Military lands are a valuable resource in recovery of threatened, endangered, and at-risk species worldwide and have the highest density of threatened and endangered species of all major land management agencies in the United States. Many red-cockaded woodpeckers (Picoides borealis) that reside on federal lands occur on 15 military installations in the southeastern United States. This close association has increased concern over potential conflicts between conservation requirements of endangered species and the military's mission of combat readiness. Our objectives were to 1) determine if military training operations affect behavior, reproductive success, and productivity of red-cockaded woodpeckers; 2) develop a frequency-weighting function to assess woodpecker hearing sensitivity; 3) identify factors that affect woodpecker responses to military training operations; 4) develop distance and dose-response thresholds for quantifying woodpecker responses to noise levels and stimulus distances; 5) characterize military training operations through quantification of sound levels, source identification, distance from active woodpecker nests, frequency spectra, duration, and frequency of occurrence; and 6) document baseline woodpecker nesting behavior. We conducted our study on the Fort Stewart Military Installation located in southeast Georgia, USA.

Downy woodpeckers, as surrogates for red-cockaded woodpeckers, had their best hearing sensitivity within the peak range of the power spectrum of both downy and red-cockaded woodpecker vocalizations, which is at a higher frequency than that of a typical passerine. Overall, woodpeckers had a reduced auditory sensitivity relative to human hearing sensitivity and other species of small birds, especially in the frequency range >4 kHz. Woodpeckers were most sensitive in the 1.5- to 4.0-kHz range. Sensitivity appeared to drop off quickly at frequencies <1.0 kHz and >4.0 kHz. Overall, we did not find that the woodpecker-frequency-weighting function we developed provided a better predictor of woodpecker flush response compared with A-weighting. More research is needed to better understand the relationship between frequency-weighting functions and woodpecker response behavior.

Potential breeding groups of woodpeckers across the population increased from 158 in 1997 to 181 in 2000, wheras nesting groups increased from 141 in 1998 to 170 in 2000, for overall increases of 14.6% and 20.6%, respectively, over the 3 years of this project. Fledging success rates for individual nests within the overall population remained consistent from 1998 to 2000, averaging 84.4%. Mean clutch sizes for woodpecker groups for 1998 to 2000 ranged from 2.75 to 3.01 eggs/nest, brood size ranged from 2.01 to 2.22 nestlings/nest, whereas the average number of young fledged ranged from 1.57 to 1.76 young/ occupied nest. We observed no difference in reproductive success or productivity between experimental and control-tested red-cockaded woodpecker groups. Overall, experimental test groups produced an average of 2.98 eggs/nest, 1.89 nestlings/nest, and 1.54 young/occupied nest from 1999 to 2000, compared with 2.73 eggs/nest, 1.91 nestlings/nest, and 1.57 young/occupied nest at control groups.

We measured behavioral responses (nest attendance and arrivals and departures from the nest) of red-cockaded woodpeckers to military training events through direct and indirect (i.e., video surveillance) observation of 464.5 hours of woodpecker nesting behavior before and after controlled experimental events while recording and characterizing military-generated sound events using sound-recording equipment. We presented woodpeckers with actual 0.50-caliber blank machine gun fire and artillery simulators from controlled distances to develop distance and sound thresholds. We used video surveillance to document potential behavioral responses of woodp