When to start and when to stop: Effects of climate on breeding in a multi‐brooded songbird

Abstract

Climate warming has been shown to affect the timing of the onset of breeding of many bird species across the world. However, for multi‐brooded species, climate may also affect the timing of the end of the breeding season, and hence also its duration, and these effects may have consequences for fitness. We used 28 years of field data to investigate the links between climate, timing of breeding, and breeding success in a cooperatively breeding passerine, the superb fairy‐wren (Malurus cyaneus). This multi‐brooded species from southeastern Australia has a long breeding season and high variation in phenology between individuals. By applying a “sliding window” approach, we found that higher minimum temperatures in early spring resulted in an earlier start and a longer duration of breeding, whereas less rainfall and more heatwaves (days > 29°C) in late summer resulted in an earlier end and a shorter duration of breeding. Using a hurdle model analysis, we found that earlier start dates did not predict whether or not females produced any young in a season. However, for successful females who produced at least one young, earlier start dates were associated with higher numbers of young produced in a season. Earlier end dates were associated with a higher probability of producing at least one young, presumably because unsuccessful females kept trying when others had ceased. Despite larger scale trends in climate, climate variables in the windows relevant to this species’ phenology did not change across years, and there were no temporal trends in phenology during our study period. Our results illustrate a scenario in which higher temperatures advanced both start and end dates of individuals’ breeding seasons, but did not generate an overall temporal shift in breeding times. They also suggest that the complexity of selection pressures on breeding phenology in multi‐brooded species may have been underestimated.

Fitness consequences of longer breeding seasons of a migratory passerine under changing climatic conditions

“…Similar results were described in a few passerine species studied in areas where climatic conditions did not change throughout the study period (Lv et al, 2019;Tarwater & Arcese, 2018;Townsend et al, 2013; see Section 1 for details). Research analysing links between season duration and fecundity, in the areas where climate has changed, are very rare.…”

“…In a population of song sparrows, the duration of breeding period strongly affected female annual fecundity, but no consistent temporal pattern in laying dates and the duration of breeding season has been found (Tarwater & Arcese, 2018 ). Likewise, in Australian fairy‐wrens earlier laying dates were associated with a higher number of young produced in a season by successful females but the authors did not find temporal trends in phenology or climate during their study period (Lv et al., 2019 ). Female prothonotary warblers increased the proportion of second broods across time, but it was not related to changes in laying dates (Bulluck et al., 2013 ).…”

Do the ages of parents or helpers affect offspring fitness in a cooperatively breeding bird?

“…As each parent has a single measure for lifespan, but may have produced offspring at multiple ages across their lifespan, fitting lifespan and parental ages therefore effectively models between-versus within-individual differences [38]. Julian incubation date (the number of days counted from 1 January of the calendar year of the cohort) was included in order to control for any potential changes in chick performance across the breeding season [47,48]. Julian incubation date was ztransformed (to zero mean and unit standard deviation) in all models.…”

Aging and Senescence across Reproductive Traits and Survival in Superb Fairy-Wrens (Malurus cyaneus)

“…For every year that a female was alive, the Julian day on which she began incubating her first clutch of the breeding season was determined (number of days counted from January 1 st ). In order to control for effects of annual weather and food conditions (Lv et al 2019), the median population-level breeding start date for that given year was subtracted from each breeding start date. Thus, breeding start date was a relative value that compared each breeding female to the population median in that year.…”

“…Since the fairy-wren breeding season can last up to six months of the year (Lv et al 2019), it is fairly common for an individual to die part way through a breeding season. Death during the breeding season could plausibly lessen the estimate of annual reproductive success of both males and females solely due to the decreased window of time they have to breed in that year, irrespective of any intrinsic decline in reproductive potential.…”