Skip to content

Water, ice, and the importance of the ocean in global climate change

Facebooktwitterredditpinterestlinkedin

We had an unexpected and unwelcome snowfall yesterday.  It was a brief reminder that we are not in control, but temperatures are again above freezing and headed higher.  I THINK our savage winter is over.  I see that NOAA is now reporting that globally, land temperatures in February were only the 21st warmest on record – it WAAS a cold month.  The global average temperature, land and oceans combined, made it the 7th warmest February on record.

Still, back to our snow.  It got me thinking about the properties of water and how much our climate, and the current changes in that climate depend on these.  With the thaw now well under way, there is still a substantial mass of ice on the lake near my house.  Close to a meter thick and covering the lake it represents a substantial demand for heat if it is ever going to melt (as it will).  The relatively high latent heat of ice (334 kJ per kg) means that it will take a massive influx of heat to melt the ice on this lake, never mind warming the water to a swimmable range.  That ice acts like a brake on our Spring just as, on a much larger scale, the massive glaciers of Greenland and Antarctica act like a brake on climate change.

Then there is the color of ice and snow.  Imagine a world of black ice and snow.  I doubt I would be experiencing an icy winter these days if snow was black, and an excellent absorber of sunlight.  But it is highly reflective and because of this resists absorbing the energy that would be required to melt it.  Climate would be changing in a very different way if our Arctic and Antarctic were periodically covered in highly absorbent black ice and snow.  And finally there is the peculiar fact that water is most dense at 4oC and becomes less dense if cooled below this temperature, and even less dense once it becomes ice.  As a consequence, in the Fall, our lake cools down to 4oC, but only the surface layer cools further.  The ice forms at the surface, trapping warmer water beneath it.  If water was a more typical liquid, it would become increasingly dense until it froze into a solid that would sink, and lakes would freeze from the bottom up.  Presumably, we’d have enormous depths of sea ice at the bottom of the Arctic ocean and the North West Passage would have always been open (unless temperatures were cool enough to lower Arctic ocean temperature sufficiently for the whole thing to freeze solid).

These idle thoughts about the wonders of water can help explain the enormous importance of the oceans in determining our climate, and the complex ways in which the oceans are now interacting with the increased insulation due to greenhouse gases as the planet warms up.  A considerable amount of research is currently being directed to these ocean-climate interactions because they have proved challenging to model yet play important roles in how the climate will change into the future.  As a case in point, consider the reported ‘pause’ in warming that has been going on for the past 5-10 years.  When at first, climate was seen not to be getting progressively warmer, year by year, since 1998, it was put down to the natural variability of the earth’s climate system – an appropriate, cautionary approach under the circumstances.  But as the ‘pause’ has continued, there has been an active search by climate scientists to find processes that might account for what has been happening.  Basic physics says that the insulation in the atmosphere has been trending steadily upward along with increases in CO2 concentration, and that therefore the amount of heat trapped on the planet must also be increasing year by year.  If that heat is not warming our climate (which basically means warming our lower atmosphere and land and water surfaces), it must be going somewhere else.  Among various possibilities, it might be melting glaciers at faster rates than anticipated, and it might be warming the deeper waters of the ocean to a greater degree than expected.  Recent research reports suggest that both these possibilities appear to be correct.

In the April 2014 issue of Nature Climate Change, Shfaqat Khan of the National Space Institute of the Technical University of Denmark, and 12 colleagues from Europe and the USA, reported on the rate of loss of mass in the glaciers of north-eastern Greenland.  Over the past twenty years, the melting of ice from Greenland glaciers has been responsible for slightly more than 15% of the observed rise in global sea level.  It has been known that much of this was due to speed-up of glaciers draining to the southeast and the northwest coasts of the island, however, the substantially larger, northeast ice stream, which extends more than 600 km into the center of the island and drains a sizeable portion of the icefield, had been stable in size (i.e. gaining new ice mass through precipitation as rapidly as it lost ice mass through calving at the coast) at least through the last 25 years.

Using sensitive, satellite-based GPS readings of elevation and of seaward extent, Khan and colleagues were able to show that, after 25 years of relative balance, these northeastern glaciers have been rapidly thinning (losing elevation) and moving more rapidly coastward.  From a rate of loss of ice mass of 1.4 gigatonnes (Gt) per year during the period 1974 to 2003,  the rate increased to 4.3 Gt per year during 2003-2006, 19.6 Gt per year during 2006-2009, and 20.2 Gt per year during 2009-2012.  The overall loss of ice from Greenland’s ice sheet was approximately 172.4 Gt per year in 2006, and more than doubled to 359.8 Gt per year during 2009 to 2012.

Khan and colleagues suggest that thinning of sea ice due to warming has been responsible for the acceleration in loss of glacial mass, and therefore that this process is likely to continue in the near-term future with resulting impacts on sea level and on surface salinity in the vicinity.  The latter may have important consequences for the rate of movement of the ocean conveyor that takes cold surface waters down to become the deep oceanic water.  And of course, any unaccounted for increase in ice sheet melting represents heat in the planetary system that does not show up as climate warming.

Khan et al Fig 1 nclimate2161-f1

Loss of ice mass on Greenland as measured by ice surface elevation loss in meters per year, for the three periods 2003-6, 2006-9 and 2009-12.  The extent of the northeastern glaciers can be seen in the left-hand panel.  Figure © as Fig 1 from Khan et al, Nature Climate Change, April 2014.

