Since the first detection of the African swine fever (ASF) virus in the Republic of Korea in 2019, the Korean government has applied interventions, including fencing, increasing the biosecurity level at domestic pig farms, and the capture-and-removal of wild boars. In particular, wild boars are an important risk factor for ASF control because they can spread disease among susceptible animals, such as wild boars or domestic pigs. A capture-and-removal method aims to reduce the likelihood of ASF transmission from wild boars to domestic boars or among wild boars by decreasing the number of susceptible wild boars. This study estimated the required number of wild boars captured and removed for ASF control using population viability analysis. Population factors, such as a life span, sex ratio, or an inbreeding depression with different capture-and-removal proportions of wild boars, were included in the analysis. Ten scenarios with different capture-and-removal proportions of wild boars and different periods of culling were considered. According to the results, a method in which 75% of wild boars are captured-and-removed for at least three years showed long-term effectiveness for more than ten years. The current ASF control method, in which 33% of wild boars are captured-and-removed, decreased the number of wild boars for three years, after which the wild boar population increased to more than its initial number. Given the limited human and material resources for controlling ASF in the Republic of Korea, it is recommended that resources be prioritized to increase the capture-and-removal proportion of wild boars to take full advantage of the ASF-control effectiveness.

In this paper, a mathematical model of regionalization based on graph theory to investigate the patterns induced by movements of livestock vehicles in cities under outbreaks of highly pathogenic avian influenza (HPAI) is proposed. We then compare the results of simulation from the regionalization model to actual HPAI outbreaks in 2016/2017 to evaluate the validity of the model. Specifically, we (1) configured a complex network structure with analytic tools and properties in graph theory to abstract the paths among farms and livestock facilities; (2) employed statistical methods to estimate the possibility of propagation between two clusters; (3) applied the developed method to an actual HPAI outbreak in Korea in 2016 and conducted a simulation to determine if the proposed modeling for regionalization is an effective prediction measure. The clustered regions proposed by the simulation correctly reflected the regional clustering of actual cases, while simultaneously contain the cities exposed to potential damage when separated. Based on these findings, we conclude that our proposed regionalization model is suitable for making policy judgments to establish a preemptive biosecurity system.