Population structure and genetic connectivity in the endangered Pectis imberbis: addressing conservation and genetic gaps in the Arizona Sky Islands

Population structure and genetic connectivity in the endangered Pectis imberbis: addressing conservation and genetic gaps in the Arizona Sky Islands | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Population structure and genetic connectivity in the endangered Pectis imberbis: addressing conservation and genetic gaps in the Arizona Sky Islands Scott Gilb, Karen Haubensak, Clare Aslan, Liza Holeski, Trevor Faske, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7022625/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The conservation of rare and endangered plants requires understanding of genetic diversity and connectivity to mitigate the effects of habitat fragmentation and environmental change. Pectis imberbis (Gray), a perennial herb endemic to the Arizona Sky Islands, is listed as endangered due to recent extirpations, population declines, habitat loss, and restricted range. To address critical knowledge gaps in P. imberbis conservation, we assessed population structure, genetic diversity, and connectivity across its range using single nucleotide polymorphisms (SNPs) generated through RADseq. We identified three genetically distinct population clusters, with limited gene flow among populations located in the Huachuca, Santa Rita, and Atascosa-Pajarito Mountain ranges. Estimated effective migration surfaces revealed barriers to gene flow, particularly around Montezuma Pass and the Patagonia Mountains, which corresponded with demographic declines and recent extirpations. Pollinator visitation and floral network analyses showed consistent overlap of key pollinator taxa across populations but suggested limited pollen transfer over large distances. These findings highlight the need for targeted restoration efforts to enhance genetic connectivity, such as establishing stepping-stone populations in regions of limited migration. Future research should focus on testing adaptive variation to guide restoration actions taken to increase connectivity. By integrating genetic, demographic, and pollinator data, this work directly informs P. imberbis conservation, and more generally, contributes to understanding of rare species conservation in fragmented landscapes. Conservation Genetic Population Genetics – Empirical Population Ecology Habitat Degradation Quantitative Genetics Climate Change Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Nearly 40% of vascular plants are considered at risk of extinction, necessitating broadscale efforts to conserve biodiversity (Nic Lughadha et al. 2020). Rare plant conservation planning is predicated on understanding the distribution of genetic variation across the landscape (Dennis et al. 1991 , Dibner et al. 2019 , Lande & Shannon 1996 ). Land managers and conservationists apply this knowledge to select conservation targets (Hedrick & Miller, 1992 ), identify critical corridors to maintain gene flow (Segelbacher et al. 2010 ), and introduce novel genotypes across the landscape via managed relocation or assisted evolution (Holderegger et al. 2019). Integrating genetic approaches into conservation and land management is becoming increasingly important, as rapid human-driven global change introduces new selective pressures and threatens both genetic connectivity and genetic diversity in already imperiled species (Aguilar et al. 2008 ; Pauls et al. 2013 ) Rare plants, by definition, are characterized by small population sizes that shape demographic and evolutionary trajectories and thus conservation approaches (Ellstrand & Elam, 1993). Small populations are subject to demographic and genetic stochasticity that increase the likelihood of random allele loss, leading to non-adaptive shifts in gene frequencies (i.e., genetic drift) (Honnay & Jacquemyn, 2007). When small population size leads to increased inbreeding, individual fitness may decrease due to inbreeding depression via the expression of deleterious recessive alleles (Hamabata et al. 2019 ). Other factors such as the Allee effect – a decline in fitness linked to diminishing populating sizes -impacts small populations by lowering per capita reproductive rates and reinforcing patterns of rarity (Stephens et al., 1999 ). Smaller patch sizes of flowering plants are less effective at attracting pollinators due to pollen limitation and diminished visual cues, driving Allee effects while also reducing the potential for genetic or demographic rescue from neighboring populations (Anic et al. 2015 ). These factors — inbreeding depression, Allee effects, and genetic drift — can create a feedback loop termed ‘the extinction vortex’ that drives populations toward extinction (Nordstrom et al. 2023 ; Pruett et al. 2025 ). Global change and habitat loss interacts with rarity to accelerate extinction (Brook et al., 2008 ). Plant species with small and isolated populations may be unable to adapt to novel environmental conditions when these factors reduce genetically-based phenotypic variation (Boyd et al. 2022 ). When populations are connected rather than isolated, gene flow maintains genetic diversity through admixture with populations in other selective environments (Frankham et al., 2011 ). For this reason, rare species conservation often focuses on maintaining genetic diversity both within and across populations and may target populations with distinct genetic signatures that could represent unique adaptations to the environment as well as those with high levels of genetic diversity (Coates et al., 2018 ). Moreover, rapid climatic and environmental changes can outpace the adaptive potential of rare plants (Jump & Peñuelas, 2005), which often possess life history traits associated with low rates of reproduction and slow growth (Adler et al. 2014 ). This is particularly true for dryland plants with restricted ranges that exist at the edge of physiological tolerances to aridity such that even small decreases in precipitation can cause widespread mortality (Laikre et al. 2010 , Richardson et al. 2013). The Madrean Sky Archipelago ecoregion of southern Arizona (Sky Islands) hosts a broad array of endemic and relic species, with over 100 species protected under the Endangered Species Act (ESA). Like many rare species, these species are threatened by multiple interacting factors including climate change, competition with non-native species, herbivory by overabundant wild ungulates, and habitat fragmentation (Buchmann 1994 , Yanahan & Moore 2019 ). This region is also subjected to a host of land use pressures, such as erosion due to livestock grazing, encroachment by woody plants, and historical fire suppression (Villarreal et al., 2019 ), that may accelerate decline of rare species that depend on rapid establishment for survival (McPherson & Weltzin, 2000). The Sky Islands are mountains with mid-to-high elevation woodland communities separated by expanses of desert and arid grasslands. While among-population isolation is a natural feature of the Sky Island landscape (Yanahan & Moore, 2019 ), habitat fragmentation may have a disproportionate effect on gene flow for rare species in this already isolated habitat mosaic (Honnay & Jacquemyn, 2007). However, under prolonged isolation, one possible outcome is the evolution of self-compatibility—and the recurrent self-fertilization it permits—if populations purge enough partially recessive deleterious alleles, potentially rendering them more resilient to inbreeding depression (Byers & Waller, 1999 ; Busch, 2005 , Frye & Neel, 2017 ). Pectis imberbis (Gray) is a perennial herb in the Asteraceae family endemic to the Arizona Sky Island ecoregion. Pectis imberbis was granted status as an endangered species on June 15th, 2021 under the ESA, due to the putative extirpation of 9 populations since the 1970s and the restricted range of this species (Crawford, 2023 ). The largest populations occur in and around the Coronado National Memorial (CNM), located in the Huachuca Mountain Range, with an array of smaller, isolated populations occurring across other Sky Islands, including the Baboquivari, Sierrita, Santa Rita, Rincon, Patagonia, Whetstone, and Dragoon Mountain ranges (Wilson, 2024 ) (Fig. 1 a /b) . Populations in and near the CNM demonstrate population growth above replacement levels, while populations peripheral to the Memorial are in decline (Souther et al. 2025 ). Known extrinsic threats include competition from introduced species, grazing, road maintenance, and mining operations, each posing localized risks that could drive small populations toward extirpation (Crawford, 2023 , Souther et al. 2025 ). These disturbances likely interfere with critical reproductive processes, including flowering, fruiting, seed dispersal, and pollinator visitation. Furthermore, the species' inherent rarity may compound these threats by making the small yellow flowers less conspicuous to pollinators, thus reducing visitation frequency and successful cross-pollination events, and potentially increasing dependence on autogamy (Courchamp et al. 1999 ; Souther et al. 2022 ). Consequently, clarifying patterns and effectiveness of pollinator visitation across populations is essential for identifying vulnerabilities and informing targeted conservation management (Souther et al. 2025 ). Extrinsic disturbances and rarity-driven pollinator limitation can ultimately intensify genetic risks, including reduced gene flow and increased potential for inbreeding depression (Aguilar et al., 2008 ). Gene flow in plants relies heavily on pollinator visitation and seed dispersal, processes that facilitate genetic connectivity among fragmented populations (Kwak et al., 1998 ). Estimating these genetic exchanges is therefore critical for determining appropriate conservation interventions aimed at maintaining connectivity and genetic variation (Liu et al., 2015). Reduced connectivity is frequently linked to declines in overall genetic diversity—such as lower expected heterozygosity and nucleotide diversity—which may lead to diminished fitness and reproductive viability (Gentili et al. 2018, Mhemmed et al. 2008 ; Wright, 1965 ). Genetic bottlenecks can often accompany landscape fragmentation, amplifying Allee effects and further compromising the resilience of rare plant populations (Hannah, 2022 ; Souther et al. 2022 ). Here, we use SNP markers to estimate effective migration surfaces to delineate population structure and gene flow across populations of an endangered plant, P. imberbis , endemic to the Arizona Sky Island ecoregion. To contextualize migration surfaces, we relate findings to pollinator observations as well as to recently published demographic performance data (Souther et al. 2025 ). To our knowledge, this is the first genetic analysis of this species at the population level. Based on the natural isolation imposed by the Sky Island landscape, the small and declining population sizes in peripheral areas of P. imberbis ’s range, and the ecological pressures of habitat fragmentation, we hypothesize that gene flow will be heterogeneously restricted among small, more isolated populations. We expect reduced connectivity will be reflected in pronounced population genetic structure, with Sky Island ranges acting as partial barriers to gene migration. Such uneven gene flow patterns may in turn influence population persistence via interactions with pollinator communities and demographic performance, ultimately informing targeted conservation interventions. This study addresses a critical knowledge gap identified by the Species Recovery Report, which highlights the need to understand patterns of genetic variation within and among populations of P. imberbis (Crawford 2023 , Souther et al. 2025 ). To contribute to the conservation and management goals of this species, our objectives for this study were to 1) delineate the population genetic structure of P. imberbis through SNP markers; 2) determine potential barriers to gene flow through regions of reduced migration; and 3) identify key pollinators for the species and elaborate on their role in the exchange of genetic material for P. imberbis . Methods DNA Extraction and RADseq. In summer 2022, we collected foliar samples from 281 individuals across 32 P. imberbis subpopulations representing its known range in this region, as determined in three seasons of previous demographic surveys (Souther et al. 2022 , Souther et al. 2025 ). We defined subpopulations as clusters of plants at least 150 meters from conspecifics or separated by other significant natural or manmade barriers. We randomly selected up to 10 fresh leaves per individual, which were then desiccated using silicon beads in paper envelopes and frozen at -80° C until DNA extraction. Some sample locations (i.e. Peña Blanca) maintained very few individuals, thus not every site could be sampled equally (up to 10 individuals) without the potential risk of damaging fitness at small, threatened populations. We sourced the sampled individuals from sites in proximity to three different sky-island mountain ranges (the Huachuca Mountains, Santa Rita Mountains, and Atascosa-Pajarito Mountains), and recorded sample coordinates for each population cluster. Specifically, we extensively sampled P. imberbis at the following six sites: Coronado National Memorial (CNM), Anne Tank Wash (ATW), Scotia Canyon (SCO), O’Donnell Canyon (ODC), Peña Blanca (PBC), and the Santa Ritas, which consists of two subpopulations, Wasp Canyon and McCleary Canyon (WSP & MCC) ( Fig. 1 , SI Table I). We then transferred the dried leaf material samples in Qiagen Extraction tubes (Qiagen, Valencia, CA, USA) and shipped to the University of Minnesota Genomics facility for DNA extraction and Next-Generation genomic sequencing. Following DNA extraction, the sequencing facility created 2 pools of inline barcoded SbfI RADseq libraries, with samples grouped by subpopulation but randomly divided between pools. Each pool was sequenced on a separate lane of a NovaSeq S1 platform (Illumina Inc., UK) using a 1 × 100 bp single-end run, generating 91 bp reads. Genome size was estimated at ~ 13 Gb based on flow cytometry conducted by Ag-Biotech (Ag-Biotech, Inc., Colton, CA, USA), and assuming the rare-cutting SbfI enzyme (8 bp recognition site), we expected to recover ~ 200,000 RAD loci. This design targeted ~ 26× per-sample coverage. Pool 1 yielded ~ 536 million reads across 138 samples (~ 3.88 million per sample), and Pool 2 generated ~ 732 million reads across 144 samples (~ 5.08 million per sample), closely aligning with the expected sequencing depth. Quality control using FastQC v0.11.7 (Andrews, 2010 ) within the UMGC’s Gopher-pipelines v2.4 confirmed high-quality sequencing (Phred scores ≥ 30), with adapter content ranging from 2–8%. Notably, low-yield DNA extraction and differences in sequencing efficiency between lanes introduced a strong lane effect: Pool 1 exhibited higher missing data due to lower sequencing depth. To maximize statistical power and data quality, only Pool 2 (n = 141) was retained for downstream analysis due to its lower percentage of missing data. After applying a 40% missing data threshold per individual, three additional samples were removed, resulting in a final dataset of 137 individuals representing 17 subpopulations nested within six broader populations ( SI Table I ). RAD tags and SNP calling. We analyzed FASTQ files containing RAD tags using the Stacks v2.66 de novo pipeline (Catchen et al. 2013). We used the denovo_map.pl wrapper to assemble orthologous tags into stacks, generate a catalog of putative RAD loci, transpose the data by locus (Dang et al. 2022), and call single nucleotide polymorphisms (SNPs). We set the minimum depth of coverage required to form a stack (m), the maximum distance between stacks within an individual (M), and the number of mismatches allowed between stacks when building the catalog (n) each to 2, following common parameter choices for moderate-coverage datasets (Paris et al. 2017 ). Following catalog assembly, we applied additional filtering to retain high confidence SNPs for downstream analyses. Using VCFtools v0.1.16 (Danecek et al., 2011), we first removed low-quality individuals and retained only biallelic SNPs present in at least 40% of individuals, applying a minor allele frequency (MAF) threshold of 0.05 (5%) to exclude rare variants (Nazareno & Knowles, 2021). Individuals with greater than 70% missing data were excluded based on visual inspection of a histogram of missing genotype rates, which showed relatively few samples exceeded this threshold (Faske et al. 2021 ). This filtering resulted in 523 SNPs across 138 individuals. To minimize linkage among markers from the same RAD locus, we applied a thinning filter (--thin 91) based on our 91 bp RAD-tag length, retaining only one SNP per locus (Danecek et al., 2011). The final dataset included 337 SNPs, which were converted to PLINK-format .012 matrices for downstream analysis. Genetic Diversity and Population Structure. To assess between-population genetic differentiation, we performed principal component analysis (PCA) to summarize SNP variation and visualize population structure across Sky Island regions (Jombart et al. 2010; Faske et al. 2021 ). We applied a K-means clustering algorithm using the kmeans function in the R stats package (v3.6.2) to identify genetic clusters (Truelove et al. 2015 ). We assigned three centroids based on minimum Bayesian information criterion (BIC) to the mapped populations to investigate population structure (Hartigan & Wong, 1979 ; Milano et al. 2020 ). We performed a dispersion analysis to test for significant differences in dispersion values for three K-means clusters in PCA space. After assigning groups and testing for dispersion, we conducted a permanova test with the ‘ adonis2 ’ function in the Vegan package (Evans et al. 2023) (permutations = 10,000) to test for significant differences between three genetic clusters determined from K-means. To further determine genetic distances between populations at a higher resolution, we calculated pairwise genetic differentiation (F ST and Nei’s Diversity) among 17 subpopulations and 6 populations using the hierfstat R package (Goudet & Jombart, 2020 ; Faske et al. 2021 ). To compare F ST to another estimate of between-population genetic diversity, we calculated Nei’s D estimates of genetic diversity using custom code following Faske et al. ( 2021 ). Finally, we estimated migration of alleles between populations using Wright’s island model of migration, using F ST as an estimate parameter (Wright 1951, Slatkin, 1985). For within-population genetic diversity, we used expected heterozygosity (Hₑ) and nucleotide diversity (π) as our primary metrics. Subpopulation Hₑ was computed with custom R scripts (Faske et al. 2021 ), incorporating per-locus sample-size weighting and Nei’s unbiased correction to accommodate uneven sample sizes (2–10 individuals) and variable missing data (Sopniewski & Catullo 2024 ). Nucleotide diversity (π) and gene flow (Nₘ) were then estimated in Stacks v2.0 (Catchen et al. 2013) under an 80% per-population locus-presence filter and minor-allele-frequency ≥ 0.05. Finally, Hₑ and π were summarized at the population level to produce diversity estimates that more accurately reflect true biological variation rather than artifacts of sampling bias (Khatri & Burt 2019 ). Estimated Effective Migration Surfaces across Isolated Populations We used SNP genetic distances and geographical distances across populations to estimate effective migration surfaces across the region of study (EEMS; Petkova et al. 2016). For assessing potential barriers to gene flow between sky islands and into Mexico, we included all 137 individuals and the associated high-quality SNPs that were filtered for population genetic analyses to visualize similarities between population structure and estimated gene flow. The 337 high-quality SNPs extracted for the population structure analyses were used to calculate a genetic distance matrix with the bed2diff_v1 program (Petkova et al. 2016). Demes were defined across a habitat grid that extended beyond the sampled populations, including a buffer around the species' range to reduce edge effects during the Markov chain simulations and to explore potential migration surfaces extending into northern Mexico, where the species’ distribution and genetic diversity remain poorly understood. To assess the robustness of inferred migration patterns, we ran the EEMS algorithm ( runeems_snps ; Petkova et al. 2016) using both 300 and 400 demes. This allowed us to evaluate the sensitivity of spatial inferences to deme resolution, as shown in EEMS documentation and prior studies (Jones et al. 2021 ; Li et al. 2020; Petkova et al. 2016). Each run used 1 million burn-in iterations and 12 million sampling iterations to ensure convergence toward a stationary distribution of migration rates. We inspected the results after multiple runs for convergence success after each run (Herman et al.2022). We then used the reemsplots2 package in R to visualize the results of the program, plotting effective migration rates ( m) and effective genetic diversity rates ( q) on a log 10 scale after inspection of correlation between observed versus expected genetic dissimilarity (Jones et al. 2021 ). Pollination network and breeding system To identify primary pollinators and assess the breeding system of P. imberbis , we conducted insect visitation surveys and pollinator exclusion trials between August and October from 2019–2023. Surveys were conducted at three subpopulation sites: Coronado National Memorial (CNM Visitor Center Maintenance Shed Front), Anne Tank Wash (Upper ATW), and Scotia Canyon Populations from August to October, 2019–2023. Detailed methodological protocols for both pollinator observations and mesh bag exclusion experiments are published in Souther et al. ( 2025 ). In brief, insect visitors were observed during timed sampling sessions and categorized into functional groups based on citizen science categories (Ullmann et al., 2011). To distinguish likely pollinators from incidental visitors, only functional groups with voucher specimens confirmed to transport pollen via fuchsin gel staining (Kearns & Inouye, 1993 ) were retained for analysis. Here, we analyzed functional group importance across sites using a combination of visitation frequency and observed resource use behaviors, such as legitimate pollen collection, or nectar robbing, in which the visitor does not appear to contact reproductive parts of the flower when feeding. Importance values were used to assess variation in pollinator community structure among sites via variance analysis and visualized in a weighted bipartite network, with scaled importance values serving as edge weights linking taxa and sites (Castillo et al. 2024). Results RADseq Determination of Population Structure Reduced dimensionality of SNP frequencies across individual samples revealed evidence of population structure in P. imberbis ( Fig. 2 ) . Principal component analysis of 137 individuals and 337 quality-filtered SNPs indicated three genetically distinct clusters: (1) Coronado National Memorial, (2) the western edge of the Huachuca Mountains (with Peña Blanca as an outlier), and (3) the Santa Rita Mountains (Wasp Canyon and McCleary Canyon). K-means clustering on the first two principal