Network topology of stable isotope interactions in a sub-arctic raptor guild

Predation is an ecologically important process, and intra-guild interactions may substantially influence the ecological effects of predator species. Despite a rapid expansion in the use of mathematical graph theory to describe trophic relations, network approaches have rarely been used to study interactions within predator assemblages. Assemblages of diurnal raptors are subject to substantial intra- and interspecific competition. Here we used the novel approach of applying analyzes based on network topology to species-specific data on the stable isotopes (13)C and (15)N in feathers to evaluate patterns of relative resource utilization within a guild of diurnal raptors in northern Sweden. Our guild consisted of the golden eagle (Aquila chrysaetos), the gyrfalcon (Falco rusticolus), the peregrine falcon (Falco peregrinus) and the rough-legged buzzard (Buteo lagopus). We found a modular trophic interaction structure within the guild, but the interactions were less nested than expected by chance. These results suggest low redundancy and hence a strong ecological importance of individual species. Our data also suggested that species were less connected through intra-guild interactions than expected by chance. We interpret our results as a convergence on specific isotope niches, and that body size and different hunting behaviour may mediate competition within these niches. We finally highlight that generalist predators could be ecologically important by linking specialist predator species with disparate dietary niches.

Recreational trails as a source of negative impacts on the persistence of keystone species and facilitation

Hiking trails, which are among the most common forms of infrastructure created for nature-based tourism, can alter key ecological processes. Trails can damage plants that facilitate the establishment and growth of other species leading to changes in community and functional composition. This can be a particular concern in harsh alpine ecosystems where plant communities are often dominated by one or two keystone species that provide shelter to a suite of beneficiary species. We analysed how a hiking trail affects interspecific facilitation by a dominant trampling-sensitive nurse shrub in the highest National Park in Australia. First we assessed the effects of the trail on the abundance, size and density of the nurse shrub at different distances from the trail. We then compared species richness and composition between areas in, and out, of the nurse shrub's canopy at different distances from the trail. To better understand why some species may benefit from facilitation and any effects of the trail on the quality of facilitation we compared functional composition between quadrats using community trait weighted means calculated by combining plant composition with species functional traits (canopy height, leaf area, % dry weight of leaves and specific leaf area). The abundance, size and density of nurse shrubs was lower on the trail edges than further away, particularly on the leeward edge, where there was more bare ground and less shrub cover. There were differences in species richness, cover, composition and functional composition in and outside the nurse shrub canopy. The shrubs appeared to facilitate species with more competitive, but less stress tolerant traits (e.g. taller plants with leaves that were larger, had high specific leaf area and low dry matter content). However, despite reductions in nurse shrubs near the trail, where they do exist, they appear to provide the same 'quality' of facilitation as nurse shrubs further away. However, longer-term effects may be occurring as the loss of nurse shrubs alters the wind profile of the ridgeline and therefore succession. The use of a steel mesh walkway along the trail may facilitate the regeneration of nurse shrubs and other plants that require protection from wind. Our results highlight the importance of diversifying recreation ecology research to assess how trails affect important ecological processes.

Multivariate product-shot-noise Cox point process models

We introduce a new multivariate product-shot-noise Cox process which is useful for modeling multi-species spatial point patterns with clustering intra-specific interactions and neutral, negative, or positive inter-specific interactions. The auto- and cross-pair correlation functions of the process can be obtained in closed analytical forms and approximate simulation of the process is straightforward. We use the proposed process to model interactions within and among five tree species in the Barro Colorado Island plot.