The Importance of Landscape Structure for the Long-Term Conservation of Species
0 likes268 views
This study uses individual-based population models to evaluate how landscape structure affects animal population recovery after large-scale disturbances. Four model landscapes with varying fragmentation were simulated. Population dynamics of four species with different dispersal abilities - beetle, vole, skylark, spider - were modeled in the landscapes. Recovery to pre-disturbance population sizes and return times were measured and found to be slower for short-dispersing species in fragmented landscapes where their habitat was less abundant and more isolated. The results underline the importance of including spatial context and all relevant habitat types when studying population dynamics.
![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 University
Introduction Beetle Vole Skylark Spider
Fig. 2. Population recovery after 13 11 10 12
The arrangement of patches in a landscape and the degree removing 95% of the individuals
to which they are connected by corridors are thought to be (real landscape). Population 12
10
numbers (N) were counted yearly 9 11
important for the long-term conservation of many species, 9
for 170 years. Perturbations took
and 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 Times
In fragmented landscapes subpopulations are isolated and 7 perturbed
less likely to be maintained through continuous immigra- 7 9 1 6
9 2 7
6
tion. Recolonization of empty habitat patches may also 3 8
4 9
take longer, resulting in lower population sizes on the 8 5 6 8 5 10
landscape 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 () 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
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 1010 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 AC [3]. 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 (), 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: nabe@dmu.dk](https://image.slidesharecdn.com/nabe-nielsenetal2010iccbposter-130225235133-phpapp01/85/The-Importance-of-Landscape-Structure-for-the-Long-Term-Conservation-of-Species-1-320.jpg)
The Importance of Landscape Structure for the Long-Term Conservation of Species
- 1. 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 University Introduction Beetle Vole Skylark Spider Fig. 2. Population recovery after 13 11 10 12 The arrangement of patches in a landscape and the degree removing 95% of the individuals to which they are connected by corridors are thought to be (real landscape). Population 12 10 numbers (N) were counted yearly 9 11 important for the long-term conservation of many species, 9 for 170 years. Perturbations took and 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 Times In fragmented landscapes subpopulations are isolated and 7 perturbed less likely to be maintained through continuous immigra- 7 9 1 6 9 2 7 6 tion. Recolonization of empty habitat patches may also 3 8 4 9 take longer, resulting in lower population sizes on the 8 5 6 8 5 10 landscape 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 () 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 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 1010 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 AC [3]. 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 (), 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: nabe@dmu.dk