Introduction
Isolation and subsequent local adaptation of populations are considered common processes that lead to speciation. Changes in habitat, barriers to dispersal, or stochastic demographic events can cause population isolation and diversification (Slatkin 1987; Steinberg et al.2000). In this context, environmental variation plays an important role in facilitating or hindering connectivity, and thus in promoting the persistence of populations, their vicariance, or even extinction (Waitset al. 2016). On one hand, increased structural and functional connectivity facilitates the persistence of small populations that are highly susceptible to demographic stochasticity, genetic drift and density-dependent effects (Hanski 1998; Lopes & de Freitas 2012; Wittmann et al. 2018). On the other hand, high levels of connectivity result in more genetically homogenous populations, with less propensity for local adaptation (Kawecki & Ebert 2004). With loss of connectivity between populations, allele frequencies tend to diverge due to genetic drift, ultimately leading to neutral genetic differentiation (Rundell & Price 2009). Additionally, spatial variation in local selection pressures within a species’ range, particularly when there is habitat fragmentation, can lead to changes in allele frequencies and fixation of new adaptive mutations, resulting in the emergence of adaptive differences (Orsini et al. 2013). This is especially important for small and isolated populations that are restricted to increasingly unfavorable habitat, for which studies have shown that local adaptation tends to occur more rapidly (Wood et al. 2016; Hoffmann et al. 2017). Recently diverged populations represent a great opportunity to study the process of genetic differentiation (Fišer et al. 2018; Fletcher et al. 2019; Marques et al. 2019). The comparison of neutral and adaptive variation should provide evidence for distinguishing which processes are contributing most to differentiation, and what can be done to circumvent or sustain that diversification, depending on specific conservation goals (Orsini et al. 2013).
Northern and southern Idaho ground squirrels (Urocitellusbrunneus and U. endemicus , respectively) are a recently diverged pair of sister species, which currently have an allopatric distribution (Figure 1) (Yensen 1991). Northern Idaho ground squirrels (hereafter NIDGS) and southern Idaho ground squirrels (hereafter SIDGS) were formerly considered two subspecies of Spermophilus brunneus and have been distinguished as separate species on the basis of ecological niche modelling, morphology and genetics (Gill & Yensen 1992; Yensen & Sherman 1997; Helgen et al. 2009; Hoisington-Lopez et al.2012; USFWS 2015). Both species are rare, endemic to Idaho, and are of high conservation concern (IUCN 2000, 2018). Ecologically, both species are semi-colonial and patchily distributed, representing classic examples of metapopulation structure whereby dispersal among populations is uncommon and tends to occur in a ‘stepping stone’ manner (Yensen 1991; Yensen & Sherman 1997; USFWS 2003). NIDGS live in open meadows, grassy scabs and small rocky outcroppings at an elevation of 1100 to 2300 m within coniferous forests of central Idaho (Burak 2011; Goldberg, Conway, Mack, et al. 2020), and they persist within only a small fraction of their former range likely due to habitat loss and reduced population connectivity, mostly as a result of forest encroachment (Sherman & Runge 2002; Suronen & Newingham 2013; Yensen & Dyni 2020; Helmstetter et al. 2021). SIDGS live in sagebrush steppe and rolling hill slopes at an elevation of 630 to 1400 m in southwestern Idaho, and are currently threatened by urban and agricultural development, as well as the spreading of invasive annual plants (USFWS 2000; Lohr et al. 2013). Morphological differences between the two species include coat color, which tends to mimic differences in soil color between the species’ geographic ranges (Yensen 1991), pelage (longer in SIDGS), and baculum characteristics (longer with more spines in SIDGS) (Yensen 1991). Previous genetic work on NIDGS and SIDGS estimated that the divergence between NIDGS and SIDGS occurred about 32.5 (18.3-63.5) thousand years ago, during the Quaternary climate cycles, and found no subsequent gene flow between the two species (Hoisington-Lopez et al. 2012). Vicariant events of this magnitude have been found to be sufficient for distinct evolutionary lineages to become different species, a pattern frequently found in several North American small mammals (Hope et al. 2014, 2016, 2020).
