Antaky Et Al. 2019 Nesting Ecology Band-Rumped Storm Petrels.Pdf

Home , Hawaiian petrel

Nesting ecology in the Hawaiian population of an endangered seabird, the Band-rumped Storm-Petrel (Oceanodroma castro) Authors: Carmen C. Antaky, Nicole K. Galase, and Melissa R. Price Source: The Wilson Journal of Ornithology, 131(2) : 402-406 Published By: The Wilson Ornithological Society URL: https://doi.org/10.1676/18-123

BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access titles in the biological, ecological, and environmental sciences published by nonprofit societies, associations, museums, institutions, and presses.

Your use of this PDF, the BioOne Complete website, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/terms-of-use.

Usage of BioOne Complete content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

Downloaded From: https://bioone.org/journals/The-Wilson-Journal-of-Ornithology on 30 Sep 2019 Terms of Use: https://bioone.org/terms-of-use Access provided by University of Hawaii at Manoa The Wilson Journal of Ornithology 131(2):402–406, 2019

Nesting ecology in the Hawaiian population of an endangered seabird, the Band- rumped Storm-Petrel (Oceanodroma castro)

Carmen C. Antaky,1* Nicole K. Galase,2 and Melissa R. Price1

ABSTRACT—The first confirmed nesting location of the en peligro, se descubrio´ recientemente en la isla de Hawai’i despues´ Hawaiian population of the Band-rumped Storm-Petrel de decadas´ de busqueda.´ Posterior a su descubrimiento, analizamos (Oceanodroma castro), an endangered seabird, was recently las preferencias de sitio-nido de este pa´ıno˜ usando un diseno˜ pareado. Los pa´ınos˜ prefieren grietas ma´s profundas comparadas con discovered on Hawai‘i island after decades of searching. las disponibles a menos de 100 m de los sitios de anidacion.´ Las Following the discovery, we analyzed nest site preferences of caracter´ısticas f´ısicas y ambientales de sus sitios-nido podr´ıan ayudar the Band-rumped Storm-Petrel at this site using a paired design. a sus esfuerzos de conservacion,´ incluyendo busquedas´ sobre el Band-rumped Storm-Petrels preferred deeper crevices terreno, remocion´ de mam´ıferos invasores depredadores, compared with those available within 100 m of the nest sites. identificacion´ de sitios potenciales de translocacion´ y la Physical and environmental characteristics of Hawaiian Band- restauracion´ del ha´bitat de esta especie en peligro. rumped Storm-Petrel nest sites may aid conservation efforts Palabras clave: conservacion,´ especie en peligro, evasion´ de including on-the-ground searches, removal of invasive depredadores, islas del Pac´ıfico, Procellariiformes, seleccion´ de sitio- mammalian predators, identification of potential translocation nido sites, and habitat restoration for this endangered species. Received 25 July 2018. Accepted 29 September 2018.

Key words: conservation, endangered species, nest-site selection, PaciﬁcIslands, predator avoidance, Procellariiformes Hawai‘i is known for its rich biodiversity and spectacular endemic species radiations, but Ha- Ecolog´ıa de la anidacion´ de la poblacion´ hawaiana de un waiian biota are at risk due to climate change, light ave marina en peligro, el pa´ıno˜ Oceanodroma castro pollution, and invasive species (Fortini et al. 2015, Curry et al. 2016, Raine et al. 2017b). In the RESUMEN (Spanish)—La primera localidad de anidacion´ de la poblacion´ hawaiana del pa´ıno˜ Oceanodroma castro, un ave marina Hawaiian Islands, the arrival of humans and associated species introductions over the last 1 University of Hawai‘iatManoa,¯ Department of 1,100 yr led to seabird population declines due Natural Resources and Environmental Management, to a lack of coevolved adaptations (Jones et al. Honolulu, HI, USA 2008). The Procellariiformes, which include alba- 2 Colorado State University, Center for Environmental trosses, petrels, shearwaters, storm petrels, and Management of Military Lands funded by Pohakuloa¯ Training Area Natural Resources Ofﬁce, Hilo, HI, USA diving petrels, have declined not only in Hawai‘i * Corresponding author: [email protected] but globally as well (Spatz et al. 2014).

