Analysis of Nest Record Cards for the Buzzard

Transcription

1 Bird Study ISSN: (Print) (Online) Journal homepage: Analysis of Nest Record Cards for the Buzzard C.R. Tubbs To cite this article: C.R. Tubbs (1972) Analysis of Nest Record Cards for the Buzzard, Bird Study, 19:2, , DOI: / To link to this article: Published online: 23 Jun Submit your article to this journal Article views: 64 View related articles Citing articles: 4 View citing articles Full Terms & Conditions of access and use can be found at Download by: [ ] Date: 26 November 2017, At: 14:25

2 Analysis of Nest Record Cards for the Buzzard by C. R. Tubbs Downloaded by [ ] at 14:25 26 November 2017 The data amassed by the B.T.O. Nest Records Scheme for the Buzzard were used by the author to fortify the results of a personal study of Buzzards in the New Forest, Hampshire. Analysis of all the cards, and comparison of the results on a regional basis, showed that variations in clutchsize and fledging success exist. DURING A STUDY of the population of Buzzards Buteo buteo in the New Forest, Hampshire, between 1961 and 1970, I found that the average clutch-size and output of young was consistently lower than appeared from fragmentary evidence to be usual in some other regions. I therefore used the British Trust for Ornithology Nest Record Scheme to compare the success of the New Forest population with success elsewhere. At the same time it seemed logical to extend the analysis back in time to see if regional or national trends in clutch-size, hatching and fledging success were discernible over a longer term. However, the cards proved to have considerable limitations. Not only were the numbers returned annually small, but complete histories from eggl aying to fledging were recorded for comparatively few nests. Information about hatching success proved to be inadequate. Despite these limitations some features emerged from the analysis which it is felt are worth while recording. SIZE AND DISTRIBUTION OF THE SAMPLE A total of 874 nest record cards yielded information about clutchsize and fledging success. Of these, 852 referred to the period and 24 (mostly from Westmorland and northwest Yorkshire) to the period Not until 1948 did the annual total achieve double figures, and not until 1964 were more than 50 cards returned for any one year. The maximum annual return was 91 cards in 1968 (Table I). There are great distributional disparities in the origin of cards and the map (Figure 1) shows the numbers returned by counties. Of the 874 cards, 534 (about 60%) are from only five areas the New Forest (158 cards), Pembrokeshire (65), Westmorland (157), the North and West Ridings of Yorkshire (79), and Inverness-shire (75). These areas afford the only lengthy 'runs' of substantial annual numbers of cards. Of the 158 New Forest cards, 151 refer to the period and were filled in during the course of the population study some early results of which were published in Tubbs (1967). Of the Pembrokeshire cards, 34 were from Skomer Island during , the period of a study published by 97

3 BIRD STUDY Davis and Saunders (1965). The Westmorland and Yorkshire cards extend over the whole period and were returned mainly by Sedbergh School Ornithological Society, R. W. Robson and D. C. Bishop. Those for Inverness-shire were returned mainly by the Hon. Douglas Weir for the years Downloaded by [ ] at 14:25 26 November 2017 TABLE I-NATIONAL ANALYSIS OF CLUTCH SIZE AND BREEDING SUCCESS OF BUZZARDS Total Cards Clutch Size No. Nests Average Outcome of Breeding Known Average No. Cards Nests No.* Young Used Failed Reared Year TOTALS * This is the average for all nests of known outcome, including total failures. ANALYSIS OF THE CARDS Clutch-size and fledging success-national analysis. Annual averages for clutch-size and fledging success are given in Table I. Certain assumptions made in obtaining the figures are outlined below in an explanation of the Table. Average clutch-size was calculated from cards on which clutch-size was recorded, or could be reasonably inferred from brood-size (plus remaining eggs, if any) whilst the young were still completely in down. Experience in the New Forest (and indeed the evidence of the cards) has shown that addled eggs are likely to remain in the nest for up to three weeks after the young hatch, whilst losses among the young are minimal until the oldest of the brood is more than two weeks old. Though the figures are minimal in view of the assumptions made, so too is the margin of error. The figure in the last column of Table I refers to the average number of young reared per 98

