Voedsel van Nederlandse Torenvalken Falco tinnunculus in de afgelopen eeuw

Kestrels are the proverbial vole predators, but little information on the Kestrel’s diet has been published in The Netherlands since the extensive studies of Bouma (1931), Tinbergen (1940) and Cave (1968). In the present study, pellet analyses and prey remains from nests and sitting posts are used to determine the diet of Dutch Kestrels since the early 1970s, and to compare these data with the previous studies. All together, 1468 pellets were analysed, mostly from Veluwe (68%) and Drenthe (25%), separately for summer (March-August) and winter (September-February). The 338 prey remains were mostly collected during summer in the Veluwe region (Table 1). Both Veluwe and Drenthe are characterised by sandy soils, and are largely forested and interspersed with heathland and farmland. Consequently, the study is strongly biased towards regions where densities of Common Voles are likely well below those in prime habitats like grasslands in the peat district. Microtus arvalis was the most common prey species taken, although more so in winter than in summer. Expressed in biomass, Common Voles amounted to 51% of the prey of Kestrels breeding on the Veluwe, compared to 76% in Drenthe; in winter, respectively 83% and 89% of the prey were Common Voles. Alternative prey of Kestrels in the Veluwe region mostly consisted of birds (in 18 species), and in Drenthe of other voles species (Microtus agrestis, Myodes glareolus), mice (Apodemus sylvaticus) and birds (Table 2). Insects, lizards and frogs were of minor importance. Within the same region (Drenthe), large dietary differences were found in a – presumably the same – male Kestrel, with Common Voles responsible for 93% of prey biomass in late winter and only 38% in the following autumn and winter; in both seasons, local Common Vole densities were poor, though. When comparing Kestrel diets across the 20th and early 21 st century, the proportion of Common Voles was remarkably consistent at an average of 69% of prey biomass. The proportion of Common Voles in the Kestrel’s diet was smaller in the 1980s, when the Common Vole cycle was strongly depressed, and a bit higher in the 1990s, when vole numbers improved and a 3-year cycle took effect (as it had been in the 1970s). However, overall Common Vole numbers on the sandy soils must have declined since the 1970s, as evident from censuses of active burrows in fixed plots on farmland on the Veluwe (since 1968, Figure 5) and in Drenthe (since 1990). Also, Kestrel numbers have steeply declined in both regions; in fact, Kestrels have all but disappeared as breeding birds. The fact that Common Voles are still the single most important prey species in combination with the still high proportion of this prey species in Kestrel diets, shows that Kestrels on sandy soils nowadays are only present when vole numbers are at a peak (but still well below vole densities in the 1970s and earlier in the 20th century). Not surprisingly, reproductive performance (expressed as clutch size and number of fledglings) of Kestrels is positively correlated with vole index (Figure 6). On sandy soils, Kestrels have become scarce breeding birds (although provisioning nestboxes may – to some extent – counter the decline; the impact of ‘compensation areas’ is not known but presumably negligible). Compared to the rest of Europe and North Africa, Dutch Kestrels are typical central European representatives with a high proportion of voles and small numbers of invertebrates in their diet. Voles are (much) less important in southern Europe and North Africa, where invertebrates take the major share in Kestrel diets (Fig. 3). These trends are consistent, even considering large variations in methodologies deployed in dietary studies, seasonal and annual variations in diets and duration of studies.