Stepping up from wildlife disease surveillance to integrated wildlife monitoring in Europe

The influence of social and spatial processes on the epidemiology of environmentally transmitted pathogens in wildlife: implications for management

Social and spatial structures of host populations play important roles in pathogen transmission. For environmentally transmitted pathogens, the host space use interacts with both the host social structure and the pathogen's environmental persistence (which determines the time-lag across which two hosts can transmit). Together, these factors shape the epidemiological dynamics of environmentally transmitted pathogens. While the importance of both social and spatial structures and environmental pathogen persistence has long been recognized in epidemiology, they are often considered separately. A better understanding of how these factors interact to determine disease dynamics is required for developing robust surveillance and management strategies. Here, we use a simple agent-based model where we vary host mobility (spatial), host gregariousness (social) and pathogen decay (environmental persistence), each from low to high levels to uncover how they affect epidemiological dynamics. By comparing epidemic peak, time to epidemic peak and final epidemic size, we show that longer infectious periods, higher group mobility, larger group size and longer pathogen persistence lead to larger, faster growing outbreaks, and explore how these processes interact to determine epidemiological outcomes such as the epidemic peak and the final epidemic size. We identify general principles that can be used for planning surveillance and control for wildlife host-pathogen systems with environmental transmission across a range of spatial behaviour, social structure and pathogen decay rates. This article is part of the theme issue 'The spatial-social interface: a theoretical and empirical integration'.

An old unknown: 40 years of crayfish plague monitoring in Switzerland, the water tower of Europe

The oomycete Aphanomyces astaci is the causative agent of crayfish plague, a disease threatening susceptible freshwater crayfish species in Europe. To detect its spatiotemporal occurrence in Switzerland, we reviewed (1) the literature regarding occurrence of crayfish plague and North American crayfish carrier species and (2) the necropsy report archive of the Institute for Fish and Wildlife Health (FIWI) from 1968 to 2020. In the past, crayfish plague was diagnosed through several methods: conventional PCR, culture, and histology. When available, we re-evaluated archived Bouin's or formalin-fixed, paraffin-embedded samples collected during necropsies (1991-2020) with a recently published quantitative PCR. Literature research revealed putative reports of crayfish plague in Switzerland between the 1870s and 1910s and the first occurrence of three North American crayfish species between the late 1970s and 1990s. Finally, 54 (28.1%) cases were classified as positive and 9 (4.7%) cases as suspicious. The total number of positive cases increased by 14 (14.7%) after re-evaluation of samples. The earliest diagnosis of crayfish plague was performed in 1980 and the earliest biomolecular confirmation of A. astaci DNA dated 1991. Between 1980-1990, 1991-2000 and 2001-2010 crayfish plague spread from one to two and finally three catchment basins, respectively. Similar to other European countries, crayfish plague has occurred in Switzerland in two waves: the first at the end of the 19th and the second at the end of the 20th century in association with the first occurrence of North American crayfish species. The spread from one catchment basin to another suggests a human-mediated pathogen dispersal.

COVID-19 could accelerate the decline in recreational hunting: A natural experiment from Northern Italy

Although many studies highlighted the potential of COVID-19 to reshape existing models of wildlife management, empirical research on this topic has been scarce, particularly in Europe. We investigated the potential of COVID-19 pandemic to accelerate the ongoing decline in an aging population of recreational hunters in Italy. Namely, we modelled spatiotemporal trends between 2011 and 2021 in the number of recreational hunters in 50 Italian provinces with a varying incidence of COVID-19, and temporally delayed waves of infection. Compared to projections from 2011-2019 data, we detected a lower number of hunters who enrolled for the hunting season, both in 2020 (14 provinces) and in 2021 (15 provinces). The provinces with the highest incidence of COVID-19 in the Lombardy and Emilia-Romagna regions were also those experiencing the most marked decrease in hunting participation. Our findings revealed that a wildlife management system based on recreational hunting can be rapidly destabilized by epidemics and their associated public health measures, particularly when the average age of hunters is high, like in Italy. Considered the high incidence attained by COVID-19 in many European countries, where hunters are pivotal for the management of large ungulates and where they were already declining before the pandemic, our findings call for further large-scale research about the impact of COVID-19 on hunting participation.

