Declining demographic performance and dispersal limitation influence the geographic distribution of the perennial forb Astragalus utahensis (Fabaceae)

Abstract

1.A central goal of ecology is understanding the determinants of species' distributions. "Metapopulation" models for the existence of distributional boundaries predict that species' geographic ranges arise from the landscape-scale deterioration of habitat suitability towards the range edge (i.e., niche mechanisms), which simultaneously hinders demographic performance and limits dispersal to suitable habitat beyond the edge (i.e., dispersal limitation). However, few studies have examined both of these mechanisms for the same species by examining abundance and comprehensive measures of demographic performance across the distribution and beyond its boundary.2. We tested the predictions of metapopulation models for range limits by contrasting abundance, demographic performance, and population growth (λ) of the perennial forb Astragalus utahensis in central and northern range edge populations.We tested for dispersal limitation by transplanting individuals at and beyond the northern range boundary and monitoring their demographic performance.3. Astragalus abundance and stochastic λ decreased from the range centre to the northern range edge, with stochastic λ falling to or below replacement in range edge populations. However, transplants at some sites beyond the northern range edge survived and reproduced at levels similar to transplants within the range.Thus, in addition to deteriorating conditions at the range edge, dispersal limitation appears to contribute to limiting A. utahensis' northern distribution. Synthesis.Our results support metapopulation models for range limits by suggesting that decreased demographic performance constrains the viability of range edge populations while contributing to dispersal limitation of Astragalus utahensis' distribution. Thus, the tandem action of niche and dispersal processes appears to play an important role in constraining this species' northern latitudinal range. K E Y W O R D Sabundant centre hypothesis, Astragalus utahensis, biogeography and macroecology, dispersal limitation, integral projection model, metapopulation model, niche limit, population growth, range limit, transplant

Ecological niche models display nonlinear relationships with abundance and demographic performance across the latitudinal distribution of Astragalus utahensis (Fabaceae)

“…To do this, we placed multiple transects at randomly selected locations through each study population and quantified the density of A. utahensis in 1 m 2 quadrats placed every 2 m on alternating sides of each transect. We meas- (Baer & Maron, 2019). We also esti- precipitation in the driest month (bio14) was the most important predictor of suitability in the ensemble model, followed in importance by temperature seasonality (bio4), with a general pattern of higher suitability in areas of higher annual variation in temperature ( Figure S2.1).…”

“…Few of these potential explanations for the lack of a suitabilityabundance relationship apply to A. utahensis. For example, while A. utahensis is dispersal limited beyond its northern range boundary (Baer & Maron, 2019), this does not explain the pattern of low density in populations of intermediate suitability that yielded a J-shaped rather than linear suitability-abundance relationship. Moreover, A. utahensis generally grows in sparsely vegetated areas where negative intraspecific density dependence is rare or absent, although interspecific competition may impact density by excluding the species from more benign microsites in some populations (K.C.…”

“…Our sampling frame for the construction of ENMs was a polygonal convex hull that buffered presence points by 100 km. We used a sampling frame larger than the distribution of presence records because a previous study found that habitat suitable for the presence of A. utahensis exists more than 200 km beyond the northern edge of its current distribution (Baer & Maron, 2019).…”

“…The ensemble model's specificity value indicates an absence was predicted where a presence had been recorded in roughly twelve percent of cases, meaning that it may often underestimate suitability for this species. Furthermore, evidence of dispersal limitation in A. utahensis (Baer & Maron, 2019) suggests that the current distribution may not fully reflect the range of environmental conditions suitable for its presence and that estimates of suitability based upon climatic conditions within the current distribution may be somewhat biased.…”

“…The J-shaped relationship between λ s and the suitability of central and northern edge study populations appears attributable to a combination of similar relationships in cumulative recruitment and mean annual survival and a positive linear relationship between recruitment rates from the seed bank and suitability. We previously found that significant declines in λ s between central and northern edge populations were primarily attributable to three demographic rates, of which germination rates from the seed bank and total seed production at the population scale were two (Baer & Maron, 2019). While per capita seed production was not significantly correlated with the suitability of central and northern edge study sites, reproductive plant density was positively associated with suitability in those sites, which likely led to increased total seed production at the population scale in more suitable populations (K.C.…”