Duur van de eifase en berekening van het legbegin bij de Sperwer Accipiter nisus

In 2003, 24 nests of Eurasian Sparrowhawks in the northern Netherlands were visited at least once during the egg laying period, during which 8x 1 egg, 8x 2 eggs, 5x 3 eggs and 3x 4 eggs were recorded. As timing of nest visits was randomly distributed over the day, it was impossible to exactly pinpoint start of laying (possible variation 0-2 days, assuming a laying interval of two days). Mid-point assumption was used to calculate laying date of the first egg (error ± 1 day). From 37 days after onset of laying onwards, nests were daily (eggs pipping) or every other day (no life sign yet) visited in order to establish hatching day (maximum error one day). All 24 females were in their second year of life or older (based on observations and findings of moulted feathers); the same applied to 11 males of which age could be recorded. Consequently, effect of age upon length of the egg stage could not be quantified. The duration of the egg stage varied between 37 and 42 days (mean 39.96 days), depending on clutch size: 38.50 ± 2.12 days in C/4 (n=2), 39.63 ±1.31 days in C/5 (n=16) and 41.33 ± 0.52 days in C/6 (n=6) (Fig. 1). This increase in incubation period was significantly correlated with clutch size (Pearson, r=0.589, n=24, p=0.002). The length of the egg stage was not correlated with lay date of the first egg (Fig. 2), nor when corrected for clutch size. Also, no correlation was found between laying date and clutch size, presumably because our small sample of Sparrowhawks (except one) started laying rather early and within a period of 10 days. Comparing these data with published accounts of back-calculating onset of laying (Bijlsma 1997, van Diermen & Donkers 2002), a rather wide range of incubation periods emerges (defined as the number of days elapsed between laying of first egg and hatching of first chick). This range is further widened by errors in the age- estimation of chicks (up to 2 days; Vedder & Dekker in prep.). It is therefore suggested not to back-calculate onset of laying, but instead use hatching date of the first chick in studies on seasonal trends in clutch and brood size, sex ratio and survival.