The demography and ecology of the European Honey-buzzard (Pernis apivorus) in southern Africa

Abstract

Globally, there have been significant declines in long-distance migratory birds. Over the past three to four decades, this trend has been well documented in Palaearctic migrants, which breed in Europe and winter in Africa. In the case of Palaearctic migrants, there has been a strong research focus on the population trends and breeding statistics in Europe but relatively little research in the African non-breeding regions, particularly with regards to habitat loss and transformation. This is despite evidence from the Neotropics that indicates that the non-breeding season of migratory birds is critical to both their survival and breeding success. It is, therefore, essential to understand the non-breeding period of Palaearctic migrants in Africa to aid in their conservation. European Honey-buzzard (Pernis apivorus) is a Palaearctic migrant that is declining globally but displays an increase in records in the southern African sub-region. Honey-buzzard is an unusual raptor species that specializes in preying upon Hymenoptera larvae. In addition, the species is also a forest specialist, breeding in mature woodland in Europe and spending the non-breeding season in tropical rainforest in Africa. In this thesis, I aimed to examine the causes and consequences of the apparent increase in European Honey-buzzards in southern Africa, in order to assess how both Palaearctic migrants and specialist species may be able to adapt to a rapidly changing world. To begin (in Chapter 2), I examined the drivers of the increase in records in European Honey-buzzard in southern Africa. As there was no evidence of an increase in the breeding population in Eastern Europe and Russia (where tracked South African honey-buzzards bred), I addressed two hypotheses: i) the growth in birdwatching and citizen science effort in southern Africa was driving the increase in records, and ii) a loss of suitable habitat further north in the honey-buzzards non-breeding range was pushing honey-buzzards further south, and therefore, causing the proliferation of records in southern Africa. There was evidence that increased birdwatching effort accounted for a portion of the growth in European Honey-buzzard records but did not explain the majority of the increase. When assessing both southern African and East African honey-buzzard records, there was a positive correlation over time between forested area in East Africa and the number of East African honey-buzzard records, and a negative correlation between East African forested area and southern African honey-buzzard records. This indicates that with the loss of forest further north in Africa, European Honey-buzzards have shifted their non-breeding range further south, and the population of honey-buzzards in southern Africa has grown. In Chapter 3, I assessed the sex and age class ratios of the southern African European Honey-buzzard population over the period of its range shift. The southern African adult honey-buzzard population was overwhelmingly female (93.3%). This supports the arrival time hypothesis which predicts that the sex responsible for establishing the breeding territory (male in the honey-buzzard) will winter closer to the breeding grounds than the opposite sex (female in the honey-buzzard). The southern African population had a high proportion of juvenile and second year birds (41.6%). Young birds were more likely to be found in southerly and low elevation locations, likely due to the funnelling effect of Africa on migration. The high proportion of young birds may indicate that they are the drivers of the range shift. In addition to age and sex ratios, I explored the colour morph ratios of the southern African European Honey-buzzard population in Chapter 4. There were significant differences in the colour morph ratios between ages and sexes. The differences between age classes support the hypothesis that there is temporal morph instability between age classes in European Honey-buzzard. Juvenile birds were disproportionately likely to be of the dark morph while adults were disproportionately likely to be of the dark intermediate morph. There may be a selective advantage for the dark morph for juvenile birds who spend a minimum of two years in Africa. For Chapter 5, a large tracking dataset (n = 32) of European Honey-buzzards was used to assess which habitats were preferred on the non-breeding grounds in Africa, and how the availability and fragmentation of these habitats affected honey-buzzard movement. Adult birds were found to have much higher site fidelity and fewer home ranges than juvenile birds, supporting the hypothesis that juvenile birds are likely to be driving the southwards range shift. As expected, honey-buzzards strongly favoured treed habitats over all other habitats. Lastly, honey-buzzards with a large patch of treed habitat had smaller home ranges than those with a smaller treed patch. At the same time, individuals with higher edge density (a measure of fragmentation) also had smaller home ranges. This may reveal that honey-buzzards have some tolerance for fragmented habitat if they have a large forest patch in which to roost and feed. While the birds are able to survive in these fragmented habitats, they may also be at greater risk of detection by predators, and greater daily energy expenditure due to longer daily flight distances between patches. Lastly, European Honey-buzzard primary moult timing and location were assessed in Chapter 6, using stable isotope analysis. Photographs of honey-buzzards in southern Africa showed that outer five primaries (P6-P10) were commonly moulted in the sub-region. This aligns with the period that tracked birds are found in southern Africa. The δ15N, δ13C, and δ2H values of the primary feathers were highly variable between the seven individuals tested. To begin, only two adult birds grew primaries in Europe (two and three feathers respectively). Between the outer two and seven of the adult birds’ primaries appeared to be grown in southern Africa. All other primaries were grown in other parts of Africa, implying that southern African honey-buzzards are growing primaries while migrating, which may result in increased energy expenditure over this period as well as reduced flight effiency. In conclusion, the southwards range shift in European Honey-buzzards has highlighted the complexity of problems facing Palaearctic migrants during the non-breeding season. This specialist has adapted to a novel environment in southern Africa but there are costs involved including longer migration, increased fragmentation of non-breeding habitat, and mistiming of moult. Juveniles are likely driving the movement south; during their two or more years on the African non-breeding grounds, they range widely. This study emphasises the importance of studying migrants on their non-breeding grounds in order to better understand the causes of their declines as well as to develop potential conservation actions.