The origin, connectivity, and individual specialization of island wolves after deer extirpation.

Wolves are assumed to be ungulate obligates, however, a recently described pack on Pleasant Island, Alaska USA, is persisting on sea otters and other marine resources without ungulate prey, violating this long-held assumption. We address questions about these wolves regarding their origin and fate, degree of isolation, risk of inbreeding depression, and diet specialization by individual and sex. We applied DNA metabarcoding and genotyping by amplicon sequencing using 957 scats collected from 2016 to 2022, and reduced representation sequencing of tissue samples to establish a detailed understanding of Pleasant Island wolf ecology and compare them with adjacent mainland wolves. Dietary overlap was higher among individual wolves on Pleasant Island (Pianka's index mean 0.95 ± 0.03) compared to mainland wolves (0.70 ± 0.21). The individual diets of island wolves were dominated by sea otter, ranging from 40.6% to 63.2% weighted percent of occurrence (wPOO) (mean 55.5 ± 8.7). In contrast, individual mainland wolves primarily fed on ungulates (42.2 ± 21.3) or voles during a population outbreak (31.2 ± 23.2). We traced the origin of the Pleasant Island pack to a mainland pair that colonized around 2013 and produced several litters. After this breeding pair was killed, their female offspring and an immigrant male became the new breeders in 2019. We detected 20 individuals of which 8 (40%) were trapped and killed while two died of natural causes during the 6-year study. Except for the new breeding male, the pedigree analysis and genotype results showed no additional movement to or from the island, indicating limited dispersal but no evidence of inbreeding. Our findings suggest wolves exhibit more flexible foraging behavior than previously believed, and hunting strategies can substantially differ between individuals within or between packs. Nevertheless, anthropogenic and natural mortality combined with limited connectivity to the mainland may inhibit the continued persistence of Pleasant Island wolves.

Patterns of Wolf Dispersal Respond to Harvest Density across an Island Complex.

Wolves are highly mobile predators and can disperse across a variety of habitats and over long distances. However, less is known about dispersal capabilities across water and among islands. The biogeography of island systems fosters spatially structured local populations, and their degree of connectivity may influence the dynamics and long-term viability of the regional population. We sought to quantify wolf dispersal rate, distance, and dispersal sex bias throughout Prince of Wales Island, a 6670 km2 island in southeast Alaska, and the surrounding islands that constitute the wildlife management unit (9025 km2). We also investigated patterns of dispersal in relation to hunting and trapping intensity and wolf population density. We used DNA data collected during 2012-2021 long-term monitoring efforts and genotyped 811 wolves, 144 of which (18%) were dispersers. Annual dispersal rates were 9-23% and had a weakly positive relationship with wolf density. Wolves dispersed 41.9 km on average (SD = 23.7 km), and males and females did not disperse at different rates. Of the dispersing wolves, 107 died, and the majority (n = 81) died before they were able to settle. The leading manner of death was trapping (97% of mortalities), and wolves tended to disperse from areas with low harvest density to areas where harvest density was relatively higher. Dispersal occurred both to and from small islands and the larger Prince of Wales Island, indicating bidirectional as opposed to asymmetrical movement, and the genetic overlap of wolf groups demonstrates connectivity throughout this naturally patchy system. Island ecosystems have different predator-prey dynamics and recolonization processes than large, intact systems due to their isolation and restricted sizes; thus, a better understanding of the degree of population connectivity including dispersal patterns among islands in the Prince of Wales archipelago could help inform the management and research strategies of these wolves.

Variation in adult and pup wolf diets at natal den sites is influenced by forest composition and configuration.

Although wolves are wide-ranging generalist carnivores throughout their life cycle, during the pup-rearing season wolf activity is focused on natal den sites where pup survival depends upon pack members provisioning food. Because prey availability is influenced by habitat quality within the home range, we investigated the relative importance of prey species for adults and pups and further examined the relationship between habitat characteristics, wolf diet, and litter size on Prince of Wales Island (POW) in Southeast Alaska. During 2012-2020, we detected 13 active den sites within the home ranges of nine wolf packs. We estimated minimum pup counts using motion-detecting cameras and individual genotypes from noninvasive samples (hair: n = 322; scat: n = 227) and quantified wolf diet composition using fecal DNA metabarcoding (n = 538). We assessed habitat composition, configuration, and connectivity within denning and annual home ranges estimated using wolf GPS-collar data. Contrary to expectations, wolves had a more constricted diet during denning season (April 15-July 31), and within this season pups had a narrower dietary niche (species richness [S] = 4) focused more on deer (relative frequency of occurrence [O/I] = 0.924) than adults (S = 15; deer O/I = 0.591). Litter size had a positive relationship with the relative frequency of deer in a wolf pack's diet. Wolf consumption of deer was positively associated with the proportion of young-growth forest (≤25 years old) within denning and annual home ranges. High levels of vegetation patch interspersion, and the density of closed logging roads were also important predictors, suggesting these habitat qualities were influential for increasing the availability of deer to wolves. Our results contrast with previous research indicating wolf pup diets included more alternate prey (i.e., beaver) than adults and emphasize the importance of deer to wolf viability on POW, especially during denning season.

