Achtergrond

Tracing the upper ocean’s 'missing heat'

Against expectations, the upper ocean (from 0 to 700 meter depth) has not warmed since 2003. A recent KNMI study shows that an eight-year interruption of the rise expected from global warming is not exceptional.

It can be explained by natural variability of the climate, in particular the climate oscillation El Niño in the Pacific Ocean and changes in ocean currents in the North Atlantic Ocean. Recent observations point to an upcoming resumption of the heating of the upper ocean. 

ARGO buoy
ARGO buoy

Observations of the sea water temperature show that the upper ocean has not warmed since 2003. This is remarkable as it is expected the ocean would store that the lion's share of the extra heat retained by the Earth due to the increased concentrations of greenhouse gases. The observation that the upper 700 meter of the world ocean have not warmed for the last eight years gives rise to two fundamental questions:

What is the probability that the upper ocean does not warm for eight years as greenhouse gas concentrations continue to rise?
As the heat has not been not stored in the upper ocean over the last eight years, where did it go instead?

These question cannot be answered using observations alone, as the available time series are too short and the data not accurate enough. We therefore used climate model output generated in the ESSENCE project, a collaboration of KNMI and Utrecht University that generated 17 simulations of the climate with the ECHAM5/MPI-OM model to sample the natural variability of the climate system. When compared to the available observations, the model describes the ocean temperature rise and variability well.

Eight years trends in heat content
Observations of the temperature of the upper few hundred meters of the ocean go back to the 1960s. Up to ten years ago most measurements were taken by simple thermometers that were thrown overboard and sent back the temperature as they fell down through a wire (expandable bathythermographs, XBTs). Since about ten years these have been superseded by fully automatised ARGO floats that measure temperature down to 2000 m depth and send the data home every ten days. Starting from these raw observations the global temperature distribution down to 700 meter is reconstructed, filling in the gaps in the coverage. Using the heat capacity of water this enables the estimation of the amount of heat stored in the world ocean.

Over the last decades the heat content of the upper ocean has risen on average, but the rate of increase shows large variability (Figure 1). The model calculations show a very similar behaviour. In the 17 different model computations the eight-year trends in heat content were even negative 11% of the time between 1969 and 1990, implying that he ocean cooled on average for eight years in spite of the warming background trend due to global warming. In the model, the fraction of negative eight-year trends decrease as the warming trend accelerates, but between 1990 and 2020 (31 years around 2005) 3% of the trends still is negative. This implies a one in three chance of at least one eight-year period with a negative trend in these 31 years. An eight-year pause in the rise of the upper ocean heat content is therefore not at all rare, even in a warming climate.

Figure 1: Increase of the amount of heat stored in the upper layers of he ocean (0-700m). Three different analyses are denoted in blue. The measurements are not very reliable before 1970. The warming trend has been simulated 17 times with a climate model
Figure 1: Increase of the amount of heat stored in the upper layers of he ocean (0-700m). Three different analyses are denoted in blue. The measurements are not very reliable before 1970. The warming trend has been simulated 17 times with a climate model

Where does the heat go?
The amounts of CO2 and other greenhouse gases in the atmosphere are steadily increasing. The increased absorption of thermal radiation by these gases causes the radiation to space to emanate from higher in the atmosphere on average, where it is colder. The colder air emits less thermal radiation, so that the incoming solar energy is no longer balanced by outgoing radiation. The excess heat is absorbed by the ocean, slowly warming the water from the top down.

If the upper ocean does not warm for a few years the excess heat from the imbalance between incoming and outgoing radiation has to go elsewhere. The ocean temperature has only risen 0.02 ºC less than expected, but due to the size of the ocean and the large heat capacity of water this represents a huge amount of heat. If this heat would have been used to heat the atmosphere, the air temperature would have increased by 5 ºC. This obviously did not happen, so the heat was not stored in the atmosphere. The ground has a larger heat capacity, but heat penetrates only slowly down. Storing the heat missing from the upper ocean in the ground would have raised its temperature by about 1.5 ºC. This also was not observed, so we can conclude that the bulk of the heat did not go into the ground. If the heat would have been absorbed by land or sea ice it would also have had large consequences that have not been detected, for example a sea level rise of 20 cm if the heat would have been used to melt land ice.

By elimination, only two possibilities remain. Either the Earth radiates more energy to space during these periods of no increase in upper ocean heat content or the heat content of the deep ocean (below 700 meter) increases temporarily. Both possibilities were found to play a role in the climate model.

