Weighing the relative potential impacts of climate change and land-use change on an endangered bird

Abstract

Climate change and land-use change are projected to be the two greatest drivers of biodiversity loss over the coming century. Land-use change has resulted in extensive habitat loss for many species. Likewise, climate change has affected many species resulting in range shifts, changes in phenology, and altered interactions. We used a spatially explicit, individual-based model to explore the effects of land-use change and climate change on a population of the endangered Red-cockaded Woodpecker (RCW; Picoides borealis). We modeled the effects of land-use change using multiple scenarios representing different spatial arrangements of new training areas for troops across Fort Benning. We used projected climate-driven changes in habitat and changes in reproductive output to explore the potential effects of climate change. We summarized potential changes in habitat based on the output of the dynamic vegetation model LPJ-GUESS, run for multiple climate change scenarios through the year 2100. We projected potential changes in reproduction based on an empirical relationship between spring precipitation and the mean number of successful fledglings produced per nest attempt. As modeled in our study, climate change had virtually no effect on the RCW population. Conversely, simulated effects of land-use change resulted in the loss of up to 28 breeding pairs by 2100. However, the simulated impacts of development depended on where the development occurred and could be completely avoided if the new training areas were placed in poor-quality habitat. Our results demonstrate the flexibility inherent in many systems that allows seemingly incompatible human land uses, such as development, and conservation actions to exist side by side.

Keywords: Development; HexSim; Picoides borealis; Red‐cockaded Woodpecker; environmental stress; habitat change; habitat loss; population model; precipitation; spatially explicit individual‐based model.

Figure 1

(A) Map of training ranges added to the landscape at Fort Benning. This map formed the baseline range development scenario from which all other range development scenarios were generated. Values indicate the effect of each range on Red‐cockaded Woodpeckers (RCW). The habitat ratings in the initial habitat probability map were multiplied by the values shown to downgrade habitat quality where new ranges were placed. Lower values indicate a larger negative effect of the range on RCW habitat. (B) Baseline development habitat map for RCW using proposed training ranges shown in A. Each range shown in A was placed on the landscape in the location shown in A and habitat within the range was reduced based on estimated impact of the range on Red‐cockaded Woodpecker habitat quality. Values in the legend represent likelihood of RCW presence, where higher values indicate higher likelihood of RCW presence.

Figure 2

Red‐cockaded Woodpecker annual cycle as implemented in HexSim. At the beginning of each year, 1 year was added to the age of each individual present in the landscape (“increment age”). All movement routines, survival, and reproductive rates were modeled using the values described in the text and Table 3. In these scenarios, “adults” are individuals surviving the first year.

Figure 3

Red‐cockaded Woodpecker (RCW) population projections under three scenarios and baseline conditions. Lines represent means of 20 replicate model runs plus 1 standard error. (A) Conservation land‐use scenarios. RCW population projections are not different from baseline development (no further development) in all Conservation scenarios. (B) Convenience land‐use scenarios. Population projections are slightly lower than baseline development simulations (no further development) in all Convenience scenarios. (C) Worst‐case land‐use scenarios. Population projections are much lower than baseline (no further development) in all Worst‐case scenarios.