4 3- Average Clutch Size NEST RECORD CARDS FOR BUZZARD nest taking into account all nests (including the failures recorded in the previous column) the outcome of which was known or could reasonably be inferred. The sample was enlarged by including broods seen feathered or almost feathered but not yet ready to fly. The figure in the last column is thus the average number of young fledged or strongly presumed to have fledged. There is no direct relationship in Table I between the number of clutches whose size was known and the number of nests for which the outcome was known because, on the one hand, many nests were found with eggs but were not revisited late enough in the fledgling-period, and, on the other hand, many nests were found only when they had large young. An attempt was made to obtain annual averages for hatching success but, except for instances in which clutch and brood sizes were similar, the cards failed to yield a satisfactory sample because of the irregularity with which most nests were visited. To assist interpretation, the data for average clutch-size and fledging success are reproduced in a graph (Figure 2) which discards the few data for the years before The main feature of Figure 2 is perhaps the relative inconsistency of annual average clutch-sizes and breeding success, the former varying between 1.9 and 3.0 eggs per clutch and the latter between 0.9 and 1.6 young reared per nest. It will be shown later that clutches in the south tended to be smaller than in the north and it was thought at first that the inconsistencies might have arisen as a result of annual variations in the regional distribution of the sample. For example in years of low average clutch-size there might have been a predominance of cards from the south. Regional analyses, however, showed that this was not so, and in 1962, when average clutch-size was lowest, 19 out of 24 cards used were from Wales and northwest England Year Figure 2. National Analysis of clutch size and breeding success of Buzzard,

5 BIRD STUDY A second feature of Figure 2 is the slight, progressive, decline in average clutch-size from and the equally slight increase from It would be hazardous, however, to attempt positive interpretation of these data in view of the small size and irregular distribution of the annual sample. Certainly no major slump in either clutch-size or breeding success, such as might be expected to mark the myxomatosis episode after 1953, is readily apparent from Figure 2, though does coincide with one of the 'lows' in both curves. It is, however, important to remember that nest record cards were only filled in for pairs of Buzzards which laid eggs; they give no information on the incidence of non-breeding. The average size of 641 clutches between was 2.56 eggs and the average number fledged from the 645 nests for which the outcome was known was 1.37 young. For the shorter period referred to above, the averages are 2.46 (620 clutches) and 1.33 (625 nests) respectively. Single addled eggs in otherwise fertile clutches, and losses of one (occasionally more) young from broods during the fledgling period were frequent but the difference between average clutch-size and average numbers fledged is largely accounted for by the numbers of complete clutches and broods which failed. Of the 645 nests whose outcome was known between , 146 failed, most of them during the incubation period. Clutches Lost Circumstances TABLE II-CLUTCH AND BROOD LOSSES OF BUZZARDS Clutches or Broods Lost No. Circumstances No. Broods Lost Circumstances Human predation 34 Human predation 8 Human predation 7 Found addled or deserted 32 Disappeared 12 Found dead on nest 7 Found broken 21 Tree blown down 1 Disappeared 7 Disappeared 14 Tree felled 1 Tree felled 1 Nest flooded 1 TOTALS NOTE: The middle column lists instances in which it was uncertain whether eggs of young had been lost. These failures are analysed in Table II, from which it is apparent that human predation was the most often recorded cause of losses. Moreover, human predation is almost certainly under-represented in Table II: only instances where this was definitely the cause of failure are recorded there, but the circumstances in which numerous other clutches and broods disappeared or were deserted strongly suggested the intervention of man. Thirty-nine of the 49 clutches and broods recorded in the Table as lost to human predation were in northwest England and Wales; only six were in Scotland, three in southwest England and one in the New Forest. In addition to losses of eggs and young there are a number of recorded instances in which the adult Buzzards were shot or trapped at the nest. It is fairly certain that gamekeepers and farmers were responsible for most of the losses due to man, though egg-collectors also took their toll. No. 100

