Irodalom | Középiskola » Winter Predation by Goshawks in Lowland Britain

Winter predation by Goshawks in lowland Britain R. E Kenward The G o s h a w k i s recolonising Britain. What will be i t s diet, a n d i t s effect on an e c o n o m i c a l l y i m p o r t a n t prey, the Woodpigeon? N ow becoming re-established as a breeding species in Britain (Sharrock 1976), the Goshawk Accipiter gentilis is likely to increase the predation on a n u m b e r of medium-sized birds and mammals. O n e interesting aspect of this is its possible effect on the Woodpigeon Columba palumbus, a species which can cause crop damage a n d is sometimes an important food for Goshawks. I n the breeding season, for instance, Woodpigeons can account for as much as 41 %, or less than 10%, of prey brought to Goshawk nests in central Europe (Tinbergen 1936, Uttendorfer 1952). Goshawk diet outside the breeding season has been studied by searching for prey remains (Briill 1964, Thiollay 1967). I n a D u t c h study area one winter, 7 2 % appeared to be Woodpigeons ( O p d a m 1975);

but this continued. 64 [Brit. Birds 72; 64-73, F e b r u a r y 1979] Predation by Goshawks in Britain 65 technique probably overestimates the proportion of pale-coloured prey (e.g pigeons) because their remains are particularly eye-catching Bias also occurs in other winter diet analysis techniques (Errington 1932). Pellet analysis tends to overestimate the proportion of large prey, which are represented in more t h a n one pellet if a carcase is used for several meals; and to bias against the recording of small prey whose bones are completely digested, unless correction factors can be applied (Dare 1961). Gut-content analyses can be biased by variation in prey digestibility and by samples originating mainly from poultry farms and game-rearing areas. Brown (1975) has suggested that the value of a diet analysis depends on the number of prey items recorded, but a small quantity of unbiased d a t a may be more useful in a predation study than any amount of biased material. This

paper presents d a t a obtained by following radio-tagged Goshawks to obtain accurate records of their diet, their rate of predation on Woodpigeons, and their ability to survive in an area of lowland Britain. T h e study was conducted in an area of rolling farmland (primarily arable, but with some pasture) between 80 m and 250 m above sea level, near the town of Chipping Norton, Oxfordshire. Woodland, which is scattered and mainly deciduous, accounts for about 8 % of the land use; streams occur in the valleys, with the occasional small p o n d ; and there are larger artificial lakes of u p to 1.5 ha in private parks T h e avifauna is typical of central English lowland farmland (see Williamson 1967). T h e most a b u n d a n t larger (200-1,000 g) wintering species are Woodpigeons, Rooks Corvus frugilegus, Jackdaws C. monedula, Lapwings Vanellus vanellus, Common Larus canus and Black-headed Gulls L. ridibundus, and Grey Perdix perdix a n d Red-legged Partridges Alectoris rufa. In parts

of the area, Pheasants Phasianus colchicus are reared and are locally a b u n d a n t ; and Moorhens Gallinula chloropus are common along the waterways. T h e predominant m a m m a l weighing more than 200 g is the rabbit Oryctolagus cuniculus, which is moderately a b u n d a n t locally but still suffers from outbreaks of myxomatosis. Individual h a w k s and radio-telemetry W i t h no wild Goshawk population in Oxfordshire, it was necessary to study released birds. Since the past history of a hawk might influence its hunting, four birds with different backgrounds were used. Falconry terminology denotes each ones origin concisely. T h e eyass ( = obtained from the nest) was brought from Finland, trained, and flown at quarry using falconry techniques (see Woodford i960). T h e other three had all been trapped after learning to h u n t for themselves. O n e of two passagers ( = obtained as first-winter birds) came from the German Federal Republic and the other from Finland, while a

