Barriers to dispersal and the challenges facing the southern expansion of bobcats in New Jersey
Date
2021
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
In urban environments, the threats of habitat fragmentation and loss, barriers to dispersal, and anthropogenic causes of mortality affect the recolonization potential of extirpated species. One such species, the bobcat (Lynx rufus), historically occurred throughout the state of New Jersey, but due to increased urbanization and agricultural expansion has been confirmed to occur almost exclusively in the northern portions of the state since the 1960’s. In this study I examine current barriers to bobcat dispersal and the possibility of the establishment of central and southern New Jersey bobcat populations. First, I developed a statewide habitat suitability index for bobcats and validated it with known bobcat locations. I then identified discrete suitable habitat patches and calculated landscape resistance values across the state from the inverse of the habitat suitability values with some adjustment for road-based metrics. To evaluate landscape connectivity throughout New Jersey, I applied circuit theory using the program Circuitscape within a GIS framework. I ran “current,” representing movement potential, through the resistance landscape between each pair of previously identified habitat patches. Using circuit theory in combination with least-cost path analysis, I next identified bottlenecks to bobcat movement throughout New Jersey. I identified barriers to movement using the change in least-cost path value based on a hypothetical barrier reduction. Finally, I repeated all these analyses for a case study in central New Jersey, where all stream crossing structures under roadways were assigned the lowest resistance value of 1, representing the ideal scenario that all structures had been modified to allow bobcat passage. I found that habitat patches were connected with some redundancy in the north, but with less redundancy throughout central New Jersey. The regions of high cumulative current flow around the Trenton area indicate that dispersing bobcats would have to pass through west central New Jersey to move from northern to southern habitat patches. Of the four least-cost paths connecting northern patches to southern patches, the shortest was approximately 25 km in length. While bobcats can easily disperse that distance, the movement corridors I identified cross through a highly urbanized landscape and contain several potential barriers. The resultant maps I produced will allow state managers to target specific areas for connectivity maintenance and improvement work within the CHANJ framework that are important for bobcat movement. ☐ I next used spatially explicit agent-based simulation models to examine the possibility of the establishment of central and southern New Jersey bobcat populations. The simulations consisted of a primary population model using life history statistics as well as two spatially explicit sub-models that examined dispersal and home range formation through a rasterized landscape representing movement resistances for bobcats. I evaluated the impacts of the following management actions: 1) status quo, 2) barrier reduction, and 3) single translocation event. I ran each scenario ten times to project population size and distribution at the end of 1, 5, 10, and 25 years. There was no significant difference in population size at the 1-, 5-, 10-, and 25-year time intervals for both the status quo and barrier reduction scenario. The translocation scenario revealed that the model needs further refinement as the population decreased between the 1- and 5-year time intervals, with no differences found between any other time interval pairs. Results from all scenarios indicated high probability of bobcat occupancy in habitat north of Route 80 over the course of 25 years. Bobcats more consistently occupied territories south of Route 80 but north of Interstate 95 under the barrier reduction scenario than either the status quo or translocation scenarios. The translocation scenario revealed higher probabilities of occupancy in the south at the 5-year time step likely due to the continued presence of translocated bobcats in their original territories. However, this high probability decreased by the 10-year time step and occupied territories largely disappeared by year 25. The surprising results of the translocation scenario highlighted the need to refine how the model identifies reproducing females, as it is currently too strict. Additionally, the impact of multiple small translocations on the persistence of a southern bobcat population should be evaluated before any decisions are made regarding translocation. Overall, my results suggested that bobcats are unlikely to recolonize central and southern New Jersey under current conditions within the next quarter century. Mitigation of barriers, through the implementation of modified culverts and crossing structures appropriate for bobcat movement, may allow for more reliable establishment of bobcat presence in the region south of Route 80, but still little occupation of territory south of Interstate 95. Given limitations in the model and its need for further refinement, no conclusions can be made regarding the translocation scenario. Translocation, especially in conjunction with barrier mitigation and the conservation of corridor habitat, may still be a viable management opportunity for NJDFW as it has previously been successful with bobcats in both in Georgia and northern New Jersey. However, further research should be pursued before any decisions are made. Once properly calibrated, spatially explicit agent-based models can be useful tools for wildlife managers to evaluate the potential impacts and results of different management actions. Overall, this research has expanded the scientific understanding of species-specific connectivity in an increasingly urban world, thus informing the implementation of mitigation methods to maintain and increase permeability within the landscape.
Description
Keywords
Agent-based models, Bobcat, Circuit theory, Dispersal barriers, Habitat suitability index, Recolonization