components supported these groupings (K = 3). A permanova confirmed significant differences among clusters (F₍2,135₎ = 98.72, R² = 0.594, p < 0.001), and analysis of multivariate homogeneity of dispersions indicated no significant differences among group variances, validating assumptions for the test. A PCA highlighting population means further showed distinct separation, particularly between Coronado and Santa Rita groups ( Fig. 2 ; SI Fig. III). For population genetic diversity estimates, pairwise F ST values for 6 aggregated populations ranged from 0.077 to 0.421 and Nei’s diversity coefficient ranged from 0.026 to 0.343 (SI Fig. II) . The global F ST value for all pairwise groups was 0.21. Gene flow coefficients ranged from 0.343 to 2.978, with a mean of 1.21 (SI Table II, SI Table III) . For within population genetic diversity estimates for 17 subpopulations, expected heterozygosity ranged from 0.09 to 0.25, with a global He of 0.20. Nucleotide diversity (π) for 17 subpopulations ranged from 0.04 to 0.10, with a mean of 0.07 ( SI Table IV ). Nucleotide diversity was highest in Scotia Canyon followed by ATW and O’Donnell Canyon (SI Table IV). Expected heterozygosity was highest in O’Donnell Canyon (SI Table IV). For populations in which pollinator observations were conducted, gene flow was highest between CNM and ATW (Nm = 2.978), whereas Scotia Canyon had lower values of 0.926 between ATW and 0.903 between CNM (SI Table V) . Migration and Gene Flow Between Populations The EEMS resulted in positive relationships between expected and observed genetic dissimilarity in simulations with both 300 and 400 predefined demes, with runs of 400 demes resulting in higher R 2 values for dissimilarities between pairs of sampled demes, indicating a stronger fit for observed versus predicted values for the 400 deme models (Jones et al. 2021 , Petkova et al. 2016) ( SI Fig. I) . MCMC simulations of migration rates and genetic diversity through EEMS mirrored the geography of the region and demographic patterns of the species, with low effective migration surfaces coinciding with Sky Island edges (Fig. 3 ., Fig. 4 ) . Specifically, migration resistance was evident on the western edge of the Huachuca Mountains and through the Patagonia Mountains. The pattern of relatively low effective migration coincided with locations of populations thought to be extirpated ( Fig. 3 ) , which suggests that this region is less conducive to dispersal and establishment. Scotia and O'Donnell Canyon sites, where populations are declining, appeared in the center of the low migration pathway. Simulations with the courser 300 deme parameter showed greater migration connectivity between Coronado National Memorial and the neighboring Anne Tank Wash populations, but increasing the deme count to 400 showed a distinct reduction a log(m) coinciding with the geographic feature of Montezuma pass ( Fig. 3 , SI Fig. I). Genetic diversity calculations from EEMS showed a distinct limitation of Q diversity rates in the most distant, isolated population of Peña Blanca. Populations across the Huachuca Mountains to the Santa Rita Mountains showed similar rates of diversity, with Santa Rita and O’Donnell Canyon populations demonstrating higher diversity than the larger populations of Anne Tank Wash and CNM ( Fig. 4 a ). Pollination network and breeding system Importance values used to parameterize the bipartite network were derived from 361 ten-minute observations of flower visitation across populations and subpopulations distributed across Coronado National Memorial, Scotia Canyon, and Anne Tank Wash. Observers detected 226 legitimate flower visits by a total of 17 visitor taxa/functional groups (Fig. 5 ) . Visitors common to all three locations included Megachilid bees and Bombyliid flies (Fig. 5 ) , the two most frequent and most important visitor groups recorded overall for P. imberbis (Souther et al. 2025 ). Other important visitors occurred at two of the three sites (e.g., Steniolia sp. wasps were recorded at Scotia and CNM; syrphid flies were recorded at Anne Tank Wash and CNM; and Halictidae bees were recorded at Scotia and Anne Tank Wash) (Fig. 5 ) . Such commonality of visitors suggests that the pollinator community is at least somewhat consistent across the landscape, which may have important implications for pollen transfer among disconnected plant populations. Discussion Estimating population structure and gene flow can aid in conservation efforts of small, isolated populations of rare plants by identifying barriers to the exchange of genetic material across the landscape (Laikre et al.2010, Zhou et al.2023). Overall, the results of this analysis indicate distinct population structure among subpopulations of P. imberbis , suggesting that distances between populations (90 km) is enough to hinder gene flow. Moreover, estimated migration surfaces indicate restricted gene flow patterns align with both landscape features and demographic processes observed in other studies (Souther et al. 2022 , Souther et al. 2025 ). Specifically, mountainous barriers of the Patagonia Mountain range appear to overlap with predictive models of low migration surfaces, which do not take environmental variability into account (Petkova et al., 2016). This pattern, in part, could be explained by prevailing wind patterns. Pectis imberbis seeds and pollen are dispersed by wind, thus directionality of wind patterns influence genetic structure (Kling & Ackerly, 2021 ). Moreover, wind conditions alter pollinator foraging patterns and flight efficiency, thus shaping the efficacy and directionality of pollen transfer (Burnett et al. 2021 , Wang et al. 2016 ). The predominant southwesterly winds that occur during P. imberbis flower and seed production (July - October) may prevent seeds and pollen produced by CNM populations from moving west over the Huachuca mountains into the historical range of the species. The developed area of Sierra Vista is located to the east of CNM and the recently constructed border wall to the south, thus suitable habitat for the establishment of new populations in these directions is limited. While population growth rate analyses indicate positive growth of populations located in and in proximity to the CNM, isolated populations on the range periphery are in decline (Souther et al. 2025 ) This pattern is likely in part explained by underlying habitat suitability, but also may reflect the consequences of fragmentation and genetic isolation – when population crashes occur there is limited opportunity for recolonization or supplementation of genetic diversity (Ellstrand & Elam, 1993). Patterns of genetic diversity and estimated effective migration Overall, F ST values and gene flow coefficients indicate high levels of genetic differentiation across populations. F ST values ranged from 0.077 to 0.421, which can be considered moderate to very high genetic variation (Hartl & Clark, 1997). Across all pairwise groups, the global F ST value was 0.21, suggesting high genetic variation among the sampled populations of P. imberbis (SI Table III) . For comparison, a study on the speciation of an endemic Hawaiian plant found an F ST of 0.57 for inter-island populations, and 0.027 for two populations on Maui (Filatov & Burke, 2004 ). One study of a range-restricted plant endemic to the Great Basin Desert found a global F ST of 0.158, which was considered high (Borokini et al. 2021 ). These findings highlight that P. imberbis exhibits notable genetic differentiation across its populations, with F ST values that are moderate to very high relative to other species found in island chains or that are characterized by rarity or range restriction. These F ST values underscore the fragmented nature of P. imberbis habitat, shaped by both natural barriers inherent to the Sky Island landscape and anthropogenic habitat degradation (Yanahan & Moore, 2019 ). The CNM, which contains the largest population of P. imberbis , was found to be moderately genetically distinct from the nearby Anne Tank Wash population, despite being only 4 kilometers apart. These two populations exhibited moderately high genetic differentiation, with the lowest pairwise F ST value between them being 0.077 ( SI Table III ). Although they are the largest populations, they are separated by Montezuma Pass (400 m), which may act as a partial barrier to gene flow. The EEMS analysis with a finer scale (400 demes) showed reduced migration rates near Montezuma Pass, while a coarser model (300 demes) did not capture this potential barrier as clearly ( Fig. 4 , SI Fig. I) . The population sampled from Peña Blanca in the Atascosa-Pajarito Mountains had a small sample size (n = 4), which likely limited its distinct geographic isolation in the PCA ( Fig. 2 ) . However, it still showed relatively high F ST values, indicating genetic dissimilarity. Both Peña Blanca and Scotia Canyon were the most genetically isolated populations, with mean F ST values of 0.29 and 0.27, respectively. The greatest genetic separation was observed between these two populations, with a pairwise FST value of 0.421 and a low gene flow coefficient of 0.343 (SI Table V) . Analysis of estimated effective migration mirrored the patterns of population structure observed in the PCA of SNP frequencies in which CNM populations were shown to be more connected via migration to Anne Tank Wash, Santa Rita, and Peña Blanca populations than to the nearby Scotia Canyon and O’Donnell Canyon groups ( Fig. 2 , Fig. 3 ) . The relatively higher rates of gene flow from the southern Huachuca populations to northern and western populations is possibly influenced by a variety of factors aside from geographic distance alone. EEMS demonstrates a path of reduced migration from the southwestern edge of the Huachuca range up to the western Santa Rita range. A combination of disturbance factors, climate influences, wind patterns, and the self-crossing ability of P. imberbis may contribute to the population level partitioning of genetic variation (Aguilar et al., 2008 ; Souther et al., 2025 ; Yanahan & Moore, 2019 ). Intriguingly, gene flow patterns may in part explain the low λ-values observed for the Scotia Canyon population, which is in close proximity to high performing CNM and Anne Tank populations (Souther et al., 2025 ). Both Scotia Canyon and O’Donnell canyon demonstrated the lowest population growth rates (0.92 and 0.74, respectively) (Souther et al., 2025 : table 1)., and are in the gene flow barrier zone as demonstrated by the EEMS surface. Again, isolation can influence demographic patterns in several key ways: isolated populations are more frequently subjected to Allee effects, inbreeding depression, and reduced genetic diversity (Ellstrand & Elam, 1993). Furthermore, following population crashes or loss due to demographic stochasticity, isolated populations are less likely to experience genetic or demographic rescue from other populations (Bontrager & Angert, 2019 , Jones et al. 2021 ). These population structure data, paired alongside the spatial context of estimated effective migration and evidence of the extirpation of multiple populations in the Patagonia Mountains area (Fig. 3 ) , suggest that diminished connectivity between the Huachuca and Atascosa populations (via the Patagonia Mountains) is perhaps not a recent development but may have been exacerbated by landscape changes over the past century (Love et al., 2023 ). Specifically, all populations apart from CNM subpopulations exist on grazing allotments within Coronado National Forest. Historic grazing in the vicinity of nearly all monitored populations is suspected to be a primary driver of initial decline of P. imberbis , due to direct impact of herbivory as well as indirect effects, such as erosion, soil degradation, and proliferation of non-native species that accompany overgrazing (McPherson & Weltzin, 2000). Demographic monitoring suggests that grazing cessation may have contributed to the recovery of some populations, but herbivory from native ungulates has also been shown to negatively impact demographic vital rates in CNM (Crawford 2023 , Souther et al. 2022 , Souther et al. 2025 ). At the same time, opportunities for transport by wild ungulates may be one mechanism of dispersal that has allowed for the exchange of genetic material between the two populations over the geographic barrier of Montezuma Pass, which separates CNM subpopulations from ATW subpopulations on the other side of the Huachuca range ( Fig. 3 ) . Relatively low F ST values and high Nm values for CNM and ATW suggest that these two populations may be the greatest source of genetic variation for this species and could potentially be the primary contributors to gene flow to other extant populations. On the other hand, EEMS suggest the migration of alleles is limited outside of the Southern Huachuca populations and that future gene flow may be hindered by the prevalence of disturbed habitat surrounding the Huachucas. Moreover, historically observed populations in Mexico have not been relocated by researchers in the past decade. The recently constructed border wall, in conjunction with more extensive grazing impacts just south of the border near CNM, are likely limiting migration and gene flow with southern populations. At the northern edge of the species range, on the other hand, the Santa Rita Mountains are under threat from copper mining activities. The populations located there represent the northernmost edge of the range and exhibit high levels of genetic variation from other populations (F ST = 0.18), and combined subpopulation sizes total less than 100 individuals (SI Table I, SI Table III). This population has anecdotally been observed to have higher seed set and more robust phenotypes (larger and more stems), thus the extirpation of this population could constitute a reduction in genetic diversity and seed availability at the leading edge of the species range (Souther et al. 2025 ). Floral visitors and Movement of Genetic Material The gene flow coefficients observed here indicate that gene flow between groups is medium to high according to Wright’s definitions (Wright 1978), but EEMS surfaces provide more context to these values by highlighting a large region of restricted gene flow on the western edge of the Huachuca Mountains in which migration rates (m) range from 0.1–0.3 on a log scale ( Fig. 3 ) . The low migration rates and restricted gene flow overlaps with relatively undisturbed National Forest land, making it a potentially suitable region to increase genetic connectivity (via pollen transport) through the establishment of new populations (Kwak et al., 1998 ). While mining, grazing, and geographic barriers all serve as extrinsic threats to rare plants like P. imberbis , the species also faces intrinsic stressors due to its inherently small population sizes (Luijten et al. 2000 ). These stressors include potential genetic bottlenecks and factors like the Allee effect, where limited pollen availability for floral visitors hampers growth within groups of plants maintaining low population densities (Hannah, 2022 ; Souther et al. 2022 ). An adaptive strategy observed in P. imberbis to counteract this effect is the formation of dense clustering patterns on steep slopes, which may enhance floral visibility and facilitate pollen transfer amongst individuals, potentially increasing genetic diversity at the population level (Birzu et al. 2019 ; Chi & Molano-Flores, 2014; Dibner et al. 2019 ). At the same time population growth rates have been shown to be lower on slopes (Souther et al. 2022 , Souther et al. 2025 ), perhaps suggesting slopes serve as a refugia from herbivory or competition, though seedling survival may be lower overall due to higher rates of soil erosion. While clustering and persistence on less crowded slopes may be an advantage at the microhabitat scale, this clustering also makes populations particularly vulnerable to disturbance events such as fires, which could rapidly eliminate entire clusters, posing a significant risk to their survival. Despite being self-compatible—with effective population size (N e ) in selfing lineages potentially eroding at twice the rate of strict outcrossers (Pollak, 1987; Willi et al. 2020 ) — Pectis imberbis appears to rely on native bees (e.g. Megachile) and Bombyliidae flies for outcrossing (Souther et al. 2022 ). Megachile and Bombyliidae were the only two visitor taxa present at all three populations with confirmed pollen transport on voucher specimens, placing them as two groups with higher potential for the exchange of genetic material between populations. However, foraging range for solitary bees such as Megachile has been shown to be limited to 150-600m, with Megachile on the lower edge of the range (Gathmann & Tscharntke 2002 , Hoffman et al.2020). The body size of these pollinators is thought to limit their foraging distance (Greenleaf et al. 2007 ), thereby restricting gene flow between populations due to a combination of small pollinator ranges and habitat fragmentation (Gamba & Muchhala 2022, Warzecha et al. 2016 ). Observations indicate that most floral visitors to various P. imberbis populations have foraging distances of less than 1 km (Souther et al.2022; USFWS, 2023), significantly less than the global mean distance of 47.84 km between populations studied ( SI Table IV ). Consequently, if intra-population pollen transfer is severely limited as observations indicate, P. imberbis likely has greater reliance on seed dispersal for gene flow. Contribution of seed dispersal to gene flow Pectis imberbis seeds, characterized by few spreading awns under 5mm, suggest a limited capacity for wind dispersal (Fishbein and Warren, 1994 ). The extent of wind (anemochorous) versus animal (zoochorous) dispersal remains uncertain, but evidence suggests native Coues deer ( Odocoileus virginianus couesi ) may be an important vector given high levels of herbivory observed (Souther et al. 2022 ). Coues deer may act as passive transporters of P. imberbis ; however, feeding trials would be necessary to determine if viability persists following passage through Coues deer’s digestive tract. Given the species’ distribution and the current understanding of wind dispersal models, it appears that P. imberbis is primarily wind dispersed within a roughly 1 km range, with longer-distance dispersal potentially dependent on animal movement (Tackenberg et al. 2003 ). Moreover, multiple populations exhibit a pattern of stochastic clustering following the downward slope of shallow canyons or arroyos, indicating a third flood-driven (hydrochorous) means of dispersal. The ability of P. imberbis to potentially utilize a diverse array of dispersal strategies such as wind, animal interactions, and water flow highlights one aspect of the species adaptive resilience in the face of fragmented habitat. These varied dispersal mechanisms are critical in maintaining gene flow across fragmented landscapes, enabling the species to adapt to environmental changes and potentially enhance survival and reproductive success in different ecological niches (Pauls et al., 2013 ). In contrast, isolated populations are likely to continue the observed demographic decline if intervention is not undertaken. The role of gene flow in species persistence This interplay of dispersal strategies with environmental challenges emphasizes the need for integrated conservation strategies that address both intrinsic and extrinsic threats to long-term persistence of this species (Pruett et al., 2025 ). The dynamics of pollen exchange and seed dispersal are crucial for understanding the magnitude of gene flow among the peripheral populations at the northern edge of the species' range, particularly as these populations confront the uncertainties of climate change (Love et al., 2023 ). The nature of gene flow at these range edges can have mixed impacts on fitness, depending on the context; high rates of gene flow from diverse environmental conditions may either impede or enhance the ability of leading-edge populations to adapt rapidly to their local habitats (Bridle and Vines, 2007). Conversely, when considering climatic variations, natural selection might favor the survival of northern peripheral populations that maintain higher genetic diversity through effective gene flow (Bontrager & Angert, 2019 ). Overall, dispersal limitation appears to be a driver of the observed population structure of P. imberbis , whereas climate variability may reduce habitat suitability for populations located in southern or lowland regions of the species range. Demographic analyses have indicated that populations in or near CNM have positive population growth rates, while populations farther from this epicenter are in decline (Souther et al. 2025 ). Resurvey efforts, in which teams have attempted to relocate populations, reflect this finding and generally confirm that extirpations have occurred outside the CNM in the past few decades (Crawford 2023 ). Several important environmental factors have been linked to performance, including deer browse, competition with non-native species, and climate change; yet, observed demographic patterns are not fully explained by these drivers alone. One hypothesis is that populations located far from CNM are experiencing these stressors, but due to isolation, do not receive sufficient gene flow that could buffer against disturbances. Thus, demographic and genetic losses following drought events, high-browsing years, or other perturbations are not offset by immigration and gene flow. Dwindling numbers then reinforce decline via feedback loops driven by Allee effects, inbreeding depression, and genetic drift. Increasing habitat connectivity on the landscape can ameliorate these potential effects by facilitating the movement of pollinators and seed dispersing ungulates between habitat patches (Kwak et al., 1998 ; Mhemmed et al., 2008 ). Breeding system trials confirmed that P. imberbis is both self-compatible and autogamous, capable of producing achenes without pollinator visitation (Souther et al. 2022 ). While autogamy may ensure reproductive success under pollinator-limited conditions, it does not protect against inbreeding or genetic isolation, particularly in small or fragmented populations. Our population structure and EEMS analyses revealed reduced effective migration rates among several peripheral populations (e.g., Wasp Canyon and Peña Blanca), suggesting barriers to gene flow that may be shaped by both landscape features and biotic interactions. Regarding the latter, our weighted bipartite pollination network ( Fig. 5 ) highlights both overlap and variation in pollinator assemblages: while Megachilidae bees and bombyliid flies were frequent, high-importance visitors shared across all sites, other taxa were more localized. For example, Steniolia wasps were observed at Scotia and CNM but not at Anne Tank Wash, while syrphid flies and halictid bees showed inconsistent presence. CNM generally exhibited greater diversity, which may not necessarily indicate more effective pollen transfer but could reflect a habitat more conducive to a broad suite of plant-animal interactions in a wider landscape subjected to drought, fire, and invasive species. Moreover, adaptive foraging by generalist pollinators (e.g. shifting their visits toward the most specialized plant partners) can reverse the effects of network structure on community