Relationship between historical sea-surface temperature variability and climate change-induced coral mortality in the western Indian Ocean M. Ateweberhan a, * , Tim R. McClanahan b a Wildlife Conservation Society, Coral Reef Conservation Project, P.O. Box 99470, Mombasa, Kenya b Wildlife Conservation Society, Marine Programs, Bronx, NY 10460-1099, USA article info Keywords: 1998 Climatic oscillation Climate change Climatic disturbance Coral bleaching Coral cover change El Nino/Southern Oscillation (ENSO) SST distributions Temperature variation abstract Many of the world’s coral reefs suffered high coral mortality during the 1998 ENSO, with the highest mor- tality in the western Indian Ocean (WIO). A meta-analysis of field data on change in coral cover across the 1998 ENSO event was conducted for 36 major reef areas in the WIO, and relationship of the change with the historical sea-surface temperature (SST) variability investigated. WIO reefs were categorized into three major SST groups of differing coral cover change. Cover change was negatively associated with stan- dard deviation (SD) SST until about SD 2.3, with increasing flatness of the SST frequency distributions. It increased with further increase in SD as the SST distributions became strongly bimodal in the Arabian/ Persian Gulf area. The study indicates that environmental resistance/tolerance to extreme anomalous events could be predicted and management priorities directed accordingly for a warmer and more vari- able future climate. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Coral reefs have become one of the Earth’s first ecosystems to directly suffer large-scale mortality from climate change, por- trayed by widespread bleaching-induced mortality of corals and consequent changes in the patterns of distribution, abundance, species diversity, and general ecological function (Bellwood et al., 2004; Hoegh-Guldberg, 1999; Hughes et al., 2003). Many coral reef areas in the world have already experienced significant declines due to heavy resource extraction, coastal development, and pollu- tion (Bellwood et al., 2004; Gardner et al., 2003; Hughes et al., 2003; McClanahan et al., 2002). Climate change is expected to compound the impacts by reducing coral growth and reproduction (Brown and Ogden, 1993; Bruno and Selig, 2007; Marshall and Baird, 2000; Pandolfi et al., 2003) and their ability to resist, toler- ate, or recover from other stresses and disturbances (Hughes et al., 2003). It is feared that the continued threats will jeopardize their existence and the ecological services they provide to millions of coastal people in tropical countries. Widespread bleaching events occur primarily as a result of ex- treme warm sea-surface temperatures (SST), often in combination with other environmental factors, including high irradiance (Brown et al., 2002; Fitt et al., 2001), making temperature a critical variable to consider in studies of climate change. The 1998 climatic oscillation caused the strongest oceanic warming in recorded mod- ern human history, and it was associated with a severe coral bleaching and mortality globally, with the WIO suffering the worst (Enfield, 2001; Goreau et al., 2000; Sheppard, 2003; Wilkinson, 1998). SST distribution and bleaching are patchy in their spatial distribution and vary within a reef, and more commonly among reefs (Berkelmans and Oliver, 1999; Marshall and Baird, 2000; McClanahan et al., 2005). One of the main challenges in under- standing the effects of climate change is, therefore, to seek patterns from this high spatial variability and uncover causative relation- ships that can help predict future changes. The main objective of this study was to determine if responses to extreme anomalous events, such as the 1998 ENSO, could be predicted based on regional historical SST patterns, and if so, to determine the relative contribution of specific SST properties to the predictions. The relationships can be useful in identifying reefs with high environmental tolerance or vulnerability, which can be crucial for prioritizing decisions concerning the levels and types of management. By analyzing the relationship of the bleaching mortality with local environmental properties (SST), we make inferences about large-scale responses and designed to improve the capacity to predict the effects of future extreme events. We also highlight the importance of local and regional SST properties (background historical SST) in influencing the outcome of strong stress events, such as the 1998 climatic oscillation. We show that despite differences in local and regional hydrodynamics, responses to extreme stress events are related, possibly causally, to a few uni- fying SST statistical properties. 0025-326X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpolbul.2010.03.033 * Corresponding author. E-mail addresses: m.ateweberhan@warwick.ac.uk, mateweberhan@wcs.org (M. Ateweberhan). Marine Pollution Bulletin 60 (2010) 964–970 Contents lists available at ScienceDirect Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul