We initiated a 3-year study in mid-summer 2005 to determine patterns of bat activity and evaluate the use of acoustic monitoring to predict fatality of bats at a proposed wind energy facility in south-central Wisconsin. The study area represents a situation where a large bat hibernaculum is within proximity to the proposed turbine locations. The primary objectives of this study are to 1) combine these data with other similar studies to evaluate whether indices of pre-construction bat activity can be used to predict relative risk of postconstruction bat fatalities at proposed wind facilities; 2) determine the levels and patterns of activity of different species groups of bats using the area of the proposed wind facility; 3) determine temporal and spatial patterns of bat activity at turbine locations across the wind facility and vertically using detectors positioned at ground level (2m), within the turbine rotor-swept zone (48 m above the ground), and at an intermediate level (22 m above the ground); 4) determine if patterns of post-construction bat fatality are related to and may be predicted by pre-construction activity, weather conditions, and other environmental conditions. The goal of the study is to provide information to minimize mortality of bats migrating to Neda Mine and through the area (foliage roosting bats heading south). In addition, results of our study will help evaluate the efficacy of acoustic monitoring to assess risk of bat fatality at proposed wind power sites.
We used broadband ultrasound detectors during the 2005 fall migration period (19 July through September) to assess the spatial and temporal distribution of bat activity across the proposed wind farm location, two reference areas, and two sites located near the Neda Mine. We used 50 m meteorological towers and 22 m tall, portable, telescoping towers to vertically array detectors for acoustic sampling during this study. We recorded bat echolocation calls at proposed turbine locations using detectors arranged on 3 meteorological towers (one detector at 2, 22, and 48 m high at each tower) and rotated 5 mobile towers (one detector at 2 and 22m high at each tower) throughout the study period to sample the 33 proposed turbine locations.
We recorded a total of 26,495 bat passes at all towers throughout the study. Feeding-type activity was identified in 3998 passes (15 % of total bat passes); and feeding activity appeared to occur throughout the night and at all heights sampled. Bat activity was highly variable throughout the study and varied considerably among the towers. Recorded bat activity was highest in August with secondary peaks in late July and September. By October bat activity had declined considerably.
Two species groups were defined using the average minimum call frequency of each bat pass recorded. High frequency group included species with the average minimum call frequency ≥35 kHz (Little brown bat, Myotis lucifugus; Northern long-eared bat, Myotis septentrionalis; Eastern pipistrelle, Pipistrellus subflavus; and Eastern Red bat, Lasiurus borealis) and low frequency group as <35 kHz (Big brown bat, Eptesicus fuscus; Silverhaired bat, Lasionycteris noctivagans; and Hoary bat, Lasiurus cinereus).
The two species groups tended to fly at different heights at the Wisconsin site. Activity of high frequency bats was estimated to be 3.2–5.5 times higher at 2 m than at 22 m, and 3.8–7 times higher at 2 m than at 48 m. There was no detectable difference in activity at any height for the low frequency bats, nor at the higher altitudes (22 vs. 48 m) for the high frequency bats. We estimated that activity of high frequency bats was 2–7.3 times higher than that of low frequency bats at 2 m. At higher altitudes (22 and 48 m), the activity of low frequency bats was not detectably different than that of high frequency bats.
Myotis species (primary inhabitants of the Neda Mine) were treated as a subgroup within the high frequency species group to assess bat activity in relation to a turbine’s distance from the mine and habitat features. We found that relative activity of Myotis bats at 2 m decreased by between 6 and 28% for every kilometer increase in distance of a tower from the Neda Mine. There was no detectable relationship of relative activity with distance from the mine for low frequency bats.
Temperature and wind speed affected bat activity rates at our site, but the effect of temperature differed for the two species groups. Temperature during this study ranged from 7–26 ºC. The effect of temperature was very strong, and differed for the two groups. For each 1ºC increase in temperature, the activity rate of the high frequency group was estimated to increase by 3-9% and the activity rate of the low frequency group was estimated to increase by 7-13%. Average nightly wind speed during this study ranged from 1.0–9.6 m/s and strongly affected bat activity. For each increase in wind speed of 1 m/s, the activity rate of bats was estimated to decrease by 4-13%. Bat activity decreased with increasing wind speed, but there was still some activity, even at the highest wind speeds measured in this study.