stability, and suggests that population-specific differences in visitor assemblages may drive spatially variable pollen‐mediated gene flow (Kwak et al., 1998 ; Valdovinos et al., 2016). Thus, the geographic heterogeneity in effective migration we observe may arise from both abiotic resistance and biotic filtering by the pollinator community (Stankowski et al. 2015 ). Future work integrating pollinator abundance, behavior, and pollen transport capacity with fine-scale genetic data—such as parentage or paternity analysis—could directly test the extent to which pollinator community composition shapes gene flow across the Sky Island landscape (Gigant et al. 2016 ). Implications for conservation and management We found high levels of genetic isolation of populations at the edge of P. imberbis ’s current range. A cross-population demographic study of P. imberbis found that population growth rates of these populations are below replacement levels and speculated that they are on a trajectory to extirpation, in part driven by rarity which reinforces population decline (Souther et al. 2025 ). One possible solution is to increase connectivity across populations, possibly by reintroducing populations into suitable habitat across the species’ range. Increasing connectivity could increase the likelihood of genetic and demographic rescue (Richards, 2000), critical in this highly heterogeneous environment, characterized by disturbance events like drought that can result in stochastic population loss (Love et al. 2023 ). However, it is unclear whether reestablished populations would be sufficient bolster to cross population gene flow in the continental “islands” of the Madrean archipelago. A recent investigation of suitable habitat for P. imberbis used an ensemble species distribution modelling approach to aid in future survey or restoration efforts (Wilson 2024 ). Of the suite of 43 predictor variables and 11 hectares used to train the ensemble models, the final model predictions identified 5 key predictor variables (spring precipitation, elevation, solar radiation, EPA ecoregion, and USGS/SWRG data) and a potential suitable habitat extent of 35,505 hectares. When compared to the EEMS migration surfaces produced in this study, a narrow band of suitable habitat overlaps a region of what is determined to be an area of historically limited migration. However, we note that EEMS assumes a stepping-stone, migration–drift equilibrium and constant local population sizes, so spatially correlated changes in demographic history (e.g. recent bottlenecks or expansions) can produce the same genetic discontinuities that EEMS would call ‘low migration’ (Petkova et al. 2016). Therefore, some of our inferred barriers (e.g. distances between Sky Islands) might reflect past declines in census size as much as true dispersal limits (Ravinet et al., 2017 ). Nonetheless, these areas of apparent low effective migration still emphasize populations at risk of losing diversity and so remain high-priority targets for any connectivity or rescue-based intervention. One approach to enhancing genetic connectivity could be to establish restoration garden sites along intervals of suitable habitat to allow for pollen transfer between higher diversity or larger populations, such as Scotia and Anne Tank Wash populations, respectively. An alternative approach would be to directly introduce individuals or pollen from outside populations to increase genetic diversity, but such interventions are not without risk (Willi et al., 2007). If genetic signatures reflect underlying adaptations to local conditions, introduction of non-local genotypes may result in outbreeding depression (Schierup & Christiansen, 1996 ). By defining population clusters and barriers to gene flow this study revealed broad gene flow patterns, yet it does not determine the influence of local adaptation on survival – a critical next step for conservation planning. As a counterpoint to reestablishing new populations, rapid climate change may render formerly suitable habitats maladaptive (Souther & McGraw, 2011 , Souther et al. 2012 ). Thus, interventions, such as increasing connectivity and introduction of propagules from other populations may be warranted even when adaptive differentiation is detected (Crémieux et al., 2010; Willi et al., 2007). In cases where local conditions are changing, the benefits of greater genetic diversity and population numbers outweigh the risks of outbreeding depression (Souther & McGraw 2014). Conclusions This study described patterns of gene flow and isolation of P. imberbis , an endangered plant species in the Sky Island Region. We found evidence of three distinct population clusters across the populations we sampled, but the degree to which these groups are influenced by genetic drift or local adaptation is less clear. We found a distinct reduction of estimated effective migration rates, which we used as an indicator of gene flow, from the largest population at Coronado National Memorial to all other populations. Populations in the Huachuca and Santa Rita Mountains demonstrated higher genetic diversity compared to the distant and isolated Peña Blanca population, as confirmed by nucleotide diversity, expected heterozygosity, and Q diversity from EEMS. We found pollinator visitation and diversity was higher at CNM subpopulations, suggesting a potential mechanism explaining EEMS findings though there was overlap in visitation by Megachile spp. across all populations. Ultimately, this study provides fine-scale genetic and environmental context that can inform future demographic and genetic analyses for an endangered plant. More broadly, the insights generated by this study can be applied to other endemic species in the Sky Islands, specifically when determining whether isolation should be considered a natural feature of adaptation or a threat to survival (Love et al., 2023 ). Future studies should attempt to quantify gene flow and migration patterns directly through observations of seed dispersal strategies, as well as attempt to garner higher taxonomic resolution for floral visitors to ascertain more accurate estimates of pollen transfer (Pornon et al.2017). Furthermore, tests for adaptive variation through a reciprocal common garden experiment are needed to determine the extent of local adaptation across populations, as well as to investigate the role of inbreeding depression in influencing phenotypic variation and fitness (Schierup & Christiansen, 1996 ). By understanding both the role of inbreeding depression and whether the genetic differences observed reflect local adaptations to site conditions, conservationists and land managers can determine whether introduction of non-local genotypes is warranted, as well as establish parameters around these actions that are non-harmful to focal populations. The patterns of genetic isolation and population decline exhibited by P. imberbis serve as an early warning of the threats facing other species in the Madrean Archipelago as anthropogenic change progresses and as the climate continues to shift. In this naturally fragmented habitat, as connectivity erodes, the performance and adaptive capacity of this species—and many others within these biologically rich desert islands—are at risk of collapse. Identifying conservation pathways to maintain connectivity and thus demographic and genetic resilience, is critical for preserving the diversity of this unique landscape. Declarations Competing Interests The authors have no relevant financial or non-financial interests to disclose. Funding We thank the Coronado National Memorial and the Coronado National Forest for their support of this project. This work was funded by National Park Service funding PMIS #316237 awarded to S. Souther. Pollination analyses were funded by USGS award #G21AC10137-00 to S. Souther, C. E. Aslan, and USGS collaborator L. Norman. Author Contribution SPG prepared the material, designed and performed the genetic analysis and wrote the manuscript. SS contributed to the study conception and design, secured funding, and assisted with the writing and revisions of the manuscript. CA designed the pollinator study, secured funding, and contributed to the revisions of the pollinator methodology section. TF prepared python and R scripts and contributed to the analysis and paper revisions. KH and LH advised and commented on several drafts of the manuscript. Acknowledgement We thank the Audubon Appleton Whittell Research Ranch for housing support. We thank Julie Crawford at USFWS for her support of endangered species conservation and for her development of the species recovery plan for P. imberbis. We thank Martha Sample and Morgan Andrews for their administration, logistics, and field support. We thank Matthew Weiss for the suggestions, conversations, and conceptual guidance regarding bioinformatic methodologies. Finally, we would like to thank Laura Nicholson, Mia Brann, Megan Quinn, Alex Croydon, Hanna Ryder, Alia Raderstorf, Alex Wakefield, and many others who contributed diligence and passion to the demographic surveys of the Beardless Chinchweed. Data Availability Data supporting our findings will be archived in Dryad Digital Repository, and/or are available from the corresponding author on reasonable request. References Adler, P. B., Salguero-Gomez, R., Compagnoni, A., Hsu, J. S., Ray-Mukherjee, J., Mbeau-Ache, C., et al. (2014). Functional traits explain variation in plant life history strategies. Proceedings of the National Academy of Sciences 111, 740–745. doi: 10.1073/pnas.1315179111 Aguilar, R., Quesada, M., Ashworth, L., Herrerias-Diego, Y., & Lobo, J. (2008). Genetic consequences of habitat fragmentation in plant populations: Susceptible signals in plant traits and methodological approaches. Molecular Ecology , 17 (24), 5177–5188. https://doi.org/10.1111/j.1365-294X.2008.03971.x Andrews, S. (2010). FastQC: A Quality Control Tool for High Throughput Sequence Data [Online]. http://www.bioinformatics.babraham.ac.uk/projects/fastqc/ Anic, V., C. A. Henríquez, S. R. Abades, and R. O. Bustamante. “Number of Conspecifics and Reproduction in the Invasive Plant Eschscholzia Californica (Papaveraceae): Is There a Pollinator-Mediated Allee Effect?” Plant Biology 17, no. 3 (2015): 720–27. https://doi.org/10.1111/plb.12293. Birzu, Gabriel, Sakib Matin, Oskar Hallatschek, and Kirill S. Korolev. “Genetic Drift in Range Expansions Is Very Sensitive to Density Dependence in Dispersal and Growth.” Ecology Letters 22, no. 11 (2019): 1817–27. https://doi.org/10.1111/ele.13364. Bontrager, Megan, and Amy L. Angert. “Gene Flow Improves Fitness at a Range Edge under Climate Change.” Evolution Letters 3, no. 1 (2019): 55–68. https://doi.org/10.1002/evl3.91. Borokini, Israel T., Kelly B. Klingler, and Mary M. Peacock. “Life in the Desert: The Impact of Geographic and Environmental Gradients on Genetic Diversity and Population Structure of Ivesia Webberi.” Ecology and Evolution 11, no. 23 (2021): 17537–56. https://doi.org/10.1002/ece3.8389. Boyd, J. N., Anderson, J. T., Brzyski, J., Baskauf, C., & Cruse-Sanders, J. (2022). Eco-evolutionary causes and consequences of rarity in plants: A meta-analysis. New Phytologist , 235 (3), 1272–1286. https://doi.org/10.1111/nph.18172 Bridle, Jon R., and Timothy H. Vines. “Limits to Evolution at Range Margins: When and Why Does Adaptation Fail?” Trends in Ecology & Evolution 22, no. 3 (March 2007): 140–47. https://doi.org/10.1016/j.tree.2006.11.002. Brook, B. W., Sodhi, N. S., & Bradshaw, C. J. A. (2008). Synergies among extinction drivers under global change. Trends in Ecology & Evolution , 23 (8), 453–460. https://doi.org/10.1016/j.tree.2008.03.011 Burnett, N. P., Badger, M. A., and Combes, S. A. (2021). Wind and route choice affect performance of bees flying above versus within a cluttered obstacle field. doi: 10.1101/2021.10.08.463704 Buchmann, S. L. (1994). Diversity and importance of native bees from the Arizona/Mexico Madrean archipelago. In L. F. DeBano, P. F. Ffolliott, A. Ortega-Rubio, G. J. Gottfried, R. H. Hamre, & C. B. Edminster (Eds.), Biodiversity and management of the Madrean archipelago: The sky islands of southwestern United States and northwestern Mexico (pp. 301–310). Fort Collins, CO: U. S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station Burnett, Nicholas P., Marc A. Badger, and Stacey A. Combes. “Wind and Route Choice Affect Performance of Bees Flying above versus within a Cluttered Obstacle Field.” PLOS ONE 17, no. 3 (March 24, 2022): e0265911. https://doi.org/10.1371/journal.pone.0265911. Busch, J. W. “Inbreeding Depression in Self-Incompatible and Self-Compatible Populations of Leavenworthia Alabamica.” Heredity 94, no. 2 (February 2005): 159–65. https://doi.org/10.1038/sj.hdy.6800584. Byers, D. L., and D. M. Waller. “Do Plant Populations Purge Their Genetic Load? Effects of Population Size and Mating History on Inbreeding Depression.” Annual Review of Ecology, Evolution, and Systematics 30, no. Volume 30, 1999 (November 1, 1999): 479–513. https://doi.org/10.1146/annurev.ecolsys.30.1.479. Catchen, Julian, Paul A. Hohenlohe, Susan Bassham, Angel Amores, and William A. Cresko. “Stacks: An Analysis Tool Set for Population Genomics.” Molecular Ecology 22, no. 11 (June 2013): 3124–40. https://doi.org/10.1111/mec.12354. Castillo, William J., Laura A. Burkle, and Carsten F. Dormann. “Dynamics of a Plant–Pollinator Network: Extending the Bianconi–Barabási Model.” Applied Network Science 9, no. 1 (December 2024): 1–12. https://doi.org/10.1007/s41109-024-00636-0. Courchamp, F., T. Clutton-Brock, and B. Grenfell. “Inverse Density Dependence and the Allee Effect.” Trends in Ecology & Evolution 14, no. 10 (October 1999): 405–10. https://doi.org/10.1016/s0169-5347(99)01683-3. Chi, Katherine, and Brenda Molano-Flores. “Can Floral Display Size Compensate for Allee Effects Caused by Low Population Abundance and Density in Synthyris Bullii (Plantaginaceae), a Rare Species?” American Journal of Botany 101, no. 3 (2014): 428–36. https://doi.org/10.3732/ajb.1300457. Coates, D. J., Byrne, M., & Moritz, C. (2018). Genetic Diversity and Conservation Units: Dealing With the Species-Population Continuum in the Age of Genomics. Frontiers in Ecology and Evolution , 6 . https://doi.org/10.3389/fevo.2018.00165 Crawford, Julie. (2023). U.S. Fish and Wildlife Service. 2023. Beardless Chinchweed (Pectis imberbis) Final Recovery Plan. Southwest Region, Albuquerque, NM, USA. 37 Dang, Zhenhua, Jiabin Li, Yanan Liu, Miaomiao Song, Peter J. Lockhart, Yunyun Tian, Miaomiao Niu, and Qinglang Wang. “RADseq-Based Population Genomic Analysis and Environmental Adaptation of Rare and Endangered Recretohalophyte Reaumuria Trigyna.” The Plant Genome n/a, no. n/a (n.d.): e20303. https://doi.org/10.1002/tpg2.20303. Dennis, Brian, Patricia L. Munholland, and J. Michael Scott. “Estimation of Growth and Extinction Parameters for Endangered Species.” Ecological Monographs 61, no. 2 (1991): 115–43. https://doi.org/10.2307/1943004. Dibner, Reilly R., Megan L. DeMarche, Allison M. Louthan, and Daniel F. Doak. “Multiple Mechanisms Confer Stability to Isolated Populations of a Rare Endemic Plant.” Ecological Monographs 89, no. 2 (2019): e01360. https://doi.org/10.1002/ecm.1360. Ellstrand, Norman C., and Diane R. Elam. “Population Genetic Consequences of Small Population Size: Implications for Plant Conservation.” Annual Review of Ecology, Evolution, and Systematics 24, no. Volume 24, 1993 (November 1, 1993): 217–42. https://doi.org/10.1146/annurev.es.24.110193.001245. Evans, Simone A., Dennis F. Whigham, Ida Hartvig, and Melissa K. McCormick. “Hybridization in the Fringed Orchids: An Analysis of Species Boundaries in the Face of Gene Flow.” Diversity 15, no. 3 (March 2023): 384. https://doi.org/10.3390/d15030384. Faske, Trevor M., Alison C. Agneray, Joshua P. Jahner, Lana M. Sheta, Elizabeth A. Leger, and Thomas L. Parchman. “Genomic and Common Garden Approaches Yield Complementary Results for Quantifying Environmental Drivers of Local Adaptation in Rubber Rabbitbrush, a Foundational Great Basin Shrub.” Evolutionary Applications 14, no. 12 (2021): 2881–2900. https://doi.org/10.1111/eva.13323. Filatov, D., Burke, S. DNA diversity in Hawaiian endemic plant Schiedea globosa . Heredity 92 , 452–458 (2004). https://doi.org/10.1038/sj.hdy.6800440 Fishbein, M. and P. Warren. 1994. Population studies of sensitive plants of the Lone Mountain Ecosystem Management Area, Coronado National Forest, Arizona. Frankham, R., Ballou, J. D., Eldridge, M. D. B., Lacy, R. C., Ralls, K., Dudash, M. R., & Fenster, C. B. (2011). Predicting the Probability of Outbreeding Depression. Conservation Biology , 25 (3), 465–475. https://doi.org/10.1111/j.1523-1739.2011.01662.x Frye, Christopher T., and Maile C. Neel. “Benefits of Gene Flow Are Mediated by Individual Variability in Self-Compatibility in Small Isolated Populations of an Endemic Plant Species.” Evolutionary Applications 10, no. 6 (2017): 551–62. https://doi.org/10.1111/eva.12437 Gathmann, Achim, and Teja Tscharntke. “Foraging Ranges of Solitary Bees.” Journal of Animal Ecology 71 (September 6, 2002): 757–64. https://doi.org/10.1046/j.1365-2656.2002.00641.x. Gamba, Diana, and Nathan Muchhala. “Pollinator Type Strongly Impacts Gene Flow within and among Plant Populations for Six Neotropical Species.” Ecology 104, no. 1 (2023): e3845. https://doi.org/10.1002/ecy.3845. Gentili, Rodolfo, Aldo Solari, Martin Diekmann, Cecilia Duprè, Gianna Serafina Monti, Stefano Armiraglio, Silvia Assini, and Sandra Citterio. “Genetic Differentiation, Local Adaptation and Phenotypic Plasticity in Fragmented Populations of a Rare Forest Herb.” PeerJ 6 (June 13, 2018): e4929. https://doi.org/10.7717/peerj.4929. Gigant, R. L., A. De Bruyn, T. M’sa, G. Viscardi, L. Gigord, A. Gauvin-Bialecki, T. Pailler, L. Humeau, M. Grisoni, and P. Besse. “Combining Pollination Ecology and Fine-Scale Spatial Genetic Structure Analysis to Unravel the Reproductive Strategy of an Insular Threatened Orchid.” South African Journal of Botany 105 (July 1, 2016): 25–35. https://doi.org/10.1016/j.sajb.2016.02.205. Goudet, J. and Jombart, T. (2020) Hierfstat: Estimation and Tests of Hierarchical F-Statistics. R Package Version 0.5-7. https://CRAN.R-project.org/package=hierfstat Greenleaf, Sarah S., Neal M. Williams, Rachael Winfree, and Claire Kremen. “Bee Foraging Ranges and Their Relationship to Body Size.” Oecologia 153, no. 3 (September 2007): 589–96. https://doi.org/10.1007/s00442-007-0752-9. Hannah, Lee. “Chapter 4 - Ecological, Evolutionary, and Biogeographic Implications of Climate Change.” In Climate Change Biology (Third Edition) , edited by Lee Hannah, 77–94. Academic Press, 2022. https://doi.org/10.1016/B978-0-08-102975-6.00004-2. Hamabata, T., Kinoshita, G., Kurita, K., Cao, P.-L., Ito, M., Murata, J., Komaki, Y., Isagi, Y., & Makino, T. (2019). Endangered island endemic plants have vulnerable genomes. Communications Biology , 2 (1), Article 1. Hartigan, J. A., and M. A. Wong. “Algorithm AS 136: A K-Means Clustering Algorithm.” Journal of the Royal Statistical Society. Series C (Applied Statistics) 28, no. 1 (1979): 100–108. https://doi.org/10.2307/2346830. Hartl, Daniel L., and A. G. Clark. “Principles of Population Genetics, 4 th Edition”. Journal of Heredity , Volume 98, Issue 4, July 2007, Page 382. (2007). https://doi.org/10.1093/jhered/esm035 Hedrick, P. W., & Miller, P. S. (1992). Conservation genetics: techniques and fundamentals. Ecological Applications, 2(1), 30-46 Herman, Rachael W, Benjamin M Winger, Donna L Dittmann, and Michael G Harvey. “Fine-Scale Population Genetic Structure and Barriers to Gene Flow in a Widespread Seabird (Ardenna Pacifica).” Biological Journal of the Linnean Society 137, no. 1 (September 1, 2022): 125–36. https://doi.org/10.1093/biolinnean/blac091 Hedrick, Philip W., and Philip S. Miller. “Conservation Genetics: Techniques and Fundamentals.” Ecological Applications 2, no. 1 (1992): 30–46. https://doi.org/10.2307/1941887. Hofmann, Michaela M., Andreas Fleischmann, and Susanne S. Renner. “Foraging Distances in Six Species of Solitary Bees with Body Lengths of 6 to 15 Mm, Inferred from Individual Tagging, Suggest 150 m-Rule-of-Thumb for Flower Strip Distances.” Journal of Hymenoptera Research 77 (June 29, 2020): 105–17. https://doi.org/10.3897/jhr.77.51182. Holderegger, Rolf, Niko Balkenhol, Janine Bolliger, Jan O. Engler, Felix Gugerli, Axel Hochkirch, Carsten Nowak, Gernot Segelbacher, Alex Widmer, and Frank E. Zachos. “Conservation Genetics: Linking Science with Practice.” Molecular Ecology 28, no. 17 (September 2019): 3848–56. https://doi.org/10.1111/mec.15202. Honnay, Olivier, and Hans Jacquemyn. “Susceptibility of Common and Rare Plant Species to the Genetic Consequences of Habitat Fragmentation.” Conservation Biology 21, no. 3 (2007): 823–31. https://doi.org/10.1111/j.1523-1739.2006.00646.x. Jombart, Thibaut, Sébastien Devillard, and François Balloux. “Discriminant Analysis of Principal Components: A New Method for the Analysis of Genetically Structured Populations.” BMC Genetics 11, no. 1 (October 15, 2010): 94. https://doi.org/10.1186/1471-2156-11-94. Jones, Matthew R., Daniel E. Winkler, and Rob Massatti. “Demographic Modeling Informs Functional Connectivity and Management Interventions in Graham’s Beardtongue.” Conservation Genetics 22, no. 6 (December 1, 2021): 993–1003. https://doi.org/10.1007/s10592-021-01392-9. Jump, Alistair S., and Josep Peñuelas. “Running to Stand Still: Adaptation and the Response of Plants to Rapid Climate Change.” Ecology Letters 8, no. 9 (2005): 1010–20. https://doi.org/10.1111/j.1461-0248.2005.00796.x. Kearns, C. A., & Inouye, D. W. (1993). Techniques for pollination biologists Khatri, Bhavin S., and Austin Burt. “Robust Estimation of Recent Effective Population Size from Number of Independent Origins in Soft Sweeps.” Molecular Biology and Evolution 36, no. 9 (September 1, 2019): 2040–52. https://doi.org/10.1093/molbev/msz081. Kling, M. M., and Ackerly, D. D. (2021). Global wind patterns shape genetic differentiation, asymmetric gene flow, and genetic diversity in trees. Proc. Natl. Acad. Sci. U.S.A. 118, e2017317118. doi: 10.1073/pnas.2017317118 Kwak, Manja M., Odilia Velterop, and Jelte van Andel. “Pollen and Gene Flow in Fragmented Habitats.” Applied Vegetation Science 1, no. 1 (1998): 37–54. https://doi.org/10.2307/1479084. Laikre, Linda, Fred W. Allendorf, Laurel C. Aroner, C. Scott Baker, David P. Gregovich, Michael M. Hansen, Jennifer A. Jackson, et al. “Neglect of Genetic Diversity in Implementation of the Convention of Biological Diversity.” Conservation Biology 24, no. 1 (2010): 86–88. Lande, Russell, and Susan Shannon. “The Role of Genetic Variation in Adaptation and Population Persistence in a Changing Environment.” Evolution 50, no. 1 (1996): 434–37. https://doi.org/10.2307/2410812. Li, Yao, Amol C. Shetty, Chanthap Lon, Michele Spring, David L. Saunders, Mark M. Fukuda, Tran Tinh Hien, et al. “Detecting Geospatial Patterns of Plasmodium Falciparum Parasite Migration in Cambodia Using Optimized Estimated Effective Migration Surfaces.” International Journal of Health Geographics 19, no. 1 (April 10, 2020): 13. https://doi.org/10.1186/s12942-020-00207-3. Liu, Min, Stephen G. Compton, Fo-En Peng, Jian Zhang, and Xiao-Yong Chen. “Movements of Genes between Populations: Are Pollinators More Effective at Transferring Their Own or Plant Genetic Markers?” Proceedings of the Royal Society B: Biological Sciences 282, no. 1808 (June 7, 2015): 20150290. https://doi.org/10.1098/rspb.2015.0290. Love, S. J., Schweitzer, J. A., Woolbright, S. A., & Bailey, J. K. (2023). Sky Islands Are a Global Tool for Predicting the Ecological and Evolutionary Consequences of Climate Change. Annual Review of Ecology, Evolution, and Systematics , 54 (Volume 54, 2023), 219–236. https://doi.org/10.1146/annurev-ecolsys-102221-050029 Luijten, Sheila H., Angelo Dierick, J. Gerard, B. Oostermeijer, Léon E. L. Raijmann, and Hans C. M. Den Nijs. “Population Size, Genetic Variation, and Reproductive Success in a Rapidly Declining, Self-Incompatible Perennial (Arnica Montana) in The Netherlands.” Conservation Biology 14, no. 6 (2000): 1776–87. https://doi.org/10.1111/j.1523-1739.2000.99345.x. McPherson, Guy R., and Jake F. Weltzin. “Disturbance and Climate Change in United States/Mexico Borderland Plant Communities: A State-of-the-Knowledge Review.” Gen. Tech. Rep. RMRS-GTR-50. