Predicting the Foraging Habitats of Sympatrically Breeding Gadfly Petrels in the South Pacific Ocean

Abstract

Gadfly petrels (genus Pterodroma) are one of the most threatened groups of birds. They are exceptionally well adapted to forage over enormous areas to maximize chances of encountering prey. Their wide-ranging travel, extensive use of oceanic habitats beyond national jurisdictions (the high seas), and limited information on their at-sea distributions and foraging ecology pose several management challenges. Here, we examined the foraging distributions and habitat preferences of three gadfly petrels that breed on Phillip Island (Norfolk Island Group), in the southwest Pacific Ocean, and tested the ability of species distribution models (SDMs) to predict important marine habitats. GPS loggers were deployed in 2018 and 2019 on chick-provisioning black-winged petrels (P. nigripennis) and white-necked petrels (P. cervicalis) and in 2020 on Kermadec petrels (P. neglecta), and hidden Markov models (HMMs) were used to estimate behavioral states across 387 foraging trips. SDMs were built using six algorithms and the predictive performance of models constructed using conventional random cross-validation (CV) was compared to those constructed with spatially independent CV. All three species demonstrated dual-foraging strategies with short trips closer to the colony and longer, presumably self-provisioning, trips with maximum distances from the colony of several thousand kilometers for black-winged and white-necked petrels. Foraging areas of each species were distinctly partitioned across the Tasman Sea during long trips, but there was high overlap during short trips. Black-winged and white-necked petrels exhibited area-restricted search foraging behavior throughout their foraging ranges which spanned almost the entire Tasman Sea and into the western Pacific, whereas the foraging range of Kermadec petrels was restricted closer to the colony. Approximately half of each species’ foraging range extended into the high seas. Response curves and variable importance between the two SDM CV approaches were similar, suggesting that model fitting was robust to the CV approach. However, evaluation using spatially independent CV indicated that generalizability of ensemble SDMs to new data ranged from poor to fair for all three species. This suggests that the maximal-area foraging strategy of gadfly petrels (whereby they search opportunistically for resources across expansive oceanic habitats) results in weak or wide associations with environmental features making predicting important habitats extremely challenging.

Fine-scale movement data of the endangered Bermuda petrel highlights surface foraging and greater nocturnal flight activity

“…We analysed movement data, activity patterns and diving behaviour to quantify fine scale at-sea behaviours of incubating Bermuda petrels. Similar to previous studies [16, 31], GPS tracking showed that Bermuda petrels undertook extended foraging trips to the NNE of Bermuda, apparently targeting the path of the Gulf Stream, and confirming a reliance on vast oceanic areas for successful foraging as in other Pterodroma petrels [14, 16, 49, 50]. However, in contrast to many other Procellariiformes [10, 27], which can perform dives of considerable duration and depth (see references in [28, 51]), our TDR data suggest that the Bermuda petrel is a surface feeder with limited capacity to dive.…”

“…Bermuda petrels foraged thousands of kilometers away from the breeding colony (average 1200 km) spending on average 10 days at-sea, similar to other Pterodrom species [14,49,50,52]. They were consistent in their departing direction from the colony, as has been observed in the Murphy's petrels (P. ultima) [49], showing a clear preference for the NNE (Fig.…”

Marine distribution and habitat use by Snow Petrels Pagodroma nivea in East Antarctica throughout the non-breeding period

“…It is possible that Snow Petrels spend the majority of their seatime in productive areas where they forage successfully, but this interpretation is limited to nesting birds (a common limitation in seabird tracking studies (Priddel et al, 2014), as those are the individuals accessible to researchers (i.e., juveniles, pre-breeders, and non-breeders are generally not constrained to a nest and are therefore extremely difficult to retrieve from the same location when attaching/detaching tracking devices). State-space modelling of GPS tracking data would allow researchers to identify specific "foraging only" regions [i.e., environmental features actively targeted by the birds; Halpin et al (2022)]. Moreover, with the emergence of circumpolar databases that increasingly provide information on mesopelagic fish species (Woods et al, 2022), and krill (Atkinson et al, 2017), relevant seabird studies whose data overlap with prey data could gain valuable insight on which prey species are most commonly associated with seabird presence.…”

A review of records and research actions for the poorly known Vanuatu Petrel Pterodroma [cervicalis] occulta

“…Given the extreme difficulty of reliably separating Vanuatu Petrels from White-necked Petrels at sea (Shirihai and Bretagnolle 2010), most at-sea records are inferred by proximity to this site, likely underrepresenting the true distribution of the taxon (Figure 1). Recent tracking of provisioning trips by breeding White-necked Petrels from Phillip Island (-29.12, 167.95; one of two current breeding sites for this species) recorded mean “short” trips of 234 km (range Phillip Island 25–433) and 206 km (30–480) and mean “long” trips of 1,063 km (546–2,680) and 810 km (520–1,451) (Halpin et al 2022). It is possible that the Vanuatu Petrel is similar.…”

“…This is a key factor in many seabird breeding colony failures (Ramirez et al 2018). Gadfly petrels are generally less susceptible to this process than other procellariiform species, as they forage over long distances actively searching for prey, rather than targeting specific areas of known high productivity (Halpin et al 2022). With an unknown pelagic range, this process may affect Vanuatu Petrels more acutely than other gadfly petrels.…”