2.3 Fence Survey
In February 2018, a pilot fence survey was conducted to identify portions of the fence with holes used by wild suids. We term these holes as sites. The fence was divided into four main sections to relate fence crossings by pigs to the localities of farms. Thirty-two sites were identified on the western part of the reserve, 57 on the South-western side, 46 on the south-eastern and 6 on the eastern side. Two bouts of fence surveys were conducted to identify wild suid activities at the respective sites: the wet season survey was conducted on 27 consecutive days of February 2018 and the dry season survey was conducted in August 2018 for 30 consecutive days. The number and location of all the sites were the same for both surveys. On each day, each site was inspected for the presence of bushpig and/or warthog tracks. Once tracks were observed, the researcher, with the help of experienced tracker game scouts, took note of the species responsible, identity of the site, and the number of tracks counted. For each observed set of tracks, the species was identified, based on the footprints (on a few occasions droppings). Three items were recorded for each site: a) whether tracks indicated wild suids crossing the fence (= a crossing event); b) the species of suid; c) an estimate of the number of wild suids that had crossed at that point, based on the tracks leading to and from the site. A crossing event refers to an occasion when one or more warthogs/bushpigs crossed the fence at a specific site. The mean number of crossing events/site/day during a survey (fc) is the mean daily number of crossing events/site for a specific section of the fence during the survey (Table 5). Similarly, the mean number of wild suids/site/day during a survey (fp) represents the mean daily number of wild suids/site for a specific section of the fence during the survey (Table 5). Since the statistical distribution of the number of crossings was similar to that described by a negative exponential function, we performed analyses on the natural log-transformed values of fc and fp. For detecting differences among the four sections of the fence, a 1-way ANOVA was separately performed on fc as well as on fp classified by fence section. For detecting differences between the dry and wet seasons, a repeated-measures ANOVA was separately performed on fc as well as on fp categorised by season. Standard errors of the estimates (S.E.M.) for the number of wild suids crossing the whole fence each day were generated by performing 1000 bootstrap samples of the observations at each fence site and finding the S.E.M. of the 1000 estimates of the wild suid crossing rate. R V3.4.4 (R Core Team, 2018) was used for all statistical analysis and statistical significance was set for P-value lower or equal to 0.05.