BIJLSMA RG (1998) Breeding biology and population trend of Hawfinches Coccothraustes coccothraustes in Flevoland. LIMOSA 71 (4): 137-148.

The breeding biology of Hawfinches in the recently reclaimed polders Noordoostpolder (woodland planted in 1944-55), Oostelijk Flevoland (planted in 1958-68) and Zuidelijk Flevoland (planted in 1987-96) was studied in 1989-97 (Tab. 1). Woodlands in Flevoland have mostly been planted on clayey and loamy soils, thus permitting the use of a large variety of deciduous trees (Tab. 1, 2). This contrasts with woodlands elsewhere in The Netherlands, which have been planted on poor sandy soils and mainly consist of coniferous trees (mostly Scots pine). During large-scale surveys in 1989-97, each area was covered with five complete visits between mid-March and late June/early July, on average spending 9.5 min/ha in territory mapping scarce and rare breeding birds. Special attention was paid to behaviour and nesting biology of Hawfinches. Densities thus obtained should be read as relative densities.
      Polder woodlands became occupied by Hawfinches between 8 and 18 years after planting (year of planting = 0). Initial settlement was slowest in the 1970s, when the Dutch Hawfinch population was not yet booming (started in mid-1980s), and fastest in the 1980s (countrywide increase). In the latter case, Hawfinches already reached relative densities of >1 pair/10 ha within a few years after settlement. The increase in density levelled off some 20 years after tree planting, at 2.0-2.6 pairs/lO ha in deciduous woodland (Fig. 1). The overall density in woodlands in Flevo1and reached 1.5 territories/lO ha (Tab. 2), but locally much higher densities were found. For example, in 199026 nests were located on 26.8 ha of Carpinus betulus, Quercus robur, Fraxinus excelsior, Picea abies and P. sitchensis in Voorsterbos, i.e. 9.7 nests/lO ha. Flocks of up to 120 Hawfinches are a typical feature of woodlands in the Flevopolders in winter, mainly foraging on seeds of Carpinus betulus and Acer campestre. Flocks started to disperse from early March onwards, and settled in loose colonies in nearby suitable habitats. Hawfinches seemed to prefer Populus spec., Acer pseudoplatanus and mixed deciduous/coniferous stands, apparently avoiding coniferous stands (but not Larix spec.) and Salix spec. (Tab. 2). Nest site choice did not always reflect habitat choice, because Hawfinches preferentially nested at heights of <15 m in dense deciduous and mixed stands and shrubs (Tab. 3), whereas foraging, displaying and pair formation often took place in more open stands (poplars) up to 200 m away from the nest site. Moreover, so-called territorial behaviour in Hawfinches is actually mate-guarding, with no specific relation to the nest site. Exploitation of food sources showed clear seasonal trends, with seeds of Carpinus, Acer and Prunus being important throughout the year, buds of Populus, Fraxinus, Acer and Fagus becoming important in spring and summer and invertebrate prey being taken in May and June (food for nestlings).
      Onset of laying starts in the second half of April, peaks in early May (average onset of laying 12 May in Flevoland, median date 5 May; table 4) and trails off into early July. A similar skewed distribution oflaying dates was found in Drenthe (northern Netherlands) and Veluwe (central Netherlands). Most layings from the second half of May onwards might have been repeat layings (Fig. 2). Second layings were probably rare. Annual variations in onset of laying, as depicted by the first five clutches per year (to exclude repeat and second layings), were significantly correlated with mean April temperature in °C (Fig. 3). An exception to this rule was found in 1977, when normal April temperatures coincided with a very early start of laying. This was caused by widespread breeding in coniferous woodland (nest coverage in April not yet possible in deciduous trees!) following a heavy cone crop in Pseudotsuga menziesii. Onset of laying was not correlated with the Unsen frost index of the preceding winter, apparently because severe winters in The Netherlands rarely coincide with prolonged snow cover. Clutch size was on average 4.68 eggs, very much the same as found elsewhere in The Netherlands (table 4). Clutch size showed a clear seasonal decline, with C/5 and C/6 restricted to April and May and C/3 and C/4 to May and June (Fig. 4). Brood size was relatively large, on average 4.37 in Flevoland and slightly smaller in Drenthe and on the Veluwe (Tab. 4). Nest success in several regions in The Netherlands varied between 38.9 and 58.8%, using the Mayfield-method (Tab. 4). In Flevoland, causes and timing of failure were known in 29 out 34 nests: 21x egg stage and 8x nestling stage. Including circumstantial evidence, most nest failures could be attributed to predation (26x, with only three desertions of clutches), mainly from Jays Garrulus glandarius. Some predation was also noticed by Goshawk Accipiter gentilis (l Hawfinch on 690 prey items), Sparrowhawk A. nisus (21 on 1063 prey items) and Common Buzzard Buteo buteo (Ion 432 prey items). Semi-colonial breeding in Hawfinches not only synchronizes onset of laying (Bijlsma 1979), but is probably also an effective predator deterrent. The proportion of successful nests decreased with increasing nest height, being 80% for nests at heights of 1-5 m (N=15), 68% for nests at heights of 6-10 m (N=57), 61% for nests at heights of 11-15 m (N=31) and 50% in 2 nests at heights of>15 m. Low nests had better coverage, probably resulting in smaller predation rates.
      It is argued that the fast colonisation of, and high overall density in the Flevopolders have been caused by the planting of a large variety of seed and berry carrying trees and shrubs on fertile soils, resulting in excellent feeding and breeding conditions, high fecundity, good breeding success and probably a high survival rate in winter. Some of these conditions may also apply to other parts of The Netherlands, where woodland diversity has improved during the last few decades. Altogether, the Dutch Hawfinch population has at least quadrupled since the mid-1970s, especially after 1985, and has reached a minimum of 15 000 pairs in the late 1990s.

[free pdf] [dutch summary]