Habitat fragmentation can adversely impact plant populations through its impact on various demographic mechanisms (including fecundity and recruitment), yet individuals vary in their responses to fragmentation. However, it is unclear how landscape mechanisms may govern why some individuals become demographically important, i.e., contribute the majority of new recruits to the population. Individual responses to fragmentation may vary based on particular fragmentation factors (edge proximity, patch isolation, edge:area) and/or local environmental conditions. Untangling these processes requires experiments, given the co-varying nature of fragmentation effects (e.g., high edge:area patches may also have more individuals near edges) and how local environmental conditions may be governed by or independent of fragmentation effects.
We explored how fragmentation and local environmental conditions affect recruitment of six plant species. These were transplanted into seven replicated experimentally-fragmented landscapes designed to independently test effects of edge proximity, isolation, and edge:area. Additionally, at each of 448 transplant plots, we surveyed recruitment microsite availability, canopy cover, soil water holding capacity (WHC), parental performance (number of reproductive structures), and number of recruits from each transplanted plant. We used structural equation modelling to examine which effects best predicted individual plant performance. We hypothesized that recruitment is positively correlated with distance from edge, WHC, microsite availability, and parental performance and negatively correlated with increased edge:area, patch isolation, and canopy cover.
Results/Conclusions
Our models explained 35-60% of the variation in recruitment across our six species (partial R2). We found that edge:area was not significant, while isolation and canopy cover significantly negatively impacted 1/6 species. Distance from edge showed a significantly negative relationship with recruitment for 2/6 species. However, distance from edge often had a significant indirect effect on recruitment (4/6 species), through its relationship with factors, such as canopy cover. WHC significantly explained recruit number for 2/6 species, but the relationship direction was species-specific. Parental performance was significantly positively correlated with recruitment in 5/6 species, and was itself influenced by canopy cover and edge proximity. Surprisingly, microsite availability did not significantly affect plant recruitment, likely because microsite availability is structured by proximity to the edge of the patch. Overall, among our three fragmentation effects, edge proximity most strongly affected plant recruitment. Local environmental conditions were also often important, with these effects sometimes structured by edge proximity and other times unrelated to fragmentation. Our results emphasize the importance of understanding both landscape and local effects on plant recruitment when managing populations at the landscape scale.