Recovery of a marine keystone predator transforms terrestrial predator-prey dynamics.

Sea otters (Enhydra lutris) and wolves (Canis lupus) are two apex predators with strong and cascading effects on ecosystem structure and function. After decades of recovery from near extirpation, their ranges now overlap, allowing sea otters and wolves to interact for the first time in the scientific record. We intensively studied wolves during 2015 to 2021 in an island system colonized by sea otters in the 2000s and by wolves in 2013. After wolf colonization, we quantified shifts in foraging behavior with DNA metabarcoding of 689 wolf scats and stable isotope analyses, both revealing a dietary switch from Sitka black-tailed deer (Odocoileus hemionus), the terrestrial in situ primary prey, to sea otters. Here we show an unexpected result of the reintroduction and restoration of sea otters, which became an abundant marine subsidy for wolves following population recovery. The availability of sea otters allowed wolves to persist and continue to reproduce, subsequently nearly eliminating deer. Genotypes from 390 wolf scats and telemetry data from 13 wolves confirmed island fidelity constituting one of the highest known wolf population densities and upending standardly accepted wolf density predictions based on ungulate abundance. Whereas marine subsidies in other systems are generally derived from lower trophic levels, here an apex nearshore predator became a key prey species and linked nearshore and terrestrial food webs in a recently deglaciated and rapidly changing ecosystem. These results underscore that species restoration may serve as an unanticipated nutrient pathway for recipient ecosystems even resulting in cross-boundary subsidy cascades.

Diet analysis using generalized linear models derived from foraging processes using R package mvtweedie.

Diet analysis integrates a wide variety of visual, chemical, and biological identification of prey. Samples are often treated as compositional data, where each prey is analyzed as a continuous percentage of the total. However, analyzing compositional data results in analytical challenges, for example, highly parameterized models or prior transformation of data. Here, we present a novel approximation involving a Tweedie generalized linear model (GLM). We first review how this approximation emerges from considering predator foraging as a thinned and marked point process (with marks representing prey species and individual prey size). This derivation can motivate future theoretical and applied developments. We then provide a practical tutorial for the Tweedie GLM using new package mvtweedie that extends capabilities of widely used packages in R (mgcv and ggplot2) by transforming output to calculate prey compositions. We demonstrate this approach and software using two examples. Tufted Puffins (Fratercula cirrhata) provisioning their chicks on a colony in the northern Gulf of Alaska show decadal prey switching among sand lance and prowfish (1980-2000) and then Pacific herring and capelin (2000-2020), while wolves (Canis lupus ligoni) in southeast Alaska forage on mountain goats and marmots in northern uplands and marine mammals in seaward island coastlines.

Patterns and processes of pathogen exposure in gray wolves across North America.

The presence of many pathogens varies in a predictable manner with latitude, with infections decreasing from the equator towards the poles. We investigated the geographic trends of pathogens infecting a widely distributed carnivore: the gray wolf (Canis lupus). Specifically, we investigated which variables best explain and predict geographic trends in seroprevalence across North American wolf populations and the implications of the underlying mechanisms. We compiled a large serological dataset of nearly 2000 wolves from 17 study areas, spanning 80° longitude and 50° latitude. Generalized linear mixed models were constructed to predict the probability of seropositivity of four important pathogens: canine adenovirus, herpesvirus, parvovirus, and distemper virus-and two parasites: Neospora caninum and Toxoplasma gondii. Canine adenovirus and herpesvirus were the most widely distributed pathogens, whereas N. caninum was relatively uncommon. Canine parvovirus and distemper had high annual variation, with western populations experiencing more frequent outbreaks than eastern populations. Seroprevalence of all infections increased as wolves aged, and denser wolf populations had a greater risk of exposure. Probability of exposure was positively correlated with human density, suggesting that dogs and synanthropic animals may be important pathogen reservoirs. Pathogen exposure did not appear to follow a latitudinal gradient, with the exception of N. caninum. Instead, clustered study areas were more similar: wolves from the Great Lakes region had lower odds of exposure to the viruses, but higher odds of exposure to N. caninum and T. gondii; the opposite was true for wolves from the central Rocky Mountains. Overall, mechanistic predictors were more informative of seroprevalence trends than latitude and longitude. Individual host characteristics as well as inherent features of ecosystems determined pathogen exposure risk on a large scale. This work emphasizes the importance of biogeographic wildlife surveillance, and we expound upon avenues of future research of cross-species transmission, spillover, and spatial variation in pathogen infection.