Analysis of a European general wildlife health surveillance program: Chances, challenges and recommendations

In a One Health perspective general wildlife health surveillance (GWHS) gains importance worldwide, as pathogen transmission among wildlife, domestic animals and humans raises health, conservation and economic concerns. However, GWHS programs operate in the face of legal, geographical, financial, or administrative challenges. The present study uses a multi-tiered approach to understand the current characteristics, strengths and gaps of a European GWHS that operates in a fragmented legislative and multi-stakeholder environment. The aim is to support the implementation or improvement of other GWHS systems by managers, surveillance experts, and administrations. To assess the current state of wildlife health investigations and trends within the GWHS, we retrospectively analyzed 20 years of wildlife diagnostic data to explore alterations in annual case numbers, diagnosed diseases, and submitter types, conducted an online survey and phone interviews with official field partners (hunting administrators, game wardens and hunters) to assess their case submission criteria as well as their needs for post-mortem investigations, and performed in-house time estimations of post-mortem investigations to conduct a time-per-task analysis. Firstly, we found that infectious disease dynamics, the level of public awareness for specific diseases, research activities and increasing population sizes of in depth-monitored protected species, together with biogeographical and political boundaries all impacted case numbers and can present unexpected challenges to a GWHS. Secondly, we found that even a seemingly comprehensive GWHS can feature pronounced information gaps, with underrepresentation of common or easily recognizable diseases, blind spots in non-hunted species and only a fraction of discovered carcasses being submitted. Thirdly, we found that substantial amounts of wildlife health data may be available at local hunting administrations or disease specialist centers, but outside the reach of the GWHS and its processes. In conclusion, we recommend that fragmented and federalist GWHS programs like the one addressed require a central, consistent and accessible collection of wildlife health data. Also, considering the growing role of citizen observers in environmental research, we recommend using online reporting systems to harness decentrally available information and fill wildlife health information gaps.

AIMS for wildlife: Developing an automated interactive monitoring system to integrate real-time movement and environmental data for true adaptive management

To effectively manage species and habitats at multiple scales, population and land managers require rapid information on wildlife use of managed areas and responses to landscape conditions and management actions. GPS tracking studies of wildlife are particularly informative to species ecology, habitat use, and conservation. Combining GPS data with administrative data and a diverse suite of remotely sensed, geo-referenced environmental (e.g., climatic) data, would more comprehensively inform how animals interact with and utilize habitats and ecosystems and our goal was to create a conceptual model for a system that would accomplish this - the 'Automated Interactive Monitoring System (AIMS) for Wildlife'. Our objective for this study was to develop a Customized Wildlife Report (CWR) - the first AIMS for Wildlife deliverable product. CWRs collate and summarize our 8-year GPS tracking dataset of ∼11 million locations from 1338 individual (16 species) avifauna and make actionable, real-time data on animal movements and trends in a specific area of interest available to managers and stakeholders for rapid application in day-to-day management. The CWR exemplar presented in this paper was developed to address needs identified by habitat managers of Sacramento National Wildlife Refuge and illustrates the highly specific, information offered and how it contributes to assessing the efficacy of conservation actions while allowing for near real-time adaptive management. The report can be easily customized for any of the thousands of wildlife refuges or regional areas of interest in the United States, emphasizing the broad application of an animal movement data stream. Utilizing diverse, extensive telemetry data streams through scientific collaboration can aid managers and conservation stakeholders with short and long-term research and conservation planning and help address a cadre of issues from local-scale habitat management to improving the understanding of landscape level impacts like drought, wildfire, and climate change on wildlife populations.

A web-based geographic information system monitoring wildlife diseases in Abruzzo and Molise regions, Southern Italy

BACKGROUND

Nowadays there is a worldwide consensus on the importance of conducting wildlife disease surveillance. Indeed, 60% of emerging infectious diseases are zoonotic in nature, and the majority of these (71.8%) originate in wildlife. Surveillance of wildlife diseases is crucial to prevent negative effects on human and animal health. Data digitization and sharing are among the main goals for the present and coming years. Geographic Information Systems (GIS) are increasingly used to analyze the geographical distribution of diseases and the relationships between pathogenic factors and their geographic environments.

METHODS

Wild animal's samples collected in the Abruzzo and Molise regions and delivered to our laboratory are entered in our Laboratory Information System and processed to be displayed through the Web-GIS mash-up presented in this paper. We built it using both open source and proprietary solutions, to produce data driven interactive maps, charts and tables to help to understand the epidemiology of wild animal diseases, their spread and trend.

