Use of SST for monitoring coral stress

Looking forward while keeping an eye on the past.

William Skirving, Benjamin Marsh, Gang Liu, Jacqueline De La Cour,

Andrew Harris, Eileen Maturi, Christopher Merchant, Jonathan Mittaz,

Erick Geiger, Craig Steinberg, Roxana Vasile, Mark Eakin.

Talk Overview

• The DHW algorithm and why climatologies are important• Introducing CoralTemp• Bleaching extent through time – CoralTemp vs CCI• Which of CoralTemp and CCI are likely to be more consistent• Effect of cool CoralTemp bias on MMM• The extent of global coral bleaching through time

Degree Heating Week ProductMMM = Maximum of the Monthly Means

(1985 – 2012)

HotSpot = daily SST – MMMwhere Hotspot ≥ 0

DHW = ∑ HotSpot, for HotSpot ≥ 1

Previously

• Climatology from AVHRR Pathfinder

• Near-Real Time SST (NOAA Goes/Poes Blended SST)

• HotSpot and DHW• derived with Apples and Pears• accumulated bias

1985 2019OSTIA Reanalysis Blended Reanalysis

Blended NRT2002 2016

CoralTemp

A consistent SST product spanning 1985-presentDaily, gap-free 0.05 degree resolution SST

Temporal consistency provided via OSTIA

1985 2019OSTIA Reanalysis Blended Reanalysis

Blended NRT2002 2016

CoralTemp

A consistent SST product spanning 1985-presentDaily, gap-free 0.05 degree resolution SST

Bias correction of Blended SST

OSTIA Reanalysis

OSTIA NRT

1980 1985 1990 1995 2000 2005 2010 2015 2020

0

5

10

15

20

25

30

35

40

45

50

55

Perc

enta

ge o

f re

efs

with D

HW

4

Year

Extent of Global Coral Reef Heat Stress Through Time

1980 1985 1990 1995 2000 2005 2010 2015 2020

0

5

10

15

20

25

30

35

40

45

50

55

Perc

enta

ge o

f re

efs

with D

HW

4

Year

CCI

CoralTemp

Extent of Global Coral Reef Heat Stress Through Time

1985 1990 1995 2000 2005 2010 2015 2020-0.4

-0.2

0.0

0.2

0.4

0.6

1 2 3

OSTIA Blended

Ave Bias

All: -0.095OSTIA: -0.117Blended: -0.070

Number of Buoys

1) <2002) 400 – 7003) >1000

Daily Mean Bias (oC)Against Drifting Buoys + Moored Arrays using SQUAM

CoralTemp MMM – CCI MMM1985-2012

Pixels with Coral Reefsover the Great Barrier Reef, Australia

CoralTemp vs CCI1985-2016 (Great Barrier Reef)

1986 1990 1994 1998 2002 2006 2010 2014

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

Co

ralT

em

p -

CC

I (o

C)

Year

1985 1990 1995 2000 2005 2010 2015

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

Co

ralT

em

p -

CC

I (o

C)

Year

All Months February-only(the MMM month)

1985 to 2012 average diff = -0.11 oC 1985 to 2012 average diff = -0.19 oC

1980 1985 1990 1995 2000 2005 2010 2015 2020

0

5

10

15

20

25

30

35

40

45

50

55

Perc

enta

ge o

f re

efs

with D

HW

4

Year

CCI

CoralTemp

Extent of Global Coral Reef Heat Stress Through Time

1982 1986 1990 1994 1998 2002 2006 2010 2014

0

2

4

6

8

10

12

14

16

18

20

% o

f R

eef P

ixels

with D

HW

4

year

Extent of Global Coral Reef Heat Stress Through Time

Take Home Messages

• Consistent bias through time

• New products need to be related to historic products or shouldbe able to be reprocessed back through time

• CCI looks very consistent through time compared to CoralTemp

• Satellite SST is EXTREMELY important for Coral Reef ManagementWhy? Because mass coral bleaching only began in the 80s!So, satellite SST data covers the entire history of mass bleaching.