Distance to seed sources and land‐use history affect forest development over a long‐term heathland to forest succession

Abstract

Questions: Is there a spatial pattern in the community structure (stem densities, species richness and species composition) of trees and shrubs during more than 100yr of heathland to forest succession? To what extent is community structure influenced by land-use history and distance to seed sources? Do these effects change in time? Location: A 350-ha heathland (NOrholm) in southwest Denmark was abandoned in 1895 and left for free succession. Prior to abandonment the heathland was under traditional management for centuries. Method: Trees and shrubs were recorded and measured in ten surveys spanning 91yr (1921-2012). In the first nine surveys, complete censuses were used, whereas 116 randomly placed plots (10-m radius) were used in the most recent survey. We used mixed models and different multivariate techniques (non-metric multidimensional scaling and permutational multivariate ANOVA) to analyse spatio-temporal patterns in stem densities, species richness and species composition and the effects of land-use history and distance to seed sources. Results: Tree and shrub densities increased exponentially over time and were consistently lower at longer distance from seed sources. Land-use history affected the rate of increase in tree densities, with lower encroachment rates at previously cultivated soils. Initially, species richness increased rapidly but leveled off in later successional stages. Richness was marginally higher at areas closer to seed sources, whereas it was not affected by land-use history. The succession showed a shift from early to mid-successional species over time; but distance to seed source determined which species were colonizing at early stages. Conclusion: Distance to seed source and land-use history can differentially affect the structure and spatial patterns of developing forest communities. Although both factors had long-lasting effects on rates of colonization, the spatial patterns of colonization and species composition were mainly determined by distance to seed sources. The importance of distance to seed sources became less over time, suggesting that dispersal is a stronger driver at early stages. After more than a century since abandonment, forest covers <30% of the area, indicating that lack of heathland management did not strongly promote tree colonization at this site

Interactive effects of land‐use history, tree encroachment and distance to edge on species richness in an unmanaged heathland

“…The same effect was found in a successional study of this area, demonstrating that species richness and composition of trees were not in equilibrium with soil conditions (Kepfer‐Rojas et al. ). We cannot rule out, however, that this could be an effect mediated by the grass cover.…”

“…However, previous work on this heathland attributed this negative effect to suppression of tree colonization by increased grass cover (particularly Deschampsia flexuosa ), mediated by nutrient availability (Kepfer‐Rojas et al. , ), which highlights the importance of the abiotic environment in regulating biotic interactions.…”

“…The rationale for this is that distance to external seed sources is the main determinant of tree colonization rates in this heathland (Kepfer‐Rojas et al. ), and several studies have demonstrated that distance to edge is a good proxy for colonization rates, an important determinant of vegetation structure (Bartha et al. ; Matsumura & Takeda ; Knappová et al.…”

Dating lava flows of tropical volcanoes by means of spatial modeling of vegetation recovery

“…Distance to lava flow edge is considered as a proxy parameter for the proximity to seed sources, which is important for the start of vegetation recovery (Kepfer‐Rojas et al ., ). In general, NDVI is lower at longer distances from the lava flow edge (Figures and ); this could be explained by the fact that it is more difficult for seeds to reach bare lava surfaces – the number of seeds in a location decreases with the distance from the source (Howe and Smallwood, ; McClanahan, ; Nathan and Muller‐Landau, ).…”

Do not neglect surroundings in restoration of disturbed sites

“…If a population of some species establishes in a restored site, then it can become a part of a metapopulation with possible continuous interactions with the surroundings (Hanski ). In the case of vascular plants in some terrestrial habitats, it was shown that proximity (up to 100 m) to populations of target plant species dictated the degree to which restoration occurred spontaneously (del Moral et al ; Řehounková & Prach ), but some species may arrive from much longer distances (Kirmer et al ; Kepfer‐Rojas et al ).…”