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 24 p. 050 (2000). https://doi.org/10.2737/RMRS-GTR-50. Mhemmed, Gandour, Hessini Kamel, and Abdelly Chedly. “Does Habitat Fragmentation Reduce Genetic Diversity and Subpopulation Connectivity?” Ecography 31, no. 6 (2008): 751–56. https://doi.org/10.1111/j.1600-0587.2008.05622.x. Milano, Elizabeth R., Margaret R. Mulligan, Jon P. Rebman, and Amy G. Vandergast. “High-Throughput Sequencing Reveals Distinct Regional Genetic Structure among Remaining Populations of an Endangered Salt Marsh Plant in California.” Conservation Genetics 21, no. 3 (June 1, 2020): 547–59. https://doi.org/10.1007/s10592-020-01269-3. Nazareno, Alison G., and L. Lacey Knowles. “There Is No ‘Rule of Thumb’: Genomic Filter Settings for a Small Plant Population to Obtain Unbiased Gene Flow Estimates.” Frontiers in Plant Science 12 (October 14, 2021). https://doi.org/10.3389/fpls.2021.677009. Nic Lughadha, Eimear, Steven P. Bachman, Tarciso C. C. Leão, Félix Forest, John M. Halley, Justin Moat, Carmen Acedo, et al. “Extinction Risk and Threats to Plants and Fungi.” PLANTS, PEOPLE, PLANET 2, no. 5 (2020): 389–408. https://doi.org/10.1002/ppp3.10146. Nordstrom, S. W., Hufbauer, R. A., Olazcuaga, L., Durkee, L. F., & Melbourne, B. A. (2023). How density dependence, genetic erosion and the extinction vortex impact evolutionary rescue. Proceedings of the Royal Society B: Biological Sciences , 290 (2011), 20231228. https://doi.org/10.1098/rspb.2023.1228 Paris, Josephine R., Jamie R. Stevens, and Julian M. Catchen. “Lost in Parameter Space: A Road Map for Stacks.” Methods in Ecology and Evolution 8, no. 10 (2017): 1360–73. https://doi.org/10.1111/2041-210X.12775. Pauls, S. U., Nowak, C., Bálint, M., & Pfenninger, M. (2013). The impact of global climate change on genetic diversity within populations and species. Molecular Ecology , 22 (4), 925–946. https://doi.org/10.1111/mec.12152 Petkova, Desislava, John Novembre, and Matthew Stephens. “Visualizing Spatial Population Structure with Estimated Effective Migration Surfaces.” Nature Genetics 48, no. 1 (January 2016): 94–100. https://doi.org/10.1038/ng.3464. Pornon, André, Christophe Andalo, Monique Burrus, and Nathalie Escaravage. “DNA Metabarcoding Data Unveils Invisible Pollination Networks.” Scientific Reports 7, no. 1 (December 4, 2017): 16828. https://doi.org/10.1038/s41598-017-16785-5. Pruett, C. L., Stroupe, J., & Peterson, C. L. (2025). Habitat fragmentation influences the population genetics of a Florida endemic: Implications for recovery of a critically endangered plant. Conservation Genetics . https://doi.org/10.1007/s10592-025-01708-z Ravinet, M., Faria, R., Butlin, R. K., Galindo, J., Bierne, N., Rafajlović, M., Noor, M. A. F., Mehlig, B., & Westram, A. M. (2017). Interpreting the genomic landscape of speciation: A road map for finding barriers to gene flow. Journal of Evolutionary Biology , 30 (8), 1450–1477. https://doi.org/10.1111/jeb.13047 Richardson, Andrew D., Trevor F. Keenan, Mirco Migliavacca, Youngryel Ryu, Oliver Sonnentag, and Michael Toomey. “Climate Change, Phenology, and Phenological Control of Vegetation Feedbacks to the Climate System.” Agricultural and Forest Meteorology 169 (February 15, 2013): 156–73. https://doi.org/10.1016/j.agrformet.2012.09.012. Schierup, M. H., & Christiansen, F. B. (1996). Inbreeding depression and outbreeding depression in plants. Heredity , 77 (5), 461–468. https://doi.org/10.1038/hdy.1996.172 Segelbacher, G., Cushman, S. A., Epperson, B. K., Fortin, M. J., Francois, O., Hardy, O. J., ... & Manel, S. (2010). Applications of landscape genetics in conservation biology: concepts and challenges. Conservation genetics, 11, 375-385. Slatkin, Montgomery, and Naoyuki Takahata. “The Average Frequency of Private Alleles in a Partially Isolated Population.” Theoretical Population Biology 28, no. 3 (December 1, 1985): 314–31. https://doi.org/10.1016/0040-5809(85)90032-2. Sopniewski, Jarrod, and Renee A. Catullo. “Estimates of Heterozygosity from Single Nucleotide Polymorphism Markers Are Context-Dependent and Often Wrong.” Molecular Ecology Resources 24, no. 4 (2024): e13947. https://doi.org/10.1111/1755-0998.13947 Souther, S., Lechowicz, M. J., & McGraw, J. B. (2012). Experimental test for adaptive differentiation of ginseng populations reveals complex response to temperature. Annals of Botany , 110 (4), 829-837. Souther, S., & McGraw, J. B. (2011). Evidence of local adaptation in the demographic response of American ginseng to interannual temperature variation. Conservation Biology , 25 (5), 922-931. Souther, Sara, James B. McGraw, John D. Souther, and Donald M. Waller. "Effects of altered climates on American ginseng population dynamics." Population Ecology 64, no. 1 (2022): 47-63. Souther, Sara, Martha Sample, Genevieve Conley, and Clare Aslan. “Demographic Analysis of the Endangered Plant Pectis Imberbis Highlights Tradeoffs between Deer Browse and Interspecific Competition.” Natural Areas Journal 42, no. 3 (July 2022): 230–41. https://doi.org/10.3375/22-3. Souther, S., Sample, M., and Aslan, C. (2025). Interacting stressors drive landscape variation in demographic response of the endangered plant, Pectis imberbis (Gray). Journal of Arid Environments in press. Stankowski, S., Sobel, J. M., & Streisfeld, M. A. (2015). The geography of divergence with gene flow facilitates multitrait adaptation and the evolution of pollinator isolation in Mimulus aurantiacus. Evolution , 69 (12), 3054–3068. https://doi.org/10.1111/evo.12807 Stephens, P. A., W. J. Sutherland, and R. P. Freckleton. 1999. What is the Allee effect? Oikos 87 : 185–190. Tackenberg, Oliver, Peter Poschlod, and Susanne Bonn. “Assessment of Wind Dispersal Potential in Plant Species.” Ecological Monographs 73, no. 2 (2003): 191–205. https://doi.org/10.1890/0012-9615(2003)073[0191:AOWDPI]2.0.CO;2. Truelove, Nathan K., Kim Ley-Cooper, Iris Segura-García, Patricia Briones-Fourzán, Enrique Lozano-Álvarez, Bruce F. Phillips, Stephen J. Box, and Richard F. Preziosi. “Genetic Analysis Reveals Temporal Population Structure in Caribbean Spiny Lobster ( Panulirus Argus ) within Marine Protected Areas in Mexico.” Fisheries Research 172 (December 1, 2015): 44–49. https://doi.org/10.1016/j.fishres.2015.05.029. Villarreal, M. L., Haire, S. L., Iniguez, J. M., Cortés Montaño, C., & Poitras, T. B. (2019). Distant neighbors: Recent wildfire patterns of the Madrean Sky Islands of southwestern United States and northwestern Mexico. Fire Ecology , 15 (1), 2. https://doi.org/10.1186/s42408-018-0012-x Warzecha, Daniela, Tim Diekötter, Volkmar Wolters, and Frank Jauker. “Intraspecific Body Size Increases with Habitat Fragmentation in Wild Bee Pollinators.” Landscape Ecology 31, no. 7 (September 1, 2016): 1449–55. https://doi.org/10.1007/s10980-016-0349-y. Wang, Z.-F., Lian, J.-Y., Ye, W.-H., Cao, H.-L., Zhang, Q.-M., and Wang, Z.-M. (2016). Pollen and seed flow under different predominant winds in wind-pollinated and wind-dispersed species Engelhardia roxburghiana. Tree Genetics & Genomes 12, 19. doi: 10.1007/s11295-016-0973-3 Willi, Yvonne, Marco Fracassetti, Olivier Bachmann, and Josh Van Buskirk. “Demographic Processes Linked to Genetic Diversity and Positive Selection across a Species’ Range.” Plant Communications 1, no. 6 (November 2020): 100111. https://doi.org/10.1016/j.xplc.2020.100111. Wilson, N.R., 2024, Species Distribution Models for Pectis imberbis, a Rare Plant Species in Southeastern Arizona: U.S. Geological Survey data release, https://doi.org/10.5066/P13VMRBC. Wright, S. “The Genetical Structure of Populations.” Annals of Eugenics 15, no. 4 (March 1951): 323–54. https://doi.org/10.1111/j.1469-1809.1949.tb02451.x. Wright, Sewall. “The Interpretation of Population Structure by F-Statistics with Special Regard to Systems of Mating.” Evolution 19, no. 3 (1965): 395–420. https://doi.org/10.2307/2406450. Yanahan, A. D., & Moore, W. (2019). Impacts of 21st-century climate change on montane habitat in the Madrean Sky Island Archipelago. Diversity and Distributions , 25 (10), 1625–1638. https://doi.org/10.1111/ddi.12965 Zhou, Chengchuan, Shiqi Xia, Qiang Wen, Ying Song, Quanquan Jia, Tian Wang, Liting Liu, and Tianlin Ouyang. “Genetic Structure of an Endangered Species Ormosia Henryi in Southern China, and Implications for Conservation.” BMC Plant Biology 23, no. 1 (April 26, 2023): 220. https://doi.org/10.1186/s12870-023-04231-w. Additional Declarations No competing interests reported. Supplementary Files GilbSupplementaryInformation.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7022625","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":487673627,"identity":"b2b54ef8-b269-4164-af25-8195022a8a9a","order_by":0,"name":"Scott Gilb","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzklEQVRIiWNgGAWjYBAC+QYwZcfDJ8F8ACJ0gIAWgwMMjA0HGJJ52CTYEojUwgDWcoCBTYLHgEgt7IePP/7AcECGTbrn24OfbQxyfDcS8GuR70lLBNnCwyZzdrthbxuDsSQhLQw3eAyBWo4D/ZK7TYK3jSFxA5FaDgO15DyT/NvGUE+SFjZpoC0JBoS0GJxJS5xxxgAYyDLHzKRlzkkYzjzzAL8W+fbDBz5UVNjZ80s3P5N8U2Yjz3eckMMgdsFZEsQoHwWjYBSMglFACAAAtgtDhbwmQeAAAAAASUVORK5CYII=","orcid":"","institution":"Northern Arizona University","correspondingAuthor":true,"prefix":"","firstName":"Scott","middleName":"","lastName":"Gilb","suffix":""},{"id":487673628,"identity":"2ab912c6-721b-4b3f-a20b-400166a8b4d1","order_by":1,"name":"Karen Haubensak","email":"","orcid":"","institution":"Northern Arizona University","correspondingAuthor":false,"prefix":"","firstName":"Karen","middleName":"","lastName":"Haubensak","suffix":""},{"id":487673629,"identity":"401e037d-d2ef-448e-a772-3678ecfd1860","order_by":2,"name":"Clare Aslan","email":"","orcid":"","institution":"Northern Arizona University","correspondingAuthor":false,"prefix":"","firstName":"Clare","middleName":"","lastName":"Aslan","suffix":""},{"id":487673630,"identity":"eb9ffa4e-7188-4ace-875d-b1b02af59a89","order_by":3,"name":"Liza Holeski","email":"","orcid":"","institution":"Northern Arizona University","correspondingAuthor":false,"prefix":"","firstName":"Liza","middleName":"","lastName":"Holeski","suffix":""},{"id":487673631,"identity":"8084534b-e567-4312-a84b-4f95c684432b","order_by":4,"name":"Trevor Faske","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Trevor","middleName":"","lastName":"Faske","suffix":""},{"id":487673632,"identity":"cf437394-a645-4acb-a4c7-5eb7ff4c7e64","order_by":5,"name":"Sara Souther","email":"","orcid":"","institution":"Northern Arizona University","correspondingAuthor":false,"prefix":"","firstName":"Sara","middleName":"","lastName":"Souther","suffix":""}],"badges":[],"createdAt":"2025-07-01 17:23:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7022625/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7022625/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87197830,"identity":"ca17cf39-0ff1-40b5-9821-efc03f54fccd","added_by":"auto","created_at":"2025-07-21 12:48:46","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1496871,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ea)\u003c/strong\u003eDistribution of historically monitored populations including populations thought to be recently extirpated or not found during recent surveys. \u003cstrong\u003eb)\u003c/strong\u003eColor coded populations sampled for genetic analysis. Sky island mountain ranges have been highlighted in individual colors for comparison to PCA cluster designations.\u003cstrong\u003e c)\u003c/strong\u003e Photo of a floral visitor (\u003cem\u003eHalictidae) \u003c/em\u003evisiting an open flower on \u003cem\u003eP. imberbis. \u003c/em\u003e\u003cstrong\u003ed)\u003c/strong\u003e Photo of Coronado National Memorial viewed from the rim of Montezuma Pass.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7022625/v1/55d44428ad61fe06b307cc3a.jpeg"},{"id":87197829,"identity":"d90f32f3-b498-44ef-852c-7e8103201a8a","added_by":"auto","created_at":"2025-07-21 12:48:46","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":326123,"visible":true,"origin":"","legend":"\u003cp\u003ePrincipal components analysis (PCA) with k-means genetic clusters derived from genetic (SNP) variation. Permanova results provide evidence of significant differences among k-means clusters (F₍2,135₎ = 98.72, \u003cem\u003eR²\u003c/em\u003e = 0.594, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001). Population genetic clusters broadly reflect geographic distinctions between populations with patterns derived from their proximities to individual Sky Island ranges \u003cstrong\u003e(Figure 1b).\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7022625/v1/3e8a077e5c2d51a83ccb92cd.jpeg"},{"id":87197836,"identity":"512ebe26-a751-469a-9c5c-895804583a0d","added_by":"auto","created_at":"2025-07-21 12:48:47","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":555531,"visible":true,"origin":"","legend":"\u003cp\u003eMap of Estimated Effective Migration Rates calculated from SNP dataset. Green dots represent populations sampled for RADseq analysis, from which SNP data was calculated and used to estimate effective migration rates across the range. Higher probability of migration, or gene flow, is shown in blue, whereas limited migration rates (or potential barriers to allelic migration) are shown in orange. Red circles indicate populations thought to be extirpated, and grey dots represent populations that have not been rediscovered in recent surveys.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7022625/v1/1784ba60ccad37fd4b860293.png"},{"id":87197832,"identity":"9087fec4-8c9a-495c-9014-bb19615a635d","added_by":"auto","created_at":"2025-07-21 12:48:47","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":999976,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a) \u003c/strong\u003eEEMS posterior mean distribution maps of effective genetic diversity rates for simulations with 300 demes compared to 400 demes. \u003cstrong\u003e(b)\u003c/strong\u003e EEMS posterior mean distribution maps of effective genetic diversity rates for simulations with 300 demes compared to 400 demes.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7022625/v1/4f91bad235e420da918c62a2.jpeg"},{"id":87199581,"identity":"5b995410-a784-476c-9d06-2b05129d0716","added_by":"auto","created_at":"2025-07-21 13:04:47","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":359444,"visible":true,"origin":"","legend":"\u003cp\u003eBipartite pollinator network diagram of observed floral visitors on \u003cem\u003eP. imberbis \u003c/em\u003eacross three sites. Floral visitors were categorized to functional group and citizen science categories, based on visual visitation observation in combination with voucher specimen collection. Lines correspond to observed visits at each population and line width indicates relative importance value of each visitor taxon.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7022625/v1/d7299e897b6c30df76e38edc.jpeg"},{"id":91970058,"identity":"145972ac-070c-4f84-be41-f5b4131a287f","added_by":"auto","created_at":"2025-09-23 09:02:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4736098,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7022625/v1/fa386363-d6ad-432b-985f-fc6d68aacfeb.pdf"},{"id":87199225,"identity":"0763a2ac-1172-4316-83c5-3b3621919f15","added_by":"auto","created_at":"2025-07-21 12:56:47","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":931580,"visible":true,"origin":"","legend":"","description":"","filename":"GilbSupplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7022625/v1/eedd72074812f6a66cc02969.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Population structure and genetic connectivity in the endangered Pectis imberbis: addressing conservation and genetic gaps in the Arizona Sky Islands","fulltext":[{"header":"Introduction","content":"\u003cp\u003eNearly 40% of vascular plants are considered at risk of extinction, necessitating broadscale efforts to conserve biodiversity (Nic Lughadha et al. 2020). Rare plant conservation planning is predicated on understanding the distribution of genetic variation across the landscape (Dennis et al. \u003cspan class=\"CitationRef\"\u003e1991\u003c/span\u003e, Dibner et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e, Lande \u0026amp; Shannon \u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e). Land managers and conservationists apply this knowledge to select conservation targets (Hedrick \u0026amp; Miller, \u003cspan class=\"CitationRef\"\u003e1992\u003c/span\u003e), identify critical corridors to maintain gene flow (Segelbacher et al. \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e), and introduce novel genotypes across the landscape via managed relocation or assisted evolution (Holderegger et al. 2019). Integrating genetic approaches into conservation and land management is becoming increasingly important, as rapid human-driven global change introduces new selective pressures and threatens both genetic connectivity and genetic diversity in already imperiled species (Aguilar et al. \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e; Pauls et al. \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e\n\u003cp\u003eRare plants, by definition, are characterized by small population sizes that shape demographic and evolutionary trajectories and thus conservation approaches (Ellstrand \u0026amp; Elam, 1993). Small populations are subject to demographic and genetic stochasticity that increase the likelihood of random allele loss, leading to non-adaptive shifts in gene frequencies (i.e., genetic drift) (Honnay \u0026amp; Jacquemyn, 2007). When small population size leads to increased inbreeding, individual fitness may decrease due to inbreeding depression via the expression of deleterious recessive alleles (Hamabata et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). Other factors such as the Allee effect \u0026ndash; a decline in fitness linked to diminishing populating sizes -impacts small populations by lowering per capita reproductive rates and reinforcing patterns of rarity (Stephens et al., \u003cspan class=\"CitationRef\"\u003e1999\u003c/span\u003e). Smaller patch sizes of flowering plants are less effective at attracting pollinators due to pollen limitation and diminished visual cues, driving Allee effects while also reducing the potential for genetic or demographic rescue from neighboring populations (Anic et al. \u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e). These factors \u0026mdash; inbreeding depression, Allee effects, and genetic drift \u0026mdash; can create a feedback loop termed \u0026lsquo;the extinction vortex\u0026rsquo; that drives populations toward extinction (Nordstrom et al. \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e; Pruett et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eGlobal change and habitat loss interacts with rarity to accelerate extinction (Brook et al., \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). Plant species with small and isolated populations may be unable to adapt to novel environmental conditions when these factors reduce genetically-based phenotypic variation (Boyd et al. \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e). When populations are connected rather than isolated, gene flow maintains genetic diversity through admixture with populations in other selective environments (Frankham et al., \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e). For this reason, rare species conservation often focuses on maintaining genetic diversity both within and across populations and may target populations with distinct genetic signatures that could represent unique adaptations to the environment as well as those with high levels of genetic diversity (Coates et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). Moreover, rapid climatic and environmental changes can outpace the adaptive potential of rare plants (Jump \u0026amp; Pe\u0026ntilde;uelas, 2005), which often possess life history traits associated with low rates of reproduction and slow growth (Adler et al. \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e). This is particularly true for dryland plants with restricted ranges that exist at the edge of physiological tolerances to aridity such that even small decreases in precipitation can cause widespread mortality (Laikre et al. \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e, Richardson et al. 2013).