RESULTS

Since 2013, 9.606 samples collected from wild animals have been analyzed in the laboratories of the IZS-Teramo and have been recorded in the system, facilitating the reporting to the judicial authorities and the identification of highly risky areas to set up control and repression measures. Moreover, thanks to the monitoring health protocol, a canine distemper epidemic in wolves has been detected and monitored in its temporal and spatial evolution, as well as cases of bovine tuberculosis in wild boars.

CONCLUSIONS

While it is more evident that the starting point is to choose the right sampling method, it is for sure less obvious that the information system in which data is stored is equally important. In fact, it should give the possibility to consult it in an easy and instructive way. GIS allows immediately grasping the spatial relationships between the data itself and those between the data and the territory; it is an important tool to support veterinary services in managing epidemic and non-epidemic emergencies and performing epidemiological investigations, but also to examine control plans and identify new gaps and challenges.

Further investigation of lead exposure as a potential threatening process for a scavenging marsupial species

There is a growing recognition of the harmful effects of lead exposure on avian and mammalian scavengers. This can lead to both lethal and non-lethal effects which may negatively impact wildlife populations. Our objective was to assess medium-term lead exposure in wild Tasmanian devils (Sarcophilus harrisii). Frozen liver samples (n = 41), opportunistically collected in 2017-2022, were analysed using inductively coupled plasma mass spectrometry (ICP-MS) to determine liver lead concentrations. These results were then used to calculate the proportion of animals with elevated lead levels (>5 mg/kg dry weight) and examine the role of explanatory variables that may have influenced the results. The majority of samples analysed were from the south-east corner of Tasmania, within 50 km of Hobart. No Tasmanian devil samples were found to have elevated lead levels. The median liver lead concentration was 0.17 mg/kg (range 0.05-1.32 mg/kg). Female devils were found to have significantly higher liver lead concentrations than males (P = 0.013), which was likely related to lactation, but other variables (age, location, body mass) were not significant. These results suggest that wild Tasmanian devil populations currently show minimal medium-term evidence of exposure to lead pollution, although samples were concentrated in peri-urban areas. The results provide a baseline level which can be used to assess the impact of any future changes in lead use in Tasmania. Furthermore, these data can be used as a comparison for lead exposure studies in other mammalian scavengers, including other carnivorous marsupial species.

Diseases of Iberian ibex (Capra pyrenaica)

Iberian ibex (Capra pyrenaica) is an ecologically and economically relevant medium-sized emblematic mountain ungulate. Diseases participate in the population dynamics of the species as a regulating agent, but can also threaten the conservation and viability of vulnerable population units. Moreover, Iberian ibex can also be a carrier or even a reservoir of pathogens shared with domestic animals and/or humans, being therefore a concern for livestock and public health. The objective of this review is to compile the currently available knowledge on (1) diseases of Iberian ibex, presented according to their relevance on the health and demography of free-ranging populations; (2) diseases subjected to heath surveillance plans; (3) other diseases reported in the species; and (4) diseases with particular relevance in captive Iberian ibex populations. The systematic review of all the information on diseases affecting the species unveils unpublished reports, scientific communications in meetings, and scientific articles, allowing the first comprehensive compilation of Iberian ibex diseases. This review identifies the gaps in knowledge regarding pathogenesis, immune response, diagnostic methods, treatment, and management of diseases in Iberian ibex, providing a base for future research. Moreover, this challenges wildlife and livestock disease and wildlife population managers to assess the priorities and policies currently implemented in Iberian ibex health surveillance and monitoring and disease management.

DNA virome composition of two sympatric wild felids, bobcat (Lynx rufus) and puma (Puma concolor) in Sonora, Mexico

