One of the goals of the CRACAB project is to understand how contemporary climate variability influences the reproductive success of Ascension Island’s nesting seabirds, and thus how their conservation status might be affected by longer-term climate change. As a first step in this process, researchers from the University of Exeter and Ascension Island Government Conservation & Fisheries Directorate have been analysing data collected from the Seabird Monitoring Programme to measure annual variability in fledging success and explore some of the key oceanographic processes that might influence it.

A female Ascension frigatebird with young chick

Initial analyses have focussed on the endemic Ascension frigatebird, which is the Island’s most numerous large-bodied seabird and is of high conservation significance. Monitoring data are available for 1078 nests monitored since 2012 when an accessible colony first established on the main island, although only in more recent years has this colony been sufficiently large to reliably estimate fledging success. A preliminary analysis of these data confirmed that nesting of Ascension frigatebirds is highly seasonal, with a peak in egg-laying in early October and 95% of observed nesting attempts initiated over a 5-month period between 27th July and 1st January. For the purposes of measuring reproductive performance, we can therefore divide monitoring data into discrete ‘nesting seasons’ with egg laying beginning on 1st May and ending on 30th April. Based on this measure, fledging success varied from 43-65% between 2015-16 and 2019-2020 (seasons with > 100 nests monitored) and tended to decline over the course of each season with earlier nests performing better than later nests (Figure 1).

Figure 1. Summary of available frigatebird productivity monitoring data. A) shows the overall seasonal distribution of egg-laying based on 1078 nests monitored between 2012 and 2020. The broken lines enclose the core season during which 95% of nesting attempts were initiated. B) shows the seasonal trend in fledging success for nests initiated at different points over the core season between 2015 and 2020 (years with > 100 nests monitored); C) shows inter-annual variability in mean fledging success between 2015 and 2020, represented as deviations from the overall mean (56.8%). Illustration (c) Shirley Merrell.

Why is nesting seasonal, and why does breeding success vary?

In many seabirds (and birds in general) breeding is timed to coincide with seasonal peaks in food availability, and reproductive performance is closely linked to the availability of prey. While climate is less variable in the tropics than in temperate regions, we hypothesised that similar constraints may influence the breeding of Ascension frigatebirds.

Frigatebirds feed primarily on small fish and squid (collectively known as ‘micronekton’) whose seasonal and inter-annual availability may in turn be influenced by primary and secondary production lower down the food chain. To quantify variation in prey availability, we analysed satellite-derived sea surface chlorophyll concentrations – a measure of primary productivity – along with SEAPODYM model predictions of zooplankton and micronekton biomass within the known foraging range of Ascension frigatebirds across the period for which nest monitoring has been carried out.

Figure 2. Comparison of seasonality in marine food web productivity around Ascension Island and observed nesting seasonality of Ascension frigatebirds. A) shows seasonal trends in average sea-surface chlorophyll A concentration (a measure of primary production at the base of the food chain) and SEAPODYM model predictions of zooplankton and epipelagic micronekton biomass within a 315 km foraging range of Ascension Island between 2012 and 2020. Biomass estimates are expressed relative to the seasonal maxima for each taxa. Illustrations: public domain via Wikimedia.com; B) shows the overall seasonal distribution of frigatebird egg-laying and chick emergence (assuming an average incubation period of 44 days).

A preliminary analysis of these data shows that egg-laying peaks approximately 1 month after a seasonal peak in primary production driven by upwelling of cool, nutrient rich waters over the equator during the austral winter (see Figure 2). This seasonally-elevated primary production is associated with a corresponding increase in zooplankton biomass that accumulates up the food chain, fuelling a peak in epipelagic micronekton biomass in late November. Thus, the peak emergence of Ascension frigatebird chicks in mid-November occurs at a time when prey is expected to be at its most abundant, supporting the hypothesis that nesting seasonality is food-dependent.

The magnitude of the seasonal peak in primary production has varied considerably over recent years and there is some evidence of a positive correlation with fledging success, although this relationship is driven by single poor year in 2018-19 when both reproductive performance and upwelling-induced chlorophyll enhancement was exceptionally low (Figure 3). In contrast a very productive year in 2015-16 did not result in unusually high fledging success.

Figure 3. Inter-annual variation in primary productivity within the foraging range of chick-rearing Ascension frigatebirds (A) and its relationship with fledging success (B). (A) shows smoothed seasonal trends of mean sea surface chlorophyll A concentration within 315 km of Ascension Island between 2015 and 2020. (B) shows the relationship between average sea surface chlorophyll concentration across each nesting season and the overall fledging success anomaly from Figure 1. The linear regression is plotted (broken line) but is not statistically significant (p = 0.2)

Further analyses are planned incorporating data from the current nesting season and other seabird species, along with a wider range of environmental predictors. However, early results suggest that there may be some predictability in seabird breeding success arising from regional oceanographic variability, which could form the basis for evaluating long-term climate change impacts.