6 NEST RECORD CARDS FOR BUZZARD Ratcliffe (1970) postulated a casual relationship between organochlorine contamination, decreased egg-shell thickness and egg-breaking in some raptors, commencing in the 1940s. However, he found no sign of egg-shell change and only occasional egg-breaking in the Buzzard. It is therefore of interest that 21 clutches were recorded on the cards as having been broken, and in 14 cases the descriptions of the broken clutches are consistent with egg-breaking by adult birds as described by Ratcliffe comminuted shell, often held together by the lining membranes; and occasionally complete shells save for single large punctures. In one instance Speyside in 1965 an adult was actually disturbed whilst eating its eggs. In three other cases the progressive depletion of a clutch over several days is recorded. In most of the 14 cases, breakages occurred in the later part of the incubation period. In the remaining seven cases, making up the total of 21, the clutches are simply recorded as 'broken', though in four instances the opinion is given that crows Corvus corone were responsible. 60- Year ----Total Percentage Failures Percentage Failures Due To Human Predation Figure 3. Percentages of Buzzard nests which failed, Figure 3 shows the annual percentage of nests which failed during It also shows the percentage which failed owing to human predation. The most conspicuous feature of the graph is the sharp fall in the percentage of failures after This is difficult to interpret other than in terms of reduced persecution by man, which implies, as suggested earlier, that human predation may have been a more important cause of failure than the actual figures for definite losses due to man suggest. It is at least encouraging that there is evidence of reduced direct pressure from man on breeding Buzzards in the past decade. 101

7 BIRD STUDY Clutch-size and fledging success-regional analysis In order to explore whether the national analysis masked regional characteristics, annual clutch-size and fledging success graphs were produced for each of the following regions, the boundaries of which are shown in Figure 1: 1. The New Forest, Hampshire 2. Southwest England 3. Wales and the Marches 4. Northwest England 5. Scottish Highlands and Islands Space does not permit reproduction of the graphs and they have been deposited with the British Trust for Ornithology. They can be interpreted only with great caution because of the small and fluctuating nature of the annual samples. It is not possible to distinguish with confidence any clear regional trends in clutch-size or output of young, though in most graphs there is some suggestion of larger clutches before the mid-1950s than since, and the graph for the Scottish Highlands does show a definite increase in the average size of clutches from 2.0 in 1962 to 3.1 in The graphs for average fledging success exhibit comparatively wild fluctuations (except in the New Forest) which probably reflect the inherent inadequacies of the sample rather than the real situation. The sample is smaller and more varied in size from year to year than that for clutch-size, with the increased chance of 'freak' results. In particular, extremes in the curves are noticeably associated with very small numbers of nests-often a single nest which happened to be a failure. TABLE ID-AVERAGE CLUTCH SIZE AND FLEDGING SUCCESS OF BUZZARDS Average Clutch Size Average Fledged! Nest* Average Fledged! Successful Nest Region New Forest SW England Wales and the Marches NW England Scottish Highlands * i.e. total number of nests the outcome of which was known. These regional data, however, do suggest a latitudinal increase in clutch-size and breeding success from south to north. Table III gives average clutch-size for the period (before which the data are mainly from northwest England) and also for the period (to which most of the New Forest data are confined). The larger differences between average clutch-size and average fledging success in Wales, northwest England and Scotland, than in the south are largely the result of higher predation (mainly human predation), and in order to show regional differences in potential reproduction more clearly the Table also gives averages fledged per successful nest. The Table confirms that 102