With viruses often having devastating effects on wildlife population fitness and wild mammals serving as pathogen reservoirs for potentially zoonotic diseases, determining the viral diversity present in wild mammals is both a conservation and One Health priority. Additionally, transmission from more abundant hosts could increase the extinction risk of threatened sympatric species. We leveraged an existing circular DNA enriched metagenomic dataset generated from bobcat (Lynx rufus, n = 9) and puma (Puma concolor, n = 13) scat samples non-invasively collected from Sonora, Mexico, to characterize fecal DNA viromes of each species and determine the extent that viruses are shared between them. Using the metaWRAP pipeline to co-assemble viral genomes for comparative metagenomic analysis, we observed diverse circular DNA viruses in both species, including circoviruses, genomoviruses, and anelloviruses. We found that differences in DNA virome composition were partly attributed to host species, although there was overlap between viruses in bobcats and pumas. Pumas exhibited greater levels of alpha diversity, possibly due to bioaccumulation of pathogens in apex predators. Shared viral taxa may reflect dietary overlap, shared environmental resources, or transmission through host interactions, although we cannot rule out species-specific host-virus coevolution for the taxa detected through co-assembly. However, our detection of integrated feline foamy virus (FFV) suggests Sonoran pumas may interact with domestic cats. Our results contribute to the growing baseline knowledge of wild felid viral diversity. Future research including samples from additional sources (e.g., prey items, tissues) may help to clarify host associations and determine the pathogenicity of detected viruses.

Tuberculosis Epidemiology and Spatial Ecology at the Cattle-Wild Boar Interface in Northern Spain

Tuberculosis (TB) is a contagious chronic disease due to infection with Mycobacterium tuberculosis complex (MTC) bacteria. Monitoring of wildlife, especially potential reservoirs, is important for detecting changes in disease occurrence and assessing the impact of interventions. Here, we examined whether wild boar (Sus scrofa) may contribute to the re-emergence of TB in Asturias (10,604 km2), northern Spain. Although this province was declared free of TB in cattle in November 2021, MTC bacteria remain prevalent in several “hotspots,” with the European badger (Meles meles) suggested as a TB potential wild reservoir. Drawing on data from the Spanish National Bovine Tuberculosis Eradication Program and the Government of the Principality of Asturias covering the period 2014–2020, we analyzed the prevalence of TB in cattle and wild boar in this region. In hotspots (592 km2), we also investigated the ranging behavior and habitat use of five cows that belonged to farms with a history of TB and six trapped sympatric wild boar. During the observation period, TB prevalence was 0.14% among cattle overall and 0.13–0.41% in hotspots, which was much lower than the prevalence in wild boar, which was 3.15% overall and 5.23–5.96% in hotspots. Infected cattle and infected wild boar in hotspots shared the same strains of M. bovis, and GPS tracking showed spatiotemporal overlap between the species, mainly around pastures during sunrise (06:00–07:00 h) and sunset (19:00–20:00 h). Our results suggest that in addition to cattle and badgers, wild boar possibly help maintain TB in northern Spain, increasing the host richness that influences TB transmission risk in the area, which should be taken into account in monitoring and eradication efforts.

When Vibrios Take Flight: A Meta-Analysis of Pathogenic Vibrio Species in Wild and Domestic Birds

Of the over 100 species in the genus Vibrio, approximately twelve are associated with clinical disease, such as cholera and vibriosis. Crucially, eleven of those twelve, including Vibrio cholerae and Vibrio vulnificus, have been isolated from birds. Since 1965, pathogenic Vibrio species have been consistently isolated from aquatic and ground-foraging bird species, which has implications for public health, as well as the One Health paradigm defined as an ecology-inspired, integrative framework for the study of health and disease, inclusive of environmental, human, and animal health. In this meta-analysis, we identified 76 studies from the primary literature which report on or examine birds as hosts for pathogenic Vibrio species. We found that the burden of disease in birds was most commonly associated with V. cholerae, followed by V. metschnikovii and V. parahaemolyticus. Meta-analysis wide prevalence of our Vibrio pathogens varied from 19% for V. parahaemolyticus to 1% for V. mimicus. Wild and domestic birds were both affected, which may have implications for conservation, as well as agriculturally associated avian species. As pathogenic Vibrios become more abundant throughout the world as a result of warming estuaries and oceans, susceptible avian species should be continually monitored as potential reservoirs for these pathogens.