The Band-rumped Storm-Petrel (Oceanodroma seabird within the Pohakuloa¯ Training Area (PTA) castro) in the Pacific region was recently listed on the northern slopes of Mauna Loa between 2,090 under the Endangered Species Act (USFWS and 2,210 m elevation. The climate around this 2015). Once widespread along the Hawaiian island newly discovered Band-rumped Storm-Petrel colo- chain, as evidenced by midden sites on the main ny is relatively cool, dry, and tropical. The colony is Hawaiian Islands (Harrison 1990), its range is now found in 5,000- to 11,000-year-old pahoehoe lava limited (Olson and James 1982, Raine et al. with relatively low light pollution due to the terrain 2017a). and its remote location (Galase et al. 2016). Nest-site selection by species in the order Procellariiformes, such as the Band-rumped Study design Storm-Petrel, is influenced by competition, light The PTA Natural Resource Office (NRO) staff pollution, conspecific attraction, and predator identified nesting locations based on direct and density (Schramm 1986, Podolsky and Kress indirect observations during the 2015–2017 breeding 1989, Buxton et al. 2015). Globally, Band-rumped seasons (Fig. 1). The PTA NRO staff observed more Storm-Petrels nest in a variety of habitat types active nests in the subsequent field seasons following including rock, soil with vegetation, and mixed the discovery of the first active nest in 2015 (Galase rock and soil (Ramos et al. 1997, Slotterback 2019). However, the first active nest site found in 2002), but little is known about their nesting 2015 was also active in 2017. We defined nests of the ecology in the Hawaiian Islands. Band-rumped Storm-Petrel in this study as lava This pelagic seabird spends most of its life at crevices utilized to lay eggs and raise chicks during sea and only comes to land during the summer the breeding season. We defined direct observations breeding season to nest in steep, heavily vegetated of nests (N ¼ 2) by the visual presence of seabird cliffs and high-elevation volcanic terrain (Slotter- activity via camera traps. We defined indirect back 2002). In this study, we explored nest-site observations of nests (N 9) by multiple indicators selection and associated ecological variables in the ¼ only known nesting colony of the Hawaiian of presence including feathers, scent detection, and population of the Band-rumped Storm-Petrel, acoustic evidence of nesting in a crevice. which was recently discovered (Galase et al. Using a paired design, we evaluated factors 2016). We predicted that crevices selected for potentially influencing nest-site selection at the 11 nesting would be deeper and larger compared with sites by comparing characteristics of nest sites with nearby crevices not selected for nesting, consistent those of 4 corresponding randomly selected crev- with findings on nesting preferences of other ices within 100 m of each nest site (Table 1). We Procellariiformes (Bourgeois and Vidal 2007, generated the corresponding random points using Fricke et al. 2015). QGIS software ‘Random Points’ tool based on the GPS locations of the observed nests (QGIS Development Team 2016). We evaluated character- Methods istics including width and height of the outer Study area crevice opening, width and height of the inner The first confirmed Band-rumped Storm-Petrel crevice opening, percent cover of dirt or lava rock nest in the Hawaiian Islands was recently discovered on the ground of the crevice entrance, percent cover using acoustic monitoring, a search dog, and night of vegetation on the ground of the crevice entrance, vision surveillance while conducting surveys for the percent cover of vegetation within a 2 m radius of U.S. Army at Pohakuloa¯ Training Area (PTA) on the crevice entrance, depth, and elevation, for each Hawai‘i island (Galase et al. 2016). Although nest site and 4 corresponding randomly located Hawaiian Petrels (Pterodroma sandwichensis)also points. At the random points, we measured the nest in high-elevation volcanic terrain on Hawai’i closest crevice opening to the point that was island, they are found within the confines of a physically accessible in the pahoehoe lava. We predator-proof fence between 2,440 and 2,900 m then calculated the mean values for the 4 associated elevation on the southeast flank of Mauna Loa in random points and compared them to paired values Hawaii Volcanoes National Park (Hu et al. 2001). for the respective storm petrel nest site. We The Band-rumped Storm-Petrel is the only nesting classified depth as deep (.1 m) or shallow (,1

