BREEDING BIOLOGY OF WOOD DUCKS USING NATURAL CAVITIES IN SOUTHERN ILLINOIS

Information on breeding biology of wood ducks (Air sponsa) acquired la rgely through studies of artificial nest boxes may not represent natur al cavities where most wood ducks are produced. We studied nesting bio logy of wood ducks using natural cavities in southern Illinois during springs 1993-95. Forty-four radiomarked females were tracked to nest s ites, but these represented only 56% of resident females alive during the breeding season. Most (82%) nests were in upland forest beyond the Mississippi River floodplain where birds were captured. Six of 23 (26 %) nests were parasitized, which is a rate higher than preciously repo rted for natural cavities (12%), but is less than in nest boxes (40%). Nest success (64%) was higher than reported elsewhere in natural cavi ties (40%), but below that in nest boxes (66-75%). Success rates did n ot differ between floodplain (80%) and upland (59%) nests (P = 0.205). Duckling survival to fledging was high (42%), despite lengthy ((x) ov er bar = 1.4-3.2 lan) movements over mostly upland terrain. In our stu dy 2.9 ducklings fledged per nest attempt, compared to 3.8 ducklings p er nest attempt from boxes in Tennessee. Low nesting effort and high r ates of nest parasitism could have resulted from suboptimal habitat co nditions in 1994 and 1995. In 1994, dry conditions during prenesting, followed by sudden flooding during egg laying and early incubation, we re associated with a shorter nesting season (91 vs. 132 days), smaller proportions of nesting birds (42 vs. 57%), more nest parasitism (43 v s. 18%), and lower survival (55 vs. 81%) compared to 1995, when condit ions were dry, but water levels were stable. Upland forests may be imp ortant sources of wood duck production in other areas where extensive loss and fragmentation of floodplain forest have occurred.