SARS-CoV-2 in animals: susceptibility of animal species, risk for animal and public health, monitoring, prevention and control

The epidemiological situation of SARS-CoV-2 in humans and animals is continually evolving. To date, animal species known to transmit SARS-CoV-2 are American mink, raccoon dog, cat, ferret, hamster, house mouse, Egyptian fruit bat, deer mouse and white-tailed deer. Among farmed animals, American mink have the highest likelihood to become infected from humans or animals and further transmit SARS-CoV-2. In the EU, 44 outbreaks were reported in 2021 in mink farms in seven MSs, while only six in 2022 in two MSs, thus representing a decreasing trend. The introduction of SARS-CoV-2 into mink farms is usually via infected humans; this can be controlled by systematically testing people entering farms and adequate biosecurity. The current most appropriate monitoring approach for mink is the outbreak confirmation based on suspicion, testing dead or clinically sick animals in case of increased mortality or positive farm personnel and the genomic surveillance of virus variants. The genomic analysis of SARS-CoV-2 showed mink-specific clusters with a potential to spill back into the human population. Among companion animals, cats, ferrets and hamsters are those at highest risk of SARS-CoV-2 infection, which most likely originates from an infected human, and which has no or very low impact on virus circulation in the human population. Among wild animals (including zoo animals), mostly carnivores, great apes and white-tailed deer have been reported to be naturally infected by SARS-CoV-2. In the EU, no cases of infected wildlife have been reported so far. Proper disposal of human waste is advised to reduce the risks of spill-over of SARS-CoV-2 to wildlife. Furthermore, contact with wildlife, especially if sick or dead, should be minimised. No specific monitoring for wildlife is recommended apart from testing hunter-harvested animals with clinical signs or found-dead. Bats should be monitored as a natural host of many coronaviruses.

Genome Characterization and Spaciotemporal Dispersal Analysis of Bagaza Virus Detected in Portugal, 2021

In September 2021, Bagaza virus (BAGV), a member of the Ntaya group from the Flavivirus genus, was detected for the first time in Portugal, in the heart and the brain of a red-legged partridge found dead in a hunting ground in Serpa (Alentejo region; southern Portugal). Here we report the genomic characterization of the full-length sequence of the BAGV detected (BAGV/PT/2021), including phylogenetic reconstructions and spaciotemporal analyses. Phylogenies inferred from nucleotide sequence alignments, complemented with the analysis of amino acid alignments, indicated that the BAGV strain from Portugal is closely related to BAGV strains previously detected in Spain, suggesting a common ancestor that seems to have arrived in the Iberia Peninsula in the late 1990s to early 2000s. In addition, our findings support previous observations that BAGV and Israel turkey meningoencephalitis virus (ITV) belong to the same viral species.

Using integrated wildlife monitoring to prevent future pandemics through one health approach

In the One Health context, Integrated Wildlife Monitoring (IWM) merges wildlife health monitoring (WHM) and host community monitoring to early detect emerging infections, record changes in disease dynamics, and assess the impact of interventions in complex multi-host and multi-pathogen networks. This study reports the deployment and results obtained from a nationwide IWM pilot test in eleven sites representing the habitat diversity of mainland Spain. In each study site, camera-trap networks and sampling of indicator species for antibody and biomarker analysis were used to generate information. The results allowed identifying differences in biodiversity and host community characteristics among the study sites, with a range of 8 to 19 relevant host species per point. The Eurasian wild boar (Sus scrofa) was the most connected and central species of the host communities, becoming a key target indicator species for IWM. A negative relationship between biodiversity and disease risk was detected, with a lower number and prevalence of circulating pathogens in the sites with more species in the community and larger network size. However, this overall trend was modified by specific host-community and environmental factors, such as the relative index of wild boar - red deer interactions or the proximity to urban habitats, suggesting that human-driven imbalances may favour pathogen circulation. The effort of incorporating wildlife population monitoring into the currently applied WHM programs to achieve effective IWM was also evaluated, allowing to identify population monitoring as the most time-consuming component, which should be improved in the future. This first nationwide application of IWM allowed to detect drivers and hotspots for disease transmission risk among wildlife, domestic animals, and humans, as well as identifying key target indicator species for monitoring. Moreover, anthropogenic effects such as artificially high wildlife densities and urbanisation were identified as risk factors for disease prevalence and interspecific transmission.

SARS-CoV-2 and West Nile Virus Prevalence Studies in Raccoons and Raccoon Dogs from Germany

Unlike farm animals, wild animals are not subject to continuous health surveillance. Individual projects designed to screen wildlife populations for specific pathogens are, therefore, also of great importance for human health. In this context, the possible formation of a reservoir for highly pathogenic zoonotic pathogens is a focus of research. Two of these pathogens that have received particular attention during the last years are the novel severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2), due to its fast global spread and high impact to the human health, and, since its introduction into Germany, the flavivirus West Nile virus (WNV). Especially in combination with invasive vertebrate species (e.g., raccoons (Procyon lotor) and raccoon dogs (Nyctereutes procyonoides) in Germany), risk analysis must be done to enable health authorities to assess the potential for the establishment of new wild life reservoirs for pathogens. Therefore, samples were collected from raccoons and raccoon dogs and analyzed for the presence of SARS-CoV-2 and WNV infections in these populations. Molecular biological and serological data obtained imply that no SARS-CoV-2 nor WNV reservoir has been established in these two wild life species yet. Future investigations need to keep an eye on these invasive carnivore populations, especially since the close contact of these animals to humans, mainly in urban areas, would make animal-human transmission a challenge for human health.