Downloaded From: https://bioone.org/journals/The-Wilson-Journal-of-Ornithology on 30 Sep 2019 Terms of Use: https://bioone.org/terms-of-use Access provided by University of Hawaii at Manoa 404 The Wilson Journal of Ornithology Vol. 131, No. 2, June 2019

Figure 1. Map of Band-rumped Storm-Petrel nests (N ¼ 11) in Pohakuloa¯ Training Area (PTA), Hawai‘i island.

m) due to measuring constraints in the cavernous applied the Bonferroni correction to determine the and narrow lava system. signiﬁcance of characteristics, as multiple charac- Parametric tests (paired t-tests) were used to teristics were tested (a ¼ 0.05/11 ¼ 0.0045). We identify characteristics that signiﬁcantly differed also computed the standardized difference by between nests and paired random locations. We using Cohen’s d, to enable comparison with other

Table 1. Nest-site selection by the Band-rumped Storm-Petrel, with comparisons of nest sites and randomly selected locations (N ¼ 11) via paired t-tests.

Nest-site scale Nestx ¯ 6 SE Randomx ¯ 6 SE Test statistic P Standardized difference

Width of opening (cm) 104.20 6 30.78 38.82 6 8.09 t10 ¼ 1.57 0.156 d ¼ 0.523 Height of opening (cm) 28.50 6 8.07 15.47 6 1.39 t10 ¼ 1.61 0.138 d ¼ 0.487 Width of inner opening (cm) 93.55 6 22.61 41.11 6 8.76 t10 ¼ 1.29 0.239 d ¼ 0.455 Height of inner opening (cm) 23.50 6 4.05 12.98 6 1.73 t10 ¼ 2.53 0.032 d ¼ 0.801 % of dirt at opening 30.27 6 11.69 10.34 6 4.65 t10 ¼ 1.51 0.162 d ¼ 0.455 % of lava at opening 69.55 6 11.65 89.66 6 4.65 t10 ¼ –1.53 0.158 d ¼ 0.460 % of dirt at 2 m 19.55 6 6.11 8.27 6 2.11 t10 ¼ 1.68 0.124 d ¼ 0.506 % of lava at 2 m 80.45 6 6.11 91.73 6 2.11 t10 ¼ –1.67 0.124 d ¼ 0.506 % of vegetation at opening 1.91 6 1.21 0.46 6 0.46 t10 ¼ 1.07 0.309 d ¼ 0.323 % of vegetation at 2 m 5.73 6 3.62 1.80 6 0.74 t10 ¼ 1.27 0.233 d ¼ 0.383 Elevation (m) 2,145.00 6 11.61 2,145.00 6 11.26 t10 ¼ 0.05 0.962 d ¼ 0.015 a b Depth (frequency) 1.00 6 0.00 0.22 6 0.147 t10 ¼ 11.02 ,0.001 d ¼ 1.764

a Frequency of Deep Nests (.1 m). b Indicates signiﬁcance under the Bonferroni correction (a ¼ 0.05/11 ¼ 0.0045).

