Richard S. Turner, Helen L. Osmond, Robert D. Magrath, Andrew Cockburn, Loeske E. B. Kruuk

1. Nest predation is frequently the primary cause of early-life mortality in wild avian populations, generating selection for optimising the timing of reproduction to reduce predation risk. Investigating temporal patterns of nest predation is therefore necessary for understanding the intricate relationships between birds and their predators.
2. In this study, we considered the role of temporal variation in nest predation in a wild population of cooperatively breeding superb fairy-wrens Malurus cyaneus in southeastern Australia, using data collected from nearly 4000 nests over a 27-year period (1994 to 2020). In this species, more than half of all nest attempts end in failure, mostly due to nest predation, with females sometimes initiating as many as ten clutches over their long breeding season.
3. We analysed temporal variation in daily nest predation risk over three temporal scales in relation to: (i) the age of the young within the nest; (ii) the timing of nesting within the breeding season; and (iii) differences between years. For each of these temporal scales, we considered predation during the overall nesting period and for three specific stages of development: (i) the incubation stage (1 to 13 days from the onset of incubation); (ii) the early nestling stage (1 to 5 days post-hatching); and (iii) the late nestling stage (6 to 11 days post-hatching).
4. We found that the average daily risk of predation was lowest during the incubation stage (0.016 ± 0.124 SD), intermediate during the early nesting stage (0.025 ± 0.158 SD) and highest during the late nestling stage (0.066 ± 0.248 SD). Predation increased with the age of the clutch during the incubation stage and with the age of the brood during the early nestling stage, but there was no further increase during the late nestling stage.
5. Throughout the breeding season, daily nest predation rates varied quadratically, with a peak approximately mid-season. There was no evidence that these within-season trends differed between years, and we also found little evidence of any longer-term directional change in daily nest predation rates over the study period. Neither within nor between-year variation in nest predation was related to changes in nest density (i.e., the proportion of active nests at a given time). Instead, within-season patterns closely mirrored the breeding behaviour of pied currawongs Strepera graculina, a large corvid-like passerine that is a common predator of superb fairy-wren eggs and nestlings in our study area.
6. In addition to the temporal variation, we found higher daily rates of nest predation for females assisted by fewer helpers, for younger females, and for nests built at lower heights. However, the significance and magnitude of these effects varied across the different development stages. Furthermore, we found mixed effects of clutch and brood size. Our results therefore indicate a close association between temporal patterns of nest predation in superb fairy-wrens and this seemingly important avian nest predator.