\u003c/p\u003e\n\u003cp\u003eThe Madrean Sky Archipelago ecoregion of southern Arizona (Sky Islands) hosts a broad array of endemic and relic species, with over 100 species protected under the Endangered Species Act (ESA). Like many rare species, these species are threatened by multiple interacting factors including climate change, competition with non-native species, herbivory by overabundant wild ungulates, and habitat fragmentation (Buchmann \u003cspan class=\"CitationRef\"\u003e1994\u003c/span\u003e, Yanahan \u0026amp; Moore \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). This region is also subjected to a host of land use pressures, such as erosion due to livestock grazing, encroachment by woody plants, and historical fire suppression (Villarreal et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e), that may accelerate decline of rare species that depend on rapid establishment for survival (McPherson \u0026amp; Weltzin, 2000). The Sky Islands are mountains with mid-to-high elevation woodland communities separated by expanses of desert and arid grasslands. While among-population isolation is a natural feature of the Sky Island landscape (Yanahan \u0026amp; Moore, \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e), habitat fragmentation may have a disproportionate effect on gene flow for rare species in this already isolated habitat mosaic (Honnay \u0026amp; Jacquemyn, 2007). However, under prolonged isolation, one possible outcome is the evolution of self-compatibility\u0026mdash;and the recurrent self-fertilization it permits\u0026mdash;if populations purge enough partially recessive deleterious alleles, potentially rendering them more resilient to inbreeding depression (Byers \u0026amp; Waller, \u003cspan class=\"CitationRef\"\u003e1999\u003c/span\u003e; Busch, \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e, Frye \u0026amp; Neel, \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePectis imberbis\u003c/em\u003e (Gray) is a perennial herb in the Asteraceae family endemic to the Arizona Sky Island ecoregion. \u003cem\u003ePectis imberbis\u003c/em\u003e was granted status as an endangered species on June 15th, 2021 under the ESA, due to the putative extirpation of 9 populations since the 1970s and the restricted range of this species (Crawford, \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e). The largest populations occur in and around the Coronado National Memorial (CNM), located in the Huachuca Mountain Range, with an array of smaller, isolated populations occurring across other Sky Islands, including the Baboquivari, Sierrita, Santa Rita, Rincon, Patagonia, Whetstone, and Dragoon Mountain ranges (Wilson, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e) (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003ea\u003cstrong\u003e/b)\u003c/strong\u003e. Populations in and near the CNM demonstrate population growth above replacement levels, while populations peripheral to the Memorial are in decline (Souther et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). Known extrinsic threats include competition from introduced species, grazing, road maintenance, and mining operations, each posing localized risks that could drive small populations toward extirpation (Crawford, \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e, Souther et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). These disturbances likely interfere with critical reproductive processes, including flowering, fruiting, seed dispersal, and pollinator visitation. Furthermore, the species\u0026apos; inherent rarity may compound these threats by making the small yellow flowers less conspicuous to pollinators, thus reducing visitation frequency and successful cross-pollination events, and potentially increasing dependence on autogamy (Courchamp et al. \u003cspan class=\"CitationRef\"\u003e1999\u003c/span\u003e; Souther et al. \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e). Consequently, clarifying patterns and effectiveness of pollinator visitation across populations is essential for identifying vulnerabilities and informing targeted conservation management (Souther et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eExtrinsic disturbances and rarity-driven pollinator limitation can ultimately intensify genetic risks, including reduced gene flow and increased potential for inbreeding depression (Aguilar et al., \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). Gene flow in plants relies heavily on pollinator visitation and seed dispersal, processes that facilitate genetic connectivity among fragmented populations (Kwak et al., \u003cspan class=\"CitationRef\"\u003e1998\u003c/span\u003e). Estimating these genetic exchanges is therefore critical for determining appropriate conservation interventions aimed at maintaining connectivity and genetic variation (Liu et al., 2015). Reduced connectivity is frequently linked to declines in overall genetic diversity\u0026mdash;such as lower expected heterozygosity and nucleotide diversity\u0026mdash;which may lead to diminished fitness and reproductive viability (Gentili et al. 2018, Mhemmed et al. \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e; Wright, \u003cspan class=\"CitationRef\"\u003e1965\u003c/span\u003e). Genetic bottlenecks can often accompany landscape fragmentation, amplifying Allee effects and further compromising the resilience of rare plant populations (Hannah, \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e; Souther et al. \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eHere, we use SNP markers to estimate effective migration surfaces to delineate population structure and gene flow across populations of an endangered plant, \u003cem\u003eP. imberbis\u003c/em\u003e, endemic to the Arizona Sky Island ecoregion. To contextualize migration surfaces, we relate findings to pollinator observations as well as to recently published demographic performance data (Souther et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). To our knowledge, this is the first genetic analysis of this species at the population level. Based on the natural isolation imposed by the Sky Island landscape, the small and declining population sizes in peripheral areas of \u003cem\u003eP. imberbis\u003c/em\u003e\u0026rsquo;s range, and the ecological pressures of habitat fragmentation, we hypothesize that gene flow will be heterogeneously restricted among small, more isolated populations. We expect reduced connectivity will be reflected in pronounced population genetic structure, with Sky Island ranges acting as partial barriers to gene migration. Such uneven gene flow patterns may in turn influence population persistence via interactions with pollinator communities and demographic performance, ultimately informing targeted conservation interventions. This study addresses a critical knowledge gap identified by the Species Recovery Report, which highlights the need to understand patterns of genetic variation within and among populations of \u003cem\u003eP. imberbis\u003c/em\u003e (Crawford \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e, Souther et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). To contribute to the conservation and management goals of this species, our objectives for this study were to 1) delineate the population genetic structure of \u003cem\u003eP. imberbis\u003c/em\u003e through SNP markers; 2) determine potential barriers to gene flow through regions of reduced migration; and 3) identify key pollinators for the species and elaborate on their role in the exchange of genetic material for \u003cem\u003eP. imberbis\u003c/em\u003e.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eDNA Extraction and RADseq.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn summer 2022, we collected foliar samples from 281 individuals across 32 \u003cem\u003eP. imberbis\u003c/em\u003e subpopulations representing its known range in this region, as determined in three seasons of previous demographic surveys (Souther et al. \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e, Souther et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). We defined subpopulations as clusters of plants at least 150 meters from conspecifics or separated by other significant natural or manmade barriers. We randomly selected up to 10 fresh leaves per individual, which were then desiccated using silicon beads in paper envelopes and frozen at -80\u0026deg; C until DNA extraction. Some sample locations (i.e. Pe\u0026ntilde;a Blanca) maintained very few individuals, thus not every site could be sampled equally (up to 10 individuals) without the potential risk of damaging fitness at small, threatened populations. We sourced the sampled individuals from sites in proximity to three different sky-island mountain ranges (the Huachuca Mountains, Santa Rita Mountains, and Atascosa-Pajarito Mountains), and recorded sample coordinates for each population cluster. Specifically, we extensively sampled \u003cem\u003eP. imberbis\u003c/em\u003e at the following six sites: Coronado National Memorial (CNM), Anne Tank Wash (ATW), Scotia Canyon (SCO), O\u0026rsquo;Donnell Canyon (ODC), Pe\u0026ntilde;a Blanca (PBC), and the Santa Ritas, which consists of two subpopulations, Wasp Canyon and McCleary Canyon (WSP \u0026amp; MCC) \u003cstrong\u003e(\u003c/strong\u003eFig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e, \u003cstrong\u003eSI Table I).\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe then transferred the dried leaf material samples in Qiagen Extraction tubes (Qiagen, Valencia, CA, USA) and shipped to the University of Minnesota Genomics facility for DNA extraction and Next-Generation genomic sequencing. Following DNA extraction, the sequencing facility created 2 pools of inline barcoded SbfI RADseq libraries, with samples grouped by subpopulation but randomly divided between pools. Each pool was sequenced on a separate lane of a NovaSeq S1 platform (Illumina Inc., UK) using a 1 \u0026times; 100 bp single-end run, generating 91 bp reads. Genome size was estimated at ~\u0026thinsp;13 Gb based on flow cytometry conducted by Ag-Biotech (Ag-Biotech, Inc., Colton, CA, USA), and assuming the rare-cutting SbfI enzyme (8 bp recognition site), we expected to recover\u0026thinsp;~\u0026thinsp;200,000 RAD loci. This design targeted\u0026thinsp;~\u0026thinsp;26\u0026times; per-sample coverage. Pool 1 yielded\u0026thinsp;~\u0026thinsp;536\u0026nbsp;million reads across 138 samples (~\u0026thinsp;3.88\u0026nbsp;million per sample), and Pool 2 generated\u0026thinsp;~\u0026thinsp;732\u0026nbsp;million reads across 144 samples (~\u0026thinsp;5.08\u0026nbsp;million per sample), closely aligning with the expected sequencing depth. Quality control using FastQC v0.11.7 (Andrews, \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e) within the UMGC\u0026rsquo;s Gopher-pipelines v2.4 confirmed high-quality sequencing (Phred scores\u0026thinsp;\u0026ge;\u0026thinsp;30), with adapter content ranging from 2\u0026ndash;8%. Notably, low-yield DNA extraction and differences in sequencing efficiency between lanes introduced a strong lane effect: Pool 1 exhibited higher missing data due to lower sequencing depth. To maximize statistical power and data quality, only Pool 2 (n\u0026thinsp;=\u0026thinsp;141) was retained for downstream analysis due to its lower percentage of missing data. After applying a 40% missing data threshold per individual, three additional samples were removed, resulting in a final dataset of 137 individuals representing 17 subpopulations nested within six broader populations (\u003cstrong\u003eSI Table I\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRAD tags and SNP calling.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe analyzed FASTQ files containing RAD tags using the Stacks v2.66 de novo pipeline (Catchen et al. 2013). We used the denovo_map.pl wrapper to assemble orthologous tags into stacks, generate a catalog of putative RAD loci, transpose the data by locus (Dang et al. 2022), and call single nucleotide polymorphisms (SNPs). We set the minimum depth of coverage required to form a stack (m), the maximum distance between stacks within an individual (M), and the number of mismatches allowed between stacks when building the catalog (n) each to 2, following common parameter choices for moderate-coverage datasets (Paris et al. \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e). Following catalog assembly, we applied additional filtering to retain high confidence SNPs for downstream analyses. Using VCFtools v0.1.16 (Danecek et al., 2011), we first removed low-quality individuals and retained only biallelic SNPs present in at least 40% of individuals, applying a minor allele frequency (MAF) threshold of 0.05 (5%) to exclude rare variants (Nazareno \u0026amp; Knowles, 2021). Individuals with greater than 70% missing data were excluded based on visual inspection of a histogram of missing genotype rates, which showed relatively few samples exceeded this threshold (Faske et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). This filtering resulted in 523 SNPs across 138 individuals. To minimize linkage among markers from the same RAD locus, we applied a thinning filter (--thin 91) based on our 91 bp RAD-tag length, retaining only one SNP per locus (Danecek et al., 2011). The final dataset included 337 SNPs, which were converted to PLINK-format .012 matrices for downstream analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenetic Diversity and Population Structure.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo assess between-population genetic differentiation, we performed principal component analysis (PCA) to summarize SNP variation and visualize population structure across Sky Island regions (Jombart et al. 2010; Faske et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). We applied a K-means clustering algorithm using the kmeans function in the R \u003cem\u003estats\u003c/em\u003e package (v3.6.2) to identify genetic clusters (Truelove et al. \u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e). We assigned three centroids based on minimum Bayesian information criterion (BIC) to the mapped populations to investigate population structure (Hartigan \u0026amp; Wong, \u003cspan class=\"CitationRef\"\u003e1979\u003c/span\u003e; Milano et al. \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e). We performed a dispersion analysis to test for significant differences in dispersion values for three K-means clusters in PCA space. After assigning groups and testing for dispersion, we conducted a permanova test with the \u0026lsquo;\u003cem\u003eadonis2\u003c/em\u003e\u0026rsquo; function in the Vegan package (Evans et al. 2023) (permutations\u0026thinsp;=\u0026thinsp;10,000) to test for significant differences between three genetic clusters determined from K-means. To further determine genetic distances between populations at a higher resolution, we calculated pairwise genetic differentiation (F\u003csub\u003eST\u003c/sub\u003e and Nei\u0026rsquo;s Diversity) among 17 subpopulations and 6 populations using the hierfstat R package (Goudet \u0026amp; Jombart, \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e; Faske et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). To compare F\u003csub\u003eST\u003c/sub\u003e to another estimate of between-population genetic diversity, we calculated Nei\u0026rsquo;s D estimates of genetic diversity using custom code following Faske et al. (\u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). Finally, we estimated migration of alleles between populations using Wright\u0026rsquo;s island model of migration, using F\u003csub\u003eST\u003c/sub\u003e as an estimate parameter (Wright 1951, Slatkin, 1985).\u003c/p\u003e\n\u003cp\u003eFor within-population genetic diversity, we used expected heterozygosity (Hₑ) and nucleotide diversity (\u0026pi;) as our primary metrics. Subpopulation Hₑ was computed with custom R scripts (Faske et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e), incorporating per-locus sample-size weighting and Nei\u0026rsquo;s unbiased correction to accommodate uneven sample sizes (2\u0026ndash;10 individuals) and variable missing data (Sopniewski \u0026amp; Catullo \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e). Nucleotide diversity (\u0026pi;) and gene flow (Nₘ) were then estimated in Stacks v2.0 (Catchen et al. 2013) under an 80% per-population locus-presence filter and minor-allele-frequency\u0026thinsp;\u0026ge;\u0026thinsp;0.05. Finally, Hₑ and \u0026pi; were summarized at the population level to produce diversity estimates that more accurately reflect true biological variation rather than artifacts of sampling bias (Khatri \u0026amp; Burt \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEstimated Effective Migration Surfaces across Isolated Populations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe used SNP genetic distances and geographical distances across populations to estimate effective migration surfaces across the region of study (EEMS; Petkova et al. 2016). For assessing potential barriers to gene flow between sky islands and into Mexico, we included all 137 individuals and the associated high-quality SNPs that were filtered for population genetic analyses to visualize similarities between population structure and estimated gene flow. The 337 high-quality SNPs extracted for the population structure analyses were used to calculate a genetic distance matrix with the bed2diff_v1 program (Petkova et al. 2016). Demes were defined across a habitat grid that extended beyond the sampled populations, including a buffer around the species\u0026apos; range to reduce edge effects during the Markov chain simulations and to explore potential migration surfaces extending into northern Mexico, where the species\u0026rsquo; distribution and genetic diversity remain poorly understood. To assess the robustness of inferred migration patterns, we ran the EEMS algorithm (\u003cem\u003eruneems_snps\u003c/em\u003e; Petkova et al. 2016) using both 300 and 400 demes. This allowed us to evaluate the sensitivity of spatial inferences to deme resolution, as shown in EEMS documentation and prior studies (Jones et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e; Li et al. 2020; Petkova et al. 2016). Each run used 1\u0026nbsp;million burn-in iterations and 12\u0026nbsp;million sampling iterations to ensure convergence toward a stationary distribution of migration rates. We inspected the results after multiple runs for convergence success after each run (Herman et al.2022). We then used the reemsplots2 package in R to visualize the results of the program, plotting effective migration rates (\u003cem\u003em)\u003c/em\u003e and effective genetic diversity rates (\u003cem\u003eq)\u003c/em\u003e on a log\u003csub\u003e10\u003c/sub\u003e scale after inspection of correlation between observed versus expected genetic dissimilarity (Jones et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePollination network and breeding system\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo identify primary pollinators and assess the breeding system of \u003cem\u003eP. imberbis\u003c/em\u003e, we conducted insect visitation surveys and pollinator exclusion trials between August and October from 2019\u0026ndash;2023. Surveys were conducted at three subpopulation sites: Coronado National Memorial (CNM Visitor Center Maintenance Shed Front), Anne Tank Wash (Upper ATW), and Scotia Canyon Populations from August to October, 2019\u0026ndash;2023. Detailed methodological protocols for both pollinator observations and mesh bag exclusion experiments are published in Souther et al. (\u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). In brief, insect visitors were observed during timed sampling sessions and categorized into functional groups based on citizen science categories (Ullmann et al., 2011). To distinguish likely pollinators from incidental visitors, only functional groups with voucher specimens confirmed to transport pollen via fuchsin gel staining (Kearns \u0026amp; Inouye, \u003cspan class=\"CitationRef\"\u003e1993\u003c/span\u003e) were retained for analysis. Here, we analyzed functional group importance across sites using a combination of visitation frequency and observed resource use behaviors, such as legitimate pollen collection, or nectar robbing, in which the visitor does not appear to contact reproductive parts of the flower when feeding. Importance values were used to assess variation in pollinator community structure among sites via variance analysis and visualized in a weighted bipartite network, with scaled importance values serving as edge weights linking taxa and sites (Castillo et al. 2024).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eRADseq Determination of Population Structure\u003c/b\u003e\u003c/p\u003e\u003cp\u003eReduced dimensionality of SNP frequencies across individual samples revealed evidence of population structure in \u003cem\u003eP. imberbis\u003c/em\u003e \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. Principal component analysis of 137 individuals and 337 quality-filtered SNPs indicated three genetically distinct clusters: (1) Coronado National Memorial, (2) the western edge of the Huachuca Mountains (with Pe\u0026ntilde;a Blanca as an outlier), and (3) the Santa Rita Mountains (Wasp Canyon and McCleary Canyon). K-means clustering on the first two principal components supported these groupings (K\u0026thinsp;=\u0026thinsp;3). A permanova confirmed significant differences among clusters (F₍2,135₎ = 98.72, \u003cem\u003eR\u0026sup2;\u003c/em\u003e = 0.594, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and analysis of multivariate homogeneity of dispersions indicated no significant differences among group variances, validating assumptions for the test. A PCA highlighting population means further showed distinct separation, particularly between Coronado and Santa Rita groups \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; \u003cb\u003eSI Fig. III).\u003c/b\u003e For population genetic diversity estimates, pairwise F\u003csub\u003eST\u003c/sub\u003e values for 6 aggregated populations ranged from 0.077 to 0.421 and Nei\u0026rsquo;s diversity coefficient ranged from 0.026 to 0.343 \u003cb\u003e(SI Fig. II)\u003c/b\u003e. The global F\u003csub\u003eST\u003c/sub\u003e value for all pairwise groups was 0.21. Gene flow coefficients ranged from 0.343 to 2.978, with a mean of 1.21 \u003cb\u003e(SI Table II, SI Table III)\u003c/b\u003e. For within population genetic diversity estimates for 17 subpopulations, expected heterozygosity ranged from 0.09 to 0.25, with a global He of 0.20. Nucleotide diversity (π) for 17 subpopulations ranged from 0.04 to 0.10, with a mean of 0.07 (\u003cb\u003eSI Table IV\u003c/b\u003e). Nucleotide diversity was highest in Scotia Canyon followed by ATW and O\u0026rsquo;Donnell Canyon \u003cb\u003e(SI Table IV).\u003c/b\u003e Expected heterozygosity was highest in O\u0026rsquo;Donnell Canyon \u003cb\u003e(SI Table IV).\u003c/b\u003e For populations in which pollinator observations were conducted, gene flow was highest between CNM and ATW (Nm\u0026thinsp;=\u0026thinsp;2.978), whereas Scotia Canyon had lower values of 0.926 between ATW and 0.903 between CNM \u003cb\u003e(SI Table V)\u003c/b\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eMigration and Gene Flow Between Populations\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe EEMS resulted in positive relationships between expected and observed genetic dissimilarity in simulations with both 300 and 400 predefined demes, with runs of 400 demes resulting in higher R\u003csup\u003e2\u003c/sup\u003e values for dissimilarities between pairs of sampled demes, indicating a stronger fit for observed versus predicted values for the 400 deme models (Jones et al. \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, Petkova et al. 2016) (\u003cb\u003eSI Fig. I)\u003c/b\u003e. MCMC simulations of migration rates and genetic diversity through EEMS mirrored the geography of the region and demographic patterns of the species, with low effective migration surfaces coinciding with Sky Island edges (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e., Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. Specifically, migration resistance was evident on the western edge of the Huachuca Mountains and through the Patagonia Mountains.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe pattern of relatively low effective migration coincided with locations of populations thought to be extirpated \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e, which suggests that this region is less conducive to dispersal and establishment. Scotia and O'Donnell Canyon sites, where populations are declining, appeared in the center of the low migration pathway. Simulations with the courser 300 deme parameter showed greater migration connectivity between Coronado National Memorial and the neighboring Anne Tank Wash populations, but increasing the deme count to 400 showed a distinct reduction a log(m) coinciding with the geographic feature of Montezuma pass \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, \u003cb\u003eSI Fig. I).\u003c/b\u003e Genetic diversity calculations from EEMS showed a distinct limitation of Q diversity rates in the most distant, isolated population of Pe\u0026ntilde;a Blanca. Populations across the Huachuca Mountains to the Santa Rita Mountains showed similar rates of diversity, with Santa Rita and O\u0026rsquo;Donnell Canyon populations demonstrating higher diversity than the larger populations of Anne Tank Wash and CNM \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea\u003cb\u003e).