three- to four-year-old haggard ( = obtained as an adult) was also Finnish. These last three had received some handling a n d had become fairly tame, but had not been trained. All four were males T h e German passager was released at dawn on five days in J a n u a r y 1974, and was recovered to the fist from a kill, or to the lure, on four of 66 Predation by Goshawks in Britain 23. Radio-telemetry equipment used for tracking Goshawks Acdpiter gentilis Clockwise from right: receiver, earphones, transmitter and receiving antenna {Edward Grey Institute) t h e m ; on the fifth, he was lost. T h e Finnish haggard was similarly released on four days between late September and early November 1974; after release on a fifth, he remained free for 11 days. Between early December 1974 and mid M a r c h 1975, the Finnish passager and the Finnish eyass were each released on two occasions and left free for totals of 37 and 32 days respectively. Each hawk was equipped with a 173-MHz

transmitter, weighing between 7 g a n d 15 g, attached proximally to the upper surface of a central rectrice using thread and epoxy-resin adhesive. T h e quarter-wave antenna was bound for about three-quarters of its length to the feather shaft, and was therefore unlikely to be damaged by preening or to tangle in vegetation. (Keeping the antenna straight and off the ground enhances signal transmission) This form of transmitter attachment appears not to interfere with a hawks prey-capture, nutritional condition, or tendency to leave an area (Kenward 1978a). T h e portable receiver was operated with a three-element unidirectional hand-held Yagi antenna (plate 23). This equipment gave a range of several kilometres when a hawk was perched in a tree, but occasionally as little as 500 m when one was on the ground in thick cover. Results and discussion Diet Radio-telemetry was used to complement vision in locating and following the hawks without getting close enough to disturb them or their

prey. T h e lack of leaves in the deciduous woodland assisted observation. T h e Goshawks usually hunted from inconspicuous perches in trees; they 67 Predation by Goshawks in Britain changed position by flying a n average of 100 m in woodland, or 200 m in open country, most frequently at intervals of two to six minutes. Preycapture was rarely seen; it was usually detected when the radio signal indicated that a hawk had ceased hunting flights, was on the ground and, on some occasions, was making rhythmic feeding movements (see Kenward 1976 for details). Feeding hawks were approached to identify their prey. Those remaining free had their pigeon kills removed for analysis and replaced by similar food. T h e eyass stayed t a m e enough to be coaxed from his prey to the substitute food, but the other hawks became shy of h u m a n beings within two days of release and h a d to be frightened away. If they left their kill, they returned within a n hour, unless they h a d already h a d a

substantial meal. Kills they flew off with could be identified from the pluckings Prey the size of a thrush Turdus, weasel Mustela nivalis or larger took from 15 minutes to several hours to consume, but smaller prey m a y sometimes have been eaten so rapidly that it was not recorded. Nevertheless, close observation of the eyass, which did not differ from the other hawks in his hunting although he was more approachable, suggested that small mammals and small passerines were captured too infrequently to provide more t h a n a tiny proportion of the winter diet. T h e prey captured by the four Goshawks between October and M a r c h is shown in fig. 1 Individuals did not differ significantly in the proportion of their diet which was Woodpigeons, an average of 4 0 % ; nor did this proportion differ significantly during the three two-month periods Other Moorhen Rabbit Woodpigeon Fig. 1 Prey captured by four Goshawks Accipiter gentilis released in Oxfordshire during winter. Numbers above

columns show total prey items Hawks are denoted by letters below the columns: F H had been trapped in Finland as adult; FE had been obtained from nest in Finland; FP and GP had been trapped in their first winters in Finland and Germany respectively. Other prey included: 3 Pheasants Phasianus colchicus, 3 large crows Corvus, 1 Stock Dove Columba oenas, 1 Kestrel Falco tinnunculus, 1 Blackbird Turdus tnerula, 1 leveret Lepus capensis, 1 common rat Rattus nowegicus, 2 weasels Mustela nivalis, 1 field vole Microtus agrestis and one unidentified small mammal 68 Predation by Goshawks in Britain 514, Captive adult male Goshawk Accipikr gentilis, Chvyd, October 1971 (Dennis Green) October-November 1974, December-January 1974/75, and FebruaryMarch 1975. Moorhens and rabbits (both 19%) were the next most frequent food items: all four hawks captured at least one Moorhen, but neither the haggard nor the German passager killed a rabbit. Since the haggard subsequently refused to pursue

