The study and analysis of ecological connectivity has gained considerable attention in ecology, landscape genetics, and conservation biology. For conservation planning, governmental and non-governmental organizations are devoting considerable resources to mapping areas that facilitate movement to maintain population connectivity, promote climate adaptation, and achieve other conservation goals. In contrast, there has been little focus on identifying important barriers to movement. Yet knowing where barriers such as roads most strongly affect connectivity can complement traditional maps showing best movement routes. Barrier maps could inform decisions on trade-offs between restoration and protection; for example, purchasing an intact corridor may be substantially more costly than restoring a barrier that blocks an alternative corridor. We introduce a novel method to identify important barriers, i.e., those landscape features which significantly alter best movement routes connecting habitat patches, whether complete (impermeable) or partial (impeding but still allowing some movement). The method uses GIS neighborhood analyses in conjunction with either cost-distance or circuit-based connectivity analyses to identify areas that significantly degrade corridors and which, if restored, would significantly improve connectivity. We tested our methods in three ecoregions in the western USA: Columbia Plateau, High Plains, and Mojave Basin and Range.
Results/Conclusions
We found that barrier mapping provided a strong complement to corridor mapping by broadening the range of connectivity conservation alternatives from outright protection of existing corridors to include different types of restoration. Our analyses identified numerous areas offering restoration opportunities, including potential road crossings and candidate areas for the restoration of agricultural lands. In many cases, small restoration efforts would entirely re-route the most efficient movement route between habitat patches. An additional benefit is that, by identifying influential barriers, the method also helped to identify where connectivity analyses may be particularly sensitive to errors in base data. Identifying which modeled barriers have the greatest impact—e.g., those which change the best movement routes or decrease availability of alternative pathways—can help prioritize areas for error checking. Such applications can help to identify locations on the landscape where field data will most improve a connectivity model. Finally, models that predict important barriers may be more straightforward to validate with genetic data than those that predict movement routes because landscape genetic analyses often lend themselves to detecting barrier effects. In conclusion, barrier identification provide a different way to view the landscape, increasing conservation options while broadening thinking about connectivity and fragmentation.