\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003ePollination network and breeding system\u003c/b\u003e\u003c/p\u003e\u003cp\u003eImportance values used to parameterize the bipartite network were derived from 361 ten-minute observations of flower visitation across populations and subpopulations distributed across Coronado National Memorial, Scotia Canyon, and Anne Tank Wash. Observers detected 226 legitimate flower visits by a total of 17 visitor taxa/functional groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. Visitors common to all three locations included Megachilid bees and Bombyliid flies (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e, the two most frequent and most important visitor groups recorded overall for \u003cem\u003eP. imberbis\u003c/em\u003e (Souther et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Other important visitors occurred at two of the three sites (e.g., \u003cem\u003eSteniolia\u003c/em\u003e sp. wasps were recorded at Scotia and CNM; syrphid flies were recorded at Anne Tank Wash and CNM; and Halictidae bees were recorded at Scotia and Anne Tank Wash) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. Such commonality of visitors suggests that the pollinator community is at least somewhat consistent across the landscape, which may have important implications for pollen transfer among disconnected plant populations.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eEstimating population structure and gene flow can aid in conservation efforts of small, isolated populations of rare plants by identifying barriers to the exchange of genetic material across the landscape (Laikre et al.2010, Zhou et al.2023). Overall, the results of this analysis indicate distinct population structure among subpopulations of \u003cem\u003eP. imberbis\u003c/em\u003e, suggesting that distances between populations (90 km) is enough to hinder gene flow. Moreover, estimated migration surfaces indicate restricted gene flow patterns align with both landscape features and demographic processes observed in other studies (Souther et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2022\u003c/span\u003e, Souther et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Specifically, mountainous barriers of the Patagonia Mountain range appear to overlap with predictive models of low migration surfaces, which do not take environmental variability into account (Petkova et al., 2016). This pattern, in part, could be explained by prevailing wind patterns. \u003cem\u003ePectis imberbis\u003c/em\u003e seeds and pollen are dispersed by wind, thus directionality of wind patterns influence genetic structure (Kling \u0026amp; Ackerly, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Moreover, wind conditions alter pollinator foraging patterns and flight efficiency, thus shaping the efficacy and directionality of pollen transfer (Burnett et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, Wang et al. \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The predominant southwesterly winds that occur during \u003cem\u003eP. imberbis\u003c/em\u003e flower and seed production (July - October) may prevent seeds and pollen produced by CNM populations from moving west over the Huachuca mountains into the historical range of the species. The developed area of Sierra Vista is located to the east of CNM and the recently constructed border wall to the south, thus suitable habitat for the establishment of new populations in these directions is limited. While population growth rate analyses indicate positive growth of populations located in and in proximity to the CNM, isolated populations on the range periphery are in decline (Souther et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) This pattern is likely in part explained by underlying habitat suitability, but also may reflect the consequences of fragmentation and genetic isolation \u0026ndash; when population crashes occur there is limited opportunity for recolonization or supplementation of genetic diversity (Ellstrand \u0026amp; Elam, 1993).\u003c/p\u003e\u003cp\u003e\u003cb\u003ePatterns of genetic diversity and estimated effective migration\u003c/b\u003e\u003c/p\u003e\u003cp\u003eOverall, F\u003csub\u003eST\u003c/sub\u003e values and gene flow coefficients indicate high levels of genetic differentiation across populations. F\u003csub\u003eST\u003c/sub\u003e values ranged from 0.077 to 0.421, which can be considered moderate to very high genetic variation (Hartl \u0026amp; Clark, 1997). Across all pairwise groups, the global F\u003csub\u003eST\u003c/sub\u003e value was 0.21, suggesting high genetic variation among the sampled populations of \u003cem\u003eP. imberbis\u003c/em\u003e \u003cb\u003e(SI Table III)\u003c/b\u003e. For comparison, a study on the speciation of an endemic Hawaiian plant found an F\u003csub\u003eST\u003c/sub\u003e of 0.57 for inter-island populations, and 0.027 for two populations on Maui (Filatov \u0026amp; Burke, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). One study of a range-restricted plant endemic to the Great Basin Desert found a global F\u003csub\u003eST\u003c/sub\u003e of 0.158, which was considered high (Borokini et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These findings highlight that \u003cem\u003eP. imberbis\u003c/em\u003e exhibits notable genetic differentiation across its populations, with F\u003csub\u003eST\u003c/sub\u003e values that are moderate to very high relative to other species found in island chains or that are characterized by rarity or range restriction. These F\u003csub\u003eST\u003c/sub\u003e values underscore the fragmented nature of \u003cem\u003eP. imberbis\u003c/em\u003e habitat, shaped by both natural barriers inherent to the Sky Island landscape and anthropogenic habitat degradation (Yanahan \u0026amp; Moore, \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe CNM, which contains the largest population of \u003cem\u003eP. imberbis\u003c/em\u003e, was found to be moderately genetically distinct from the nearby Anne Tank Wash population, despite being only 4 kilometers apart. These two populations exhibited moderately high genetic differentiation, with the lowest pairwise F\u003csub\u003eST\u003c/sub\u003e value between them being 0.077 (\u003cb\u003eSI Table III\u003c/b\u003e). Although they are the largest populations, they are separated by Montezuma Pass (400 m), which may act as a partial barrier to gene flow. The EEMS analysis with a finer scale (400 demes) showed reduced migration rates near Montezuma Pass, while a coarser model (300 demes) did not capture this potential barrier as clearly \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, \u003cb\u003eSI Fig. I)\u003c/b\u003e. The population sampled from Pe\u0026ntilde;a Blanca in the Atascosa-Pajarito Mountains had a small sample size (n\u0026thinsp;=\u0026thinsp;4), which likely limited its distinct geographic isolation in the PCA \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. However, it still showed relatively high F\u003csub\u003eST\u003c/sub\u003e values, indicating genetic dissimilarity. Both Pe\u0026ntilde;a Blanca and Scotia Canyon were the most genetically isolated populations, with mean F\u003csub\u003eST\u003c/sub\u003e values of 0.29 and 0.27, respectively. The greatest genetic separation was observed between these two populations, with a pairwise FST value of 0.421 and a low gene flow coefficient of 0.343 \u003cb\u003e(SI Table V)\u003c/b\u003e.\u003c/p\u003e\u003cp\u003eAnalysis of estimated effective migration mirrored the patterns of population structure observed in the PCA of SNP frequencies in which CNM populations were shown to be more connected via migration to Anne Tank Wash, Santa Rita, and Pe\u0026ntilde;a Blanca populations than to the nearby Scotia Canyon and O\u0026rsquo;Donnell Canyon groups \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. The relatively higher rates of gene flow from the southern Huachuca populations to northern and western populations is possibly influenced by a variety of factors aside from geographic distance alone. EEMS demonstrates a path of reduced migration from the southwestern edge of the Huachuca range up to the western Santa Rita range. A combination of disturbance factors, climate influences, wind patterns, and the self-crossing ability of \u003cem\u003eP. imberbis\u003c/em\u003e may contribute to the population level partitioning of genetic variation (Aguilar et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Souther et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Yanahan \u0026amp; Moore, \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Intriguingly, gene flow patterns may in part explain the low λ-values observed for the Scotia Canyon population, which is in close proximity to high performing CNM and Anne Tank populations (Souther et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Both Scotia Canyon and O\u0026rsquo;Donnell canyon demonstrated the lowest population growth rates (0.92 and 0.74, respectively) (Souther et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e: table 1)., and are in the gene flow barrier zone as demonstrated by the EEMS surface. Again, isolation can influence demographic patterns in several key ways: isolated populations are more frequently subjected to Allee effects, inbreeding depression, and reduced genetic diversity (Ellstrand \u0026amp; Elam, 1993). Furthermore, following population crashes or loss due to demographic stochasticity, isolated populations are less likely to experience genetic or demographic rescue from other populations (Bontrager \u0026amp; Angert, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, Jones et al. \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThese population structure data, paired alongside the spatial context of estimated effective migration and evidence of the extirpation of multiple populations in the Patagonia Mountains area (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e, suggest that diminished connectivity between the Huachuca and Atascosa populations (via the Patagonia Mountains) is perhaps not a recent development but may have been exacerbated by landscape changes over the past century (Love et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Specifically, all populations apart from CNM subpopulations exist on grazing allotments within Coronado National Forest. Historic grazing in the vicinity of nearly all monitored populations is suspected to be a primary driver of initial decline of \u003cem\u003eP. imberbis\u003c/em\u003e, due to direct impact of herbivory as well as indirect effects, such as erosion, soil degradation, and proliferation of non-native species that accompany overgrazing (McPherson \u0026amp; Weltzin, 2000). Demographic monitoring suggests that grazing cessation may have contributed to the recovery of some populations, but herbivory from native ungulates has also been shown to negatively impact demographic vital rates in CNM (Crawford \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003e, Souther et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2022\u003c/span\u003e, Souther et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). At the same time, opportunities for transport by wild ungulates may be one mechanism of dispersal that has allowed for the exchange of genetic material between the two populations over the geographic barrier of Montezuma Pass, which separates CNM subpopulations from ATW subpopulations on the other side of the Huachuca range \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e\u003cp\u003eRelatively low F\u003csub\u003eST\u003c/sub\u003e values and high Nm values for CNM and ATW suggest that these two populations may be the greatest source of genetic variation for this species and could potentially be the primary contributors to gene flow to other extant populations. On the other hand, EEMS suggest the migration of alleles is limited outside of the Southern Huachuca populations and that future gene flow may be hindered by the prevalence of disturbed habitat surrounding the Huachucas. Moreover, historically observed populations in Mexico have not been relocated by researchers in the past decade. The recently constructed border wall, in conjunction with more extensive grazing impacts just south of the border near CNM, are likely limiting migration and gene flow with southern populations. At the northern edge of the species range, on the other hand, the Santa Rita Mountains are under threat from copper mining activities. The populations located there represent the northernmost edge of the range and exhibit high levels of genetic variation from other populations (F\u003csub\u003eST\u003c/sub\u003e = 0.18), and combined subpopulation sizes total less than 100 individuals \u003cb\u003e(SI Table I, SI Table III).\u003c/b\u003e This population has anecdotally been observed to have higher seed set and more robust phenotypes (larger and more stems), thus the extirpation of this population could constitute a reduction in genetic diversity and seed availability at the leading edge of the species range (Souther et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cb\u003eFloral visitors and Movement of Genetic Material\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe gene flow coefficients observed here indicate that gene flow between groups is medium to high according to Wright\u0026rsquo;s definitions (Wright 1978), but EEMS surfaces provide more context to these values by highlighting a large region of restricted gene flow on the western edge of the Huachuca Mountains in which migration rates (m) range from 0.1\u0026ndash;0.3 on a log scale \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. The low migration rates and restricted gene flow overlaps with relatively undisturbed National Forest land, making it a potentially suitable region to increase genetic connectivity (via pollen transport) through the establishment of new populations (Kwak et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). While mining, grazing, and geographic barriers all serve as extrinsic threats to rare plants like \u003cem\u003eP. imberbis\u003c/em\u003e, the species also faces intrinsic stressors due to its inherently small population sizes (Luijten et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). These stressors include potential genetic bottlenecks and factors like the Allee effect, where limited pollen availability for floral visitors hampers growth within groups of plants maintaining low population densities (Hannah, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Souther et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). An adaptive strategy observed in \u003cem\u003eP. imberbis\u003c/em\u003e to counteract this effect is the formation of dense clustering patterns on steep slopes, which may enhance floral visibility and facilitate pollen transfer amongst individuals, potentially increasing genetic diversity at the population level (Birzu et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Chi \u0026amp; Molano-Flores, 2014; Dibner et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). At the same time population growth rates have been shown to be lower on slopes (Souther et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2022\u003c/span\u003e, Souther et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), perhaps suggesting slopes serve as a refugia from herbivory or competition, though seedling survival may be lower overall due to higher rates of soil erosion. While clustering and persistence on less crowded slopes may be an advantage at the microhabitat scale, this clustering also makes populations particularly vulnerable to disturbance events such as fires, which could rapidly eliminate entire clusters, posing a significant risk to their survival.\u003c/p\u003e\u003cp\u003eDespite being self-compatible\u0026mdash;with effective population size (N\u003csub\u003ee\u003c/sub\u003e) in selfing lineages potentially eroding at twice the rate of strict outcrossers (Pollak, 1987; Willi et al. \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) \u0026mdash;\u003cem\u003ePectis imberbis\u003c/em\u003e appears to rely on native bees (e.g. Megachile) and Bombyliidae flies for outcrossing (Souther et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). \u003cem\u003eMegachile\u003c/em\u003e and \u003cem\u003eBombyliidae\u003c/em\u003e were the only two visitor taxa present at all three populations with confirmed pollen transport on voucher specimens, placing them as two groups with higher potential for the exchange of genetic material between populations. However, foraging range for solitary bees such as \u003cem\u003eMegachile\u003c/em\u003e has been shown to be limited to 150-600m, with \u003cem\u003eMegachile\u003c/em\u003e on the lower edge of the range (Gathmann \u0026amp; Tscharntke \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2002\u003c/span\u003e, Hoffman et al.2020). The body size of these pollinators is thought to limit their foraging distance (Greenleaf et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), thereby restricting gene flow between populations due to a combination of small pollinator ranges and habitat fragmentation (Gamba \u0026amp; Muchhala 2022, Warzecha et al. \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Observations indicate that most floral visitors to various \u003cem\u003eP. imberbis\u003c/em\u003e populations have foraging distances of less than 1 km (Souther et al.2022; USFWS, 2023), significantly less than the global mean distance of 47.84 km between populations studied (\u003cb\u003eSI Table IV\u003c/b\u003e). Consequently, if intra-population pollen transfer is severely limited as observations indicate, \u003cem\u003eP. imberbis\u003c/em\u003e likely has greater reliance on seed dispersal for gene flow.\u003c/p\u003e\u003cp\u003e\u003cb\u003eContribution of seed dispersal to gene flow\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003ePectis imberbis\u003c/em\u003e seeds, characterized by few spreading awns under 5mm, suggest a limited capacity for wind dispersal (Fishbein and Warren, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1994\u003c/span\u003e). The extent of wind (anemochorous) versus animal (zoochorous) dispersal remains uncertain, but evidence suggests native Coues deer (\u003cem\u003eOdocoileus virginianus couesi\u003c/em\u003e) may be an important vector given high levels of herbivory observed (Souther et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Coues deer may act as passive transporters of \u003cem\u003eP. imberbis\u003c/em\u003e; however, feeding trials would be necessary to determine if viability persists following passage through Coues deer\u0026rsquo;s digestive tract. Given the species\u0026rsquo; distribution and the current understanding of wind dispersal models, it appears that \u003cem\u003eP. imberbis\u003c/em\u003e is primarily wind dispersed within a roughly 1 km range, with longer-distance dispersal potentially dependent on animal movement (Tackenberg et al. \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Moreover, multiple populations exhibit a pattern of stochastic clustering following the downward slope of shallow canyons or arroyos, indicating a third flood-driven (hydrochorous) means of dispersal. The ability of \u003cem\u003eP. imberbis\u003c/em\u003e to potentially utilize a diverse array of dispersal strategies such as wind, animal interactions, and water flow highlights one aspect of the species adaptive resilience in the face of fragmented habitat. These varied dispersal mechanisms are critical in maintaining gene flow across fragmented landscapes, enabling the species to adapt to environmental changes and potentially enhance survival and reproductive success in different ecological niches (Pauls et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). In contrast, isolated populations are likely to continue the observed demographic decline if intervention is not undertaken.\u003c/p\u003e\u003cp\u003e\u003cb\u003eThe role of gene flow in species persistence\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis interplay of dispersal strategies with environmental challenges emphasizes the need for integrated conservation strategies that address both intrinsic and extrinsic threats to long-term persistence of this species (Pruett et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The dynamics of pollen exchange and seed dispersal are crucial for understanding the magnitude of gene flow among the peripheral populations at the northern edge of the species' range, particularly as these populations confront the uncertainties of climate change (Love et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The nature of gene flow at these range edges can have mixed impacts on fitness, depending on the context; high rates of gene flow from diverse environmental conditions may either impede or enhance the ability of leading-edge populations to adapt rapidly to their local habitats (Bridle and Vines, 2007). Conversely, when considering climatic variations, natural selection might favor the survival of northern peripheral populations that maintain higher genetic diversity through effective gene flow (Bontrager \u0026amp; Angert, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Overall, dispersal limitation appears to be a driver of the observed population structure of \u003cem\u003eP. imberbis\u003c/em\u003e, whereas climate variability may reduce habitat suitability for populations located in southern or lowland regions of the species range.