rabbits when flown in falconry, his failure to kill them when released in Oxfordshire probably indicated a tendency to avoid this relatively large prey, which is not found in Finland and is difficult for a male Goshawk to subdue unless he has learned to grasp it about the head. Male Goshawks tend to capture less heavy prey than females (Hoglund 1964, Storer 1966). Opdam (1975) believed that they brought fewer Woodpigeons to the nest area than did females, but the present data do not indicate that males have difficulty catching pigeons. Perhaps they avoid catching heavy prey in the breeding season or pass it to females (which can carry it more easily) at a distance from the nest. Female Goshawks might have caught brown hares Lepus capensis, rabbits and Pheasants more easily than did the released males, but would probably not have found Woodpigeons and Moorhens {which together formed 59% of the prey) either more or less difficult to obtain. Predation by Goshawks in Britain 69 In

winter, Goshawks occupy overlapping ranges rather than discrete territories (Kenward 1977 & unpublished). Those sharing hunting areas might have less access to the most vulnerable prey, and find quarry more sensitive to their presence, than the single released hawks. It seems likely, however, that, in lowland Britain, Woodpigeons, Moorhens, rabbits and possibly hares will be the staple food of Goshawks, unless there are marked changes in the availability of these and other prey. T h e three Finnish hawks gained weight in each of the five periods of continuous release. All, including the one taken as a fledgling and trained as for falconry, were able to obtain adequate food in Oxfordshire. Frequency of kills Following single radio-tagged hawks provides an unbiased record of their predation on all but the smallest prey; the rate of capture may, therefore, be measured more accurately t h a n by combining analyses of stomach contents, pellets or prey remains with estimates of food

value and food requirements (e.g Craighead & Craighead 1956) Following single hawks is time-consuming, but, if determining the rate of predation on a few large prey is more important t h a n recording all kills, then the time per recovered prey item can be reduced by monitoring several birds at once (e.g K e n w a r d 1977) During the five periods of continuous release, prey-capture was monitored on 60 days and occurred on 42 (70%) of them. T w o kills were made in one day by the eyass, which captured a rabbit within an hour of eating a field vole Microtus agrestis. O n average, however, the Goshawks killed about twice in every three days (table 1). Table 1. Interval between kills m a d e by three Goshawks Accipiter gentilis released in Oxfordshire between October 1974 and March 1975 Same day Number of kills 1 1 day 2 days 3 days TOTAL 28 10 4 43 Effect of Goshawks on Woodpigeons Although a predator may reduce the numbers of a prey species by disruption of feeding (Murie

1944), as well as by killing, disturbance by Goshawks is unlikely to affect pigeon numbers. Pigeons were not usually scared away from feeding sites for long by Goshawk attacks, and they often resettled in the same place immediately after being flushed by a hawk (Kenward 1978b). Pigeons scared from their food return sooner the greater their food deficit; and, even at pastures where in midwinter they spend 9 5 % of the day feeding (Murton et al. 1963), they are unlikely to be disturbed by Goshawks frequently enough to reduce their food intake (Kenward 1976). T h e effect of Goshawk predation on pigeon numbers may be estimated from the n u m b e r of pigeons captured per hawk-day, the hawk density, a n d the pigeon density. Any selection for poor quality food must be taken into account, because a predator catching prey which is doomed for other reasons has no effect on its population (Jenkins et al. 1964) 70 Predation by Goshawks in Britain Table 3. Nesting densities of Goshawks