Mention of the ocean conveyor leads me logically to my next topic.  Casimir de Lavergne of McGill University and three colleagues from McGill and from University of Pennsylvania published a paper in Nature Climate Change, also in the April 2014 issue, titled’ “Cessation of deep convection in the open Southern Ocean under anthropogenic climate change.”  To appreciate its significance a brief review of the ocean conveyor system may be useful.  The ocean conveyor is a global pattern of immensely large, but quite slow oceanic currents that traverse the globe taking surface waters from the tropics towards both poles, then cooling and sinking this warm, salty, oxygen-rich water deep below the surface to become the water that exists at the very bottom of the oceans.  The conveyor then moves this bottom water around, ultimately bringing it back to the surface to continue the cycle.

ocean conveyor NASA JPL

The ocean conveyor is a complex pattern of shallow and deep currents that moves vast quantities of water around the oceans bringing shallow, well-oxygenated water to the depths and returning deep, oxygen-deficient but nutrient rich waters back to the surface.  Image courtesy NASA.

The figure shows water as descending in the North Atlantic and ascending within the central Indian and Pacific Oceans, however the reality is a bit more complex, with areas of downwelling in the north Pacific and in the Southern Ocean as well.  Movement of the ocean conveyor is driven by changes in the density of water as it changes in temperature and/or salinity.  Surface waters in the Atlantic are warm but relatively salty, and as they move north they lose heat to the atmosphere.  Cooling makes them more dense and they tend to sink below the cold but less saline waters coming out of the Arctic.  This major downwelling is the primary driver of the conveyor.  It is also potentially open to disruption of further melting of Greenland’s glaciers makes ocean waters in the North Atlantic a lot fresher.  If salinity is sufficiently reduced, even in the tropical water moving north, its cooling may become insufficient to cause it to sink, slowing the conveyor.  De Lavergne’s paper concerned the exchanges that occur in the Antarctic, where giant convection cells become established and bring cold dense water up towards the surface where it loses heat to the even colder atmosphere and to adjacent cool but less saline surface waters.  In the mid-1970s one of the largest such convection cells was apparent as the Weddell Polynya, a 250,000 km2 ice-free patch of open water within the winter sea ice of the Weddell Sea.  The continued upwelling at this site kept bringing relatively warm deep water to the surface and prevented the formation of sea ice.  Effectively, the patch of open water, which persisted in each of the two following winters, represented the top of a giant column connecting the surface waters to abyssal depths.  The polar southern ocean is normally only weakly stratified and the convection cell was able to transport substantial heat from the deep ocean to the atmosphere, while also bringing new cold water from the surface to depth, renewing the deep water of the ocean conveyor.  De Lavergne and colleagues were able to show via modelling that the Weddell Polynya and similar convection cells were a relatively frequent occurrence in the past, but have become less likely as melting of Antarctic glaciers has progressively reduced surface salinity, leading to enhanced stratification and greater resistance to convection cell formation.  The result is that not only is the production of deep water reduced because of the lack of downwelling at the convection cell, but that heat trapped in the water is not getting released to the atmosphere.

My final example concerns a paper by Matthew England of the University of New South Wales, and 9 colleagues from Australian and US institutions, that appeared in Nature Climate Change in March 2014.  Their paper, titled, “Recent intensification of wind-driven circulation in the Pacific and the ongoing warming hiatus”, moves us to the tropical ocean and yet another phenomenon that facilitates the transfer of heat from atmosphere to ocean.  Using a combination of dynamic ocean modelling and records of climate going back to the 1920s, they show how patterns of warming of surface air temperature are closely linked in time to the behavior of the Interdecadal Pacific Oscillation (IPO), a fluctuating pattern in which westerly blowing trade winds intensify and weaken on an approximately decadal cycle that ties into the el Niño – la Niña cycle also known as the el Niño Southern Oscillation or ENSO (sometimes I think climatologists just like to make life complicated, but then I realize they deal with an inherently complex system).  Their analysis shows that we have been in a period of negative IPO status for the past 20 years or so, with strong trade winds, weak el Niño events, and large amounts of deep water upwelled off the tropical Pacific coast of northern South America.  This has brought substantial quantities of cold deep water to the surface where it has readily acquired heat from the atmosphere.  As a consequence, they suggest, the warming of the climate has been less than anticipated.

Their modelling of possible scenarios into the future shows that if the IPO switches to a positive phase soon, the less strong trade winds and more powerful el Niño conditions will reduce upwellings off South America, and less atmospheric heat will be transferred to the ocean.  As a result, the rate of climate warming will increase again.  Conversely, if the current negative IPO continues, we could see a period of relative climatic stability until around 2020.  According to England and colleagues, we do not yet understand the IPO sufficiently to know how likely it is to switch over the next few years.

Kosaka Fig 1 March 2014 nclimate2138-f1

This map shows the rate of change in temperature across the globe during the recent past.  The tropical Pacific has been getting markedly cooler, and as a consequence heat has been being taken out of the atmosphere and into the ocean.  Strong trade winds over the past 20 years are responsible, but that situation could be about to change.  Figure © Kosaka, Nature Climate Change, March 2014.

Just to wind up this discussion, an article by Jeff Tollefson in the 3rd April 2014 issue of Nature reports that a strong el Niño now appears to be developing for the coming northern summer and fall.  Easterly winds were recorded in the west equatorial Pacific at the start of 2014, and it looks very much as if an el Niño event of a size comparable to that in 1997-98 may be developing.  If it does, the sink for atmospheric heat provided by the tropical Pacific over the past few years may be shut off and climate will start bumping up again.  I’m currently motivated to bet that this is in fact what will happen, simply because we are well overdue for a strong el Niño.  If that does indeed happen, I think I can be reasonably confident that next April in my part of the world, I will not be looking out on ice and snow.