TAWNY OWL (STRIX ALUCO) PREDATION IN A PRISTINE DECIDUOUS FOREST (BIALOWIEZA NATIONAL-PARK, POLAND)

1. Tawny owl Strix aluco predation on its main prey (bank voles Clethr ionomys glareolus and yellow-necked mice Apodemus flavicollis) and alt ernative prey (shrews, birds, and amphibians) was studied in the prist ine deciduous forests of eastern Poland during a 7-year period (1985/8 6-1991/92) that included 5 years of moderate densities of rodents and 2 years of outbreak and crash triggered by a heavy seed crop of oak, h ornbeam, and maple. 2. Number of resident owls was rather stable; 45-5 3 owls per 10 km(2) in years of moderate and high rodent numbers, and 33 owls per 10 km(2) in a rodent crash year. Numerical response of owl s to rodent densities was log shaped. 3. In the cold season, the dieta ry (functional) response of owls to autumn density of rodents was of t ype II (logarithmic). No functional response of owls was observed to s hrew numbers. Owl hunting rate on amphibians in winter was shaped by t he mean temperature of the cold season and, to a lesser extent, by ava ilability of rodents. Owl hunting rate on birds was inversely related to rodent abundance. 4. During the seven cold seasons (1 October-15 Ap ril), tawny owls removed 3-19 mice, 2-18 voles, 2-7 shrews, 0-2 birds, and 1-9 frogs from an average hectare of the forest. Total predation on rodents (N prey removed ha(-1)) was lowest during the crash of rode nts and highest during the outbreak; it grew logarithmically with incr easing density of rodents. On average, in the cold season, tawny owls removed 54% of autumn numbers of mice and 40% of those of bank voles. Percentage predation peaked at 22 rodents ha(-1) in autumn. At both hi gher and lower prey densities, percentage predation declined. 5. The p ercentage predation curve of tawny owl was compared to the frequency d istribution of autumn densities of rodents in Bialowieza (from 23-year trapping). The heaviest predation fell to prey densities that occurre d most frequently. 6. In summer seasons (July-September 1988-92), tawn y owls (both adults and juveniles) consumed 0.5-19 mice and 0.4-10 vol es from an average hectare. Percentage predation varied from 6% to 74% of rodent numbers recorded in July, but rodents bred rapidly and recr uitment usually more than compensated for losses due to predation. 7. Total predation by tawny owls on rodents was quite stable as a consequ ence of high stability of owl numbers. Percentage predation, although heavy in all years except for rodent outbreak, was not density-depende nt and therefore tawny owls did not regulate rodent numbers. In the de ciduous forests of Bialowieza National Park, the pattern of predation by tawny owl, a resident generalist, did not differ from that by the w easel, a resident specialist.