Download The Deep Ocean`s Contribution to the Warming Hiatus

The Deep Ocean’s Contribution to
the Warming Hiatus
Sarah Purkey, Lamont-Doherty Earth Observatory
Damien Desbruyères, National Oceanography Center,
United Kingdom
Nathalie Zilberman, Scripps Institution of Oceanography
The Deep Ocean’s Contribution to
the Warming Hiatus
Sarah Purkey, Lamont-Doherty Earth Observatory
Damien Desbruyères, National Oceanography Center,
United Kingdom
Nathalie Zilberman, Scripps Institution of Oceanography
• 
• 
• 
Outline:
How we know what we know: deep ocean data
Deep ocean temperature trends since the 1990s
The deep ocean does not appear to be
contributing to the recent warming hiatus
Introduction: Global heat budget
¤  The global mean surface temperature
trend has slowed since the early 2000s,
often referred to as a ‘Warming Hiatus’
¤  93% of the anthropogenic warming goes
into the ocean
¤  Therefore, accurately quantifying total
ocean heat content is important for
monitoring the global heat budget
(Rhein et al. 2013)
Introduction: The deep and abyssal ocean
NADW
0
0.2
0.3
4000
NADW
AABW
0.2
2000
0.6
0.1
0.3
0.4
5
4000
0.9
0.7
0.7
0.
6
3000
0.
Depth [m]
1000
0.5
6000
0
0.2
0.4
0.3
0.3
3000
0.4
0.3
0.3
0.6
0.5
2000
0.8
¤  Climatology shows
filled by two water
masses: North Atlantic
Deep Water and
Antarctic Bottom
Water
0.2
5000
(Rhein et al. 2013)
¤  The deep (below
2000m) and abyssal
(below 4000m) are
filled with dense
waters formed at high
latitudes
0.1
1
0.
Depth [m]
1000
5000
AABW
6000
−80
−60
−40
−20
0
o
Latitude [ N]
20
40
(Johnson, 2008)
60
Deep ocean data: Go-Ship
¤  Deep ocean observations
are mostly limited to ship
based hydrography work
¤  World Ocean Circulation
Experiment (WOCE):
¤  Full depth, highly
accurate temperature
(0.001 °C) and salinity
(0.003), bottom to top
oceanographic survey in
the1990s
¤  Go-Ship: CLIVAR and CO2
repeat hydrography:
¤  Repeated a key subset of
these sections allowing
for estimates of deep
changes in temperature/
salinity in time
Deep ocean trends: A16S
(Following Johnson and Doney, 2006)
¤  Repeat sections show warming (red) throughout the deep Southern Ocean
and abyssal global ocean
¤  E.g. Difference in temperature along A16 between 1989/1995 and 2005/2003
Deep ocean warming trends: 1992-2005
(Purkey & Johnson 2010)
¤  Using all repeat sections within a basin, find basin mean
temperature trend below 4000 m centered between 1992 and 2005
¤  Most trends are significant at the 95% confidence level
¤  Southern intensified warming, with a weak, but detectable,
warming signal to the north
Deep ocean warming trends: 1992-2005
¤  The abyssal ocean is
warming below 4000 m at
an average rate 5 °mC
decade-1
¤  The Southern Ocean is
warming much faster
through out the water
column, at a mean rate of
30 °mC decade-1 below
1000 m
¤  Warming accounts for an
increase in ocean heat
storage of 0.027 W m-2
below 4000 m and 0.07 W
m below 2000 m
(Purkey & Johnson 2010)
Temporal variability: The warming hiatus
YEARS
2015
2010
2005
2000
A1
¤  Sparse temporal
coverage allows
for linear trends at
best, centered
between1992 and
2005
b
6N
A1
6N
A1
ATLANTIC
a
6N
c
d
PACIFIC
6N
A1
A2
5
INDIAN
5
A0
2
A2
AR
7W
SOUTHERN
0
A2
6S
A1
¤  Using only the most
recent data, can
compare data
collected between
1981-2010 period
(centered
between
1993-2005) to data
collected between
2001-2015 period
(centered
between
2004-2013)
2
A1
0
A1
5
I0
9S
I0
2
P0
6
P0
5S
P1
6S
P1
0
P1
03
SR
01
SR
go
Ar
2000
2008
2010
2015
Year
(Purkey and Johnson 2010)
(Desbruyeres et al, in prep)
1992-2005
Temporal variability: The warming hiatus
2004-2013
(Purkey and Johnson 2010)
(Desbruyeres et al, in prep)
Temporal variability: The warming hiatus
¤  The global mean warming
rate [°C/yr]: 1992-2005 vs
2004-2013
¤  Abyssal ocean warming rate
over the 1990s – 2000s has
decreased between 2000s–
2010s
¤  Both periods show the deep
ocean heat content below
2000 m is increasing at a rate
of 0.07 ± 0.06 W m-2
¤  No evidence of an increase
in the rate of deep warming
over the hiatus period
(Purkey and Johnson 2010 and Desbruyeres et al, in prep)
Temporal variability: Cause?
¤  Deep ocean heat
content can be change
either by advection or
by isotherm heave
(Rhein et al. 2013)
Temporal variability: Cause?
¤  Deep ocean heat
content can be change
either by advection or
by isotherm heave
¤  Observations show deep
ocean warming is driven
primarily by isotherm
heave
(Rhein et al. 2013)
(Purkey & Johnson 2013)
Temporal variability: Cause?
¤  Deep ocean heat
content can be change
either by advection or
by isotherm heave
¤  Observations show deep
ocean warming is driven
primarily by isotherm
heave
¤  Isotherm heave, driven
by a reduction in bottom
volume, is
communicated around
the global oceans via
Kelvin and Rossby waves
on decade time scales
(Masuda et al 2010)
Conclusions
¤  The deep ocean has been warming,
contributing ~10% to the total ocean
heat content with some interdecadal
variability
¤  Detecting interdecadal variability is
limited owing to sparse temporal
coverage
¤  However, no evidence of an increase
in the rate of deep (below 2000 m)
warming over the hiatus period
¤  To resolve the deep ocean heat
content on decadal time scales,
need a deep observing system such
as Deep Argo Array
(Rhein et al. 2013)