Re(de)fining degree-heating week: coral bleaching variability necessitates regional and temporal optimization of global forecast model stress metrics

• Published in: Coral reefs 2024-08, Vol.43 (4), p.969-984
• Main Authors: Whitaker, Hannah, DeCarlo, Thomas
• Format: Article
• Language: English
• Subjects: Accumulation; Anthropogenic factors; Biomedical and Life Sciences; Climate change; Climate models; Climate prediction; Conservation; Coral bleaching; Coral reefs; Freshwater & Marine Ecology; Heat stress; Heat tolerance; Heating; Life Sciences; Marine conservation; Marine invertebrates; Ocean basins; Ocean models; Ocean temperature; Ocean warming; Oceanography; Regional variations; Temperature requirements; Weather forecasting
• ISSN: 0722-4028; 1432-0975
• DOI: 10.1007/s00338-024-02512-w

Tropical coral reefs are a critical ecosystem in global peril as a result of anthropogenic climate change, and effective conservation efforts require reliable methods for identifying and predicting coral bleaching events. To this end, temperature threshold-based models such as the National Oceanic and Atmospheric Administration’s (NOAA) degree-heating week (DHW) metric are useful for forecasting coral bleaching as a function of heat stress accumulation. DHW does not adequately account for regional variation in coral stress responses, however, and the current definition consistently underpredicts coral bleaching occurrence. Using a weather forecasting skill-based framework, our analysis cross-tested 1080 variations of the DHW-based bleaching occurrence (presence/absence) model against 22 years of contemporary coral bleaching observations (1998–2019) in order to optimize bleaching forecast skill at different levels of geographic specificity. On a global basis and relative to the current definition, reducing the current 1 °C warming cutoff to 0.4 °C, adjusting the accumulation window to 11 weeks, and defining a bleaching threshold of 3 DHW improved forecast skill by 70%. Allowing our new DHW definitions to vary across regions and ocean basins further doubled model skill. Our results also suggest that the most effective bleaching forecast models change over time as coral reef systems respond to a shifting climate. Since 1998, the coral bleaching threshold for the globally optimized forecast model has risen at a significant rate of 0.19 DHW/year, matching the pace of ocean warming. The bleaching threshold trajectory for each ocean basin varies. Though further work is necessary to parse the mechanism behind this trend, the dynamic nature of coral stress responses demands that our forecasting tools be continuously refined if they are to adequately inform marine conservation efforts.