Local adaptation is likely to be an important factor for ground-dwelling, hibernating, small mammals like NIDGS and SIDGS with limited dispersal abilities. Both ground squirrel species hibernate but the timing of hibernation differs, likely due to differences in elevation and climate (Yensen 1991; Goldberg & Conway 2021). This difference in emergence timing between NIDGS and SIDGS could have a genomic basis, or may simply result from a plastic response (Yensen 1991; Hut & Beersma 2011; Santos et al. 2017). Typically, adaptations are associated with the habitat variables that affect fitness the most, which in the case of the Idaho ground squirrels (hereafter IDGS) are likely variables associated with energy consumption, timing of food availability, soil temperature, forage quality, and risk of predation, which may vary between the active season and hibernation (Goldberg, Conway, Mack, et al. 2020; Goldberg, Conway, Tank, et al. 2020). Variation in ground squirrel hibernation emergence timing has been associated with food availability and snowpack and thus, site productivity appears to dictate differences within and possibly between species differentiation (French 1982; Goldberg & Conway 2021). These differences may be determinant for species divergence, but may also lead to intraspecific local adaptation if environmental differences are found across the species range (Kawecki & Ebert 2004; Savolainen et al. 2013). Previous intraspecific genetic studies on Idaho ground squirrels have found that genetic differentiation was low to moderate among NIDGS populations, with one disjunct population (Round Valley) being completely isolated (Yensen & Sherman 1997; Garner et al. 2005; Hoisington 2007; Hoisington-Lopez et al. 2012). In SIDGS, the Weiser River was documented as a barrier to dispersal between populations, but connectivity among populations on either side of the river was relatively high (Garner et al. 2005; Hoisington 2007). Additionally, there are reports of human-mediated translocations between localities east of the Weiser River, particularly from southern localities close to Van Deussen to the vicinity of the Weiser River population (Yensen et al. 2010; Yensen & Tarifa 2012). However, translocation success in SIDGS has been very limited, especially into areas without established populations, for which the majority of the translocated individuals did not survive the first winter (Busscher 2009; Yensen et al. 2010). Given the currently observed low levels of gene flow among some populations within both species, population persistence might be highly dependent on locally adapted genotypes for many isolated populations (USFWS 2000; Rundell & Price 2009). Thus, to better understand the probability of population persistence under habitat and climate change, it is essential to determine the role of neutral processes in maintaining population connectivity and overall genetic diversity, and the role of adaptive processes in improving population resilience through local adaptation (Macdonald et al. 2018).
In this study, we aimed to develop and use genomic tools to provide novel information on neutral and adaptive genetic diversity and differentiation within and among NIDGS and SIDGS to address the following five questions: 1) how do genetic patterns of adaptive and neutral variation compare between and within species? 2) are there any populations that have elevated levels of adaptive differentiation? 3) what landscape variables are associated with loci under selection? 4) can we identify specific genes under selection and what are their putative functions? and 5) can we identify conservation units based on genomic data to inform management? To address these questions, we tested the following hypotheses: (a) geographic distance will be the main driver of differentiation for neutral loci in NIDGS and SIDGS, possibly exacerbated by geographic barriers to gene flow; (b) populations within each species will exhibit signatures of adaptation to local environmental conditions ; (c) local adaptation between populations within species will be highest in NIDGS because this species occupies a more topographically diverse area (Yensen & Sherman 1997); and (d) adaptive variation will be associated with timing of hibernation, production and storage of fat, and increased metabolism at higher elevation for NIDGS (Faherty et al. 2018; Garcia-Elfring et al. 2019). Finally, we combine all of this information to identify Evolutionarily Significant Units (ESUs), Management Units (MUs) and Adaptive Units (AUs) that could warrant heightened protection due to genetic isolation or the emergence of local adaptation.