tracing the upper ocean’s “missing heat”

A pause in the rise in upper ocean heat content:how unusual is it, and where did the heat go?

Caroline KatsmanGeert Jan van Oldenborgh

Royal Netherlands Meteorological InstituteKNMI - Global Climate Division

Caroline.Katsman@knmi.nl

Tracing the upper ocean’s“missing heat”

A warming climate

% energy content change

[1993-2003, data source: IPCC (2007)]

increasing greenhouse gas concentrations affect the Earth’s energy budget

ocean expansion isresponsible for ±50%of the observed global mean sea level rise

[3 mm/yr over 1993-2003]

Observed ocean heat uptake

Levitus et al 2009

0-700 m, global meanocean heat content

Argo float program(2000-present)

International Argo float network

February 2012 - 3500 active floats NL = 37

Observed ocean heat uptake

Levitus et al 2009

0-700 m, global meanocean heat content

eXpendable BathyThermographs(XBTs) – from moving ships

observations – surface

© National Oceanographic Data Center (NODC), USA

year=2000

observations – 1000 m depth

© National Oceanographic Data Center (NODC), USA

year=2000

observations – 1000 m depth

© National Oceanographic Data Center (NODC), USA

year=1980

observations – 1000 m depth

© National Oceanographic Data Center (NODC), USA

year=1960

Upper ocean heat uptake

0-700 m global mean

upper ocean has notgained heat since 2003

Upper ocean heat uptake

“missing heat”~ 0.3 · 1022 J/yr

1. How unusual is this, in light of the ongoinggreenhouse gas forcing?

2. Where did the energy go?

Upper ocean heat uptake

“missing heat”~ 0.3 · 1022 J/yr

energy budget change of ~ 0.2 W / m2

– the imbalance in the Earth’s energy budget due to greenhouse gas forcing is ~0.85 ±0.15 W/m2 [Hansen et al 2005]

–satellites measure incoming shortwave radiation and outgoing longwave radiation at ~0.5 to 1.0 W/m2 precision

[Trenberth et al 2010]

the ESSENCE ensemble

17 coupled climate model simulations for 1950-2100

A1B emission scenario

started from slightly different initial conditions

→ natural variability versus forced trends

the ESSENCE ensemble

17 coupled climate model simulations for 1950-2100

analysis of Ocean Heat ContentOHC = ρ cp Tocean dAocean dz

the large amount of data allows for a statistically robust analysis of events and processes involved

Upper OHC in ESSENCE

OHC anomaly 0-700m

forced trend

Upper OHC in ESSENCE

OHC anomaly 0-700m

forced trend

measure for rangeof natural variability

Upper OHC in ESSENCE

OHC anomaly 0-700m

forced trend

measure for rangeof natural variability

observations

Upper OHC in ESSENCE

mean trend 1969-2000

long-term ESSENCE trends are in agreement withobservation-based estimates

variability model simulations encompasses observations

Upper OHC in ESSENCE

OHC anomaly 0-700m

calculate 8-yr trends

Upper OHC in ESSENCE

OHC anomaly 0-700m

calculate 8-yr trends

for each simulationfor a 31-year period

→ distribution

distribution of 8-yr trends in UOHC

model: 31 yrs x 17 = 527 events

distribution of 8-yr trends in UOHC

3% of distribution

25-30% chance of one or more (overlapping) 8-yr periods with a negative trend in upper ocean heat content

Where did the heat go?

missing heat8 yr x 0.3 1022 J/yr =2.4·1022 J

if absorbed in the ocean thenΔTocean ≈ +0.02 K

atmosphere continents cryosphere radiation to space deep ocean

upper ocean

deep ocean

atmosphere

cryosphere

con

tin

en

ts

space

Where did the heat go?

missing heat = 2.4·1022 J

atmosphere takes much less energy to

warm one m3 of air (vol. heat capacity x 3000)

larger volume (x 20)

ΔTatm≈ +4.8 K upper ocean

atmosphere

Where did the heat go?

missing heat = 2.4·1022 J

continents takes 40% less energy to

warm one m3 of rock (vol. heat capacity x 0.6)

heat penetrates only over ~50m in a few years

ΔTland≈ +1.2 K (upper 50m)

upper oceanco

nti

nen

ts

Where did the heat go?

missing heat = 2.4·1022 J

cryosphere [warming] takes half the energy to

warm one m3 of ice (vol. heat capacity x 0.5)

volume of the ice sheets over which heat can be absorbed is much smaller

ΔTice≈ +7.5 K (upper 100m)

upper ocean

cryosphere

Where did the heat go?

missing heat = 2.4·1022 J

cryosphere melt 7.3 ·1016 kg ice

global mean SLR ≈ +0.2m(x 8 observed)

3 x current sea ice extent Arctic Ocean

upper ocean

cryosphere

Where did the heat go?

missing heat8 yr x 0.3 1022 J/yr =2.4·1022 J

radiation to space deep ocean

plateau in upper OHC: ⇔ downward trend superposed

on longterm upward trend ⇔ 8-yr trends with respect to

the longterm trend

upper ocean

deep ocean

space

Where did the heat go?

missing heat8 yr x 0.3 1022 J/yr =2.4·1022 J

radiation to space deep ocean

plateau in upper OHC ⇔ downward trend superposed

on longterm upward trend ⇔ 8-yr trends with respect to

the longterm trend

period 1990-2020

upper ocean

deep ocean

space

Top of the Atmosphere (TOA) Radiation

45% of the variation in upper OHC is associated with a variation in the net outgoing TOA radiation

upper ocean cooling ⇔ more radiation out

Top of the Atmosphere (TOA) Radiation

CAUSE: changes in cloud cover and heat uptake associatedwith El Nino / La Nina events (Pacific Ocean)

Deep Ocean Heat Content

35% of the decrease in upper OHC (0-700 m) is associated with an increase in deep OHC (700-3000 m)

Deep Ocean Heat Content

UOHC trend as function of depth

Deep Ocean Heat Content

UOHC trend as function of depth

deep ocean convectionmixing of surface and subsurface waters down to 1000-2000 m[variable process]

Summary

A pause in the rise in upper Ocean Heat Content as recently observed is not unusual in the late 20th / early 21st century, despite greenhouse gas forcing

Such a pause is associated with increased radiation to space (~45%) and an increase in the deep ocean heat content (~35%)

The causes for these energy exchanges are two modes of natural variability (ENSO conditions and deep convection in the North Atlantic Ocean)

Summary

A pause in the rise in upper Ocean Heat Content as recently observed is not unusual in the late 20th / early 21st century, despite greenhouse gas forcing

Such a pause is associated with increased radiation to space (~45%) and an increase in the deep ocean heat content (~35%)

The causes for these energy exchanges are two modes of natural variability (ENSO conditions and deep convection in the North Atlantic Ocean)

Katsman and van Oldenborgh (2011)

Latest ocean heat content observations

Katsman and van Oldenborgh (2011)

Latest ocean heat content observations

ocean expansion is projected to result in 10-30 cm global mean sea level rise

by the end of the century