Predicting when animal populations are at risk from roads: an interactive model of road avoidance behavior
Introduction
Roads and other types of traffic lines are common occurrences everywhere humans have settled, and it is now becoming widely accepted that roads affect many aspects of ecosystems (Oxley et al., 1974, Institut für Naturschutz und Tierökologie, 1977, Ellenberg et al., 1981, Reck and Kaule, 1993, Glitzner et al., 1999, Trombulak and Frissell, 2000, Holzgang et al., 2000, Underhill and Angold, 2000, Carr et al., 2002, Spellerberg, 2002, Forman et al., 2003). For example, Forman (2000) estimates that the system of public roads affects ecologically about one-fifth of the United States land area. Based on results reported in the aforementioned literature, we infer that roads and their associated vehicular traffic affect persistence of wildlife populations in four main ways: (1) habitat loss; (2) traffic mortality; (3) resource inaccessibility; and (4) population subdivision (Fig. 1).
Habitat loss can be direct, where habitat is removed to build roads and their verges, or indirect, where habitat quality close to roads is reduced due to emissions from traffic (e.g., noise, light, pollutants). Reproduction is interrupted in areas of habitat destruction; furthermore, reproductive rates are likely reduced and mortality rates increased in lower quality habitat close to roads, leading to lowered chances of population persistence (Brody and Pelton, 1989, Reijnen and Foppen, 1994, Reijnen et al., 1995, Reijnen et al., 1996, Ortega and Capen, 1999, Forman et al., 2003, pp. 123–126).
Traffic mortality is due to collisions of individuals with vehicles on the road. If a significant proportion of a population is killed on roads, and this increased mortality is not compensated by higher birth rates, population persistence can be compromised (Fuller, 1989, Bangs et al., 1989, Andrews, 1990, Newton et al., 1991, van der Zee et al., 1992, Ferreras et al., 1992, Fahrig et al., 1995, Mumme et al., 2000, Hels and Buchwald, 2001, Gibbs and Shriver, 2002). In addition, traffic mortality contributes to population subdivision by reducing the flow of individuals between subpopulations separated by roads (Swihart and Slade, 1984, Reh and Seitz, 1990, Baker, 1998, Gerlach and Musolf, 2000, Keller and Largiadèr, 2003).
For some species, roads can act as barriers to movement and lead to resource inaccessibility. Individuals that do not cross roads cannot access resources such as food, mates, and breeding sites on the other side. Reduced access to such complementary or supplementary resources can lead to lower reproductive and survival rates, which in turn may reduce population persistence (Oxley et al., 1974, Mader, 1984, Mader et al., 1990, Dunning et al., 1992, Weidemann and Reich, 1995, Noss et al., 1996, Vos and Chardon, 1998, Clark et al., 2001). In addition, movement barriers contribute to population subdivision by reducing the flow of individuals between subpopulations separated by roads.
Population subdivision occurs when populations become separated into smaller, isolated subpopulations. As mentioned above, both traffic mortality and resource inaccessibility contribute to population subdivision. Populations living in habitat surrounded by roads are less likely to receive immigrants from other habitats, and thus may suffer from lack of genetic input and inbreeding. An increase in genetic defects may lower the probability of population persistence. Moreover, small populations are known to be particularly vulnerable to stochasticity: the smaller a population, the greater its chance of going extinct due to a random demographic, genetic, or environmental event (Wissel and Stöcker, 1991, Boyce, 1992, Remmert, 1994). Because chances of recolonization after extinction are reduced in isolated populations, extinct populations are unlikely to benefit from the rescue effect (Hanski, 1999).
Roads will affect persistence of animal populations differently depending on (1) road avoidance behavior; (2) population sensitivity to the four aforementioned road effects; (3) size of the road; and (4) traffic volume. We have created a model to predict the impact of roads on population persistence based on these population and road characteristics (to run the model on the web or to download it go to www.glel.carleton.ca/ or www.nls.ethz.ch/roadmodel/index.htm or contact the first author). In this study, we use the model to address the following questions: (1) what types of road avoidance behaviors make populations more (or less) vulnerable to roads?; (2) for given types of roads and of road avoidance behavior, does the impact of roads vary with the relative population sensitivity to the four road effects?; (3) what types of roads have the greatest (or the least) impact on population persistence?
Section snippets
Methods
Very few quantitative data are available on the impact of roads on population persistence. Therefore, our approach in creating this model was to develop relative rankings that could be used to compare the impact of roads in various combinations of population and road characteristics.
Results
Results of all 480 iterations are shown graphically in Appendix B, and a summary of the median rank values for each type of road avoidance behavior is presented in Table 5.
Our first objective was to identify behaviors that make populations more or less vulnerable to roads. Our model predicted that, in general, the most vulnerable populations are those with high noise and high surface avoidance, and secondly, those with high noise avoidance only (Fig. 3). Populations with these two behaviors
Discussion
Our results suggest that populations with high noise and high surface avoidance, and populations with high noise avoidance only are most vulnerable to roads. We also predict that populations with high car avoidance, and populations with high surface and high car avoidance are least vulnerable to roads. These patterns are consistent across road types and most combinations of population sensitivity to the four road effects (Fig. 3). Therefore, our model suggests that it is possible to predict the
Conclusion and speculation
Studies on road avoidance behavior are scarce. Some studies document a reduction in density of species in habitat near roads (Thiel, 1985, Mech et al., 1988, McLellan and Shackleton, 1988, Belden and Hagedorn, 1993, Mace et al., 1996, Mladenoff et al., 1999, Robitaille and Aubry, 2000, Nellemann et al., 2003). Such information is ambiguous because the reduced density can either be a result of avoidance behavior or a reduction in population size due to traffic mortality (Fahrig et al., 1995).
Acknowledgments
We thank Jason Nicolaides and Sarah Brown for assistance with preparing the model and the members of the Geomatics and Landscape Ecology Laboratory (GLEL) at Carleton University, Ottawa, for stimulating discussions about road effects. This work was supported through a postdoctoral scholarship from the German Academy of Natural Scientists Leopoldina to J.A.G.J. (grant number BMBF-LPD 9901/8-27) and through a Natural Sciences and Engineering Research Council of Canada grant to L.F.
References (93)
- et al.
Fuzzy rule-based models for decision support in ecosystem management
Sci. Total Environ.
(2004) - et al.
Effect of road traffic on amphibian density
Biol. Conserv.
(1995) - et al.
Rates and causes of mortality in a fragmented population of Iberian lynx Felis pardini Temminck, 1824
Biol. Conserv.
(1992) - et al.
The effect of road kills on amphibian populations
Biol. Conserv.
(2001) - et al.
Tourist vehicles as vectors of weed seeds in Kakadu National Park, northern Australia
Biol. Conserv.
(1994) Animal habitat isolation by roads and agricultural fields
Biol. Conserv.
(1984)- et al.
Linear barriers to arthropod movements in the landscape
Biol. Conserv.
(1990) Qualitative modelling with imprecise ecological knowledge: a framework for simulation
Environ. Model. Software
(2003)Assessing temporal and spatial changes of salinity using fuzzy logic, remote sensing and GIS: foundations of an expert system
Ecol. Model.
(2001)- et al.
Fire enhances weed invasion of roadside vegetation in southwestern Australia
Biol. Conserv.
(1995)