Retrospective serological and molecular survey of myxoma or antigenically related virus in the Iberian hare, Lepus granatensis

The 2018 outbreak of myxomatosis in the Iberian hare (Lepus granatensis) has been hypothesized to originate from a species jump of the rabbit-associated myxoma virus (MYXV), after natural recombination with an unknown poxvirus. Iberian hares were long considered resistant to myxomatosis as no prior outbreaks were reported. To provide insights into the emergence of this recombinant virus (ha-MYXV), we investigated serum samples from 451 Iberian hares collected over two time periods almost two decades apart, 1994-1999 and 2017-2019 for the presence of antibodies and MYXV-DNA. First, we screened all serum samples using a rabbit commercial indirect ELISA (iELISA) and then tested a subset of these samples in parallel using indirect immunofluorescence test (IFT), competitive ELISA (cELISA) and qPCR targeting M000.5L/R gene conserved in MYXV and ha-MYXV. The cut-off of iELISA relative index 10 = 6.1 was selected from a semiparametric finite mixture analysis aiming to minimize the probability of false positive results. Overall, MYXV related-antibodies were detected in 57 hares (12.6%) including 38 apparently healthy hares (n = 10, sampled in 1994-1999, none MYXV-DNA positive, and n = 28 sampled in 2017-2019 of which four were also ha-MYXV-DNA positive) and 19 found-dead and ha-MYXV-DNA-positive sampled in 2018-2019. Interestingly, four seronegative hares sampled in 1997 were MYXV-DNA positive by qPCR, the result being confirmed by sequencing of three of them. For the Iberian hares hunted or live trapped (both apparently health), seroprevalence was significantly higher in 2017-2019 (13.0%, CI95% 9.2-18.2%) than in 1994-1999 (5.4%, CI95% 3.0-9.6%) (p = .009). Within the second period, seroprevalence was significantly higher in 2019 compared to 2017 (24.7 vs 1.7% considering all the sample, p = .007), and lower during the winter than the autumn (p < .001). While our molecular and serological results show that Iberian hares have been in contact with MYXV or an antigenically similar virus at least since 1996, they also show an increase in seroprevalence in 2018-2019. The remote contact with MYXV may have occurred with strains that circulated in rabbits, or with unnoticed strains already circulating in Iberian hare populations. This work strongly suggests the infection of Iberian hares with MYXV or an antigenically related virus, at least 20 years before the severe virus outbreaks were registered in 2018.

The African swine fever modelling challenge: Model comparison and lessons learnt

Robust epidemiological knowledge and predictive modelling tools are needed to address challenging objectives, such as: understanding epidemic drivers; forecasting epidemics; and prioritising control measures. Often, multiple modelling approaches can be used during an epidemic to support effective decision making in a timely manner. Modelling challenges contribute to understanding the pros and cons of different approaches and to fostering technical dialogue between modellers. In this paper, we present the results of the first modelling challenge in animal health - the ASF Challenge - which focused on a synthetic epidemic of African swine fever (ASF) on an island. The modelling approaches proposed by five independent international teams were compared. We assessed their ability to predict temporal and spatial epidemic expansion at the interface between domestic pigs and wild boar, and to prioritise a limited number of alternative interventions. We also compared their qualitative and quantitative spatio-temporal predictions over the first two one-month projection phases of the challenge. Top-performing models in predicting the ASF epidemic differed according to the challenge phase, host species, and in predicting spatial or temporal dynamics. Ensemble models built using all team-predictions outperformed any individual model in at least one phase. The ASF Challenge demonstrated that accounting for the interface between livestock and wildlife is key to increasing our effectiveness in controlling emerging animal diseases, and contributed to improving the readiness of the scientific community to face future ASF epidemics. Finally, we discuss the lessons learnt from model comparison to guide decision making.