The authors revised ecological niche models for three putative orthoebolavirus reservoir bat species and for orthoebolavirus itself. These updated habitat suitability models identified areas environmentally suitable for virus circulation that were previously under-predicted, notably in eastern Democratic Republic of Congo (DRC), including Ituri province. The study situates these ecological findings within a landscape shaped by rainforest, food insecurity, conflict, and mining-related human mobility.
Recognising that ecological suitability alone may not capture opportunities for interspecies contact and zoonotic spillover, the authors developed an integrated spatial spillover risk framework. This framework combines ecological suitability estimates with a set of anthropogenic drivers known to modify human–wildlife contact and spillover opportunity. The anthropogenic factors explicitly included in the framework are human settlement, mining activity, bushmeat-related activities, forest loss, and conflict.
The integrated approach was tested against historical zoonotic spillover locations. When ecological suitability was combined with characteristics of the built environment, spatial predictive accuracy improved substantially. The Boyce Index, a measure of spatial predictive performance reported by the authors, increased from 0.78 for ecological suitability alone to 0.97 when ecological suitability was combined with the built environment. This result indicates stronger model alignment with observed spillover locations after inclusion of anthropogenic information.
Among the anthropogenic scenarios evaluated, mining- and bushmeat-based scenarios showed the greatest enrichment of observed spillover events. The authors highlight mining and bushmeat activities as particularly influential in increasing opportunities for wildlife-to-human contact and zoonotic emergence in the landscapes analysed. These findings support the notion that specific human activities can substantially modify spillover risk even where environmental suitability for reservoirs exists.
The study examined temporal patterns in the relative contribution of ecological suitability and anthropogenic drivers over recent periods. Mining exhibited the largest and most consistent effect on relative spillover risk over the last decade, according to the authors' analyses. Conflict was identified as a secondary amplifying factor whose apparent contribution to spillover vulnerability has grown in more recent periods. The authors emphasise this temporal association as evidence that the drivers of spillover risk can shift over time and that certain anthropogenic pressures have become more prominent.
Based on their results, the authors argue that ecological suitability maps alone do not fully characterise landscapes vulnerable to Ebola zoonotic emergence. Integrating environmental and anthropogenic information can improve identification of high-risk locations, strengthen One Health surveillance approaches, and inform epidemic preparedness and targeted public health interventions in the DRC and neighbouring countries. The combined mapping approach is presented as a tool to prioritise surveillance and interventions where both ecological and human-driven risk factors converge.
The authors state that all model estimates produced in the study are available upon reasonable request and that a public repository is in preparation. All other data used were already publicly available and are described within the manuscript. The article is a preprint posted to medRxiv and has not been certified by peer review; the authors note that the work reports new medical research that has yet to be evaluated and should not be used to guide clinical practice. The manuscript declares no competing interests and confirms adherence to relevant ethical guidelines.
The paper focuses on combining ecological niche modelling with selected anthropogenic covariates to better characterise spatial and temporal patterns of zoonotic spillover opportunity for orthoebolaviruses. Specific methodological details, data sources, and model parameter values are described in the full manuscript; where those details were not reported in the summary provided here, they are not reproduced. The principal findings reported are the improved spatial predictive performance when anthropogenic factors are included, the prominence of mining and bushmeat in explaining historical spillover observations, and the increasing contribution of conflict as an amplifying factor in recent periods.
This work demonstrates that integrating environmental suitability for putative reservoir hosts with human-driven risk factors can identify emerging Ebola spillover hotspots that would be under-predicted by ecological models alone. The authors propose that this integrated approach can guide more targeted One Health surveillance and preparedness efforts in high-risk areas such as eastern DRC.