfare in several countries. In processing plants, the hygiene focus is on product quality and food safety. Because of their close proximity to one another, COVID-19 spread rapidly between workers and the lack of sick leave and health insurance likely resulted in workers continuing to work when infectious. In the United States (U.S.) many processing plants shut down when they identified major outbreaks, putting pressure especially on pig and poultry industries. At one point, there was a 45% reduction in pig processing capacity meaning about 250,000 pigs per day were not slaughtered. This resulted in longer transport distances to plants in operation with extra capacity, but also to crowding of animals on farm. Producers were encouraged to slow growth rates, but some had to cull animals on farm in ways that likely included suffering and caused considerable upset to owners and workers. Carcass disposal was also associated with potential biosecurity risks and detrimental effects on the environment. Hence, this is a One Welfare issue, affecting human, animal, and environmental welfare and highlighting the fragility of intensive, high-throughput livestock production systems. This model needs to be re-shaped to include the animal, human, and environmental elements across the farm to fork chain. Such a One Welfare approach will ensure that food production systems are resilient, flexible, and fair in the face of future challenges.

Beyond Infection: Integrating Competence into Reservoir Host Prediction

Most efforts to predict novel reservoirs of zoonotic pathogens use information about host exposure and infection rather than competence, defined as the ability to transmit pathogens. Better obtaining and integrating competence data into statistical models as covariates, as the response variable, and through postmodel validation should improve predictive research.

Extreme mobility of the world's largest flying mammals creates key challenges for management and conservation

BACKGROUND

Effective conservation management of highly mobile species depends upon detailed knowledge of movements of individuals across their range; yet, data are rarely available at appropriate spatiotemporal scales. Flying-foxes (Pteropus spp.) are large bats that forage by night on floral resources and rest by day in arboreal roosts that may contain colonies of many thousands of individuals. They are the largest mammals capable of powered flight, and are highly mobile, which makes them key seed and pollen dispersers in forest ecosystems. However, their mobility also facilitates transmission of zoonotic diseases and brings them in conflict with humans, and so they require a precarious balancing of conservation and management concerns throughout their Old World range. Here, we analyze the Australia-wide movements of 201 satellite-tracked individuals, providing unprecedented detail on the inter-roost movements of three flying-fox species: Pteropus alecto, P. poliocephalus, and P. scapulatus across jurisdictions over up to 5 years.

RESULTS

Individuals were estimated to travel long distances annually among a network of 755 roosts (P. alecto, 1427-1887 km; P. poliocephalus, 2268-2564 km; and P. scapulatus, 3782-6073 km), but with little uniformity among their directions of travel. This indicates that flying-fox populations are composed of extremely mobile individuals that move nomadically and at species-specific rates. Individuals of all three species exhibited very low fidelity to roosts locally, resulting in very high estimated daily colony turnover rates (P. alecto, 11.9 ± 1.3%; P. poliocephalus, 17.5 ± 1.3%; and P. scapulatus, 36.4 ± 6.5%). This indicates that flying-fox roosts form nodes in a vast continental network of highly dynamic "staging posts" through which extremely mobile individuals travel far and wide across their species ranges.

CONCLUSIONS

The extreme inter-roost mobility reported here demonstrates the extent of the ecological linkages that nomadic flying-foxes provide across Australia's contemporary fragmented landscape, with profound implications for the ecosystem services and zoonotic dynamics of flying-fox populations. In addition, the extreme mobility means that impacts from local management actions can readily reverberate across jurisdictions throughout the species ranges; therefore, local management actions need to be assessed with reference to actions elsewhere and hence require national coordination. These findings underscore the need for sound understanding of animal movement dynamics to support evidence-based, transboundary conservation and management policy, tailored to the unique movement ecologies of species.

Reframing conservation physiology to be more inclusive, integrative, relevant and forward-looking: reflections and a horizon scan

Applying physiological tools, knowledge and concepts to understand conservation problems (i.e. conservation physiology) has become commonplace and confers an ability to understand mechanistic processes, develop predictive models and identify cause-and-effect relationships. Conservation physiology is making contributions to conservation solutions; the number of 'success stories' is growing, but there remain unexplored opportunities for which conservation physiology shows immense promise and has the potential to contribute to major advances in protecting and restoring biodiversity. Here, we consider how conservation physiology has evolved with a focus on reframing the discipline to be more inclusive and integrative. Using a 'horizon scan', we further explore ways in which conservation physiology can be more relevant to pressing conservation issues of today (e.g. addressing the Sustainable Development Goals; delivering science to support the UN Decade on Ecosystem Restoration), as well as more forward-looking to inform emerging issues and policies for tomorrow. Our horizon scan provides evidence that, as the discipline of conservation physiology continues to mature, it provides a wealth of opportunities to promote integration, inclusivity and forward-thinking goals that contribute to achieving conservation gains. To advance environmental management and ecosystem restoration, we need to ensure that the underlying science (such as that generated by conservation physiology) is relevant with accompanying messaging that is straightforward and accessible to end users.

Percolation models of pathogen spillover

Predicting pathogen spillover requires counting spillover events and aligning such counts with process-related covariates for each spillover event. How can we connect our analysis of spillover counts to simple, mechanistic models of pathogens jumping from reservoir hosts to recipient hosts? We illustrate how the pathways to pathogen spillover can be represented as a directed graph connecting reservoir hosts and recipient hosts and the number of spillover events modelled as a percolation of infectious units along that graph. Percolation models of pathogen spillover formalize popular intuition and management concepts for pathogen spillover, such as the inextricably multilevel nature of cross-species transmission, the impact of covariance between processes such as pathogen shedding and human susceptibility on spillover risk, and the assumptions under which the effect of a management intervention targeting one process, such as persistence of vectors, will translate to an equal effect on the overall spillover risk. Percolation models also link statistical analysis of spillover event datasets with a mechanistic model of spillover. Linear models, one might construct for process-specific parameters, such as the log-rate of shedding from one of several alternative reservoirs, yield a nonlinear model of the log-rate of spillover. The resulting nonlinearity is approximately piecewise linear with major impacts on statistical inferences of the importance of process-specific covariates such as vector density. We recommend that statistical analysis of spillover datasets use piecewise linear models, such as generalized additive models, regression clustering or ensembles of linear models, to capture the piecewise linearity expected from percolation models. We discuss the implications of our findings for predictions of spillover risk beyond the range of observed covariates, a major challenge of forecasting spillover risk in the Anthropocene. This article is part of the theme issue 'Dynamic and integrative approaches to understanding pathogen spillover'.