Rock cavity nesting as the norm: Breeding songbirds of the temperate H igh A ndes

Organisms living and breeding in alpine habitats must cope with severe environmental challenges such as temperature extremes, storms, resource limitations and, sometimes, hypoxia, resulting in short windows and reduced opportunities to reproduce (Chamberlain et al., 2023; Martin, 2001; Martin et al., 2017; Martin & Wiebe, 2004). Alpine habitats are defined as the area above the climatic treeline where vegetation growth is limited to cold-tolerant grasses and forbs, low-lying shrubs, or small patches of stunted trees <3 m in height, resulting in less diverse niche space relative to below the treeline (Körner, 2012; Körner et al., 2011). Despite these constraints, as well as limited habitat availability above treeline relative to the total landmass (Nagy & Grabherr, 2009; Testolin et al., 2020), ~12% of bird species breed in alpine habitats (de Zwaan et al., 2022a). Globally, our understanding of avian nesting biology is limited, with clutch size and nest structure documented for only 53% and 45% of species, respectively (Reynolds & Deeming, 2015). Additionally, nest descriptions tend to be biased toward low-elevation, Northern Hemisphere communities, with clear knowledge gaps for alpine breeding birds in south-temperate mountains. Here, we describe the nest and breeding traits of songbirds breeding above treeline in the temperate High Andes, many of which previously lacked detailed descriptions. We also provide natural history notes on breeding phenology, parental care, and resource competition to provide a baseline for future studies to build on within these data-deficient communities (Table 1). Finally, we compare the predominance of rock cavity-nesting species above treeline in the temperate Andes with other temperate alpine songbird communities using a global alpine breeding bird dataset (de Zwaan et al., 2022a, 2022b) and generate a hypothesis framework for investigating variation in nest traits among alpine communities. We located and monitored the nests of High-Andean or alpine breeding bird communities in the temperate Andes over two breeding seasons (November–January, 2017 and 2018) across five volcanoes within the La Araucanía Region of southern Chile (39° S, 71° W). We monitored 50 nests between 1300–1800 m above sea level for five species: Geositta rufipennis (rufous-banded miner, Furnariidae, n = 7 nests), Cinclodes oustaleti (gray-flanked cinclodes, Furnariidae, n = 8), Muscisaxicola maclovianus (dark-faced ground-tyrant, Tyrannidae, n = 3), Muscisaxicola albilora (white-browed ground-tyrant, Tyrannidae, n = 8), and Pygochelidon cyanoleuca (blue-and-white swallow, Hirundinidae, n = 24). All nests were within rock cavities (Figure 1). Despite considerable search effort, we did not locate any ground nests. Rock cavity reuse was infrequent but recorded for G. rufipennis (n = 2 nests), C. oustaleti (n = 1), and P. cyanoleuca (n = 6) within and between breeding seasons (Table 1). See Appendix S1 for detailed information on the study site, methods, and variable measurements for Table 1. Geositta rufipennis occupied the deepest rock cavities of the five focal species (Table 1), such that we were unable to observe the nest contents. This species nested primarily in open rock fields or scree and was one of the earliest species to initiate breeding. Nest materials consisted of down feathers, grass, small twigs, fur, and rootlets. Cinclodes oustaleti, another Furnariid, used narrower cavities in younger volcanic rock (Table 1). Nests were built closer to the entrance, primarily using grass with minimal other materials, including one cigarette. The nests of G. rufipennis and C. oustaleti lacked a defined cup structure, resembling a loosely assembled platform. Muscisaxicola maclovianus, a Tyranid ground-tyrant, placed their nests mainly above rivers or streams in rock crevices (Table 1). Due to accessibility issues, we were unable to record all breeding parameters for M. maclovianus, but they used feathers and fur to insulate nests. Muscisaxicola albilora also nested in cliff crevices but was not associated with water (Table 1). Their nests consisted primarily of grass and rootlet cups lined with feathers. Despite being sister species, M. maclovianus and M. albilora appeared to have partitioned their nesting habitat niche, as both species were observed nesting or feeding frequently in the same general area, but never in close proximity. Pygochelidon cyanoleuca nested in rock cavities with dimensions that varied widely among individuals relative to the other species (except perhaps G. rufipennis; Table 1), potentially indicating less reliance on a specific cavity shape. Nest cups consisted of broad-leaved grasses or sedges, with limited instances of feathers or rootlets and, in two cases, toilet paper. This species was the most colonial, with large numbers nesting in the same rock face. We recorded multiple instances of interspecific competition at nest sites. Specifically, M. albilora was highly aggressive toward P. cyanoleuca, nesting in similar