TCS Daily


Winter Weather Wonder, Part II.

By Willie Soon - February 24, 2004 12:00 AM

January 2004 was unusually cold for most of the Northern Hemisphere (see previous TCS CharTiFact). The cold prompted the UK's Independent to link the weather in a recent op/ed to increased carbon dioxide in the air saying, "Global warming will plunge Britain into new ice age 'within decades'." This is hype.

What is the idea in the Independent story? More carbon dioxide in the air would warm polar areas, melt ice there, and so intensify hydrologic cycles. That would, in turn, bring extra freshwater to the North Atlantic through increases in both precipitation and river run-off. The melting of polar ice further reduces the salt concentration, or freshens, the water of the North Atlantic.

According to some ocean models, extra columns of freshened water, formed in the Labrador Sea and North East Atlantic areas around Irminger Sea (Figure 1), would slow the deep-seated (1-3 km below the surface), southward-flowing branch of the giant ocean current system in the North Atlantic that transports water, heat, and salt called Thermohaline Circulation (THC). Some ocean models, when fed extreme scenarios, collapse THC. That would then cut off the surface branch of the THC, which transports heat from the warm tropical water, via the Gulf Stream, to the North Atlantic. An abrupt shut-down of the Gulf Stream would then quickly chill the North Atlantic. In the models, glaciation then ensues over large part of the Northern Hemisphere continents -- i.e., the first anthropogenic Ice Age.

Figure 1: Illustration of the North Atlantic Thermohaline Circulation showing both the upper (surface) branch of the "conveyor" belt as exemplified by the Gulf Stream and the lower branch of the belt maintained largely by heat-salt-freshwater balances around the region of the Labrador and Irminger Seas.

Now that is a scary forecast.

Not disclosed in this op-ed and several op-eds last month (see for example David Stipp in Fortune magazine and Thom Hartmann) is that the computer scenarios of freshening must be checked: Direct measurements of flow rate or the overturning rate of THC are either uncertain or unavailable, making the scenarios unreliable because they cannot be tested.

Ray Schmitt of the Woods Hole Oceanographic Institution has noted:

"In contrast to the 1,200 records of US land temperature used to examine climate trends ... we have only three sites with anything like a continuous deep record in all of the North Atlantic! For these few sites with rather short records. An observation once a month is often the best we have. This observation system is woefully inadequate. ... Science is the process of testing ideas against observations, and failure to make the observations is an abandonment of the scientific process."

There are also questions of the physics of energy flowing through the ocean. Carl Wunsch of MIT recently concluded that it is unclear that THC shut down during the Last Glacial Maximum, roughly 20 thousand years ago.[1] Instead, Wunsch offered an alternate idea: the much windier conditions then could produce a current in the Atlantic Ocean that rides at a shallower ocean depth than THC current presently does. This shallower current neither weakens nor shuts down during the cold, glacial periods.

Beyond the question of how THC works, let's look at the scary forecasts for glaciation, which are rooted in ocean-atmosphere climate models. Several models now suggest that THC would not weaken, even if the air's carbon dioxide content were doubled (Figure 2).

Figure 2: Early models predicted North Atlantic THC to weaken under carbon-dioxide produced global warming scenarios. But a more advanced model from Max Planck Institute for Meteorology shows THC relatively insensitive to enhanced-greenhouse gas climate, as reported in Latif et al. (Journal of Climate, vol. 13, 1809-1813, 2000).

Early computer simulations did show a weakening of the THC circulation by 20 to 50% under similar carbon dioxide loading in the air by 2100 (Figure 2, left panel). But as early as 2000, researchers from Max Planck Institute (MPI) of Meteorology in Hamburg had found no weakening of THC under carbon-dioxide produced global warming (Figure 2, right panel). The new MPI ocean model had an improved ability to resolve the tropical oceans -- at least four times better than in previous models. Also to note is that the older MPI model shown in the left panel of Figure 2 also predicted weakening of THC just like others.

The newest UK Hadley Centre model results (Figure 3) confirm[2] MPI's work. Hadley Centre's model, again employing a high-resolution ocean model with a spatial grid of 37 km in the equator -- compared to 200 km in older models -- finds little evidence on a weakening or collapse of THC as the air's carbon dioxide content is increased.

Figure 3: The new, high-resolution ocean model of UK Hadley Centre (top panel) shows no change in heat transport in the North Atlantic between 20 to 30°N for both the control (no carbon dioxide added to the air) and when carbon dioxide is added to the air. The older model (bottom panel) did show a strong decrease in the heat transport by 70-80 years later under the scenario of carbon-dioxide enhanced global warming. [Adapted from Roberts et al., Journal of Climate, vol. 17, 3-20 (2004)]

January 20's issue of Geophysical Research Letters further questions the older and immature ideas about North Atlantic THC. First note that model shows a freshening of deep ocean water (2-3 km) in the northeast Atlantic region during the latter half of the 20th century (Figure 4, top panel, which shows decreasing salinity) while at the same time a strengthening of the North Atlantic overturning (bottom panel). The authors explain that their model results -- opposite to earlier model results -- arises from an increase in the north-to-south gradient of the density in the upper oceans (top 1 km) between the sub-polar North Atlantic and the low-middle latitudes.