\u003c/p\u003e\u003cp\u003eDemographic analyses have indicated that populations in or near CNM have positive population growth rates, while populations farther from this epicenter are in decline (Souther et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Resurvey efforts, in which teams have attempted to relocate populations, reflect this finding and generally confirm that extirpations have occurred outside the CNM in the past few decades (Crawford \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Several important environmental factors have been linked to performance, including deer browse, competition with non-native species, and climate change; yet, observed demographic patterns are not fully explained by these drivers alone. One hypothesis is that populations located far from CNM are experiencing these stressors, but due to isolation, do not receive sufficient gene flow that could buffer against disturbances. Thus, demographic and genetic losses following drought events, high-browsing years, or other perturbations are not offset by immigration and gene flow. Dwindling numbers then reinforce decline via feedback loops driven by Allee effects, inbreeding depression, and genetic drift. Increasing habitat connectivity on the landscape can ameliorate these potential effects by facilitating the movement of pollinators and seed dispersing ungulates between habitat patches (Kwak et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e1998\u003c/span\u003e; Mhemmed et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eBreeding system trials confirmed that \u003cem\u003eP. imberbis\u003c/em\u003e is both self-compatible and autogamous, capable of producing achenes without pollinator visitation (Souther et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). While autogamy may ensure reproductive success under pollinator-limited conditions, it does not protect against inbreeding or genetic isolation, particularly in small or fragmented populations. Our population structure and EEMS analyses revealed reduced effective migration rates among several peripheral populations (e.g., Wasp Canyon and Pe\u0026ntilde;a Blanca), suggesting barriers to gene flow that may be shaped by both landscape features and biotic interactions. Regarding the latter, our weighted bipartite pollination network \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e highlights both overlap and variation in pollinator assemblages: while \u003cem\u003eMegachilidae\u003c/em\u003e bees and bombyliid flies were frequent, high-importance visitors shared across all sites, other taxa were more localized. For example, \u003cem\u003eSteniolia\u003c/em\u003e wasps were observed at Scotia and CNM but not at Anne Tank Wash, while syrphid flies and halictid bees showed inconsistent presence. CNM generally exhibited greater diversity, which may not necessarily indicate more effective pollen transfer but could reflect a habitat more conducive to a broad suite of plant-animal interactions in a wider landscape subjected to drought, fire, and invasive species. Moreover, adaptive foraging by generalist pollinators (e.g. shifting their visits toward the most specialized plant partners) can reverse the effects of network structure on community stability, and suggests that population-specific differences in visitor assemblages may drive spatially variable pollen‐mediated gene flow (Kwak et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e1998\u003c/span\u003e; Valdovinos et al., 2016). Thus, the geographic heterogeneity in effective migration we observe may arise from both abiotic resistance and biotic filtering by the pollinator community (Stankowski et al. \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Future work integrating pollinator abundance, behavior, and pollen transport capacity with fine-scale genetic data\u0026mdash;such as parentage or paternity analysis\u0026mdash;could directly test the extent to which pollinator community composition shapes gene flow across the Sky Island landscape (Gigant et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cb\u003eImplications for conservation and management\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe found high levels of genetic isolation of populations at the edge of \u003cem\u003eP. imberbis\u003c/em\u003e\u0026rsquo;s current range. A cross-population demographic study of \u003cem\u003eP. imberbis\u003c/em\u003e found that population growth rates of these populations are below replacement levels and speculated that they are on a trajectory to extirpation, in part driven by rarity which reinforces population decline (Souther et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). One possible solution is to increase connectivity across populations, possibly by reintroducing populations into suitable habitat across the species\u0026rsquo; range. Increasing connectivity could increase the likelihood of genetic and demographic rescue (Richards, 2000), critical in this highly heterogeneous environment, characterized by disturbance events like drought that can result in stochastic population loss (Love et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, it is unclear whether reestablished populations would be sufficient bolster to cross population gene flow in the continental \u0026ldquo;islands\u0026rdquo; of the Madrean archipelago. A recent investigation of suitable habitat for \u003cem\u003eP. imberbis\u003c/em\u003e used an ensemble species distribution modelling approach to aid in future survey or restoration efforts (Wilson \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Of the suite of 43 predictor variables and 11 hectares used to train the ensemble models, the final model predictions identified 5 key predictor variables (spring precipitation, elevation, solar radiation, EPA ecoregion, and USGS/SWRG data) and a potential suitable habitat extent of 35,505 hectares. When compared to the EEMS migration surfaces produced in this study, a narrow band of suitable habitat overlaps a region of what is determined to be an area of historically limited migration. However, we note that EEMS assumes a stepping-stone, migration\u0026ndash;drift equilibrium and constant local population sizes, so spatially correlated changes in demographic history (e.g. recent bottlenecks or expansions) can produce the same genetic discontinuities that EEMS would call \u0026lsquo;low migration\u0026rsquo; (Petkova et al. 2016). Therefore, some of our inferred barriers (e.g. distances between Sky Islands) might reflect past declines in census size as much as true dispersal limits (Ravinet et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Nonetheless, these areas of apparent low effective migration still emphasize populations at risk of losing diversity and so remain high-priority targets for any connectivity or rescue-based intervention.\u003c/p\u003e\u003cp\u003eOne approach to enhancing genetic connectivity could be to establish restoration garden sites along intervals of suitable habitat to allow for pollen transfer between higher diversity or larger populations, such as Scotia and Anne Tank Wash populations, respectively. An alternative approach would be to directly introduce individuals or pollen from outside populations to increase genetic diversity, but such interventions are not without risk (Willi et al., 2007). If genetic signatures reflect underlying adaptations to local conditions, introduction of non-local genotypes may result in outbreeding depression (Schierup \u0026amp; Christiansen, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). By defining population clusters and barriers to gene flow this study revealed broad gene flow patterns, yet it does not determine the influence of local adaptation on survival \u0026ndash; a critical next step for conservation planning. As a counterpoint to reestablishing new populations, rapid climate change may render formerly suitable habitats maladaptive (Souther \u0026amp; McGraw, \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2011\u003c/span\u003e, Souther et al. \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Thus, interventions, such as increasing connectivity and introduction of propagules from other populations may be warranted even when adaptive differentiation is detected (Cr\u0026eacute;mieux et al., 2010; Willi et al., 2007). In cases where local conditions are changing, the benefits of greater genetic diversity and population numbers outweigh the risks of outbreeding depression (Souther \u0026amp; McGraw 2014).\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study described patterns of gene flow and isolation of \u003cem\u003eP. imberbis\u003c/em\u003e, an endangered plant species in the Sky Island Region. We found evidence of three distinct population clusters across the populations we sampled, but the degree to which these groups are influenced by genetic drift or local adaptation is less clear. We found a distinct reduction of estimated effective migration rates, which we used as an indicator of gene flow, from the largest population at Coronado National Memorial to all other populations. Populations in the Huachuca and Santa Rita Mountains demonstrated higher genetic diversity compared to the distant and isolated Pe\u0026ntilde;a Blanca population, as confirmed by nucleotide diversity, expected heterozygosity, and Q diversity from EEMS. We found pollinator visitation and diversity was higher at CNM subpopulations, suggesting a potential mechanism explaining EEMS findings though there was overlap in visitation by \u003cem\u003eMegachile\u003c/em\u003e spp. across all populations. Ultimately, this study provides fine-scale genetic and environmental context that can inform future demographic and genetic analyses for an endangered plant. More broadly, the insights generated by this study can be applied to other endemic species in the Sky Islands, specifically when determining whether isolation should be considered a natural feature of adaptation or a threat to survival (Love et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Future studies should attempt to quantify gene flow and migration patterns directly through observations of seed dispersal strategies, as well as attempt to garner higher taxonomic resolution for floral visitors to ascertain more accurate estimates of pollen transfer (Pornon et al.2017). Furthermore, tests for adaptive variation through a reciprocal common garden experiment are needed to determine the extent of local adaptation across populations, as well as to investigate the role of inbreeding depression in influencing phenotypic variation and fitness (Schierup \u0026amp; Christiansen, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). By understanding both the role of inbreeding depression and whether the genetic differences observed reflect local adaptations to site conditions, conservationists and land managers can determine whether introduction of non-local genotypes is warranted, as well as establish parameters around these actions that are non-harmful to focal populations. The patterns of genetic isolation and population decline exhibited by \u003cem\u003eP. imberbis\u003c/em\u003e serve as an early warning of the threats facing other species in the Madrean Archipelago as anthropogenic change progresses and as the climate continues to shift. In this naturally fragmented habitat, as connectivity erodes, the performance and adaptive capacity of this species\u0026mdash;and many others within these biologically rich desert islands\u0026mdash;are at risk of collapse. Identifying conservation pathways to maintain connectivity and thus demographic and genetic resilience, is critical for preserving the diversity of this unique landscape.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eCompeting Interests\u003c/h2\u003e\u003cp\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eWe thank the Coronado National Memorial and the Coronado National Forest for their support of this project. This work was funded by National Park Service funding PMIS #316237 awarded to S. Souther. Pollination analyses were funded by USGS award #G21AC10137-00 to S. Souther, C. E. Aslan, and USGS collaborator L. Norman.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eSPG prepared the material, designed and performed the genetic analysis and wrote the manuscript. SS contributed to the study conception and design, secured funding, and assisted with the writing and revisions of the manuscript. CA designed the pollinator study, secured funding, and contributed to the revisions of the pollinator methodology section. TF prepared python and R scripts and contributed to the analysis and paper revisions. KH and LH advised and commented on several drafts of the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe thank the Audubon Appleton Whittell Research Ranch for housing support. We thank Julie Crawford at USFWS for her support of endangered species conservation and for her development of the species recovery plan for P. imberbis. We thank Martha Sample and Morgan Andrews for their administration, logistics, and field support. We thank Matthew Weiss for the suggestions, conversations, and conceptual guidance regarding bioinformatic methodologies. Finally, we would like to thank Laura Nicholson, Mia Brann, Megan Quinn, Alex Croydon, Hanna Ryder, Alia Raderstorf, Alex Wakefield, and many others who contributed diligence and passion to the demographic surveys of the Beardless Chinchweed.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eData supporting our findings will be archived in Dryad Digital Repository, and/or are available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAdler, P. B., Salguero-Gomez, R., Compagnoni, A., Hsu, J. S., Ray-Mukherjee, J., Mbeau-Ache, C., et al. (2014). Functional traits explain variation in plant life history strategies. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e 111, 740\u0026ndash;745. doi: 10.1073/pnas.1315179111\u003c/li\u003e\n\u003cli\u003eAguilar, R., Quesada, M., Ashworth, L., Herrerias-Diego, Y., \u0026amp; Lobo, J. (2008). Genetic consequences of habitat fragmentation in plant populations: Susceptible signals in plant traits and methodological approaches. \u003cem\u003eMolecular Ecology\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(24), 5177\u0026ndash;5188. https://doi.org/10.1111/j.1365-294X.2008.03971.x\u003c/li\u003e\n\u003cli\u003eAndrews, S. (2010). FastQC: A Quality Control Tool for High Throughput Sequence Data [Online]. http://www.bioinformatics.babraham.ac.uk/projects/fastqc/\u003c/li\u003e\n\u003cli\u003eAnic, V., C. A. Henr\u0026iacute;quez, S. R. Abades, and R. O. Bustamante. \u0026ldquo;Number of Conspecifics and Reproduction in the Invasive Plant Eschscholzia Californica (Papaveraceae): Is There a Pollinator-Mediated Allee Effect?\u0026rdquo; \u003cem\u003ePlant Biology\u003c/em\u003e 17, no. 3 (2015): 720\u0026ndash;27. https://doi.org/10.1111/plb.12293.\u003c/li\u003e\n\u003cli\u003eBirzu, Gabriel, Sakib Matin, Oskar Hallatschek, and Kirill S. Korolev. \u0026ldquo;Genetic Drift in Range Expansions Is Very Sensitive to Density Dependence in Dispersal and Growth.\u0026rdquo; \u003cem\u003eEcology Letters\u003c/em\u003e 22, no. 11 (2019): 1817\u0026ndash;27. https://doi.org/10.1111/ele.13364.\u003c/li\u003e\n\u003cli\u003eBontrager, Megan, and Amy L. Angert. \u0026ldquo;Gene Flow Improves Fitness at a Range Edge under Climate Change.\u0026rdquo; \u003cem\u003eEvolution Letters\u003c/em\u003e 3, no. 1 (2019): 55\u0026ndash;68. https://doi.org/10.1002/evl3.91.\u003c/li\u003e\n\u003cli\u003eBorokini, Israel T., Kelly B. Klingler, and Mary M. Peacock. \u0026ldquo;Life in the Desert: The Impact of Geographic and Environmental Gradients on Genetic Diversity and Population Structure of Ivesia Webberi.\u0026rdquo; \u003cem\u003eEcology and Evolution\u003c/em\u003e 11, no. 23 (2021): 17537\u0026ndash;56. https://doi.org/10.1002/ece3.8389.\u003c/li\u003e\n\u003cli\u003eBoyd, J. N., Anderson, J. T., Brzyski, J., Baskauf, C., \u0026amp; Cruse-Sanders, J. (2022). Eco-evolutionary causes and consequences of rarity in plants: A meta-analysis. \u003cem\u003eNew Phytologist\u003c/em\u003e, \u003cem\u003e235\u003c/em\u003e(3), 1272\u0026ndash;1286. https://doi.org/10.1111/nph.18172 \u003c/li\u003e\n\u003cli\u003eBridle, Jon R., and Timothy H. Vines. \u0026ldquo;Limits to Evolution at Range Margins: When and Why Does Adaptation Fail?\u0026rdquo; \u003cem\u003eTrends in Ecology \u0026amp; Evolution\u003c/em\u003e 22, no. 3 (March 2007): 140\u0026ndash;47. https://doi.org/10.1016/j.tree.2006.11.002.\u003c/li\u003e\n\u003cli\u003eBrook, B. W., Sodhi, N. S., \u0026amp; Bradshaw, C. J. A. (2008). Synergies among extinction drivers under global change. \u003cem\u003eTrends in Ecology \u0026amp; Evolution\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(8), 453\u0026ndash;460. https://doi.org/10.1016/j.tree.2008.03.011\u003c/li\u003e\n\u003cli\u003eBurnett, N. P., Badger, M. A., and Combes, S. A. (2021). Wind and route choice affect performance of bees flying above versus within a cluttered obstacle field. doi: 10.1101/2021.10.08.463704\u003c/li\u003e\n\u003cli\u003eBuchmann, S. L. (1994). Diversity and importance of native bees from the Arizona/Mexico Madrean archipelago. In L. F. DeBano, P. F. Ffolliott, A. Ortega-Rubio, G. J. Gottfried, R. H. Hamre, \u0026amp; C. B. Edminster (Eds.), \u003cem\u003eBiodiversity and management of the Madrean archipelago: The sky islands of southwestern United States and northwestern Mexico\u003c/em\u003e (pp. 301\u0026ndash;310). Fort Collins, CO: U. S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station\u003c/li\u003e\n\u003cli\u003eBurnett, Nicholas P., Marc A. Badger, and Stacey A. Combes. \u0026ldquo;Wind and Route Choice Affect Performance of Bees Flying above versus within a Cluttered Obstacle Field.\u0026rdquo; \u003cem\u003ePLOS ONE\u003c/em\u003e 17, no. 3 (March 24, 2022): e0265911. https://doi.org/10.1371/journal.pone.0265911.\u003c/li\u003e\n\u003cli\u003eBusch, J. W. \u0026ldquo;Inbreeding Depression in Self-Incompatible and Self-Compatible Populations of Leavenworthia Alabamica.\u0026rdquo; \u003cem\u003eHeredity\u003c/em\u003e 94, no. 2 (February 2005): 159\u0026ndash;65. https://doi.org/10.1038/sj.hdy.6800584.\u003c/li\u003e\n\u003cli\u003eByers, D. L., and D. M. Waller. \u0026ldquo;Do Plant Populations Purge Their Genetic Load? Effects of Population Size and Mating History on Inbreeding Depression.\u0026rdquo; \u003cem\u003eAnnual Review of Ecology, Evolution, and Systematics\u003c/em\u003e 30, no. Volume 30, 1999 (November 1, 1999): 479\u0026ndash;513. https://doi.org/10.1146/annurev.ecolsys.30.1.479.\u003c/li\u003e\n\u003cli\u003eCatchen, Julian, Paul A. Hohenlohe, Susan Bassham, Angel Amores, and William A. Cresko. \u0026ldquo;Stacks: An Analysis Tool Set for Population Genomics.\u0026rdquo; \u003cem\u003eMolecular Ecology\u003c/em\u003e 22, no. 11 (June 2013): 3124\u0026ndash;40. https://doi.org/10.1111/mec.12354.\u003c/li\u003e\n\u003cli\u003eCastillo, William J., Laura A. Burkle, and Carsten F. Dormann. \u0026ldquo;Dynamics of a Plant\u0026ndash;Pollinator Network: Extending the Bianconi\u0026ndash;Barab\u0026aacute;si Model.\u0026rdquo; \u003cem\u003eApplied Network Science\u003c/em\u003e 9, no. 1 (December 2024): 1\u0026ndash;12. https://doi.org/10.1007/s41109-024-00636-0.\u003c/li\u003e\n\u003cli\u003eCourchamp, F., T. Clutton-Brock, and B. Grenfell. \u0026ldquo;Inverse Density Dependence and the Allee Effect.\u0026rdquo; \u003cem\u003eTrends in Ecology \u0026amp; Evolution\u003c/em\u003e 14, no. 10 (October 1999): 405\u0026ndash;10. https://doi.org/10.1016/s0169-5347(99)01683-3.\u003c/li\u003e\n\u003cli\u003eChi, Katherine, and Brenda Molano-Flores. \u0026ldquo;Can Floral Display Size Compensate for Allee Effects Caused by Low Population Abundance and Density in Synthyris Bullii (Plantaginaceae), a Rare Species?\u0026rdquo; \u003cem\u003eAmerican Journal of Botany\u003c/em\u003e 101, no. 3 (2014): 428\u0026ndash;36. https://doi.org/10.3732/ajb.1300457.\u003c/li\u003e\n\u003cli\u003eCoates, D. J., Byrne, M., \u0026amp; Moritz, C. (2018). Genetic Diversity and Conservation Units: Dealing With the Species-Population Continuum in the Age of Genomics. \u003cem\u003eFrontiers in Ecology and Evolution\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e. https://doi.org/10.3389/fevo.2018.00165\u003c/li\u003e\n\u003cli\u003eCrawford, Julie. (2023). U.S. Fish and Wildlife Service. 2023. Beardless Chinchweed (Pectis imberbis) Final Recovery Plan. Southwest Region, Albuquerque, NM, USA. 37\u003c/li\u003e\n\u003cli\u003eDang, Zhenhua, Jiabin Li, Yanan Liu, Miaomiao Song, Peter J. Lockhart, Yunyun Tian, Miaomiao Niu, and Qinglang Wang. \u0026ldquo;RADseq-Based Population Genomic Analysis and Environmental Adaptation of Rare and Endangered Recretohalophyte Reaumuria Trigyna.\u0026rdquo; \u003cem\u003eThe Plant Genome\u003c/em\u003e n/a, no. n/a (n.d.): e20303. https://doi.org/10.1002/tpg2.20303.\u003c/li\u003e\n\u003cli\u003eDennis, Brian, Patricia L. Munholland, and J. Michael Scott. \u0026ldquo;Estimation of Growth and Extinction Parameters for Endangered Species.\u0026rdquo; \u003cem\u003eEcological Monographs\u003c/em\u003e 61, no. 2 (1991): 115\u0026ndash;43. https://doi.org/10.2307/1943004.\u003c/li\u003e\n\u003cli\u003eDibner, Reilly R., Megan L. DeMarche, Allison M. Louthan, and Daniel F. Doak. \u0026ldquo;Multiple Mechanisms Confer Stability to Isolated Populations of a Rare Endemic Plant.\u0026rdquo; \u003cem\u003eEcological Monographs\u003c/em\u003e 89, no. 2 (2019): e01360. https://doi.org/10.1002/ecm.1360.\u003c/li\u003e\n\u003cli\u003eEllstrand, Norman C., and Diane R. Elam. \u0026ldquo;Population Genetic Consequences of Small Population Size: Implications for Plant Conservation.\u0026rdquo; \u003cem\u003eAnnual Review of Ecology, Evolution, and Systematics\u003c/em\u003e 24, no. Volume 24, 1993 (November 1, 1993): 217\u0026ndash;42. https://doi.org/10.1146/annurev.es.24.110193.001245.\u003c/li\u003e\n\u003cli\u003eEvans, Simone A., Dennis F. Whigham, Ida Hartvig, and Melissa K. McCormick. \u0026ldquo;Hybridization in the Fringed Orchids: An Analysis of Species Boundaries in the Face of Gene Flow.\u0026rdquo; \u003cem\u003eDiversity\u003c/em\u003e 15, no. 3 (March 2023): 384. https://doi.org/10.3390/d15030384.\u003c/li\u003e\n\u003cli\u003eFaske, Trevor M., Alison C. Agneray, Joshua P. Jahner, Lana M. Sheta, Elizabeth A. Leger, and Thomas L. Parchman. \u0026ldquo;Genomic and Common Garden Approaches Yield Complementary Results for Quantifying Environmental Drivers of Local Adaptation in Rubber Rabbitbrush, a Foundational Great Basin Shrub.\u0026rdquo; \u003cem\u003eEvolutionary Applications\u003c/em\u003e 14, no. 12 (2021): 2881\u0026ndash;2900. https://doi.org/10.1111/eva.13323.\u003c/li\u003e\n\u003cli\u003eFilatov, D., Burke, S. DNA diversity in Hawaiian endemic plant \u003cem\u003eSchiedea globosa\u003c/em\u003e. \u003cem\u003eHeredity\u003c/em\u003e \u003cstrong\u003e92\u003c/strong\u003e, 452\u0026ndash;458 (2004). https://doi.org/10.1038/sj.hdy.6800440\u003c/li\u003e\n\u003cli\u003eFishbein, M. and P. Warren. 1994. Population studies of sensitive plants of the Lone Mountain Ecosystem Management Area, Coronado National Forest, Arizona.\u003c/li\u003e\n\u003cli\u003eFrankham, R., Ballou, J. D., Eldridge, M. D. B., Lacy, R. C., Ralls, K., Dudash, M. R., \u0026amp; Fenster, C. B. (2011). Predicting the Probability of Outbreeding Depression. \u003cem\u003eConservation Biology\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(3), 465\u0026ndash;475. https://doi.org/10.1111/j.1523-1739.2011.01662.x\u003c/li\u003e\n\u003cli\u003eFrye, Christopher T., and Maile C. Neel. \u0026ldquo;Benefits of Gene Flow Are Mediated by Individual Variability in Self-Compatibility in Small Isolated Populations of an Endemic Plant Species.\u0026rdquo; \u003cem\u003eEvolutionary Applications\u003c/em\u003e 10, no. 6 (2017): 551\u0026ndash;62. https://doi.org/10.1111/eva.12437\u003c/li\u003e\n\u003cli\u003eGathmann, Achim, and Teja Tscharntke. \u0026ldquo;Foraging Ranges of Solitary Bees.\u0026rdquo; \u003cem\u003eJournal of Animal Ecology\u003c/em\u003e 71 (September 6, 2002): 757\u0026ndash;64. https://doi.org/10.1046/j.1365-2656.2002.00641.x.\u003c/li\u003e\n\u003cli\u003eGamba, Diana, and Nathan Muchhala. \u0026ldquo;Pollinator Type Strongly Impacts Gene Flow within and among Plant Populations for Six Neotropical Species.\u0026rdquo; \u003cem\u003eEcology\u003c/em\u003e 104, no. 1 (2023): e3845. https://doi.org/10.1002/ecy.3845.\u003c/li\u003e\n\u003cli\u003eGentili, Rodolfo, Aldo Solari, Martin Diekmann, Cecilia Dupr\u0026egrave;, Gianna Serafina Monti, Stefano Armiraglio, Silvia Assini, and Sandra Citterio. \u0026ldquo;Genetic Differentiation, Local Adaptation and Phenotypic Plasticity in Fragmented Populations of a Rare Forest Herb.\u0026rdquo; \u003cem\u003ePeerJ\u003c/em\u003e 6 (June 13, 2018): e4929. https://doi.org/10.7717/peerj.4929.