Islands: Microcosms of socioecological change and barometers of global biodiversity trends

Le résumé fourni par la source

The recent declaration of the Christmas Island shrew (Crocidura trichura) as extinct offers a stark reminder of the vulnerability of island species (Woinarski & Burbidge, 2025). This enigmatic mammal, restricted to Christmas Island, Australia, seems to have declined sharply after the arrival of invasive black rats carrying novel pathogens and due to predation by invasive predators such as cats and snakes and habitat disturbance from yellow crazy ants (Woinarski et al., 2023). This quiet, incremental, and ultimately irreversible disappearance characterizes the global trend of islands as extinction epicenters, an outcome driven by vulnerability to well-established threats like invasive species and exacerbated by new pressures like climate change (Russell & Kueffer, 2019). The fate of the Christmas Island shrew is far from a singular loss. It is a symptomatic example of a long-standing broader global pattern of erosion of island biodiversity in the aftermath of human arrival. Islands occupy ca. 5% of Earth's land surface but hold roughly 20% of global biodiversity (Tershy et al., 2015). The magnitude of anthropogenically driven island species loss is still far from understood (see e.g., Rando et al., 2024 for a recent example of unknown extinctions in the North Atlantic Macaronesian archipelagos). Yet, from what is documented, approximately three-quarters of human-driven extinctions over the past 500 years have occurred on islands (Fernández-Palacios et al., 2021; Matthew et al., 2022). The ecological richness of islands is matched by their cultural distinctiveness, with many island societies having developed deep local ecological knowledge systems that are tightly intertwined with their environments (Hong, 2013; Kueffer & Kinney, 2017). With islands at the epicenter of global socioecological change—facing increasing challenges associated with invasive species introductions, habitat loss and degradation, pollution, climate change and human socio-political dynamics—the socioecological systems, that converge on islands with exceptional intensity, are shifting at alarming rates. Islands have been in the forefront of conservation science and practice, with some of the most high-profile conservation success stories being linked to insular species and ecosystems (Bolam et al., 2021; Jones et al., 2016), with these actions driven by consistently recorded threats to island species, particularly invasive mammals. Yet, substantial knowledge gaps persist. Basic data on lesser well known native species distributions, trends, ecological roles, and genetic diversity remain scarce (Aslan et al., 2015; Nori et al., 2022; Rato et al., 2024), while interacting threats—especially associated with invasive species—are poorly understood (Russell & Kaiser-Bunbury, 2019). Despite technological advances that facilitate biodiversity assessments at broader spatial scales and over longer periods, long-term monitoring remains limited (Brooke et al., 2018), and few studies have successfully integrated ecological and sociocultural data (but see e.g., Carvalho et al., 2015). Island freshwater ecosystems, which experience “double insularity” as islands within islands, remain poorly understood despite often supporting highly threatened species (Órfão et al., 2024). Furthermore, although island local ecological knowledge systems are being lost to ever-increasing globalization, spatially explicit information on this loss is lacking. Combined with governance constraints and research inequities, these gaps impede effective, evidence-based conservation on islands (Burt et al., 2022). In this special issue on advances, challenges, and opportunities in island conservation, we present developments that address several of these gaps, highlighting new tools, interdisciplinary approaches, and community partnerships that together are reshaping island conservation science and practice. The contributions span terrestrial and marine systems, include ecological, technological, and social dimensions, and advance our understanding of islands both as vulnerable systems and as powerful models for global change research. Although island taxa are far more prone to extinction than their mainland counterparts, they remain disproportionately overlooked in scientific research (Conenna et al., 2017; de Lima et al., 2011). Nunes et al. (2026) add to this evidence by showing that, although over one-fifth of the world's circa 12,000 reptiles are island endemics and 30% of these island-restricted species are threatened with extinction (compared to 12.1% of reptiles globally), they receive only 6.7% of all reptile-focused research. Introduced species, long recognized as major drivers of biodiversity declines, continue to reshape island ecosystems (Russell et al., 2017). In this collection, Guedes et al. (2026) show that introduced mona monkeys caused over 40% predation on artificial nests within 1 week, threatening the reproductive success of birds in Príncipe Island (Central Africa), many of which are endemic. Still focusing on mammalian predators but acknowledging that predator impacts are often temporally dynamic and involve multiple, interacting species (Russell & Kaiser-Bunbury, 2019), Fox et al. (2026) expanded density-impact functions from single to multiple species to illustrate how rats and mustelids exert distinct pressures to New Zealand petrels, revealing key management windows for predator control. Although invasive mammalian predators are key players in island extinctions and population declines (Doherty et al., 2016), harmful impacts extend well beyond this group. Philip et al. (2026) show that even relatively low densities of yellow crazy ants—a widespread invader of islands across the Pacific and Indian oceans—impact the presence and abundance of brown noddies and restructure hermit crab assemblages. Moving to the Galápagos, Hartwig and Pinto (2026) call attention to cryptic invasions with potential severe consequences by revealing the unexpected discovery of an invasive nematode infecting two endemic Galápagos rodents. They highlight the need for captive breeding as a precautionary safeguard. By assessing a broader suite of land-based threats to largely island-restricted seabird populations across the Atlantic, Kalaitzakis et al. (2026) showed that invasive alien species—particularly rodents and cats—are responsible for the greatest population losses of pelagic seabirds in the region. Furthermore, they identified that the impact of invasive mammals is particularly acute in the island systems of Bermuda and the Canary Current and that invasive species management in Tristan da Cunha and Gough is the single intervention with the greatest potential for population recovery. New technologies are increasing efficiency and effectiveness in conservation (Berger-Tal & Lahoz-Monfort, 2018). On islands these advancements include automated biodiversity monitoring, such as passive acoustic recorders and camera-trap networks (Ferreira et al., 2022; Soto et al., 2023), drone surveys (Nyberg et al., 2024), and GPS telemetry (Colosimo et al., 2022). Considering the disproportionate impact of invasive species on islands, there is increasing pressure for the identification of cost-effective tools for invasive species surveillance and management. Steibl et al. (2026) address this by showing that environmental DNA (eDNA) can be used in the early detection of rodents. They illustrate how eDNA from beach and lake sites on three atoll islands can inform about the presence of invasive rats and can even distinguish between closely related species. Other than aiding in the detection of invasive species, technological advances are also contributing to control and eradication interventions. In an insightful perspective, Smith et al. (2026) explore how drones are transforming eradication methods, offering a precise, flexible alternative to helicopter-assisted rodenticide application. These authors show that drones are particular

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.