sites despite frequent territorial disputes, with aggressive exclusion behavior between species near rock cavity entrances. In contrast, M. maclovianus was tolerant of P. cyanoleuca and often co-associated in semicolonial aggregations. Less frequently, we observed competitive interactions between M. albilora and Melanodera xanthogramma (yellow-bridled finch), and between M. maclovianus and G. rufipennis. This behavior may suggest that rock cavities are a limiting factor for High-Andean bird populations, similar to tree cavity-nesting birds (Cockle et al., 2010). In the Italian Alps, similar competition for rock and artificial cavities (i.e., snowfinch nest boxes, ski lift pylons, buildings) has been documented among Montifringilla nivalis (white-winged snowfinch), Phoenicurus ochruros (black redstart), and Motacilla alba (white wagtail; Brambilla et al., 2019). Here, cavities experience within-season interspecific reuse (i.e., between early and late broods), as well as species turnover in occupancy across years. In contrast, despite evidence of competition, we did not observe cavity takeover or interspecific reuse in the High-Andean community. Another potential limiting resource is insulative nesting materials, which can improve nest success in alpine habitats (de Zwaan & Martin, 2018). We observed both Leptastenura aegithaloides (plain-mantled tit-spinetail) and G. rufipennis taking nesting material from an active C. oustaleti nest. Nest material kleptoparasitism is rarely reported outside of colonial-nesting species (Slager et al., 2012), and never before at high elevations. At the microhabitat scale, all species nested in proximity to vegetation, predominantly Gaultheria pumila (closest for all species), Berberis empetrifolia, Cardamine chilensis, Chiliotrichum diffusum, Empetrum rubrum, Maytenus disticha, ferns, and grasses. Breeding success was relatively high across species but varied considerably (60%–100%; Table 1). Other than clutch size and provisioning rate, parental behaviors such as incubation recesses and food provisioned varied among species, potentially reflecting differences in phenology (i.e., peak in November or December) and the stage of offspring development (Table 1). Ample opportunities exist for future research to address variation in cavity dimensions, plant associations, and parental behaviors within and among species to investigate potential consequences for reproductive success. Using a global dataset of alpine breeding birds (de Zwaan et al., 2022a, 2022b), we found that the proportion of passerines nesting in rock cavities was greater in the southern Andes (this study site; 54%; 13 of 24 species) than in other major temperate alpine communities globally (excluding species that breed above treeline rarely or incidentally). In the southern ranges of Europe (e.g., Alps, Pyrenees, Carpathians), 39% (12 of 31) of alpine breeding passerines nest in rock cavities, compared with 24% (9 of 38) in the coastal and Rocky Mountains of North America, and 22% (26 of 117) on the Qinghai-Tibetan Plateau. In the New Zealand Alps, which occur at a similar latitude to the temperate Andes, only one of four (25%) alpine passerine species nest in rock cavities. Temperate mountains and alpine habitats represent 5.5% and 2.6% of the global landmass excluding Antarctica and Greenland, respectively (Nagy & Grabherr, 2009; Testolin et al., 2020). These relatively small, isolated habitats and their associated biodiversity are also threatened by a rapidly warming climate and the potential for extirpation resulting from upslope, climate-driven range shifts (Freeman et al., 2018; Scridel et al., 2018). We highlight the unique breeding biology of alpine passerines in the temperate Andes, underlining the importance of documenting basic nest traits and breeding parameters for data-deficient and climate-sensitive communities. Improving our knowledge of life-history traits for these species is a prerequisite to understanding species resilience and the future of populations under climate change. We acknowledge that this work was conducted within the Wallmapu; the traditional, unceded territory of the Mapuche Nation. We are grateful to Camila Bravo and Constanza Rivas for help finding nests in the field. Special thanks to the Chilean National Forestry Service (CONAF) for access to field sites in National Protected Areas. We also thank the Editor and two reviewers, their recommendations greatly improved the manuscript. Funding was provided by the Natural Sciences and Engineering Research Council of Canada (NSERC) and Environment and Climate Change Canada for Kathy Martin and Scott Wilson, as well as a Go Global International Learning Programs Award for Devin R. de Zwaan from the University of British Columbia. Tomás A. Altamirano was supported by a post-doctoral scholarship from CONICYT (74160073) and ANID/BASAL FB210018. The authors declare no conflict of interest. Data (de Zwaan et al., 2022b) are available in Figshare at https://doi.org/10.6084/m9.figshare.20556750. Appendix S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.