Modelling advanced knowledge of African swine fever, resulting surveillance patterns at the population level and impact on reliable exit strategy definition

Abstract

African Swine Fever (ASF) is an infectious lethal disease affecting domestic pigs and wild boar. In the EU the infection perpetuates predominantly in wild boar populations. ASF control comprises wild boar population reduction measures, e.g. pre-emptive culling in delineated zones, called white zones (WZ). These WZ are placed geographically adjacent to an area with ASF circulating in wild boar (ASF positive area). The ideal WZ would be depopulated of wild boar without possibility of recolonization. However, WZ may still harbour live wild boar after its establishment and the functionality of the WZ inherently foresees ASF entering it. But the spread of the infection is expected to stop within an effective WZ. The concept must not match legislative zones, likewise infected area, Part I, Part II, etc. In order to compare different approaches to implement a WZ (e.g. targets and speed of population reduction in the WZ, width of the WZ, and distance of the WZ from the ASF-positive area), an individual-based spatially explicit model was adjusted to four historic WZ-like situations in the EU, i.e. Estonia 2014, Latvia 2016, Czech Republic 2017, and France 2018. The model was used to simulate the reported spatio-temporal layout and targeted measures. The stochasticity of the model provided understanding of the general efficiency of these WZ. Alternatives of the local measures were simulated as scenarios to identify caveats of the settings and derive improvements in future applications. The simulation outcome suggests issues to be addressed in implementing future WZ: i) distance between ASF-positive area and the WZ was adequate if adapted to the speed of propagation according to the local wild boar density, and the time horizon of depopulation measures envisaged for the WZ; ii) the width of the WZ was adequately set if everywhere it was prevented that short infection chains already led out of the zone, iii) the WZ around focal introductions was most efficient if depopulated by culling the maximum of a defined (or fenced) population in shortest time with minimal disturbance, for instance, by trapping, sharp shooting or using silencers. Aspects of density, timing and spatial distribution in relation to the efficiency of WZ layout are explored.