\u003c/li\u003e\n\u003cli\u003eGigant, R. L., A. De Bruyn, T. M\u0026rsquo;sa, G. Viscardi, L. Gigord, A. Gauvin-Bialecki, T. Pailler, L. Humeau, M. Grisoni, and P. Besse. \u0026ldquo;Combining Pollination Ecology and Fine-Scale Spatial Genetic Structure Analysis to Unravel the Reproductive Strategy of an Insular Threatened Orchid.\u0026rdquo; \u003cem\u003eSouth African Journal of Botany\u003c/em\u003e 105 (July 1, 2016): 25\u0026ndash;35. https://doi.org/10.1016/j.sajb.2016.02.205.\u003c/li\u003e\n\u003cli\u003eGoudet, J. and Jombart, T. (2020) Hierfstat: Estimation and Tests of Hierarchical F-Statistics. R Package Version 0.5-7.\u003cbr\u003ehttps://CRAN.R-project.org/package=hierfstat\u003c/li\u003e\n\u003cli\u003eGreenleaf, Sarah S., Neal M. Williams, Rachael Winfree, and Claire Kremen. \u0026ldquo;Bee Foraging Ranges and Their Relationship to Body Size.\u0026rdquo; \u003cem\u003eOecologia\u003c/em\u003e 153, no. 3 (September 2007): 589\u0026ndash;96. https://doi.org/10.1007/s00442-007-0752-9.\u003c/li\u003e\n\u003cli\u003eHannah, Lee. \u0026ldquo;Chapter 4 - Ecological, Evolutionary, and Biogeographic Implications of Climate Change.\u0026rdquo; In \u003cem\u003eClimate Change Biology (Third Edition)\u003c/em\u003e, edited by Lee Hannah, 77\u0026ndash;94. Academic Press, 2022. https://doi.org/10.1016/B978-0-08-102975-6.00004-2.\u003c/li\u003e\n\u003cli\u003eHamabata, T., Kinoshita, G., Kurita, K., Cao, P.-L., Ito, M., Murata, J., Komaki, Y., Isagi, Y., \u0026amp; Makino, T. (2019). Endangered island endemic plants have vulnerable genomes. \u003cem\u003eCommunications Biology\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(1), Article 1.\u003c/li\u003e\n\u003cli\u003eHartigan, J. A., and M. A. Wong. \u0026ldquo;Algorithm AS 136: A K-Means Clustering Algorithm.\u0026rdquo; \u003cem\u003eJournal of the Royal Statistical Society. Series C (Applied Statistics)\u003c/em\u003e 28, no. 1 (1979): 100\u0026ndash;108. https://doi.org/10.2307/2346830.\u003c/li\u003e\n\u003cli\u003eHartl, Daniel L., and A. G. Clark. \u0026ldquo;Principles of Population Genetics, 4\u003csup\u003eth\u003c/sup\u003e Edition\u0026rdquo;. \u003cem\u003eJournal of Heredity\u003c/em\u003e, Volume 98, Issue 4, July 2007, Page 382. (2007). https://doi.org/10.1093/jhered/esm035\u003c/li\u003e\n\u003cli\u003eHedrick, P. W., \u0026amp; Miller, P. S. (1992). Conservation genetics: techniques and fundamentals. Ecological Applications, 2(1), 30-46\u003c/li\u003e\n\u003cli\u003eHerman, Rachael W, Benjamin M Winger, Donna L Dittmann, and Michael G Harvey. \u0026ldquo;Fine-Scale Population Genetic Structure and Barriers to Gene Flow in a Widespread Seabird (Ardenna Pacifica).\u0026rdquo; \u003cem\u003eBiological Journal of the Linnean Society\u003c/em\u003e 137, no. 1 (September 1, 2022): 125\u0026ndash;36. https://doi.org/10.1093/biolinnean/blac091\u003c/li\u003e\n\u003cli\u003eHedrick, Philip W., and Philip S. Miller. \u0026ldquo;Conservation Genetics: Techniques and Fundamentals.\u0026rdquo; \u003cem\u003eEcological Applications\u003c/em\u003e 2, no. 1 (1992): 30\u0026ndash;46. https://doi.org/10.2307/1941887.\u003c/li\u003e\n\u003cli\u003eHofmann, Michaela M., Andreas Fleischmann, and Susanne S. Renner. \u0026ldquo;Foraging Distances in Six Species of Solitary Bees with Body Lengths of 6 to 15 Mm, Inferred from Individual Tagging, Suggest 150 m-Rule-of-Thumb for Flower Strip Distances.\u0026rdquo; \u003cem\u003eJournal of Hymenoptera Research\u003c/em\u003e 77 (June 29, 2020): 105\u0026ndash;17. https://doi.org/10.3897/jhr.77.51182.\u003c/li\u003e\n\u003cli\u003eHolderegger, Rolf, Niko Balkenhol, Janine Bolliger, Jan O. Engler, Felix Gugerli, Axel Hochkirch, Carsten Nowak, Gernot Segelbacher, Alex Widmer, and Frank E. Zachos. \u0026ldquo;Conservation Genetics: Linking Science with Practice.\u0026rdquo; \u003cem\u003eMolecular Ecology\u003c/em\u003e 28, no. 17 (September 2019): 3848\u0026ndash;56. https://doi.org/10.1111/mec.15202.\u003c/li\u003e\n\u003cli\u003eHonnay, Olivier, and Hans Jacquemyn. \u0026ldquo;Susceptibility of Common and Rare Plant Species to the Genetic Consequences of Habitat Fragmentation.\u0026rdquo; \u003cem\u003eConservation Biology\u003c/em\u003e 21, no. 3 (2007): 823\u0026ndash;31. https://doi.org/10.1111/j.1523-1739.2006.00646.x.\u003c/li\u003e\n\u003cli\u003eJombart, Thibaut, S\u0026eacute;bastien Devillard, and Fran\u0026ccedil;ois Balloux. \u0026ldquo;Discriminant Analysis of Principal Components: A New Method for the Analysis of Genetically Structured Populations.\u0026rdquo; \u003cem\u003eBMC Genetics\u003c/em\u003e 11, no. 1 (October 15, 2010): 94. https://doi.org/10.1186/1471-2156-11-94.\u003c/li\u003e\n\u003cli\u003eJones, Matthew R., Daniel E. Winkler, and Rob Massatti. \u0026ldquo;Demographic Modeling Informs Functional Connectivity and Management Interventions in Graham\u0026rsquo;s Beardtongue.\u0026rdquo; \u003cem\u003eConservation Genetics\u003c/em\u003e 22, no. 6 (December 1, 2021): 993\u0026ndash;1003. https://doi.org/10.1007/s10592-021-01392-9.\u003c/li\u003e\n\u003cli\u003eJump, Alistair S., and Josep Pe\u0026ntilde;uelas. \u0026ldquo;Running to Stand Still: Adaptation and the Response of Plants to Rapid Climate Change.\u0026rdquo; \u003cem\u003eEcology Letters\u003c/em\u003e 8, no. 9 (2005): 1010\u0026ndash;20. https://doi.org/10.1111/j.1461-0248.2005.00796.x.\u003c/li\u003e\n\u003cli\u003eKearns, C. A., \u0026amp; Inouye, D. W. (1993). \u003cem\u003eTechniques for pollination biologists\u003c/em\u003e\u003c/li\u003e\n\u003cli\u003eKhatri, Bhavin S., and Austin Burt. \u0026ldquo;Robust Estimation of Recent Effective Population Size from Number of Independent Origins in Soft Sweeps.\u0026rdquo; \u003cem\u003eMolecular Biology and Evolution\u003c/em\u003e 36, no. 9 (September 1, 2019): 2040\u0026ndash;52. https://doi.org/10.1093/molbev/msz081.\u003c/li\u003e\n\u003cli\u003eKling, M. M., and Ackerly, D. D. (2021). Global wind patterns shape genetic differentiation, asymmetric gene flow, and genetic diversity in trees. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 118, e2017317118. doi: 10.1073/pnas.2017317118\u003c/li\u003e\n\u003cli\u003eKwak, Manja M., Odilia Velterop, and Jelte van Andel. \u0026ldquo;Pollen and Gene Flow in Fragmented Habitats.\u0026rdquo; \u003cem\u003eApplied Vegetation Science\u003c/em\u003e 1, no. 1 (1998): 37\u0026ndash;54. https://doi.org/10.2307/1479084.\u003c/li\u003e\n\u003cli\u003eLaikre, Linda, Fred W. Allendorf, Laurel C. Aroner, C. Scott Baker, David P. Gregovich, Michael M. Hansen, Jennifer A. Jackson, et al. \u0026ldquo;Neglect of Genetic Diversity in Implementation of the Convention of Biological Diversity.\u0026rdquo; \u003cem\u003eConservation Biology\u003c/em\u003e 24, no. 1 (2010): 86\u0026ndash;88.\u003c/li\u003e\n\u003cli\u003eLande, Russell, and Susan Shannon. \u0026ldquo;The Role of Genetic Variation in Adaptation and Population Persistence in a Changing Environment.\u0026rdquo; \u003cem\u003eEvolution\u003c/em\u003e 50, no. 1 (1996): 434\u0026ndash;37. https://doi.org/10.2307/2410812.\u003c/li\u003e\n\u003cli\u003eLi, Yao, Amol C. Shetty, Chanthap Lon, Michele Spring, David L. Saunders, Mark M. Fukuda, Tran Tinh Hien, et al. \u0026ldquo;Detecting Geospatial Patterns of Plasmodium Falciparum Parasite Migration in Cambodia Using Optimized Estimated Effective Migration Surfaces.\u0026rdquo; \u003cem\u003eInternational Journal of Health Geographics\u003c/em\u003e 19, no. 1 (April 10, 2020): 13. https://doi.org/10.1186/s12942-020-00207-3.\u003c/li\u003e\n\u003cli\u003eLiu, Min, Stephen G. Compton, Fo-En Peng, Jian Zhang, and Xiao-Yong Chen. \u0026ldquo;Movements of Genes between Populations: Are Pollinators More Effective at Transferring Their Own or Plant Genetic Markers?\u0026rdquo; \u003cem\u003eProceedings of the Royal Society B: Biological Sciences\u003c/em\u003e 282, no. 1808 (June 7, 2015): 20150290. https://doi.org/10.1098/rspb.2015.0290.\u003c/li\u003e\n\u003cli\u003eLove, S. J., Schweitzer, J. A., Woolbright, S. A., \u0026amp; Bailey, J. K. (2023). Sky Islands Are a Global Tool for Predicting the Ecological and Evolutionary Consequences of Climate Change. \u003cem\u003eAnnual Review of Ecology, Evolution, and Systematics\u003c/em\u003e, \u003cem\u003e54\u003c/em\u003e(Volume 54, 2023), 219\u0026ndash;236. https://doi.org/10.1146/annurev-ecolsys-102221-050029\u003c/li\u003e\n\u003cli\u003eLuijten, Sheila H., Angelo Dierick, J. Gerard, B. Oostermeijer, L\u0026eacute;on E. L. Raijmann, and Hans C. M. Den Nijs. \u0026ldquo;Population Size, Genetic Variation, and Reproductive Success in a Rapidly Declining, Self-Incompatible Perennial (Arnica Montana) in The Netherlands.\u0026rdquo; \u003cem\u003eConservation Biology\u003c/em\u003e 14, no. 6 (2000): 1776\u0026ndash;87. https://doi.org/10.1111/j.1523-1739.2000.99345.x.\u003c/li\u003e\n\u003cli\u003eMcPherson, Guy R., and Jake F. Weltzin. \u0026ldquo;Disturbance and Climate Change in United States/Mexico Borderland Plant Communities: A State-of-the-Knowledge Review.\u0026rdquo; \u003cem\u003eGen. Tech. Rep. RMRS-GTR-50. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 24 p.\u003c/em\u003e 050 (2000). https://doi.org/10.2737/RMRS-GTR-50.\u003c/li\u003e\n\u003cli\u003eMhemmed, Gandour, Hessini Kamel, and Abdelly Chedly. \u0026ldquo;Does Habitat Fragmentation Reduce Genetic Diversity and Subpopulation Connectivity?\u0026rdquo; \u003cem\u003eEcography\u003c/em\u003e 31, no. 6 (2008): 751\u0026ndash;56. https://doi.org/10.1111/j.1600-0587.2008.05622.x.\u003c/li\u003e\n\u003cli\u003eMilano, Elizabeth R., Margaret R. Mulligan, Jon P. Rebman, and Amy G. Vandergast. \u0026ldquo;High-Throughput Sequencing Reveals Distinct Regional Genetic Structure among Remaining Populations of an Endangered Salt Marsh Plant in California.\u0026rdquo; \u003cem\u003eConservation Genetics\u003c/em\u003e 21, no. 3 (June 1, 2020): 547\u0026ndash;59. https://doi.org/10.1007/s10592-020-01269-3.\u003c/li\u003e\n\u003cli\u003eNazareno, Alison G., and L. Lacey Knowles. \u0026ldquo;There Is No \u0026lsquo;Rule of Thumb\u0026rsquo;: Genomic Filter Settings for a Small Plant Population to Obtain Unbiased Gene Flow Estimates.\u0026rdquo; \u003cem\u003eFrontiers in Plant Science\u003c/em\u003e 12 (October 14, 2021). https://doi.org/10.3389/fpls.2021.677009.\u003c/li\u003e\n\u003cli\u003eNic Lughadha, Eimear, Steven P. Bachman, Tarciso C. C. Le\u0026atilde;o, F\u0026eacute;lix Forest, John M. Halley, Justin Moat, Carmen Acedo, et al. \u0026ldquo;Extinction Risk and Threats to Plants and Fungi.\u0026rdquo; \u003cem\u003ePLANTS, PEOPLE, PLANET\u003c/em\u003e 2, no. 5 (2020): 389\u0026ndash;408. https://doi.org/10.1002/ppp3.10146.\u003c/li\u003e\n\u003cli\u003eNordstrom, S. W., Hufbauer, R. A., Olazcuaga, L., Durkee, L. F., \u0026amp; Melbourne, B. A. (2023). How density dependence, genetic erosion and the extinction vortex impact evolutionary rescue. \u003cem\u003eProceedings of the Royal Society B: Biological Sciences\u003c/em\u003e, \u003cem\u003e290\u003c/em\u003e(2011), 20231228. https://doi.org/10.1098/rspb.2023.1228\u003c/li\u003e\n\u003cli\u003eParis, Josephine R., Jamie R. Stevens, and Julian M. Catchen. \u0026ldquo;Lost in Parameter Space: A Road Map for Stacks.\u0026rdquo; \u003cem\u003eMethods in Ecology and Evolution\u003c/em\u003e 8, no. 10 (2017): 1360\u0026ndash;73. https://doi.org/10.1111/2041-210X.12775.\u003c/li\u003e\n\u003cli\u003ePauls, S. U., Nowak, C., B\u0026aacute;lint, M., \u0026amp; Pfenninger, M. (2013). The impact of global climate change on genetic diversity within populations and species. \u003cem\u003eMolecular Ecology\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(4), 925\u0026ndash;946. https://doi.org/10.1111/mec.12152\u003c/li\u003e\n\u003cli\u003ePetkova, Desislava, John Novembre, and Matthew Stephens. \u0026ldquo;Visualizing Spatial Population Structure with Estimated Effective Migration Surfaces.\u0026rdquo; \u003cem\u003eNature Genetics\u003c/em\u003e 48, no. 1 (January 2016): 94\u0026ndash;100. https://doi.org/10.1038/ng.3464.\u003c/li\u003e\n\u003cli\u003ePornon, Andr\u0026eacute;, Christophe Andalo, Monique Burrus, and Nathalie Escaravage. \u0026ldquo;DNA Metabarcoding Data Unveils Invisible Pollination Networks.\u0026rdquo; \u003cem\u003eScientific Reports\u003c/em\u003e 7, no. 1 (December 4, 2017): 16828. https://doi.org/10.1038/s41598-017-16785-5.\u003c/li\u003e\n\u003cli\u003ePruett, C. L., Stroupe, J., \u0026amp; Peterson, C. L. (2025). Habitat fragmentation influences the population genetics of a Florida endemic: Implications for recovery of a critically endangered plant. \u003cem\u003eConservation Genetics\u003c/em\u003e. https://doi.org/10.1007/s10592-025-01708-z\u003c/li\u003e\n\u003cli\u003eRavinet, M., Faria, R., Butlin, R. K., Galindo, J., Bierne, N., Rafajlović, M., Noor, M. A. F., Mehlig, B., \u0026amp; Westram, A. M. (2017). Interpreting the genomic landscape of speciation: A road map for finding barriers to gene flow. \u003cem\u003eJournal of Evolutionary Biology\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e(8), 1450\u0026ndash;1477. https://doi.org/10.1111/jeb.13047\u003c/li\u003e\n\u003cli\u003eRichardson, Andrew D., Trevor F. Keenan, Mirco Migliavacca, Youngryel Ryu, Oliver Sonnentag, and Michael Toomey. \u0026ldquo;Climate Change, Phenology, and Phenological Control of Vegetation Feedbacks to the Climate System.\u0026rdquo; \u003cem\u003eAgricultural and Forest Meteorology\u003c/em\u003e 169 (February 15, 2013): 156\u0026ndash;73. https://doi.org/10.1016/j.agrformet.2012.09.012.\u003c/li\u003e\n\u003cli\u003eSchierup, M. H., \u0026amp; Christiansen, F. B. (1996). Inbreeding depression and outbreeding depression in plants. \u003cem\u003eHeredity\u003c/em\u003e, \u003cem\u003e77\u003c/em\u003e(5), 461\u0026ndash;468. https://doi.org/10.1038/hdy.1996.172\u003c/li\u003e\n\u003cli\u003eSegelbacher, G., Cushman, S. A., Epperson, B. K., Fortin, M. J., Francois, O., Hardy, O. J., ... \u0026amp; Manel, S. (2010). Applications of landscape genetics in conservation biology: concepts and challenges. Conservation genetics, 11, 375-385.\u003c/li\u003e\n\u003cli\u003eSlatkin, Montgomery, and Naoyuki Takahata. \u0026ldquo;The Average Frequency of Private Alleles in a Partially Isolated Population.\u0026rdquo; \u003cem\u003eTheoretical Population Biology\u003c/em\u003e 28, no. 3 (December 1, 1985): 314\u0026ndash;31. https://doi.org/10.1016/0040-5809(85)90032-2.\u003c/li\u003e\n\u003cli\u003eSopniewski, Jarrod, and Renee A. Catullo. \u0026ldquo;Estimates of Heterozygosity from Single Nucleotide Polymorphism Markers Are Context-Dependent and Often Wrong.\u0026rdquo; \u003cem\u003eMolecular Ecology Resources\u003c/em\u003e 24, no. 4 (2024): e13947. https://doi.org/10.1111/1755-0998.13947\u003c/li\u003e\n\u003cli\u003eSouther, S., Lechowicz, M. J., \u0026amp; McGraw, J. B. (2012). Experimental test for adaptive differentiation of ginseng populations reveals complex response to temperature. \u003cem\u003eAnnals of Botany\u003c/em\u003e, \u003cem\u003e110\u003c/em\u003e(4), 829-837.\u003c/li\u003e\n\u003cli\u003eSouther, S., \u0026amp; McGraw, J. B. (2011). Evidence of local adaptation in the demographic response of American ginseng to interannual temperature variation. \u003cem\u003eConservation Biology\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(5), 922-931.\u003c/li\u003e\n\u003cli\u003eSouther, Sara, James B. McGraw, John D. Souther, and Donald M. Waller. \u0026quot;Effects of altered climates on American ginseng population dynamics.\u0026quot; \u003cem\u003ePopulation Ecology\u003c/em\u003e 64, no. 1 (2022): 47-63.\u003c/li\u003e\n\u003cli\u003eSouther, Sara, Martha Sample, Genevieve Conley, and Clare Aslan. \u0026ldquo;Demographic Analysis of the Endangered Plant Pectis Imberbis Highlights Tradeoffs between Deer Browse and Interspecific Competition.\u0026rdquo; \u003cem\u003eNatural Areas Journal\u003c/em\u003e 42, no. 3 (July 2022): 230\u0026ndash;41. https://doi.org/10.3375/22-3.\u003c/li\u003e\n\u003cli\u003eSouther, S., Sample, M., and Aslan, C. (2025). Interacting stressors drive landscape variation in demographic response of the endangered plant, Pectis imberbis (Gray). \u003cem\u003eJournal of Arid Environments\u003c/em\u003e in press.\u003c/li\u003e\n\u003cli\u003eStankowski, S., Sobel, J. M., \u0026amp; Streisfeld, M. A. (2015). The geography of divergence with gene flow facilitates multitrait adaptation and the evolution of pollinator isolation in Mimulus aurantiacus. \u003cem\u003eEvolution\u003c/em\u003e, \u003cem\u003e69\u003c/em\u003e(12), 3054\u0026ndash;3068. https://doi.org/10.1111/evo.12807\u003c/li\u003e\n\u003cli\u003eStephens, P. A., W. J. Sutherland, and R. P. Freckleton. 1999. What is the Allee effect? \u003cem\u003eOikos\u003c/em\u003e \u003cstrong\u003e87\u003c/strong\u003e: 185\u0026ndash;190.\u003c/li\u003e\n\u003cli\u003eTackenberg, Oliver, Peter Poschlod, and Susanne Bonn. \u0026ldquo;Assessment of Wind Dispersal Potential in Plant Species.\u0026rdquo; \u003cem\u003eEcological Monographs\u003c/em\u003e 73, no. 2 (2003): 191\u0026ndash;205. https://doi.org/10.1890/0012-9615(2003)073[0191:AOWDPI]2.0.CO;2.\u003c/li\u003e\n\u003cli\u003eTruelove, Nathan K., Kim Ley-Cooper, Iris Segura-Garc\u0026iacute;a, Patricia Briones-Fourz\u0026aacute;n, Enrique Lozano-\u0026Aacute;lvarez, Bruce F. Phillips, Stephen J. Box, and Richard F. Preziosi. \u0026ldquo;Genetic Analysis Reveals Temporal Population Structure in Caribbean Spiny Lobster (\u003cem\u003ePanulirus Argus\u003c/em\u003e) within Marine Protected Areas in Mexico.\u0026rdquo; \u003cem\u003eFisheries Research\u003c/em\u003e 172 (December 1, 2015): 44\u0026ndash;49. https://doi.org/10.1016/j.fishres.2015.05.029.\u003c/li\u003e\n\u003cli\u003eVillarreal, M. L., Haire, S. L., Iniguez, J. M., Cort\u0026eacute;s Monta\u0026ntilde;o, C., \u0026amp; Poitras, T. B. (2019). Distant neighbors: Recent wildfire patterns of the Madrean Sky Islands of southwestern United States and northwestern Mexico. \u003cem\u003eFire Ecology\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(1), 2. https://doi.org/10.1186/s42408-018-0012-x\u003c/li\u003e\n\u003cli\u003eWarzecha, Daniela, Tim Diek\u0026ouml;tter, Volkmar Wolters, and Frank Jauker. \u0026ldquo;Intraspecific Body Size Increases with Habitat Fragmentation in Wild Bee Pollinators.\u0026rdquo; \u003cem\u003eLandscape Ecology\u003c/em\u003e 31, no. 7 (September 1, 2016): 1449\u0026ndash;55. https://doi.org/10.1007/s10980-016-0349-y.\u003c/li\u003e\n\u003cli\u003eWang, Z.-F., Lian, J.-Y., Ye, W.-H., Cao, H.-L., Zhang, Q.-M., and Wang, Z.-M. (2016). Pollen and seed flow under different predominant winds in wind-pollinated and wind-dispersed species Engelhardia roxburghiana. \u003cem\u003eTree Genetics \u0026amp; Genomes\u003c/em\u003e 12, 19. doi: 10.1007/s11295-016-0973-3\u003c/li\u003e\n\u003cli\u003eWilli, Yvonne, Marco Fracassetti, Olivier Bachmann, and Josh Van Buskirk. \u0026ldquo;Demographic Processes Linked to Genetic Diversity and Positive Selection across a Species\u0026rsquo; Range.\u0026rdquo; \u003cem\u003ePlant Communications\u003c/em\u003e 1, no. 6 (November 2020): 100111. https://doi.org/10.1016/j.xplc.2020.100111.\u003c/li\u003e\n\u003cli\u003eWilson, N.R., 2024, Species Distribution Models for Pectis imberbis, a Rare Plant Species in Southeastern Arizona: U.S. Geological Survey data release, https://doi.org/10.5066/P13VMRBC.\u003c/li\u003e\n\u003cli\u003eWright, S. \u0026ldquo;The Genetical Structure of Populations.\u0026rdquo; \u003cem\u003eAnnals of Eugenics\u003c/em\u003e 15, no. 4 (March 1951): 323\u0026ndash;54. https://doi.org/10.1111/j.1469-1809.1949.tb02451.x.\u003c/li\u003e\n\u003cli\u003eWright, Sewall. \u0026ldquo;The Interpretation of Population Structure by F-Statistics with Special Regard to Systems of Mating.\u0026rdquo; \u003cem\u003eEvolution\u003c/em\u003e 19, no. 3 (1965): 395\u0026ndash;420. https://doi.org/10.2307/2406450.\u003c/li\u003e\n\u003cli\u003eYanahan, A. D., \u0026amp; Moore, W. (2019). Impacts of 21st-century climate change on montane habitat in the Madrean Sky Island Archipelago. \u003cem\u003eDiversity and Distributions\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(10), 1625\u0026ndash;1638. https://doi.org/10.1111/ddi.12965\u003c/li\u003e\n\u003cli\u003eZhou, Chengchuan, Shiqi Xia, Qiang Wen, Ying Song, Quanquan Jia, Tian Wang, Liting Liu, and Tianlin Ouyang. \u0026ldquo;Genetic Structure of an Endangered Species Ormosia Henryi in Southern China, and Implications for Conservation.\u0026rdquo; \u003cem\u003eBMC Plant Biology\u003c/em\u003e 23, no. 1 (April 26, 2023): 220. https://doi.org/10.1186/s12870-023-04231-w.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Conservation Genetic, Population Genetics – Empirical, Population Ecology, Habitat Degradation, Quantitative Genetics, Climate Change","lastPublishedDoi":"10.21203/rs.3.rs-7022625/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7022625/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe conservation of rare and endangered plants requires understanding of genetic diversity and connectivity to mitigate the effects of habitat fragmentation and environmental change. \u003cem\u003ePectis imberbis\u003c/em\u003e (Gray), a perennial herb endemic to the Arizona Sky Islands, is listed as endangered due to recent extirpations, population declines, habitat loss, and restricted range. To address critical knowledge gaps in \u003cem\u003eP. imberbis\u003c/em\u003e conservation, we assessed population structure, genetic diversity, and connectivity across its range using single nucleotide polymorphisms (SNPs) generated through RADseq.\u0026nbsp;We identified three genetically distinct population clusters, with limited gene flow among populations located in the Huachuca, Santa Rita, and Atascosa-Pajarito Mountain ranges. Estimated effective migration surfaces revealed barriers to gene flow, particularly around Montezuma Pass and the Patagonia Mountains, which corresponded with demographic declines and recent extirpations. Pollinator visitation and floral network analyses showed consistent overlap of key pollinator taxa across populations but suggested limited pollen transfer over large distances. These findings highlight the need for targeted restoration efforts to enhance genetic connectivity, such as establishing stepping-stone populations in regions of limited migration. Future research should focus on testing adaptive variation to guide restoration actions taken to increase connectivity. By integrating genetic, demographic, and pollinator data, this work directly informs \u003cem\u003eP. imberbis\u003c/em\u003e conservation, and more generally, contributes to understanding of rare species conservation in fragmented landscapes.\u003c/p\u003e","manuscriptTitle":"Population structure and genetic connectivity in the endangered Pectis imberbis: addressing conservation and genetic gaps in the Arizona Sky Islands","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-21 12:48:42","doi":"10.21203/rs.3.rs-7022625/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"82b1dc98-5e0c-40a1-a988-b5dfaac7cf5a","owner":[],"postedDate":"July 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-09-23T08:53:52+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-21 12:48:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7022625","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7022625","identity":"rs-7022625","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}