Wednesday, August 6, 2008

PS 50-154: Land-use and -cover change near hibernacula of the endangered Indiana bat (Myotis sodalis): Current and projected risks to species recovery

Michael G. Just, University of Illinois at Urbana - Champaign and Matthew G. Hohmann, US Army Corps of Engineers ERDC - CERL.

Background/Question/Methods

It is estimated that 95% of all federally endangered Indiana bats (Myotis sodalis) hibernate in 71 priority 1 (>10,000 bats) and 2 (>1,000 bats) caves and mines in ten states. Given the species’ high site fidelity, seasonally heightened densities, and the apparent limited availability of suitable sites, land-use and -cover change (LUCC) near hibernacula is expected to affect wintering populations. The landscapes surrounding hibernacula not only provide critical roosting and foraging habitat during fall swarming and spring staging, they also support the highly specific microclimates Indiana bats need for successful hibernation. Unfortunately, the landscapes surrounding the vast majority of hibernacula are not afforded any land-use protections. Consequently, it is important that LUCC near hibernacula is evaluated in order to assess and prevent impacts to wintering Indiana bat populations. We characterized the current (2001) landscape composition and configuration using landscape pattern indices (LPI). We modeled potential future LUCC through 2016 using two empirically-derived, stochastic projection models. Model drivers included biophysical and socioeconomic variables. Characterization and modeling were performed at two extents (4.025 and 8.05 km radii), chosen for their relevance to the species’ biology.

Results/Conclusions

For the larger extent, we found that the percent of the landscape comprised of forest changed from a range of 7.38 - 98.99% with a mean of 74.55% (± 4.94% [95% CI]) in 1992, 4.91 - 95.142% with a mean of 65.72% (± 4.75%) in 2001, and a projected 3.33 - 78.09% with a mean of 49.67% (± 3.92%) in 2016. This represents a 24.88% decrease in the mean percent forest surrounding hibernacula projected over the 24 year period. This rate of forest cover loss is unlikely to be sustainable with regard to the species’ biology and conservation. The mean percent of urban land observed in the 71 landscapes varied from 2.37% (± 1.07% [95% CI]) in 1992 to 7.45% (± 1.46%) in 2001, and was projected to increase to 17.00% (± 1.85%) by 2016. We found a general decline in the perimeter-area fractal dimension LPI indicating a less complex landscape. Changes in the mean nearest neighbor and proximity LPI suggest that forest patches have become generally more isolated and are projected to become increasingly isolated. This loss of edge habitat and patch connectivity has the potential to impact bat habitat use near hibernacula. Results of this study address several priority research needs identified in the USFWS Revised Indiana bat Recovery Plan.