Abstract

The U.S. National Oceanic and Atmospheric Administration’s (NOAA) Coral Reef Watch (CRW) operates a global Four-Month Coral Bleaching Outlook system for shallow-water coral reefs in collaboration with NOAA’s National Centers for Environmental Prediction (NCEP). The Outlooks are generated by applying the algorithm used in CRW’s operational satellite coral bleaching heat stress monitoring, with slight modifications, to the sea surface temperature (SST) predictions from NCEP’s operational Climate Forecast System Version 2 (CFSv2). Once a week, the probability of heat stress capable of causing mass coral bleaching is predicted for four months in advance. Each day, CFSv2 generates an ensemble of 16 forecasts, with nine runs out to 45 days, three runs out to three months, and four runs out to nine months. This results in 28-112 ensemble members produced each week. A composite for each predicted week is derived from daily predictions within each ensemble member. The probability of each of four heat stress ranges (Watch and higher, Warning and higher, Alert Level 1 and higher, and Alert Level 2) is determined from all the available ensemble members for the week to form the weekly probabilistic Outlook. The probabilistic Four-Month Outlook is the highest weekly probability predicted among all the weekly Outlooks during a four-month period for each of the stress ranges. An initial qualitative skill analysis of the Outlooks for 2011-2015, compared with CRW’s satellite-based coral bleaching heat stress products, indicated the Outlook has performed well with high hit rates and low miss rates for most coral reef areas. Regions identified with high false alarm rates will guide future improvements. This Outlook system, as the first and only freely available global coral bleaching prediction system, has been providing critical early warning to marine resource managers, scientists, and decision makers around the world to guide management, protection, and monitoring of coral reefs since 2012. This has been especially valuable during the third global coral bleaching event that started in mid-2014 and extended into mid-2017. The Outlook system is an integrated component of CRW’s global decision support system for coral bleaching. Recent management actions taken in light of this system are discussed.

Highlights

  • Mass coral bleaching due to anomalously warm water temperatures has occurred with increasing frequency and severity in recent decades (Eakin et al, 2010; Heron et al, 2016a; Hughes et al, 2018)

  • We further examined sea surface temperature (SST) skill for the bleaching season of major coral reef regions globally

  • Coral Reef Watch (CRW)’s probabilistic 4-Month Coral Bleaching Outlook system is the first and only freely available global system for predicting the heat stress that leads to mass coral bleaching

Read more

Summary

INTRODUCTION

Mass coral bleaching due to anomalously warm water temperatures has occurred with increasing frequency and severity in recent decades (Eakin et al, 2010; Heron et al, 2016a; Hughes et al, 2018). Through partnership with the NOAA National Centers for Environmental Prediction (NCEP), in July 2012, CRW released its first probabilistic global subseasonal-to-seasonal-scale Coral Bleaching Outlook system (Eakin et al, 2012) It was based on 9-month SST predictions from NCEP’s operational Climate Forecast System Version 1 (CFSv1) (Saha et al, 2006). Predicting heat stress for any day within 12-weeks of the initial condition day needs historical SST values; the daily SST from the dOISSTv2 (Reynolds et al, 2007; Banzon et al, 2016) is used This widely used dOISSTv2 dataset, an operational near real-time NOAA/National Environmental Satellite, Data, and Information Service (NESDIS) SST analysis, combines satellite and in situ measurements to produce a daynight blended SST analysis. For its Outlook system, CRW used the reprocessed dOISSTv2 to develop the climatology required for deriving the historical heat stress variables for the initial condition day, and earlier days, of each near real-time CFSv2 run

METHODS
CONCLUDING REMARKS
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call