Accipiter gentilis in central Europe Location North Germany North Germany North Germany North Germany Central Germany Northeast France West Netherlands West Netherlands West Netherlands Number of pairs 12-13 20 13 5 Area per pair (ha) 1,000 1,250 3,800 8-12 5,000 2,000 -3,300 8 1,850 2 1,900 2,300 1-2,600 3,200 8-9 H Source Bednarek 1975 Bednarek 1975 Bednarek 1975 Bednarek 1975 Kramer 1955 Thiollay 1967 Tinbergen 1946 Tinbergen 1946 van Beusekom 1972 Assuming t h a t female Goshawks rate of predation on Woodpigeons would not differ very greatly from that of males, and that the rate in a normal population would be similar to that of the single released hawks, each would capture 0.27 pigeons per day (40% of the diet at 067 kills per day). Table 2 shows central European Goshawk nesting density in populations little influenced by biocide contamination. T h e highest density (one pair per 1,000 ha) was found in fertile areas with only 12-15% well-fragmented mixed woodland,

where the species frequently nested in conifer stands as small as 20 m x 50 m (Bednarek 1975). Since these Goshawks normally produce 1.6 to 28 young per pair per a n n u m (Wendland 1953, K r a m e r 1967), their food consumption in an area with one breeding pair should approximate to that of three or four hawks throughout the year. Although the post-breeding density (two adults and about two juveniles) will have fallen by spring, more than enough food for four hawks will be eaten while young are growing. Studies in eastern England showed that the Woodpigeon population increases by about 2.4 times during breeding (Murton 1965) Most mortality occurs between then and February, when the density is determined by food resources. I n spring and summer, the population is not limited by food; little further mortality occurs between February and the start of the peak breeding period, in July (Murton et al. 1964, 1966) M u r t o n (1965) pointed out that predation (shooting) between breeding a n

d midwinter would have to remove more than the post-reproductive population surplus to reduce the February density, but that any later predation would reduce the breeding density. Using a bodyweight/survival relationship derived by Dr R. K M u r t o n and R. M Sibly, it can be shown that Woodpigeons captured by those hawks released in the 1974/75 winter were 7 2 % as likely to survive one month as shot pigeons. Thus, 2 8 % of the predation would compensate mortality associated with poor pigeon condition. T h e bodyweight/survival relationship was based on Cambridgeshire pigeons subject to limited food resources; Oxfordshire pigeons, however, were in unusually good condition in the 1974/75 winter because, owing to a wet harvest and a u t u m n , much shed grain remained on unploughed stubbles (Kenward 1976). W i t h fewer pigeons likely to starve, mortality associated with prey condition might compensate that caused by predation by less t h a n 2 8 % . Predation by Goshawks in

Britain 7i T h e percentage reduction in the July Wocdpigeon density, caused by Goshawks at different densities, can be estimated assuming both 2 8 % compensation of predation mortality by selection effects a n d no compensation (table 3). Pigeon densities are one-quarter of those observed by M u r t o n et al. (1964), because of recent agricultural changes (Murton 1974). T h e population surplus between July and February is t h e difference between a 2.4-fold increase in the July population a n d t ^ a t supported by the midwinter food supply in the absence of predation (432 per 1,000 h a ) . T h e proportion of the surplus which would be taken by Goshawks, if the predation rate continued at 0.27 pigeons per hawk-day, indicates how much the rate could increase before hawks would take more than t h e surplus and therefore reduce the midwinter population. This is the time when damage to winter brassicas m a y occur; peas a n d seedlings are damaged in t h e summer. Table 3. Estimated

effect of Goshawks Accipiter gentilis on Woodpigeons Columba palumbus i n lowland Britain % REDUCTION IN JULY PIGEON POPULATION Goshawk density (per 1,000 ha) IF PREDATION MORTALITY IS Compensatory 1 I 7 2 2 14 3 3 4 4 21 % of July-February pigeon population surplus Non-compensatory which then suffers predation 12 9 3 18 28 32 27 5 64 87 37 100* 23 p l u s 12% reduction in F e b r u a r y population W i t h a Goshawk density of four per 1,000 ha, t h e pigeon breedingpopulation could be reduced by 28-37%, and remain below environment holding capacity throughout the year. T h e Goshawk density would, however, probably be no more t h a n two per 1,000 ha, producing at most a reduction in the pigeon breeding population of about 2 0 % ; it would not reduce t h e midwinter population, even with a threefold increase in the predation rate between J u l y and February. T h e rate of predation might certainly be higher in a u t u m n and winter, when at first

