This study shows that landscape connectivity studies can be a powerful tool to guide epidemiology strategies to limit the spread and persistence of diseases associated to wild population movements. Landscape connectivity predictions can identify priority areas and routes for the dispersal of wild species and the consequent spread of the diseases they carry on at landscape level. This epidemiological tool based on the connectivity approach, identify and locate hotspot areas of habitat patches and corridors with a potentially key role in the spread of ASF in wild boar populations. Identifying these hotspot areas can provide practical guidance to better focalize surveillance, prevention, and control efforts that promote early detection and more effective disease eradication measures. This is especially important for diseases with attenuated strains such as ASF that allow wildlife animals to propagate farther and contact with other populations which further complicate the disease eradication 7.
The feature that makes functional landscape connectivity studies particularly suitable to predict wildlife movements and the spread of the diseases they carry is that they explicitly consider the landscape structure and the relationship and reaction of the focal species to it 28,33, two aspects usually neglected in wildlife disease analyses. Particularly, we considered the distribution and abundance of habitat patches, the resistance that the landscape offers to the species movements, and the dispersal capacity of the species through it. Landscape connectivity studies have mainly focused on enhancing the movements of species of special conservation concern, but this study has shown the great potential of these connectivity analyses to guide measures that limit the expansion of detrimental processes such as infectious diseases. Here, we studied the spread of ASF by wild boars, but connectivity assessments can also focus on the spread of other diseases transmitted by wild boars (such as tuberculosis classical swine fever, brucella suis, hepatitis E, or Aujeszky’s disease) or for any other wildlife reservoir species.
Here, we proposed two novel factors based on connectivity assessments that help to locate habitat patches and corridors with a potentially key role in the spread of the disease. These two factors identify the countries that should prioritize the design of more restrictive strategies for preventing ASF spread by wild boars. Each of these factors provides different and critical perspectives of ASF spread that should be considered both individually and jointly. It should be noted that habitat patches with mean-high impact and risk factors have great characteristics of connectivity for wild boar and for the spread of the ASF in Europe too. This high level of connectivity is due to the topological position and configuration of habitat patches in the network at the landscape level (with high quality of habitat, high occurrence and abundance of the species). Thus, these results highlight that significant impact and risk areas of ASF could be zones where the disease can remain endemic for many years in wild boar with the risks that this entails for the rest of the European regions. Furthermore, this risk situation, due to the bidirectional characteristics of connected habitat patches, can generate a continued spread and flux of ASFv over time from the endemic areas to other free European regions. In this sense, this epidemiological tool based on the connectivity approach for wild boar, would be very useful to help to implement futures ASF surveillance and eradication programs for wild boar, as previously mentioned. These actions should focus on the most important risk and impact areas as a source of infection and the virus exit to free-ASF regions. It must be taken into account that this approach studies the entire connectivity network that exists in wild boar habitats in Europe. This European ASF network connectivity tool for wild boar at a large scale will help us to understand how wild boar populations are connected and how the disease has progressed and how it will progress in the future in the European context.
4.1. ASF Impact Factor
The disease spread and infection of the high-impact areas could seriously hinder the control of the disease in Europe, even though these areas were not necessarily within imminent threat of contagious by wild boar dispersal from the known ASF outbreaks. Given the high levels of connectivity of these areas (both habitat patches and corridors) with the entire European network of wild boars, the infection of these areas could trigger a rapid spread of infected wild boar to their many connected habitat patches. This information could be crucial for consideration in preventive and surveillance plans that seek to minimize the probability of introduction of ASF into wild populations as well as to reduce the spread of the disease in case such introduction should occur. Countries with a high impact factor (transmission hotspots) should implement additional measures to prevent the introduction of the disease in these areas both by the dispersal of wild boars from infected areas and by other anthropogenic transmission routes such as contact with ASF-contaminated food waste, domestic pigs spillover, and fomites. Furthermore, more drastic emergency actions could be more designed for these high-impact areas in case of an outbreaks such as population control, fencing, ban of people entering the affected area, and correct disposal or removal of carcasses 35. It would be also highly recommendable to minimize the wild boar - domestic pig interface and prioritize vaccination when available and approved for ASF-free but high impact areas. The identification of these high impact areas may also be useful to control other infectious diseases transmitted by wild boars.
We found that the countries with the highest mean impact factor were mostly located in the southeast and central parts of Europe. Furthermore, the countries with a high percentage of area covered by wild boar potential habitat also generally tended to have a higher impact factor, showing that the abundance of individuals and the connectivity increased with the available habitat 11. However, there are some countries such as Finland or Sweden that despite their great percentage of habitat, they presented a very low impact factor, associated with their habitat patches with low attributes (wild boar abundance) and not a prominent role for the overall connectivity. Therefore, larger habitat patches do not always equal higher impact factors, since the impact factor mainly depends on the abundance of wild boar and the topological position and configuration of habitat patches in the connectivity network. These two variables are dynamic and can rapidly shift with land use, climate, and hunting pressure changes. Therefore, it is important to update the impact factor periodically.
4.2. ASF Risk Factor
Conversely, the risk factor mainly depends on the occurrence, abundance of the species and on the position and configuration of the habitat patches in the landscape relative to the location of ASF cases in space and time. Therefore, habitat patches with a high-risk factor are in immediate danger of infection and should be deeply and continuously monitored to control the spread as soon as a new outbreak appears. Early detection is very important for the feasibility of eradication measures, as the longer the time lapse, the larger the infected area and the number of individuals affected 2. Remote telemetry or biologgers that detect behavioral changes associated to ASF infection such as declines in movement, temperature or heart-rate can be excellent tools to rapidly detect the disease in high-risk areas 36. These areas should also be subject to special strategies to prevent ASF spread, such as priority vaccination, reduced hunting pressure associated with higher movement rates 37, or movement restriction zones especially reinforced in corridor surroundings. Careful attention should be paid to regions such as Slovakia that have both high impact and risk factors, and thus the habitat patches in this country are highly likely to be infected and its infection could rapidly spread the disease to many other habitat patches connected with them. Finally, the high accuracy of the risk factor showed that (i) the risk factor is a good indicator to predict where the disease is likely to spread; and (ii) wild boars’ movements have a key role in ASF transmission in Europe as we did not considered any other mean of ASF spread. In order to predict the risk factor further into the future, it should be periodically updated with all new ASF cases detected.
Further improvements and future research
We presented here a useful standalone model that can guide ASF prevention, and control plans that include an improved wild boar management. However, only with the combination of this tool with other available management plans, the holistic approach needed to fully tackle the problematic control of the disease could be significantly improved. In that sense, the inclusion of epidemiological models for both wild boar and domestic pigs, such as transmission dynamics 7, wild boar - domestic pig interface 11,38, as well as risk of exposure assessment analyses 39–41 would be highly beneficial to improve control and prevention strategies. Other transmission routes, apart from the natural spread of the disease in wild boar studied here, should be considered for such ASF plans 42.
The applicability of this tool into other regions of the world, such as Asia where the disease is rapidly spreading and there is limited information on the affected wild boar populations 40, could improve the understanding of ASF dispersion there. However, before being applied to other continents, it is important to adapt the connectivity models to address the specific relationships of the species with different landscapes. For example, in the American continent, which is widely invaded by wild pigs, the tool should be partially adapted to the ecological and behavioral differences of those pigs and could help design emergency actions against ASF if ever introduced.