4.1 Survival
Adult survival of seabirds is generally high (Dias et al. 2019). Long-term mean adult survival rates for penguins usually exceed 0.8, with most (9 of 13 species reported) above 0.85 (Bird et al.2020). Even in a population of Magellanic penguins (S. magellanicus ), declining at ~1.3% per annum, adult survival was >0.8 in 23 of 25 study years (92%; Gownaris & Boersma 2019). Based on this, African penguin survival rates in this study appear to be relatively low compared to other penguin species, especially at Robben Island (0.77 ± 0.02) where survival was only >0.8 in 2 of 7 years (28%). However, our estimates remain consistent with previous estimates for African penguins (e.g., Wolfaardtet al. , 2008; Sherley et al. , 2014) and represent the first survival estimates from African penguins where no individuals were tagged with potentially harmful flipper bands. Within this study, we found that changes in survival are predominantly driven by changing prey (sardine) abundance, with lower abundance underpinning lower survival at Robben Island and higher survival at Stony Point. Understanding the drivers of these colony-specific differences is fundamental to successful future population management of African penguins.
The finding that survival declines along with sardine abundance at Robben Island is consistent with previous literature (Sherley et al. 2014; Robinson et al. 2015). This response underlines the recent concern for the long-term viability of the colony at Robben Island (e.g. Sherley et al . 2018), given the low and declining availability of sardine to seabirds off western South Africa (Robinsonet al. 2015, Crawford et al . 2019). However, the increasing survival rates with decreasing food abundance in individuals at Stony Point are more surprising. This may be explained by the presence of additional factors (e.g., predation, density dependence) which can impact survival differently across colonies (Weller et al . 2016); i.e., food availability may not currently be the dominant external driver of variation in survival at Stony Point. Alternatively, this may be explained by limitations within our analysis; for example, the fisheries data we used index fish abundance across a large area of South Africa’s coastal waters (Coetzee et al. 2008) and may not necessarily adequately reflect localised food availability at both colonies. Eastward displacement of sardine and anchovy in the Benguela upwelling system is driving decreased food availability for seabirds in the Western Cape (Crawford 2007, Crawford et al . 2019). With Stony Point located >70km south-east of Robben Island, localised food availability may be higher around Stony Point because of this displacement. Supporting this, a recent study found adult African penguins in Western Cape colonies situated further east (i.e., Stony Point) had a higher body mass compared to those further west (i.e., Robben Island) (Espinaze et al. 2020). However, further research with colony-specific estimates of prey availability (e.g., Campbellet al . 2019) is required to confirm this, alongside additional monitoring of external factors (e.g., predation rates) that may be driving inter-colony differences in survival.