1. In many social species, reproductive success varies between individuals within a population, resulting in socially structured populations. Social network analyses of familial relationships may provide insights on how fitness influences population-level demographic patterns. These methods have however rarely been applied to genetically-derived pedigree data from wild populations. 2. Here we use social networks to reconstruct parent-offspring relationships and create a familial network from polygamous boreal woodland caribou (Rangifer tarandus caribou) in Saskatchewan, Canada, to inform recovery efforts. We collected samples from 933 individuals at 15 variable microsatellite loci along with caribou-specific primers for sex identification. Using social network metrics, we assess the contribution of individual caribou to the population with several centrality metrics and then determine which metrics are best suited to inform on the population demographic structure. We look at the centrality of individuals from eighteen different local areas, along with the entire population. 3. We found substantial differences in centrality of individuals in different local areas, that in turn contributed differently to the full network, highlighting the importance of analyzing social networks at different scales. The full network revealed that boreal caribou in Saskatchewan form a complex, interconnected social network with strong familial ties, as the removal of edges with high betweenness did not result in distinct subgroups. Alpha, betweenness, and eccentricity centrality were the most informative metrics to characterize the population demographic structure and for spatially identifying areas of highest fitness levels and social cohesion across the range. 4. Synthesis and applications: Our results demonstrate the value of different network metrics in assessing genetically-derived familial networks. The spatial application of the familial networks identified areas of higher fitness levels and social cohesion across the range in support of population monitoring and recovery efforts.

Accurately estimating abundance is a critical component of monitoring and recovery of rare and elusive species. Spatial capture-recapture (SCR) models are an increasingly popular method for robust estimation of ecological parameters. We provide a maximum likelihood analytical framework to assess results from empirical studies to inform SCR sampling design, using both simulated and empirical data from non-invasive genetic sampling of seven boreal caribou populations (Rangifer tarandus caribou) which varied in range size and estimated population density. We use simulated population data with varying levels of clustered distributions to quantify the impact of non-independence of detections on density estimates, and empirical datasets to explore the influence of varied sampling intensity on the relative bias and precision of density estimates. Simulations revealed that clustered distributions of detections did not significantly impact relative bias or precision of density estimates. The empirical genotyping success rate was 95.1%. Empirical results indicated that reduced sampling intensity had a greater impact on density estimates in smaller ranges. The number of captures and spatial recaptures were strongly correlated with precision, but not relative bias. The best sampling designs did not differ with estimated population density but differed between large and small ranges. We provide an efficient framework implemented in R to estimate the detection parameters required when designing SCR studies. The framework can be used when designing a monitoring program to minimize effort and cost while maximizing effectiveness, which is critical for informing wildlife management and conservation.

Accurately estimating abundance is a critical component of monitoring and recovery of rare and elusive species. Spatial capture-recapture (SCR) models are an increasingly popular method for robust estimation of ecological parameters. We provide a maximum likelihood analytical framework to assess results from empirical studies to inform SCR sampling design, using both simulated and empirical data from non-invasive genetic sampling of seven boreal caribou populations (Rangifer tarandus caribou) which varied in range size and estimated population density. We use simulated population data with varying levels of clustered distributions to quantify the impact of non-independence of detections on density estimates, and empirical datasets to explore the influence of varied sampling intensity on the relative bias and precision of density estimates. Simulations revealed that clustered distributions of detections did not significantly impact relative bias or precision of density estimates. The empirical genotyping success rate was 95.1%. Empirical results indicated that reduced sampling intensity had a greater impact on density estimates in smaller ranges. The number of captures and spatial recaptures were strongly correlated with precision, but not relative bias. The best sampling designs did not differ with estimated population density but differed between large and small ranges. We provide an efficient framework implemented in R to estimate the detection parameters required when designing SCR studies. The framework can be used when designing a monitoring program to minimize effort and cost while maximizing effectiveness, which is critical for informing wildlife management and conservation.

Parallel evolution can occur through novel mutations, standing genetic variation, or adaptive introgression. Uncovering parallelism and introgressed populations can complicate management of threatened species, particularly as admixed populations are not generally considered under conservation legislations. We examined high coverage whole-genome sequences of 30 caribou (Rangifer tarandus) from across North America and Greenland, representing divergent intra-specific lineages, to investigate parallelism and levels of introgression contributing to the formation of ecotypes. Caribou are split into four subspecies and 11 extant conservation units, known as Designatable Units (DUs), in Canada. Using genomes from all four subspecies and six DUs, we undertake demographic reconstruction and confirm two previously inferred instances of parallel evolution in the woodland subspecies and uncover an additional instance of parallelism of the eastern migratory ecotype. Detailed investigations reveal introgression in the woodland subspecies, with introgressed regions found spread throughout the genomes encompassing both neutral and functional sites. Our comprehensive investigations using whole genomes highlight the difficulties in unequivocally demonstrating parallelism through adaptive introgression in non-model species with complex demographic histories, with standing variation and introgression both potentially involved. Additionally, the impact of parallelism and introgression on the designation of conservation units has not been widely considered, and the caribou designations will need amending in light of our results. Uncovering and decoupling parallelism and differential patterns of introgression will become prevalent with the availability of comprehensive genomic data from non-model species, and we highlight the need to incorporate this into conservation unit designations.

Conservation genomics is an important tool to manage threatened species under current biodiversity loss. Recent advances in sequencing technology mean that we can now use whole genomes to investigate demographic history, local adaptation, inbreeding, and more in unprecedented detail. However, for many rare and elusive species only non-invasive samples such as faeces can be obtained, making it difficult to take advantage of whole genome data. We present a method to extract DNA from the mucosal layer of faecal samples to reconstruct high coverage whole genomes using standard laboratory techniques, therefore in a cost-effective and efficient way. We use wild collected faecal pellets collected from wild caribou (Rangifer tarandus), a species undergoing declines in many parts of its range in Canada and subject to comprehensive conservation and population monitoring measures. We compare four faecal genomes to two tissue genomes sequenced in the same run. Quality metrics were similar between faecal and tissue samples with the main difference being the alignment success of raw reads to the reference genome likely due to differences in endogenous DNA content, affecting overall coverage. One of our faecal genomes was only reconstructed at low coverage (1.6X), however the other three obtained between 7 and 15X, compared to 19 and 25X for the tissue samples. We successfully reconstructed high-quality whole genomes from faecal DNA and, to our knowledge, are the first to obtain genome-wide data from wildlife faecal DNA in a non-primate species, representing an important advancement for non-invasive conservation genomics.