8 NEST RECORD CARDS FOR BUZZARD both clutch-size and fledging success in the New Forest from was significantly lower than elsewhere and that both were highest in the Scottish Highlands. In northwest England and the Scottish Highlands over the period, clutches of two or (more often) three eggs were usual, with clutches of four eggs occurring frequently (especially in the Highlands); a clutch of five eggs was recorded from Yorkshire (West Riding) in Clutches of a single egg were rare both in northwest England and Scotland. In Wales and southwest England clutches of two or three eggs were usual, occurring with approximately equal frequency, whilst clutches of four or one occurred only occasionally. In the New Forest (between ) clutches of one and two eggs occurred in an approximate ratio of 1: 3, while clutches of three eggs were infrequent. TABLE IV-SITUATIONS OF BUZZARDS' NESTS Tree in Isolated or hedge- Inland crag, cliff, Region woodland row tree quarry Sea cliff New Forest 158 SW England Wales and the Marches NW England Scottish Highlands Scotland S Uplands 3 2 Antrim 2 Surrey/Sussex 2 Nest-sites All 874 cards recorded albeit usually sketchily the situation of nests. Table IV summarises this information by regions. It should be recognised that the figures in Table IV only give a general indication of the frequency with which different types of site were used because they refer to the numbers of cards and not the number of nests many cards clearly refer to the same nest in different years. Also the Welsh sample, by virtue of the Skomer study, is biased towards sea-cliffs. Nevertheless it is possible to suggest from Table IV that the use of nest-sites other than in woodland tends to be governed more by the absence of woodland than by inherent preference. SUMMARY The main objects of the paper are to record an analysis on the British Trust for Ornithology Nest Record Cards for national or regional characteristics in the clutch-size and fledging success of the Buzzard Buteo buteo. The cards proved to have considerable limitations because few of them recorded complete breeding histories and their numbers were comparatively small and exhibited great geographical disparities in origin. The average of 641 clutches between 1937 and 1969 was 2.56 eggs and the average number of young fledged from 645 nests was Of the 645 nests whose outcome was known, 146 failed completely, mainly during the incubation period and largely through human predation. 103

9 BIRD STUDY Between annual average clutch-sizes varied between 1.9 and 3.0 eggs. The average numbers of young reared from the total number of nests for which the outcome was known varied between 0.9 and 1.6 over the same period. It was not possible to identify with certainty any trends such as might be expected to coincide with the myxomatosis episode or the intensive use of organochlorine pesticides. Analysis of the data for different regions suggested a latitudinal increase from south to north in clutch-size and fledging success, and showed that both clutchsize and output of young in the New Forest, Hampshire, were significantly lower than elsewhere. Analysis of the situation of nests suggests that the use of nest-sites other than in woodland trees depends more on the absence of such sites than on inherent preference. Downloaded by [ ] at 14:25 26 November 2017 ACKNOWLEDGEMENTS I should like to thank Mr Henry Mayer-Gross and Mr John Parslow for their many valuable comments on the first draft of this paper. Also I thank Mr T. A. W. Davis, Mr R. W. Robson and Sedbergh School Ornithological Society for answering various queries, also Mr R. A. Miller for drawing the figures. REFERENCES DAvis, T. A. W. & D. R. SAUNDERS Buzzards on Skomer Island, Nature in Wales, 9: RATCLIFFE, D. A Changes attributable to pesticides in egg breakage frequency and eggshell thickness in some British birds. J. App!. E,col., 7: TUBBS, C Population study of Buzzards in the New Forest during Brit. Birds, 60: C. R. Tubbs, The Nature Conservancy, Shrubbs Hill Road, Lyndhurst, Hampshire. 104

10 Photo by courtesy of Derick Scott Plate 4. Tawny Owl carrying a dead -r House Sparrow. See pp

11 Photo by courtesy of D. Billett Plate 5. Short-eared Owl pellets collected by D. and R. A. Billett in Farlington Marshes, Hampshire, during the 1970/71 winter. Pellets 1-2 contained short-tailed voles, pellet 3 Starling, pellets 4-6 Dunlins. Other pellets contained Dunlin rings, including a 3-years' old Swedish bird. Photo by courtesy of Eric Hoskin% Plate 6. A group of Tysties. See pp