More radiation into space
The extra radiation into space results from heat loss due to increased thermal radiation and from more clouds reflecting sunlight back to space. This happens predominantly over the Pacific Ocean and is related to the El Niño / La Niña climate oscillation. El Niño is a temporary heating of the surface waters of the ocean along the equator in the Pacific Ocean. La Niña is the opposite, a cooling of the sea surface. The two alternate in a very irregular natural cycle with episodes lasting half a year to a year, sometimes a few years in a row. During El Niño and a few months afterwards the ocean surface is warmer than normal (Figure 2). This causes an enhanced heat loss to the air, so that the ocean heat content decreases. The air cannot retain the heat for long and radiates it into space. Conversely, during and shortly after La Niña the ocean heats up. When El Niño and La Niña balance each other their effects on the ocean heat content cancel, but a series of El Niño events can give rise to a sizable loss of heat from the ocean into space.

Figure 2. Average surface temperature anomaly of the ocean three months after an El Niño episode of 1 ºC (NCDC ERSST v3b observation). The ocean cools through extra radiation and hence decreases the ocean heat content.
Figure 2. Average surface temperature anomaly of the ocean three months after an El Niño episode of 1 ºC (NCDC ERSST v3b observation). The ocean cools through extra radiation and hence decreases the ocean heat content.

Heating deeper in the ocean
The model shows that during periods that the upper layers of the ocean do not heat, the deeper layers show a stronger increase in temperature. This vertical seesaw is strongest in the North Atlantic Ocean south of Greenland. In this area the surface waters cool each winter due to cold winds from Canada. As it gets heavier than the slightly warmer but more salty water at depth the surface water sinks and the warmer water rises. This exchange therefore cools the deeper ocean. In winters with little mixing the upper ocean stays colder and the deeper layers stay warmer.

This mixing is connected with ocean currents that transport large amounts of heat to the North, the Atlantic Meridional Overturning Circulation (also called “the conveyor belt of the ocean” because of the large role it plays in the heat transport of the ocean). When this current is weak the deeper ocean, between 1 and 2 km below the surface, warms but the surface layers become colder.

Figure 3: The pattern of the decrease of heat content of the upper ocean two years after a decrease of the strength of the Atlantic Meridional Overturning Circulation (ECHAM5/MPI-OM climate model).
Figure 3: The pattern of the decrease of heat content of the upper ocean two years after a decrease of the strength of the Atlantic Meridional Overturning Circulation (ECHAM5/MPI-OM climate model).

What happened during the last eight years?
From 2002 to 2007 the Pacific Ocean experienced a series of large and small El Niño events without compensating La Niña events. These events probably cooled the upper layers of the ocean. This cooling due to extra radiation to space coincided with a period of reduced heat exchange in the Labrador Sea between Canada and Greenland, which heated the deeper ocean at the expense of the upper layers. Observations of the temperature down to 2000 meter indeed show a clear increase in temperature below 700 meter. Unfortunately these observations only go back ten years, which is too short to separate the long-term trend from natural variability.

The combination of these two random climate fluctuations contributed to the lack of heating of the upper ocean in the period 2003-2008.

Other factors that could have influenced the upper ocean heat content are the low activity of the sun over the last few years and the air pollution in East Asia. From satellite observations of the total amount of solar radiation reaching the Earth it can be shown that the lack of sunspots has contributed at most 10% to the missing upper ocean heat content. The influence of air pollution is also estimated not to be a dominant factor.

Outlook for the coming years
Since two years ago El Niño has been replaced by a series of La Niña events that should cause a heating trend in the upper ocean. The heat exchange between the upper and deep ocean in the Labrador Sea has also started again recentIy. We therefore expect that the upper ocean heat content will soon resume its upward trend.

Conclusions
In theory, the heat content of the ocean is a very good indicator of global warming. The extra heat retained by greenhouse gases simply cannot go anywhere else. As so often, reality is more complicated than the simple theory.

The first problem is that the temperature of the ocean is not observed everywhere. Below 700 meter depth the observations are too sparse to be able to compute the amount of heat stored before 2001. We found in our analysis that a heating of these deeper layers often compensates a relative cooling of the upper layers. This is coupled to a great extent to vertical heat exchange in the northern North Atlantic Ocean. The second difference between theory and practice is that there are natural fluctuations in the amount of heat that the Earth radiates to space. In our model these are largely connected to El Niño and La Niña, a climate oscillation in the Pacific Ocean that is known to influence the heat budget of the Earth.

Because of these natural fluctuations a short trend in the upper ocean heat content is not a good indicator of enhanced greenhouse warming, only the long-term trend is.

Journal reference:

C. A. Katsman en G. J. van Oldenborgh, Tracing the upper ocean’s “missing heat”, Geoph. Res. Lett, in press, doi:10.1029/2011GL048417.

The role of the deep ocean is also analysed in

M. D. Palmer, D. J. McNeall and N. J. Dunstone, Importance of the deep ocean for estimating decadal changes in Earth’s radiation balance, Geoph. Res. Lett, in press, doi:10.1029/2011GL047835.

Niet gevonden wat u zocht? Zoek meer achtergrond artikelen