newly-fledged and, later, starving pigeons are available, t h a n when they are in better condition in spring a n d early summer (the proportion of Woodpigeons in the diet of German Goshawks is highest in the a u t u m n : M u r t o n 1971). These simple calculations ignore the possibility that Woodpigeons, which hatch more young per breeding attempt at low densities (Murton & Isaacson 1964), would compensate for t h e reduction in their breeding population, or that reducing the a u t u m n population would decrease midwinter food depletion and allow more to survive to the spring. 72 Predation by Goshawks in Britain 25. Adult Goshawk Accipiter gentilis, Sweden, November 1976 (Bengt LtmdbergjM Sweden) Artificially reducing young-production of Great Tits Parus major increases the populations survival to the next breeding season (Kluyver 1970). Such effects would diminish the impact of predation on the pigeon population. While the return of the Goshawk may be welcomed on

aesthetic grounds, it is unlikely to produce a marked reduction in crop damage by Woodpigeons. Acknowledgements 1 thank the many Oxfordshire farmers and landowners who allowed me to track Goshawks across their property. Hawks were obtained with help from Dr U Beyerbach, K Huhtala, Dr S. Sulkava, and the British Falconers Club; they were recovered after release thanks to J. E Cooper, G N Gripper, G Hovell and J Wiggins I am grateful to M Dolan, Dr T. C Dunstan, M Fuller, Miss C Hunt, Dr H Kruuk, K T Marsland and the late Dr R. K Murton for advice and technical assistance; and to Dr N B Davies, Dr I Newton and Dr C. M Perrins for comments on the written material The work was financed by the Agricultural Research Council. Summary Four radio-tagged Goshawks Accipiter gentilis were released for a total of 89 days in Oxfordshire to study their predation and survival in winter. Three, including one taken from the nest and trained as for falconry, were released for five periods of 11-21 days

and gained weight in each. Woodpigeons Columba palumbus formed 40% of the diet; rabbits Otyctolagus cuniculus and Moorhens Gallinula cUeropus made up another 3 8 % . Estimating the effect of various Goshawk densities on the Woodpigeon population in Britain, it is concluded that any reduction of pigeon crop damage as a result of predation by Goshawks is likely to be slight. References BF.DNAREK, W 1975 Vergleichende Untersuchungen zur Populationsokoiogie des Habichts (Accipiter gentilis): Habitatbesetzung und Bestandsregulation. Deutscher Falkenordm Jahrbmk {1975): 47"53- Predation by Goshawks in Britain 73 B R O W N , L. 1976 British Birds of Prey Glasgow B R U L L , H . 1964 Das Leben Deutscher Greifvogel S t u t t g a r t C R A I G H E A D , J . J , & C R A I G H E A D , F C 1956 Hawks, Owls and Wildlife W a s h i n g t o n D A R E , P. J 1961 Ecological observations on a b r e e d i n g p o p u l a t i o n of the C o m m o n B u z z a r d , Buteo buteo (L), with

p a r t i c u l a r reference to t h e diet a n d feeding habits. P h D thesis, University of Exeter. E R R I N G T O N , P . L 1932 T e c h n i q u e of r a p t o r food habits study Condor 3 4 : 75-86 H O G L U N D , N . H 1964 U b e r die E r n a h r u n g des H a b i c h t s (Accipiter gentilis Lin) in S c h w e d e n . Viltrevy 2 : 271-328 JENKINS, D . , W A T S O N , A, & M I L L E R , G R 1964 P r e d a t i o n a n d R e d Grouse populations J.ApplEcol 1: 183-195. K E N W A R D , R . E 1976 T h e effect of p r e d a t i o n b y Goshawks, Accipiter gentilis, on W o o d pigeon, Columba palumbus, populations D P h i l thesis, University of Oxford 1977- P r e d a t i o n on released Pheasants (Phasianus colchicus) by Goshawks (Accipiter gentilis) in central Sweden. Viltrevy 10: 79-112 1978a. R a d i o transmitters t a i l - m o u n t e d on hawks Orn Scand 9 : in press 1978b. T h e influence of h u m a n a n d G o s h a w k Accipiter gentilis activity on W o o d pigeons

