Feral pig management in southern Texas: A landscape genetics approach

Feral pigs are the most abundant exotic ungulate in the United States, comprised of escaped domestic swine, Eurasian wild boar, and their hybrids. Feral pigs have been reported in 23 states, with Texas having the highest number of animals (about 2 million). Feral pigs predate on and compete for resources with native wildlife. In addition, feral pigs are a significant threat to agriculture (grain, peanut, soybean, cotton, and vegetable crops) and soils because of rooting and trampling. Feral pigs also transmit, and are reservoirs for, diseases such as pseudorabies, brucellosis, bovine tuberculosis, and foot and mouth disease that affect humans and livestock. I studied complementary ways for assessing multiple paternity, damage control effectiveness, and the influence of landscape features on genetic population structure.;I found evidence of multiple paternity in 21 of 64 litters (33%) from 7 of 9 south Texas sites sampled. Sows displayed a range of conception dates, where 2 sites had a non-concurrent estrous, sows collected at a third site had a bimodal estrous, and sows had a relatively synchronous estrous at a fourth site. Males sired from 8 to 11 offspring at 3 study sites with sufficient litters to assess paternal contribution. The mean litter size was less than the best fit value for the number of offspring sired, indicating that some males sired offspring with ≥2 females. My data could improve management strategies by providing insight into mating behavior of feral pigs. For example, the frequency of multiple paternity can be used to assess sexual contact, and therefore opportunity for disease transmission. In addition, the presence of multiple mating could influence the development of fertility control strategies, specifically in the decision of which sex should be targeted.;Lethal control methods are widely practiced but the effects of lethal control on the population dynamics of feral pigs have rarely been evaluated. I used a panel of genetic markers to assess the impact of common lethal control methods (trapping and aerial shooting) at 2 sites in southern Texas. I conducted 2-3 removals at each site and quantified population structure and differentiation among pigs sampled during the removal events. I found no evidence for immigrants from genetically differentiated populations. I conclude that lethal control at the spatial scale and level of intensity tested had minimal effect on feral pig populations. Control efforts should be conducted at greater spatial scales to achieve long-term impacts on populations of feral pigs in south Texas, perhaps through the formation of management cooperatives. My findings also emphasize the need for understanding how landscape features influence the movements and large-scale population structure of feral pigs in the south Texas region.;I used genetic data to evaluate the effect of landscape features (e.g. waterways, roads, urban area, and agricultural fields) on the populations structure and movements of feral pigs. I generated a cost-weighted distance surface for southern Texas to model the movement of feral pigs through the landscape. I validated the cost-weighted model using genetic data from 1,258 adult feral pigs collected at 24 sites in southern Texas. First, I quantified broad-scale population structure among the 24 sites by computing an overall FST value and used different Bayesian clustering algorithms implemented in Structure 2.2 and BAPS 4.2 to group individuals into genetic clusters (K), both with and without considering the spatial location of samples. Subsequently, I overlapped the cost- weighted surface with the suggested number of clusters to identify barriers and corridors among clusters. At a broad geographic scale, pig populations displayed a moderate degree of genetic structure (FST = 0.11). The best partition for number of populations in Structure 2.2, based on 2nd order rate of change of the likelihood distribution was K = 10 discrete genetic clusters. A spatially Bayesian clustering algorithm implemented in BAPS 4.2 suggested 12 genetic clusters in the region. I found evidence of past (and possibly ongoing) translocations; many populations were admixed. At finer geographic scales, it appears that predicted areas of unsuitable habitat did indeed act as barriers, while expanses of homogenous habitats may facilitate natural (not human-mediated) dispersal among sites. The cost-weighted distance surface was validated with genetic data and represents a realistic model that can be applied for other zones in Texas. Results from this study may have a significant impact feral pig control by helping define the geographic area over which control methods should be conducted to achieve long term results. (Abstract shortened by UMI.)...