Density-dependent increase in superpredation linked to food limitation in a recovering population of northern goshawks, Accipiter gentilis

S R Hoy, S J Petty, A Millon, D P Whitfield, M Marquiss, D I K Anderson, M Davison, X Lambin

Research output: Contribution to journalArticlepeer-review

16 Citations (Scopus)
8 Downloads (Pure)


A better understanding of the mechanisms driving superpredation, the killing of smaller mesopredators by larger apex predators, is important because of the crucial role superpredation can play in structuring communities and because it often involves species of conservation concern. Here we document how the extent of superpredation changed over time, and assessed the impact of such temporal variation on local mesopredator populations using 40 years of dietary data collected from a recovering population of northern goshawks (Accipiter gentilis), an archetypical avian superpredator. We then assessed which mechanisms were driving variation in superpredation, e.g., was it opportunistic, a response to food becoming limited (due to declines in preferred prey) or to reduce competition. Raptors comprised 8% of goshawk diet on average in years when goshawk abundance was high, which is higher than reported elsewhere. Additionally, there was a per capita increase in superpredation as goshawks recovered, with the proportion of goshawk diet comprising raptors increasing from 2% to 8% as the number of goshawk home-ranges increased from ≤14 to ≥25. This increase in superpredation coincided with a population decline in the most commonly killed mesopredator, the Eurasian kestrel (Falco tinnunculus), which may represent the reversal of the “mesopredator release” process (i.e., mesopredator suppression) which occurred after goshawks and other large raptors declined or were extirpated. Food limitation was the most likely driver of superpredation in this system given: 1) the substantial decline of two main prey groups in goshawk diet, the increase in diet diversity and decrease in goshawk reproductive success are all consistent with the goshawk population becoming food-limited; 2) it's unlikely to be purely opportunistic as the increase in superpredation did not reflect changes in the availability of mesopredator species; and 3) the majority of mesopredators killed by goshawks do not compete with goshawks for food or nest sites.

This article is protected by copyright. All rights reserved.
Original languageEnglish
Pages (from-to)1205-1215
Number of pages11
JournalJournal of Avian Biology
Issue number9
Early online date25 Jul 2017
Publication statusPublished - Sept 2017

Bibliographical note

We are grateful to R. Lourenço and A.K. Mueller for their helpful comments. We thank Forest Research for funding all fieldwork on goshawks during 1973-1996, Forest Enterprise for funding fieldwork after 1998 and T. Dearnley and N. Geddes for allowing and facilitating work in Kielder Forest. This work was also partly funded by a Natural Environment Research Council studentship NE/J500148/1 to SH and a grant NE/F021402/1 to XL and by Natural Research. We thank I. Yoxall and B. Little for the data they collected and their contributions to this study. Lastly, we thank English Nature and the British Trust for Ornithology for kindly issuing licences to monitor goshawk nest sites


  • food-stress hypothesis
  • intraguild predation
  • mesopredator suppression


Dive into the research topics of 'Density-dependent increase in superpredation linked to food limitation in a recovering population of northern goshawks, Accipiter gentilis'. Together they form a unique fingerprint.

Cite this