2010-09-22 19:35:10Purkey.Johnson ocean heat (Journal of Climate 2010)

(In progress) 
Billions of Blow Dryers: Missing Heat Comes Back to Haunt Us Like Bad Disco Music

"The heat will come back to haunt us sooner or later..." --Kevin Trenberth, referring to our inability over the past 5 years to locate half a watt per square meter per year of energy accumulated on Earth as a result of anthropogenic warming of the planet, approximately half of the expected warming signal.

It's a sad fact that while Earth's oceans are expected to absorb the vast majority of anthropogenically induced imbalance of the global energy budget, our physical observations of the caloric state of the deep ocean are conspicuously sparse when compared to daily remote sensing revisitations enjoyed by research subjects amenable to orbital remote sensing platforms. In some ways our instrumentation of such far-flung places as Mars and Venus is better than what we deploy here on Earth in the abyssal depths. While we have solid theoretical grounding for predicting storage of excess heat in the ocean, without the means to directly measure and accurately quantify this effect we're left missing not only heat but also a useful means of testing and validating predictions of climate sensitivity to forcing.  

As our technical capacities have risen to the challenge of dealing with an environment arguably more hostile to instrumentation than near-Earth orbital space, oceanographers at last are enjoying some of the same physical and scientific advantages as those long enjoyed by scientists working with space-based remote sensing platforms. The semi-autonomous Argo array represents a huge leap forward in our understanding of the characteristics of the upper ocean. With respect to anthropogenic climate change, of late we've been treated to increasingly dense and accurate measurements of upper ocean heat content, confirming the absorption of prodigious amounts of heat energy in the upper 2000 meters of the world ocean.

Unfortunately the present Argo implementation is depth-limited and we thus still have no automated systems in place for data retrieval from the slightly over one half of the ocean inaccessible to robotic probes. For this majority of ocean volume we still must rely on hardy investigators "going down to the sea in ships, that do business in great waters." We landlubbers wondering about "missing heat" and suspecting it may be found in the ocean can only be patient as we wait for salt-crusted mariner scientists to return to shore and write up their results.
The main reason for lamentation of  "Trenberth's Travesty" is the declining upward pace over the past 5 years of the portion of ocean heat content (OHC) we're readily able to measure. We know that sea level rise (SLR) is principally caused by both thermal expansion of the oceans and water mass contributed by continued melting of terrestrial ice. Terrestrial ice alone cannot account for the continuing sea level rise we see in the face of the slackened pace of upper ocean warming.  Juxtaposing continuing sea level rise  against OHC we don't observe, we're left with a substantial technical mystery, an inability to "close the budget" of SLR as well as an inability to specifically account for the final destination of heat we know is accumulating on the planet ( Willis 2008 ). Failing the unlikely emergence of some new mechanism able to cause SLR, we may say with reasonable confidence that continued SLR can at least partially be attributed to accumulating OHC we can't directly "see," but merely saying so is no substitute for direct measurements.

Now we may say some significant progress has been made in tracking down "missing heat." In Journal of Climate Sarah Purkey of the University of Washington and NOAA's Gregory Johnson  report on an ambitious project to quantify heat being stored in the abyssal ocean ( Warming of Global Abyssal and Deep Southern Ocean Waters Between the 1990s and 2000s: Contributions to Global Heat and Sea Level Rise Budgets ). By revisiting abyssal stations included in the World Ocean Circulation Experiment (WOCE) conducted in the 1990s, about 20% of Trenberth's famous "missing heat" appears to have been tracked down, found to be slowly traveling north from the Southern Ocean.
While integrating these new measurements into the global heat budget does not entirely close our observational gap, by producing their results Purkey and Johnson have crisply demonstrated how vast amounts of heat may have been left out of the budget for the simple reason of previously being invisible. Their work is also a compelling case for improving our capability to routinely measure with less extraordinary effort the majority of ocean volume we're presently forced to ignore when accounting for accumulation of heat. Finally, it seems reasonable to conclude that these measurements bolster our confidence in SLR as a proxy for increasing OHC. 
Purkey and Johnson's abstract: 
We quantify abyssal global and deep Southern Ocean temperature trends between the 1990s and 2000s to assess the role of recent warming of these regions in global heat and sea level budgets. We compute warming rates with uncertainties along 28 full-depth, high-quality, hydrographic sections that have been occupied two or more times between 1980 and 2010. We divide the global ocean into 32 basins defined by the topography and climatological ocean bottom temperatures and estimate temperature trends in the 24 sampled basins. The three southernmost basins show a strong statistically significant abyssal warming trend, with that warming signal weakening to the north in the central Pacific, western Atlantic, and eastern Indian Oceans. Eastern Atlantic and western Indian Ocean basins show statistically insignificant abyssal cooling trends. Excepting the Arctic Ocean and Nordic seas, the rate of abyssal (below 4000 m) global ocean heat content change in the 1990s and 2000s is equivalent to a heat flux of 0.027 (±0.009) W m–2 applied over the entire surface of the Earth. Deep (1000–4000 m) warming south of the Sub-Antarctic Front of the Antarctic Circumpolar Current adds 0.068 (±0.062) W m–2. The abyssal warming produces a 0.053 (±0.017) mm yr–1 increase in global average sea level and the deep warming south of the Sub-Antarctic Front adds another 0.093 (±0.081) mm yr–1. Thus warming in these regions, ventilated primarily by Antarctic Bottom Water, accounts for a statistically significant fraction of the present global energy and sea level budgets. 
In an interview, coauthor Gregory Johnson expressed the amount of heat identified in this study in amusingly prosaic terms: the newly located reservoir of energy is akin to what would be liberated by loading every man, woman and child on Earth with five 1,400 watt hairdryers each and running those appliances continuously for the 20 year interval between measurements. 
Purkey and Johnson's results, mapped:
"Mean local heat fluxes through 4000 m implied by abyssal warming below 4000 m from the 1990s to the 2000s within each of the 24 sampled basins (black numbers and colorbar) with 95% confidence intervals and the local contribution to the heat flux through 1000 m south of the SAF (magenta line) implied by deep Southern Ocean warming from 1000–4000 m is also given (magenta number) with its 95% confidence interval." (Purkey and Johnson, 2010)
How can Antarctic Bottom Water (AABW) influence abyssal temperatures so far north of the Antarctic? To understand this, it's helpful to grasp the huge role in deep ocean circulation played by the Southern Ocean and the Antarctic. AABW is derived from enormous quantities of chilled, relatively saline and thus dense water sinking at the extreme south of the globe, in Antarctic waters. This mass of dense water is relatively free to travel north, first plunging off the Antarctic continental shelf and then hugging the bottom as it displaces warmer water. AABW is steered by bottom topography and Coriolis forces and only ceases moving and thus influencing abyssal temperatures when it has reached equilibrium density with surrounding water. Even after traveling some 60 degrees north of its source, density differences are still large enough to drive substantial amounts of AABW past the circulation barrier imposed by the equator, thus permitting diminished but still measurable circulation effects of AABW to be measured in the abyssal depths of the Northern Hemisphere. 
A pair of illustrations of Antarctic and Southern Ocean circulation may be helpful in understanding the process of AABW transport.  