Columba palumbus at brassica feeding sites Ann Appl Biol 8 9 : 277-286 K L U Y V E R , H . N 1970 R e g u l a t i o n of n u m b e r s in p o p u l a t i o n s of G r e a t Tits (Parus m major). I n DEN B O E R , P J , & G R A D W E L L , G R (eds) Dynamics of Populations, p p 507-523. W a g e n i n g e n K R A M E R , V. 1955 Habicht und Sperber W i t t e n b e r g 1967. Die P o p u l a t i o n s d y n a m i k bei H a b i c h t u n d Sperber Der Falke 14: 4 0 - 4 1 , 78-81. M U R I E , A. 1944 The Wolves of Mount McKinley W a s h i n g t o n M U R T O N , R . K 1965 N a t u r a l a n d artificial p o p u l a t i o n control in t h e W o o d - p i g e o n Ann. Appl Biol 5 5 : 177-192 1971. Man and Birds L o n d o n 1974- T h e i m p a c t of a g r i c u l t u r e on birds. Ann Appl Biol 7 6 : 358-365 & ISAACSON, A. J 1964 P r o d u c t i v i t y a n d egg p r e d a t i o n in the W o o d p i g e o n Ardea 52: 30-47. , & W E S T W O O D , N . J 1963 T h e

feeding ecology of the W o o d p i g e o n Brit Birds 5 6 : 345-375, , 1966. T h e relationship b e t w e e n W o o d p i g e o n s a n d their clover food supply a n d the m e c h a n i s m of p o p u l a t i o n control. J Appl Ecol 3 : 55-96 , W E S T W O O D , N . J , & ISAACSON, A J 1964 A p r e l i m i n a r y investigation of t h e factors r e g u l a t i n g p o p u l a t i o n size in t h e W o o d p i g e o n Columba palumbus. Ibis 106: 482-507 O P D A M , P. 1975 I n t e r - a n d intraspecific differentiation w i t h respect to feeding ecology in two s y m p a t r i c species of the genus Accipiter. Ardea 6 3 : 30-54 SHARROCK, J . T R 1976 The Atlas of Breeding Birds in Britain and Ireland B e r k h a m s t e d S T O R E R , R . W 1966 Sexual d i m o r p h i s m a n d food habits in three N o r t h A m e r i c a n accipiters. Auk 8 3 : 423-436 T H I O L L A Y , J . M 1967 Ecologie d u n e p o p u l a t i o n d e rapaces diurnes en L o r r a i n e Terre et Vie

2 1 : 1 1 6 - 1 8 3 . T I N B E R G E N , L. 1936 Gegevens over h e t voedsel von N e d e r l a n d s e H a v i k e n (Accipiter gentilis gallinarum ( B r e h m ) ) . Ardea 2 5 : 195-200 1946. D e r S p e r w e r als roofvijand von zangvogels Ardea 3 4 : 1-213 U T T E N D O R F E R , O . 1952 Neue Ergebnisse uber die Ernahrung der Greifvogel und Eulen S t u t t g a r t VAN BEUSEKOM, C. F 1972 Ecological isolation with respect to food b e t w e e n S p a r r o w h a w k a n d Goshawk Ardea 6 0 : 72-96 W E N D L A N D , V . 1953- P o p u l a t i o n s s t u d i e n a n R a u b v o g e l n J Orn 9 4 : 103-113 WILLIAMSON, K . 1967 T h e b i r d c o m m u n i t y of f a r m l a n d Bird Study 14: 210-266 W O O D F O R D , M . H i960 A Manual of Falconry L o n d o n Dr R. E Kenward, Department of Zoology, Edward Grey Institute of Field Ornithology, South Parks Road, Oxford oxi cjps 74 Predation by Goshawks in Britain vj. Mystery photograph 27 What is this species? Answer next

month