Nest Boxes: An Effective Management Tool for Kestrels

Only three pairs of nesting kestrels (Falco sparverius) were found in a more than 20-year period on a 50,000-acre (20,243 hectare) study area in central Wisconsin. Fifty nest boxes were put up in 1968 and maintained through 1972. These boxes fledged from 8 to 12 broods per year, totaling about 204 birds or an average of 4.0 per successful box. During the same 5-year period, only two natural nests were found, producing a total of five young. J. WILDL. MANAGE. 37(3):400-403 Kestrels normally nest in old woodpecker holes and natural cavities in trees. Occasionally, they nest in barns and under eaves of buildings, even in towns and cities. Nest boxes are readily accepted. Although the kestrel is Essentially an open-field-oriented bird (Heintzelman 1970:85), the Hamerstroms found only three nests in over 20 years on the open meadows of the Prairie Chicken Study Areas of the Buena Vista and Leola Marshes in central Wisconsin (described in Hamerstrom et al. 1957, and Hamerstrom and Hamerstrom 1949). All were in abandoned buildings. Two boxes put up in 1967 each produced kestrels (one in 1967, the other in 1968). This success led us to wonder whether lack of nest sites might not have kept kestrels from nesting. In 1968 we put up 48 more kestrel boxes: we maintained 50 boxes each year from 1968 through 1972. Several factors influenced our selection of box sites. Since the early kestrel nestings had been in buildings, we put boxes on deserted barns and sheds, as well as on trees near buildings. Again, we noticed that certain areas had attracted kestrels during migration. We nailed boxes on isolated trees or buildings within these areas. We also placed boxes in fields that appeared to offer particularly good grasshopper and vole (Microtus) hunting. Finally, we set boxes in trees which had attractive looking natural holes. Many people helped with this project. In particular we would like to acknowledge the outstanding contributions of F. B. Renn. MATERIALS AND METHODS The floors of our wooden kestrel boxes were 9-12 inches (23-30.5 cm) square. The boxes were about 15 inches (38 cm) deep with a 3-inch (7.5 cm) hole near the top. They were placed about 12 feet (3.7 m) from the ground. We used a ladder to check each box to avoid leaving a scent trail for predators (Hamerstrom 1970). Although we often inspected our boxes from the road, we climbed to them only three times a year. In spring, we checked to see that the boxes were fit for use. In mid-June or early July we visited each box again to check on activity and to band young. In August, after the kestrels had fledged, we returned to clean the boxes and ready them for another season. In 1972 we made an early check during incubation. The presence of kestrels at a box is not always indicative of nesting so we have 1 Mail address: Plainfield, Wisconsin 54966. 400 J. Wildl. Manage. 37 (3):1973 This content downloaded from 207.46.13.129 on Thu, 30 Jun 2016 04:53:50 UTC All use subject to http://about.jstor.org/terms NEST BOXES FOR KESTRELS * Hamerstrom et al. 401 broken down kestrel activity into five categories (Table 1): 1. Seen near box only: We saw adults within a few feet of several boxes, but found no evidence that they went in. 2. Box accepted but no eggs: When we saw adult kestrels entering the box or found pellets or kestrel feathers, molted during incubation or rearing, we considered the box accepted. Two of our boxes were entered by kestrels that failed to lay. Since starlings (Sturnus vulgaris) carry feathers of many species into boxes, the mere presence of a kestrel feather is not always evidence that the box was accepted. (We found pellets or feathers in three boxes out of the nesting season, suggesting that kestrels had merely used the boxes as roosts. ) 3. Eggs laid, nest failed: Nine boxes contained clutches that failed. We usually collected abandoned clutches or eggs that failed to hatch by banding time. In seven boxes we were unable to determine why the eggs did not hatch. Starlings moved into one box before the kestrel eggs hatched, possibly causing desertion. Once kestrels deserted after too much human disturbance. 4. Eggs hatched, nest failed: Only three nests failed after hatching. In August 1969, we found kestrel feathers with blood quills in one box, indicating that young kestrels had died; starlings had previously hatched young in this same box. A second box with young was torn down, replaced, and later shot. No young fledged from this box. In the third box we were unable to determine why the young did not survive. 5. Young kestrels fledged: Most (44) broods were banded. We made a point of banding just before fledging time, so the number banded (plus five young in three broods that escaped us) essentially equals the number fledged in those boxes. In the Table 1. Kestrel activity at 50 boxes in central Wisconsin. Total 19681968 1969 197