SNP discovery in candidate adaptive genes using exon capture in a free-ranging alpine ungulate.

Identification of genes underlying genomic signatures of natural selection is key to understanding adaptation to local conditions. We used targeted resequencing to identify SNP markers in 5321 candidate adaptive genes associated with known immunological, metabolic and growth functions in ovids and other ungulates. We selectively targeted 8161 exons in protein-coding and nearby 5' and 3' untranslated regions of chosen candidate genes. Targeted sequences were taken from bighorn sheep (Ovis canadensis) exon capture data and directly from the domestic sheep genome (Ovis aries v. 3; oviAri3). The bighorn sheep sequences used in the Dall's sheep (Ovis dalli dalli) exon capture aligned to 2350 genes on the oviAri3 genome with an average of 2 exons each. We developed a microfluidic qPCR-based SNP chip to genotype 476 Dall's sheep from locations across their range and test for patterns of selection. Using multiple corroborating approaches (lositan and bayescan), we detected 28 SNP loci potentially under selection. We additionally identified candidate loci significantly associated with latitude, longitude, precipitation and temperature, suggesting local environmental adaptation. The three methods demonstrated consistent support for natural selection on nine genes with immune and disease-regulating functions (e.g. Ovar-DRA, APC, BATF2, MAGEB18), cell regulation signalling pathways (e.g. KRIT1, PI3K, ORRC3), and respiratory health (CYSLTR1). Characterizing adaptive allele distributions from novel genetic techniques will facilitate investigation of the influence of environmental variation on local adaptation of a northern alpine ungulate throughout its range. This research demonstrated the utility of exon capture for gene-targeted SNP discovery and subsequent SNP chip genotyping using low-quality samples in a nonmodel species.

Identification of landscape features influencing gene flow: How useful are habitat selection models?

Understanding how dispersal patterns are influenced by landscape heterogeneity is critical for modeling species connectivity. Resource selection function (RSF) models are increasingly used in landscape genetics approaches. However, because the ecological factors that drive habitat selection may be different from those influencing dispersal and gene flow, it is important to consider explicit assumptions and spatial scales of measurement. We calculated pairwise genetic distance among 301 Dall's sheep (Ovis dalli dalli) in southcentral Alaska using an intensive noninvasive sampling effort and 15 microsatellite loci. We used multiple regression of distance matrices to assess the correlation of pairwise genetic distance and landscape resistance derived from an RSF, and combinations of landscape features hypothesized to influence dispersal. Dall's sheep gene flow was positively correlated with steep slopes, moderate peak normalized difference vegetation indices (NDVI), and open land cover. Whereas RSF covariates were significant in predicting genetic distance, the RSF model itself was not significantly correlated with Dall's sheep gene flow, suggesting that certain habitat features important during summer (rugged terrain, mid-range elevation) were not influential to effective dispersal. This work underscores that consideration of both habitat selection and landscape genetics models may be useful in developing management strategies to both meet the immediate survival of a species and allow for long-term genetic connectivity.

Are inland wolf-ungulate systems influenced by marine subsidies of Pacific salmon?

Wolves (Canis lupus) in North America are considered obligate predators of ungulates with other food resources playing little role in wolf population dynamics or wolf prey relations. However, spawning Pacific salmon (Oncorhyncus spp.) are common throughout wolf range in northwestern North America and may provide a marine subsidy affecting inland wolf-ungulate food webs far from the coast. We conducted stable-isotope analyses for nitrogen and carbon to evaluate the contribution of salmon to diets of wolves in Denali National Park and Preserve, 1200 river-km from tidewater in interior Alaska, USA. We analyzed bone collagen from 73 wolves equipped with radio collars during 1986-2002 and evaluated estimates of salmon in their diets relative to the availability of salmon and ungulates within their home ranges. We compared wolf densities and ungulate:wolf ratios among regions with differing salmon and ungulate availability to assess subsidizing effects of salmon on these wolf-ungulate systems. Wolves in the northwestern flats of the study area had access to spawning salmon but low ungulate availability and consumed more salmon (17% +/- 7% [mean +/- SD]) than in upland regions, where ungulates were sixfold more abundant and wolves did or did not have salmon spawning areas within their home ranges (8% +/- 6% and 3% +/- 3%, respectively). Wolves were only 17% less abundant on the northwestern flats compared to the remainder of the study area, even though ungulate densities were 78% lower. We estimated that biomass from fall runs of chum (O. keta) and coho (O. kisutch) salmon on the northwestern flats was comparable to the ungulate biomass there, and the contribution of salmon to wolf diets was similar to estimates reported for coastal wolves in southeast Alaska. Given the ubiquitous consumption of salmon by wolves on the northwestern flats and the abundance of salmon there, we conclude that wolf numbers in this region were enhanced by the allochthonous subsidy provided by salmon and discuss implications for wolf-ungulate relations.