Contents lists available at ScienceDirect Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul Viewpoint Coral mass mortalities in the Chagos Archipelago over 40 years: Regional species and assemblage extinctions and indications of positive feedbacks Charles Sheppard a,b,⁎ , Anne Sheppard a,b , Douglas Fenner c a School of Life Sciences, University of Warwick, CV4 7AL, UK b School of Ocean Sciences, Bangor University, LL57 2DG, UK c NOAA Contractor and Consultant, Pago Pago, AS 96799, USA ARTICLEINFO Keywords: Chagos Archipelago Coral decline Warming Species extinctions ABSTRACT The global decline of reef corals has been driven largely by several marine heatwaves. This has greatly reduced coral cover but has reduced coral diversity also. While there is a lack of data in most locations to detect coral species losses, reefs of the Chagos Archipelago, central Indian Ocean, have long term monitoring data extending back to the late 1970s. Severe declines in cover have occurred since the 1970s, with regional extinctions of some species and key species assemblages. There is a severe decline in coral settlement, along with a substantial loss of habitat quality which has reduced the habitat available for settlement. This is a clear precursor to positive feedback. Regional species extinctions here occur mainly when total coral cover is <10% of pre-warming levels. Climate models predict more frequent and more severe marine heatwaves, and even if this ecosystem recovers it will contain fewer species. 1. Introduction Coral cover on reefs is in decline in all major regions (e.g. Hughes et al., 2017, 2018; Hoegh-Guldberg et al., 2017; Birkeland, 2019), in- cluding the Caribbean (Gardner et al., 2003; Schutte et al., 2010; Jackson et al., 2014), the Pacific (Bruno and Selig, 2007), Great Barrier Reef (Hughes et al., 2017, 2018), and Indian Oceans (Ateweberhan et al., 2011). However, total coral cover is not well correlated with species richness (Richards, 2013) and there are few measures of how decline in coral cover relates to decline in the component species. When the threatened status of corals was evaluated globally for the first time in 2008 it was estimated that one third of stony coral species faced extinction, based on measures of their range loss (Carpenter et al., 2008). Richards and Day (2018) highlight that the lack of data at species level increases the risk of ‘silent extinctions’, and it has been suggested that in Micronesia (southwestern Pacific) extinctions are likely to be greater than previously expected (Richards and Hobbs, 2014). The end result of a continued decline in coral cover ultimately must be extinction of the component species, but coral monitoring at species level has been limited. Marine systems may differ from terrestrial pat- terns: Hughes et al. (2014) argue that compared to terrestrial mammals or birds, the vast majority of marine plants and animals (including corals) are relatively resistant to global extinction due to high reproductive rates, wide dispersal, relatively large population sizes, and usually large geographic ranges. They suggest that the IUCN Red List criteria are not well suited to marine organisms, and present data showing that range extent and local abundances are not correlated for many corals or reef fishes (as they are for mammals and birds), so that for corals and fishes there is no ‘double jeopardy’ from having both small range size and low local abundances. It is indeed the case that many fewer global extinctions have been documented from the marine realm than the terrestrial (Carleton et al., 1999; Dulvy et al., 2003; McCauley et al., 2015) but populations of species in the marine realm are more difficult to document and much more poorly known and are probably underestimated (Carleton et al., 1999). Others have argued that the old idea that the seas are so vast that it is not possible to deplete their species is incorrect (Carleton et al., 1999; Roberts and Hawkins, 1999), and that Allee effects amongst sessile marine invertebrates, for example, make them more vulnerable to ex- tinction than previously thought (Birkeland et al., 2013) yet the IUCN Red List criteria is being used for coral reef studies. In the study by Carpenter et al. (2008) on corals, most of the species with a heightened level of extinction were only listed as “Vulnerable”, with very few being listed as either “Critically Endangered” or “Endangered.” Few recent coral species are currently considered globally extinct (Glynn, 2011). A new species of the hydrocoral Millepora was dis- covered in Pacific Panama but before it could even be named all known https://doi.org/10.1016/j.marpolbul.2020.111075 Received 24 January 2020; Received in revised form 12 March 2020; Accepted 13 March 2020 ⁎ Corresponding author at: School of Life Sciences, University of Warwick, CV4 7AL, UK. E-mail address: charles.sheppard@warwick.ac.uk (C. Sheppard). Marine Pollution Bulletin 154 (2020) 111075 Available online 10 April 2020 0025-326X/ Crown Copyright © 2020 Published by Elsevier Ltd. All rights reserved. T