Conservation of low-islands: high priority despite sea-level rise. A comment on Courchamp et al. Serge Andre ´ foue ¨ t 1 , Je ´ rome Aucan 2 , Herve ´ Jourdan 3 , Paul Kench 4 , Christophe Menkes 5 , Eric Vidal 3 , and Hiroya Yamano 6 1 Institut de Recherche pour le De ´ veloppement (IRD), Unite ´ de Recherche 227 CoRe ´ Us, LABEX CORAIL, BP A5, 98848 Noume ´ a, New Caledonia 2 Institut de Recherche pour le De ´ veloppement, Laboratoire d’Etudes en Ge ´ ophysique et Oce ´ anographie Spatiale (LEGOS), BP A5, 98848 Noume ´ a, New Caledonia 3 Institut Me ´ diterrane ´ en de Biodiversite ´ et d’E ´ cologie Marine et Continentale (IMBE), Aix-Marseille Universite ´ , Centre National de la Recherche Scientifique (CNRS), Institut de Recherche pour le De ´ veloppement (IRD), Universite ´ d’Avignon et des Pays de Vaucluse (UAPV), Centre IRD Noume ´ a, BP A5, 98848 Noume ´ a, New Caledonia 4 School of Environment, The University of Auckland, New Zealand 5 Institut de Recherche pour le De ´ veloppement, Sorbonne Universite ´ s (Universite ´ Pierre et Marie Curie; Universite ´ Paris 06)– CNRS–Muse ´ um National d’Histoire Naturelle–Institut Pierre Simon Laplace, LOCEAN Laboratory, IRD Noume ´ a, BP A5, 98848 Noume ´ a, New Caledonia 6 National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki 305-8506, Japan No triage conservation strategy for low tropical islands! Assuming that a simple drowning model is applicable to all islands facing future climate change and sea-level rise (SLR), the future existence of up to 12% of islands is said to be compromised – and therefore these should not be considered for active management and protection [1,2]. This includes tropical atolls and their low-lying islands. Howev- er, we reject the triage strategy elaborated by Courchamp et al. [1]. Evidence from geology, sedimentology, and ocean- ography, and from the ecology of invasive species, shows that island conservation, especially of low islands, should remain a priority. Geological, sedimentary, relative SLR, and tectonic evidence Simplistic drowning models over-state the risks and are inappropriate for low-lying coral reef islands. Such model- ing assumes that islands are passive geological entities that will experience permanent inundation with SLR and are unable to physically respond. This is only likely for hard rock coasts, whereas sedimentary shorelines, includ- ing coral reef islands which are composed of sand and gravel, display a diverse range of physical responses. During the mid-Holocene (6000–2000 years ago), mor- phostratigraphy and radiometric dating show that reef islands formed under rising, falling, and stable sea levels. Islands in the Maldives and Marshall Islands formed under rising sea levels (0.5–1.0 m higher than present) [3,4], whereas some Great Barrier Reef and New Caledo- nian reef islands formed at stable higher sea levels [5]. In these regions sea levels subsequently fell to current levels 2000 years ago. Over the next century the sea level will simply reoccupy the levels under which the islands formed. At decadal timescales, instead of simply eroding or shrinking, islands can display a complicated spectrum of geomorphic responses. Over the past half-century, under a SLR of 2.0 mm/year, reef islands in the Tuvalu and Marshall Islands have remained stable in size or have become larger [6]. Islands are dynamic landforms that can adjust their shape and position on reefs and build vertically. Physical island change is mediated by the sup- ply and transport of coral sediment rather than by sea level [4]. As waves and currents change, island sediments are reworked alongshore, around shorelines, or onto island surfaces via overwash processes allowing island surfaces to aggrade. In the Maldives, large sections of islands have vertically accreted by 0.3 m over the past decade, a rate much larger than anticipated sea-level rise. At centennial timescales, the 5th IPCC (Intergovern- mental Panel on Climate Change) reports that the rate of global mean SLR during the 21st century will exceed the 1971–2010 rate for all representative concentration path- way (RCP) scenarios [7]. However, SLR will also exhibit strong regional variations from the mean trend. Thus, a simple drowning model with a globally uniform sea level would be inappropriate for taking conservation decisions. Major uncertainty in centennial projections lies in decad- al variations, which can differ by more than 100% from the global long-term projected change [7]. For instance, during the past 20 years the Solomon Islands and Micronesia have experienced SLR rates above 10 mm.y 1 . Conversely, New Caledonia has experienced rates lower than 1 mm.y 1 over the past 50 years, and conclusions about future impacts are very uncertain. Stating that conservation efforts in Surprise Island (New Caledonia) are doomed therefore seems to be inappropriate [1]. Tectonic processes can also cause vertical land move- ments that have a predominant role compared to SLR [8]. For instance, in the Torres Islands (Vanuatu) the drowning rate can be much higher than that caused by SLR. Conversely, uplift (e.g., following earthquakes, as Letter 0169-5347/ ß 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tree.2014.10.001 Corresponding author: Andre ´foue ¨t, S. (serge.andrefouet@ird.fr). TREE-1873; No. of Pages 2 Trends in Ecology & Evolution xx (2014) 1–2 1