Evidence of pollinator limitation and inbreeding effects in small, warm-margin distribution limit populations of a northern forest herb, Clintonia borealis (Liliaceae)

Background/Question/Methods

Climate change is predicted to have major effects on the distributions of plant and animal species.  However, most research efforts to date have been focused on poleward range expansions or movement to higher elevations at the leading edge of species’ distributions.  In contrast, range contractions and population extinctions along species’ warm-margin distribution limits have received less attention, despite the expectation that these areas may be the principal sites of biodiversity loss in coming decades.  To begin to address these issues, this study investigated the roles of abiotic and biotic factors along an elevational and climatic gradient associated with the warm-margin distribution limit of a northern forest herb, Clintonia borealis, in southern New England.  Fifteen populations of C. borealis were surveyed for performance across a 1000 m regional elevational gradient in western Massachusetts.  Additionally, a supplemental pollination experiment including open-pollinated controls and treatments with pollen from within vs. beyond local populations was conducted at four sites, two at high elevation and two at low elevation, to test for evidence of pollen limitation and inbreeding near the species’ warm margin distribution limit.

Results/Conclusions

In the pollen supplementation experiment, plants in high elevation populations produced more flowers per plant than those at low elevation (p<0.001).  The percentage of flowers that produced mature fruits also varied significantly between the two elevation zones (p<0.001) and among the pollination treatments (p < 0.05).  In particular, low elevation plants receiving pollen supplementation produced a higher percentage of fruits than open-pollinated controls (p<0.05).  In contrast, fruit production was not influenced by these treatments at high elevation sites.  Seed mass varied significantly between elevation zones (p<0.001) and between pollination treatments (p<0.001): low elevation plants treated with pollen from outside the local population produced the largest seeds (p<0.05).   In the broader population survey, we documented that low elevation populations were restricted to mesic or wetland forest habitats, while the species had a more general distribution in forests at higher elevations, highlighting the importance of abiotic factors.  Overall, our results indicate that reproduction in low elevation, warm-margin distribution limit populations of C. borealis is substantially affected by biotic factors, including pollen limitation and inbreeding, in addition to underlying effects of the abiotic environment.