studies and datasets. All statistical analyses were Currently, the O. castro nests found on PTA are conducted in the statistical environment of R (R the only known active O. castro nests across the Core Team 2013). entire species range in the Hawaiian Islands. Surveying other areas may be useful in expanding our understanding of the species habitat use on Results Hawai‘i island. High-resolution satellite imagery We found 11 Band-rumped Storm-Petrel nests may aid in the identification of similar terrain, on PTA during the 2017 breeding season. Our suitable for the development of predator-free results indicate that lava crevices were significant- conservation areas. ly deeper (t10 ¼ 11.02, P , 0.001) than randomly Management guidelines for this species are available lava crevices (Table 1). Sample size was under development by the U.S. Fish and Wildlife relatively small, but this characteristic also had a Service. Typical management actions that are large standardized difference (depth d ¼ 1.764; known to protect breeding populations of seabirds Table 1), strengthening support for this result. No include predator control and predator-proof fenc- other variables measured showed a significant ing in suitable nesting areas. This study provides difference between lava crevices selected for guidance in identifying characteristics associated nesting and those nearby. with preferred nest sites in this species. We found that most Band-rumped Storm-Petrel Future recovery efforts for this species may nests on PTA were larger than 1 m deep and include translocation to protected areas away from consisted mostly of lava with less than 6% known threats, such as introduced predators, an vegetation surrounding them (Table 1). The mean approach successful in other seabird species (Jones width of the nest opening was 104.20 6 30.78 cm and Kress 2012). To implement this conservation and the mean height of the nest opening was 28.50 tool, the identification of source populations and 6 8.07 cm (Table 1). We included a photograph of nesting pairs with high breeding success is crucial. an active nest in the supplementary information Further research examining characteristics associ- (Fig. S1). ated with successful nesting attempts will be essential to recover the endangered populations Discussion of the Band-rumped Storm-Petrel.

Band-rumped Storm-Petrels nested in signifi- Acknowledgments cantly deeper lava crevices than those randomly We thank the United States Army Garrison, Pohakuloa¯ available in the area. Previous studies on procel- Natural Resource Office, for base access to the Band- lariiform species have identified similar preferenc- rumped Storm-Petrel breeding grounds. Funding was es for dry locations with deeper and winding nest provided by the U.S. Fish and Wildlife Service Species chambers (Bourgeois and Vidal 2007, Fricke et al. Funding (Grant Number: F16AC00833) and Hawaii Audu- 2015). Preference for depth could reflect the bon Society (Student Research Grant). pressure from predators as narrow and deep lava holes may provide protection from invasive Literature cited mammalian predators, such as cats (Felis catus) Bourgeois K, Vidal E.´ 2007. Yelkouan Shearwater nest- and rats (Rattus spp.), that are found in the area cavity selection and breeding success. Comptes Rendus (Galase et al. 2016). Biologies 330(3):205–214. Previous nesting ecology studies on procellarii- Buxton RT, Anderson D, Moller H, Jones CJ, Lyver PO. 2015. Release of constraints on nest-site selection in form species found that light intensity, predator burrow-nesting petrels following invasive rat eradica- density, and interspecies competition influence tion. Biological Invasions 17:1453–1470. nesting preferences (Schramm 1986, Ramos et al. Curry PA, Yeung NW, Hayes KA, Meyer WM, Taylor AD, 1997, Buxton et al. 2015). We did not test for these Cowie RH. 2016. Rapid range expansion of an invasive factors because there is no artificial light source in predatory snail, Oxychilus alliarius (Miller 1822), and its impact on endemic Hawaiian land snails. Biological view at the study location, no difference in Invasions 18:1769–1780. predator management across the nesting area, and Fortini LB, Vorsino AE, Amidon FA, Paxton EH, Jacobi JD. no other seabird species nesting in the study area. 2015. Large-scale range collapse of Hawaiian forest