"South (left) to north (right) section through the overturning circulation in the
Southern Ocean. South-flowing products of deep convection in the North Atlantic are
converted into upper-layer mode and intermediate waters and deeper bottom waters and
returned northward. Marked are the positions of the main fronts (PF – Polar Front; SAF –
Sub-Antarctic Front; and STF – Subtropical Front), and water masses (AABW – Antarctic
Bottom Water; LCDW and UCDW, Lower and Upper Circumpolar Deep Waters; NADW –
North Atlantic Deep Water; AAIW – Antarctic Intermediate Water and SAMW – Sub-
Antarctic Mode Water)" (Figure 1.9, SCAR "Antarctic Climate Change and the Environment" )

Role of Southern Ocean in global circulation ( Lumpkin and Speer, 2007
One might wonder, if AABW circulation is driven partly by the relative density of water chilled in the Antarctic, won't distribution of this water change as deep waters warm in response to heating by AABW circulation, thus robbing AABW of some of its physical transport impetus? This does seem to be the case; for instance, the interface between AABW and North Atlantic Deep Water (NADW) has deepened over the past few decades and as well there are indications of diminished abyssal circulation in regions of the North Pacific influenced by AABW, as would be expected in a scenario where density gradients are diminishing ( Johnson 2008, Kouketsu 2008 ). Numerous other variations in circulation behaviors controlled by thermally induced density variances may be found in Purkey and Johnson. Taken together, these indicators are broadly consistent with changes in the thermal regime of the deep ocean connected with  AABW and its source. 
It's important to note that to a greater or lesser extent the Southern Annular Mode (SAM) plays some role in controlling changes observed in Purkey and Johnson, not to the exclusion of secular changes outside of the SAM but significant nonetheless. Complications abound in forming an exact assessment of the proportionality of natural versus forced variations; the SAM itself appears to be in a process of adjusting to two anthropogenic influences, ozone depletion and greenhouse gas proliferation. 
Beyond shedding enlightenment on a specific research topic, Purkey and Johnson's work suggests some improvements we could make in the level of urgency we attach to exploring our planet. NOAA is working on upgrading our ability to sample deep and abyssal ocean water via robotic instrumentation.  As is so often the case, the pace of instrumentation improvements is set in part by budgetary limitations involving amounts of money small in the grand scheme of things. Purkey and Johnson show beyond doubt how vital better observational ability is when it comes understanding our role in shaping the climate; we're effectively blind to enormous changes in the physics of our planet because we won't make paltry expenditures for better "optics," a lamentable and unnecessary condition.  Our instrumentational inability to closely track climate change is a general problem; it's truly odd that such a important research topic so crucial to public policy should find itself lacking the equipment to quantify changes nearly everybody agrees present us with multi-trillion dollar risk and decision choices and outcomes. 

