The Importance of Landscape Structure for the Long-Term Conservation of Species
The presence of corridors and the way patches are arranged in a landscape are thought to be important for the long-term conservation of many species, and may determine whether species are able to recover from large-scale disturbances. Here we used individual-based models to investigate how population recovery was affected by landscape structure for four species in an agricultural landscape: skylark (Alauda arvensis), vole (Microtus agrestis), a ground beetle (Bimbidion lampros) and a linyphiid spider (Erigone atra). We characterized population persistence based on equilibrium population sizes (K) and the time it took populations to recover from perturbations. We separated the effects of corridors and patch arrangement by comparing results from a real landscape with results from two virtual landscapes: One where linear corridors were removed by homogenizing patch shapes, while leaving the spatial arrangement of the patches unaltered, and one where patches were shuffled around!
, while still leaving the landscape composition unaltered. Patch arrangement and the presence of corridors had a large effect on population dynamics for species whose local success depended on the identity of the neighbouring patches. The short-dispersing beetle and vole recovered slowly from perturbations in landscapes where they had low K. Our study demonstrates that it is necessary to consider the dynamics of populations in a spatially explicit context when designing landscapes for conservation of species. Poster presented at the ICCB conference in Edmonton, 2010
Published on: Mar 3, 2016
Transcripts - The Importance of Landscape Structure for the Long-Term Conservation of Species
Jacob Nabe-Nielsen1,2, Richard M. Sibly2,3, Mads C. Forchhammer1,2, Valery E. Forbes2,4 1 NERI, Aarhus University, 2 CIPE, 3 University of Reading, 4 Roskilde UniversityIntroduction Beetle Vole Skylark Spider Fig. 2. Population recovery after 13 11 10 12The arrangement of patches in a landscape and the degree removing 95% of the individualsto which they are connected by corridors are thought to be (real landscape). Population 12 10 numbers (N) were counted yearly 9 11important for the long-term conservation of many species, 9 for 170 years. Perturbations tookand may determine whether they are able to recover from Loge (N) 11 place every 17 years.large-scale disturbances [1,2]. 8 8 10 10 TimesIn fragmented landscapes subpopulations are isolated and 7 perturbedless likely to be maintained through continuous immigra- 7 9 1 6 9 2 7 6tion. Recolonization of empty habitat patches may also 3 8 4 9take longer, resulting in lower population sizes on the 8 5 6 8 5 10landscape scale, especially for short-dispersing species. 0 5 10 15 0 5 10 15 0 5 10 15 0 5 10 15 Years since perturbation Main road Fig. 3. Equilibrium population Fig 1. Model landscapes. Roadside verge sizes (K) and return times (φ) in Beetle Vole Skylark Spider Hedge 4 4 3 2 Permanent grass landscapes of decreasing Unmanaged grassland complexity (letters on x-axes 3 Rotational field (all crops) Coniferous forest correspond to Fig. 1). Error bars are 95% confidence intervals. 3 2 2 1 φ Deciduous forest 1 2 0 1 0 14 13 12 13 12 9 Loge (K) 12 11 11 10 8 10 Unperturbed Perturbed 80 pct. Perturbed 95 pct. 9 9 11 A B C D A B C D A B C D A B C D Methods Results and conclusions Here we use individual-based population models (IBMs) to evaluate to what The short-dispersing species (beetle and vole) recovered slowly from perturbations in extent animal population recovery after disturbances is affected by land- landscapes where they had low K (Fig. 3). Beetles had high K in landscapes where over- scape structure. In our models fluctuations in population sizes emerge as a wintering habitats (field boundaries) were located close to summer habitats (fields). Voles result of realistic site-specific behavior of individual animals and their inter- had high K when their key habitat (unmanaged grass) was abundant, especially if the actions with each other. edge lengths in these habitats were relatively short. We simulated population dynamics in four 10×10 km landscapes (Fig. 1): Skylark had low K in landscapes with randomized patch arrangement (C). Here forest (A) real landscape; (B) landscape without corridors; (C) landscape with patches are dispersed throughout the landscape, and skylarks avoid nesting close to trees. randomized patch arrangement; (D) landscape with randomized patch sizes. This study underlines the need to study population dynamics in a spatial context, and to Fragmentation increased in A–C . explicitly include all relevant habitat types. This is most easily done with mechanistically We selected four species with contrasting life-histories in order to study if rich IBMs. population recovery was related to dispersal. Recovery to equilibrium popu- lation size (K) after a perturbation was measured using the return time (φ), References which was calculated from the shape of the logistic curve fitted to popula- 1 Nabe-Nielsen, J., Sibly, R.M., Forchhammer, M.C., Forbes, V.E., Topping, C.J., 2010. PLoS ONE 5, e8932. tion size vs. number of years since perturbation (Fig. 2). 2 Vos, C.C., Verboom, J., Opdam, P.F.M., Ter Braak, C.J.F., 2001. Am. Nat. 157, 24-41. 3 Fahrig, L., 2003. Ann. Rev. Ecol. Evol. Syst. 34, 487-515. NATIONAL ENVIRONMENTAL RESEARCH INSTITUTE Background paper: doi:10.1371/journal.pone.0008932 AARHUS UNIVERSITY Contact: email@example.com