Figure 4: Changes in the salinity (top panel) and circulation strength (bottom panel) over the North East Atlantic Deep Water region at a depth of about 2 to 3 km during the 20th century (calculated with natural and anthropogenic forcings together) and predicted for the 21st century, under a carbon-dioxide enhanced global warming scenario. Note that as the deep water is increasingly freshened, the North Atlantic overturning circulation strengthens rather than weakens, as popularly believed. Also note that the 21st century is predicted to have increasingly more saline water (less freshening) around this critical overturning region of the THC. [Adapted from Wu et al., Geophysical Research Letters, vol. 31, L02301, 2004]

Another surprise seen in Figure 4 is for the 21st century: the model expects a decrease in freshening (increasing salinity in the top panel of Figure 4) of the deep ocean water as global warming proceeds from the air's rising carbon dioxide content. Consistent with this particular model, THC strength decreases when the deep ocean water becomes saltier (i.e., less fresh). Although those authors did not elaborate further, they carefully concluded that "our analysis based on model simulations does not seem to support an interpretation of the observed freshening trend [during the 20th century] as an early signal of climate change due to human activities."

In another paper just published,[3] Rainer Bleck of the Los Alamos National Lab and Shan Sun of the NASA Goddard Institute for Space Studies concluded that:

"[Our] analysis confirms that even regional details of the [North Atlantic] MOC [Meridional Overturning Circulation] in this experiment are rather insensitive to the climate change brought on by CO2 doubling. It is too early to state with any degree of certainty whether the decline seen in most model simulations compiled in IPCC (2001) [NB: IPCC Fig. 9.21, similar to results shown in Figure 2 here] is due to numerical errors or reflects a legitimate response to changes in the atmospheric environment. Recent publications, such as Sun and Bleck (2001b) and others referenced therein, paint a picture which makes a strong decline of the Atlantic MOC during CO2 doubling appear less likely."

So far the North Atlantic THC is not realistically represented by even the best of current generation of climate models, and models give opposite results. Canadian oceanographers Greg Holloway and Oleg Saenko cautioned earlier that:

"understanding what makes the conveyor [NB: as shown in Figure 1's cartoon here] is deficient, drawing mainly on the role of bouyancy loss leading to sinking [is] somewhat like trying to push a string. The missing dynamics are that eddies in the presence of bottom topography tend to set up mean flows that carry major circuits of the conveyors, allowing sunken water masses to 'go for a ride.' Climate models have difficulty in both these regards -- to include (if at all !) a plausible Arctic Ocean and to deal with eddies either explicitly or by parameterization."

And Woods Hole oceanographer, Ray Schmitt, continues, "We will not come to an understanding of climate by computational cycles of models with incorrect physics."

The idea that THC will weaken from excess freshening of the North Atlantic Oceans from an enhanced hydrologic cycle and polar ice melt is a popular but not scientifically accurate idea. Worse is saying that cold weather of January 2003 and 2004 -- not climate, which is averaged over several decades -- belongs to more or less anticipated, indirect effects of global warming based on the air's increased carbon dioxide content. Important as the question may be, National Center for Atmospheric Research's ocean modeler, Peter Gent, cautioned that estimating the response of the North Atlantic THC is "a very demanding question to ask of the current state-of-the-art coupled climate models."

The full complexity of ocean circulation remains to be understood. Despite that, it is no surprise to read unsubstantiated bottom-lines such as this in Fortune magazine:

"Action now matters, because we may be able to reduce its likelihood of happening ... Policymakers may even be emboldened to take steps such tightening fuel-economy standards for new passenger vehicles, a measure that would simultaneously lower emissions of greenhouse gases, reduce America's perilous reliance on OPEC oil, cut its trade deficits, and put money in consumers' pockets."

Science is to be preferred over fiction. The current generation of climate models, no matter how noble, fails to comprehend the difficult conditions of THC's past or future states -- strengthening, weakening or collapsing. Touting greenhouse gas emission reduction to save the world from a future glacial period is an unqualified prescription with no scientific vetting.



[1] Wunsch C., Quaternary Science Reviews, vol. 22, 371-385 (2003).

[2] Actually, there are at least two prominent publications that had supported the conclusion by the MPI scientists (Latif et al., 2000, Journal of Climate, vol. 13, 1809-1813) including those by Peter Gent using National Center for Atmospheric Research (NCAR) Climate Community Model (Gent, 2001, Geophysical Research Letters, vol. 28, 1023-1026) and Shan Sun and Rainer Bleck using NASA Goddard Institute for Space Studies (GISS) model (Sun & Bleck, 2001, Geophysical Research Letters, vol. 28, 4223-4226). The paper by Latif et al (2000) did not find any significant weakening of the THC because their model allows for more advection of high salinity water from the tropical Atlantic towards the sinking regions (Figure 1) to ultimately compensate for effects of local freshening and warming in the North Atlantic. Gent (2001) found in his model that because of increasing CO2, the sea surface temperatures (SST) around the Labrador Sea and East Greenland current area warm by as much as 3 to 5°C. That increase in SST and related increase in surface wind speed over those regions cause more heat and freshwater loss by evaporation which ultimately counter-acts the effects from extra freshening by other processes. So the THC remains stable in Gent's NCAR climate model. Please see also the discussion about the confirmation of a stable THC despite doubling of CO2 in the updated, February 2004's paper by Bleck & Sun in the main text.

[3] Bleck and Sun, Global and Planetary Change, vol. 40, 233-248 (2004).


Categories:
|

TCS Daily Archives