A Salute to Oceanographers

In terms of effective inaccessibility and remoteness, Earth's oceans and especially the abyssal depths have aptly been compared unfavorably to extraterrestrial space.
For researchers investigating Earth's climate, orbital space is in some ways a far friendlier environment than the oceans. Instruments aboard satellites allow researchers to collect their data while lounging in shirtsleeve comfort, facing nothing more dangerous in the daily routine than slipping and falling while taking a morning shower. Meanwhile, oceanographers often must wrest their primary information from the ocean at notable personal hazard, conducting their observations from the pitching, rolling decks of ships and directly exposed to the uncaring vagaries of fickle weather and heavy machinery. Errors in procedure, equipment failures or even inclement weather may exact the ultimate penalty on oceanographers seeing to the meticulous collection of data, career hazards not faced by many other scientists investigating climate-relate phenomena. 

Quite apart from the kinetic drama of working from small ships on the surface of such storied locales as the notoriously stormy Southern Ocean, the sea is also extraordinarily costly in terms of the personal investment of time required to wrest every few hundred kilobytes of data from the cold dark of the bottom. After dealing with a commute of thousands of miles to their laboratory enviroment, scientists are rewarded with brief spurts of information separated by long intervals of plodding, akin to "crossing the ocean at a jogging pace" as NOAA's Gregory Johnson expresses the matter. 
For all these reasons, it's worth pausing a moment in appreciation of the fanatical dedication and perseverance needed to collect data of the kind used to produce Purkey and Johnson's paper. The graphs and maps casually flung out in little essays such as the one above inadequately express countless thousands of cold and dripping hours spent in hostile conditions far from hearth and home. True enough it's an all-volunteer army, but recognition of this effort is still due. 

2010-09-22 20:22:16Looking good so far, Doug
John Cook


Always enjoy your prose, you love to spin a good yarn :-)

To me, the conundrum of ocean heat comes down to this. To properly understand climate, it's not good enough to obsess over a single metric (like upper ocean heat). We need to consider the full body of evidence and bring it all together to piece together the complete picture. There are three major pieces in the energy imbalance puzzle. Satellite measurements of ingoing and outgoing radiation, ocean heat and sea level. The satellite data has such a high uncertainty, it's not the most reliable metric. While not accurate, it does however have relatively high precision and what it does tell us is the planet's energy imbalance is increasing.

Sea level is possibly just as important a metric of global warming as ocean heat - as it tells us both how much the oceans are warming and how much ice is melting. And sea levels are still rising at around 3mm per year. This is a crucial result. If the ocean really was cooling as Pielke supposes, we wouldn't see sea level rising as much as it is (unless the ice sheets are crumbling way more than we currently think).

So a consideration of ocean heat should also keep rising sea levels in mind. We need to build a consistent picture of ocean warming and sea level rise together. Currently, scientists are having trouble closing the sea level budget. This says to me that the most likely source of the discrepancy is missing ocean heat.

Anyway, sorry Doug, I'm rambling, don't know if that helps you at all. Just thinking out aloud :-)

2010-09-22 20:41:07


10-4 on SLR, John, and it's revealing to notice P&J 2010's emphasis on how this result affects SLR budgets. Even as we've seen ocean heat accumulation appear to slow down from our  limited perspective, SLR has continued to march upward. Now we can see how that might happen; Global SLR is a handy total integrator of everything contributing to increased SLR, a proxy for thermal content, but simply leaves unquantified/unidentified contributions in a lump line item in the budget. P&J have peeled a large number off of that lump. 

I think many of us (well, me anyway) suspect that in "missing heat" we're seeing at least in part a variability effect, similar to what would happen if we were trying to derive total atmospheric heat from a single layer of troposphere. Transport variability is briefly mentioned in P&K but is a whole other ball of wax beyond the scope of the paper. Maybe I should include some doorways to that in this blog post?

2010-09-23 08:11:38nice job
Dana Nuccitelli

Very well-written, Doug.  My only comment is that I didn't even realize there was a discrepancy in the "SLR budget".  Is there a reference on this topic you could link to in the article?

I guess it just makes logical sense, if ice melt can't account for the full SLR and upper OHC has flattened out, the thermal expansion must come from deeper oceans.  But if perhaps there's a specific study on this subject, it would make for a useful reference.  Maybe something quantifying the contribution of ice melt to SLR?

2010-09-23 10:04:31


Thanks, Dana. The Willis article cited in Purkey and Johnson seems to be the most up to date and comprehensive and I think you're right that I should provide a link. I'll do that. (in the meantime, http://ecco.jpl.nasa.gov/~jwillis/willis_sl_budget_final.pdf ). Although apparently loathe to speculate, Willis seems to believe there's a combination of systematic error in one of the measurement systems acting in concert with unattributed stearic SLR causing a gap in the budget. 

I had an opportunity to talk w/Greg Johnson today; he was kind enough to  to read the draft you probably read and suggest a few tweaks. Greg emphasizes how fraught with complication the Southern Ocean circulation system is, leading me to add some caveats to the post. 

More instruments, that's what we need. Greg mentioned plans for lurking benthic measurement devices that will gather data from the seafloor,  pop up at end-of-life and squirt their data skyward for collection by satellites. As well, they're looking at creating a subpopulation of Argo floats capable of going deep. Too expensive to do 'em all that way...