Caribbean corals in crisis: record thermal stress, bleaching, and mortality in 2005

Abstract

Background: The rising temperature of the world's oceans has become a major threat to coral reefs globally as the severity and frequency of mass coral bleaching and mortality events increase. In 2005, high ocean temperatures in the tropical Atlantic and Caribbean resulted in the most severe bleaching event ever recorded in the basin.

Methodology/principal findings: Satellite-based tools provided warnings for coral reef managers and scientists, guiding both the timing and location of researchers' field observations as anomalously warm conditions developed and spread across the greater Caribbean region from June to October 2005. Field surveys of bleaching and mortality exceeded prior efforts in detail and extent, and provided a new standard for documenting the effects of bleaching and for testing nowcast and forecast products. Collaborators from 22 countries undertook the most comprehensive documentation of basin-scale bleaching to date and found that over 80% of corals bleached and over 40% died at many sites. The most severe bleaching coincided with waters nearest a western Atlantic warm pool that was centered off the northern end of the Lesser Antilles.

Conclusions/significance: Thermal stress during the 2005 event exceeded any observed from the Caribbean in the prior 20 years, and regionally-averaged temperatures were the warmest in over 150 years. Comparison of satellite data against field surveys demonstrated a significant predictive relationship between accumulated heat stress (measured using NOAA Coral Reef Watch's Degree Heating Weeks) and bleaching intensity. This severe, widespread bleaching and mortality will undoubtedly have long-term consequences for reef ecosystems and suggests a troubled future for tropical marine ecosystems under a warming climate.

Figure 1. Thermal stress and bleaching during the 2005 Caribbean bleaching event.

(A) Maximum NOAA Coral Reef Watch Degree Heating Week (DHW) values showing the highest thermal stress recorded at each 0.5-degree pixel during 2005. Values ≥4°C-weeks typically resulted in significant bleaching; ≥8°C-weeks typically resulted in widespread bleaching and significant mortality. (B) Jurisdictional means of coral bleached; marker color and size denote the severity measured as either percent live coral colonies (circles) or cover (diamonds).

Figure 2. Temporal patterns of thermal stress in the Caribbean.

Average of satellite-derived anomaly and thermal stress indices from the 0.5-degree pixels containing or nearest to reefs in the Caribbean (bounded by 35°N, 55°W, and the coast of the Americas). (A) NOAA coral bleaching HotSpots (purple) and DHW (red) in 2005. See results for explanations of (i)–(vii). Letters D–W refer to the major hurricanes of 2005: Dennis, Emily, Katrina, Rita and Wilma. (B) Average of annual maximum thermal stress (DHW) values during 1985–2006. Significant coral bleaching was reported during periods with average thermal stress above 0.5°C-weeks, and was especially widespread in 1995, 1998, and 2005.

Figure 3. Mean coral bleaching and mortality versus thermal stress.

(A) Small squares represent mean percent coral bleached (by area or colony) for each 0.5-degree pixel and twice-weekly time period plotted against observed DHW value. Solid line indicates significant linear regression (slope  = 3.41, intercept  = 26.94, DF  = 359, p<0.0001, r² = 0.24). Colored bars indicate mean (gray bar) and standard deviation of all surveys binned at 1°C-week intervals; colors correspond to low bleaching risk (DHW <4, blue), moderate risk (DHW ≥4, green), high bleaching and mortality risk (DHW ≥8, yellow), and very high risk (DHW ≥12, purple). (B) Triangles represent mean percent coral mortality (± standard deviation) reported during 25-Jul-2005 to 20-Jan-2007, plotted against the 2005 maximum DHW value recorded for each 0.5-degree pixel. Yellow and white areas correspond to the inset box where values indicate number of data points in each quadrant (quadrants defined as 0≤ DHW <8 and 0≤ mortality <8%; 0≤ DHW <8 and 8%≤ mortality; 8≤ DHW and 0≤ mortality <8%; 8≤ DHW and 8%≤ mortality).

Figure 4. Long-term temperature record in the Caribbean.

Temperature anomalies for 2.0-degree reef pixels in the tropical Caribbean computed using the NOAA Extended Reconstructed Sea Surface Temperature (ERSST) dataset. Anomalies were plotted relative to 1901–2000. The dashed line indicates the 2005 value.

Figure 5. Thermal stress and hurricanes during the 2005 Caribbean bleaching event.

Minimum observed SST anomaly for May-December 2005, overlaid with storm tracks (solid: hurricane, thickness denotes strength category; dotted: tropical storm; red: June-August; gray: September; black: October-December). Dates indicate initial date of hurricane formation. The large yellow region in the eastern Caribbean remained warmer than usual throughout this period.