Downloaded From: https://bioone.org/journals/The-Wilson-Journal-of-Ornithology on 30 Sep 2019 Terms of Use: https://bioone.org/terms-of-use Access provided by University of Hawaii at Manoa 406 The Wilson Journal of Ornithology Vol. 131, No. 2, June 2019

birds under climate change and the need 21st century QGIS Development Team. 2016. QGIS Geographic Infor- conservation options. PLOS ONE 10(10):e0140389. mation System. Open Source Geospatial Foundation Fricke EC, Blizzard KM, Gannon DP, Mauck RA. 2015. Project. http://qgis.osgeo.org Model of burrow selection predicts pattern of burrow Raine AF, Boone M, McKown MA, Holmes NI. 2017a. The switching by Leach’s Storm-Petrels. Journal of Field breeding phenology and distribution of the Band- Ornithology 86:326–336. rumped Storm-petrel Oceanodroma castro on Kaua‘i Galase NK. 2019. First confirmed Band-rumped Storm and Lehua Islet, Hawaiian Islands. Marine Ornithology 45:73–82. Petrel Oceanodroma castro colony in the Hawaiian Raine AF, Holmes ND, Travers M, Cooper BA, Day RH. Islands. Marine Ornithology 47:25–28. 2017b. Declining population trends of Hawaiian Petrel Galase NK, Doratt RE, Inman-Narahari NV, Lackey TM, and Newell’s Shearwater on the island of Kaua‘i, Schnell LD, Peshut PJ. 2016. 2015 seabird project Hawaii, USA. Condor: Ornithological Applications technical report: Band-rumped Storm Petrel (Ocean- 119:405–415. odroma castro) colony presence and flight paths at Ramos JA, Monteiro LR, Sola E, Moniz Z. 1997. Pohakuloa¯ Training Area Hawaii. Pohakuloa (HI): Characteristics and competition for nest cavities in Natural Resources Office. burrowing Procellariiformes. Condor 99:634–641. Harrison CS. 1990. Seabirds of Hawaii. Ithaca (NY): R Core Team. 2013. R: A language and environment for Cornell University Press. statistical computing. Vienna (Austria): R Foundation Hu D, Glidden C, Lippert JS, Schnell L, MacIvor JS, for Statistical Computing. http://R-project.org/ Meisler J. 2001. Habitat use and limiting factors in a Schramm M. 1986. Burrow densities and nest site population of Hawaiian Dark-rumped Petrels on Mauna preferences of petrels (Procellariidae) at the Price Loa, Hawaii. Studies in Avian Biology 22:234–242. Edwards Islands. Polar Biology 6:63–70. Jones HP, Kress SW. 2012. A review of the world’s active Slotterback JW. 2002. Band-rumped Storm-Petrel (Ocean- seabird restoration projects. Journal of Wildlife Man- odroma castro). In: Poole AF, editor. Birds of North America. Ithaca (NY): Cornell Lab of Ornithology. agement 76:2–9. https://birdsna.org/Species-Account/bna/species/barpet Jones HP, Tershy BR, Zavaleta ES, Croll DA, Keitt BS, et Spatz DR, Newton KM, Heinz R, Tershy B, Holmes ND, et al. 2008. Severity of the effects of invasive rats on al. 2014. The biogeography of globally threatened seabirds: A global review. Conservation Biology seabirds and island conservation opportunities. Con- 22:16–26. servation Biology 28:1282–1290. Olson SL, James HF. 1982. Fossil birds from the Hawaiian [USFWS] U.S. Fish and Wildlife Service. 2015. Endangered Islands: Evidence for wholesale extinction by man and Threatened Wildlife and Plants; Endangered Status before western contact. Science 217(4560):633–635. for 49 Species from the Hawaiian Islands; Proposed Podolsky RH, Kress SW. 1989. Factors affecting colony Rule. Federal Register: National Archives and Records formation in Leach’s Storm-Petrel. Auk 106:332–336. Administration. 80(189).

Downloaded From: https://bioone.org/journals/The-Wilson-Journal-of-Ornithology on 30 Sep 2019 Terms of Use: https://bioone.org/terms-of-use Access provided by University of Hawaii at Manoa