Arguments
What is the role of the ocean’s abyss in global warming?1
I’ve already posted on a study published in Nature Climate change that shows that the amount of extra global warming related heat in the Southern Oceans is greater than previously thought. There is another paper in the same journal by Llovel et al, “Deep-ocean contribution to sea level and energy budget not detectable over the past decade.” This paper verifies previous research that the oceans absorb a lot of the excess heat, but looks specifically at the ocean below 2,000 meters, which the paper referrs to in places as “deep” but that we should probably call “abyssal.”1 The paper concludes that the abyss is not warming. This is bad news, because if it was warming the total effects of global warming on the surface would be potentially less, or at least, stretched out over a longer period of time. But, it is not unexpected news. We already suspected that the abyssal ocean does not absorb much of the surface heat, while the shallower ocean absorbs quite a bit.
Research done prior to 2012 (e.g. Hansen et al 2011) parceled out the energy imbalance the Earth experiences from anthropogenic global warming. The extra heat caused by AGW from 2004 to 2010 was divided among the upper ocean (71%), the deeep ocean (5%), with the rest going various other places (only 4% over land). The new paper suggests that the abyssal ocean takes up closer to zero heat.
There are three complexities you need to be aware of to interpret this finding. First is the complexity in the climate system, second is the complexity of the research itself, and third is the relatively straight forward statistical problem of assigning meaning to specific numbers. That third one is important for journalists and regular people to pay attention to, because the climate science denial community is already exploiting it to misrepresent this study.
This is a complex and difficult problem
We know that the vast majority of the extra heat resulting from global warming ends up in the ocean, and also, we know there is a lot of interaction between the ocean and the atmosphere, with heat that might otherwise add to the atmosphere seemingly entering the ocean on a regular basis, with some of it occasionally coming out in large quantitates during El Nino events. This relationship is expected to change over time as the ocean warms, as the transfer of heat between ocean and atmosphere depends in part on the relative difference between them. At some point it is likely that the degree to which the ocean takes up net heat will decrease if the ocean warms up beyond a certain point.
Over the medium and long term this matters a lot. Because of the ocean (and polar ice and a few other things) the effect of increasing greenhouse gasses is not instantaneous. If the Earth was a simple rock with no water, but a Nitrogen atmosphere with, say, 250ppm of CO2, the greenhouse effects of the CO2 would ensure that the atmosphere was at least a little warm. If we doubled the CO2 the atmosphere would warm further, and it would do so very quickly. A new equilibrium would be reached in a geological instant (a few years?). But with the ocean, that change is much slower slower (decades, perhaps many decades), because the ocean buffers the atmospheric change.
When heat goes into the ocean, it then moves around in the ocean because it disperses across the aqueous medium, and because water is always moving in currents or mixing. An El Nino is a change in the movement of water that has been warmed with contact with the surface, so that warm water that has been building up at depth over time changes its movement pattern and moves closer to the surface (and to a different horizontal location) where heat is released. That is one (especially large and important) example of the complex dynamic of atmosphere, ocean, and heat. Currents that move through the upper ocean then dive down to depth may move some of the surface heat to the deeper waters, especially where the currents have dived not just from cooling water (hot water would tend to go up, cold water would tend to go down) but because it is driven in “conveyor” systems which may run counter to expectations of where water should go when considering only local conditions, and especially, if the water is dropping because of an increase in salinity. Again, this is an example of the complexity of the system.
If we add a lot of CO2 to the atmosphere, the atmosphere will warm up, but because of the complexities cited above, it is hard to say how much or how long it will take. The ocean serves to slow the process down. In fact, it is quite possible that if the ocean would be so kind as to absorb a certain amount of this heat permanently, maybe global warming would be somewhat reduced. The ocean is potentially a way of stretching out the effects of global warming. But this effect is likely reduced if the abyssal ocean is not in the game.
Complexities in the research
Meanwhile the measurement of heat in the ocean has been very sparse. Over the last decade more measurements have been taken using new technology, but even that is not as good as we would like to understand what is going on at depth. So, when it comes to understanding heat in the ocean, we may sometimes feel like we are at sea. The two papers in this week’s Nature Climate Change are much more important as studies that calibrate or refine the process of measuring ocean heat dynamics under global warming than they are studies that change our view of global warming. Neither paper concludes anything unexpected, both provide important refinements to key numbers, exploiting the last decade of improved data collection.
One of the complexities is in the details of the Llovel et. al study as compared to the handful of previous related studies. One of the key numbers is the energy imbalance where the ocean absorbs extra AGW produced heat. Energy imbalance is measured in terms of Watts per m–2. The present study yields a value of 0.72. A previous study reported 0.54. Other estimates have varied in this range. Llovel et al point out, however, that these differences may be due to differences in the ocean depth considered in each study and the time periods covered. At least one earlier study measured energy imbalance for the top 1,800 meters, while Llovel et al look at the top 2,000 meters, and all the studies cover somewhat different time periods.
So, we have changing quality of data, a data set that is growing incrementally over time, studies that look at slightly different time and space parameters. And, on top of this, we have the increasingly advanced methods of figuring this all out. Both of the Nature Climate Change studies used a combination of direct measurements of temperature at various depths, a measurement of the altitude of the top of the ocean (sea level) from highly accurate satellite instruments, and measures of the mass of the water in the ocean, from the GRAIL gravity research project. If the mass of the ocean stays the same (same number of water molecules) but the surface rises, that is from heat, and that allows an estimate of energy imbalance. If the ocean goes up more than it should from heat expansion, the extra may be from glacial melting. And that is the simple version.
Statistical reasoning
The statistical part of this is not really so complex. Well, it is, but the part I want to point out is not. Llovel et al concluded “Accounting for additional possible systematic uncertainties, the ocean below 2,000 m contributes −0.13 ± 0.72 mm yr−1 to global sea-level rise and −0.08 ± 0.43 W m−2 to Earth’s energy balance.” Sea level rise is close to 3 mm a year, so the abyss is decreasing sea level rise by close to 4%. And, the abyss is in negative energy balance, while the upper ocean is in positive energy balance.
But look at the numbers. –0.13 plus or minus 0.72. There is actually no way to say that the abyssal ocean is contributing negatively to sea level rise. Zero (or small positive numbers) are well within the range of statistical probability. For energy imbalance, –0.08 plus or minus 0.43. Again, zero and small positive numbers are well within the statistical range for this value.
But for some reason we see various individuals, including sadly at least one climate scientist (Judith Curry: “Evidence of deep ocean cooling?“), but mostly anti-science climate trolls, crowing that the “deep ocean” is cooling therefore we are not experiencing global warming. However, the truth is that the total amount of heat that is going into the ocean, instead of the atmosphere or other places, was thought to be large, is still known to be large, and in fact is larger than we were originally thinking (from these papers and several others that have come out recently). And, the contribution of the abyss ocean to both sea level rise and energy imbalance is statistically nil. It might be negative, it might be positive, but it is tiny either way. The deep ocean, on the other hand, is in strong positive energy balance.
________________
1The terminology used in this discussion and some of the research papers (or reporting thereof) is a bit confused. Climate scientists have repeatedly said that heat is going into the “deep ocean” and this paper seems to say it is not. But it is. It is a matter of terminology. This is a source of confusion sometimes exploited by climate science denialists. A good way to define these terms is as follows:
Shallow ocean = 0-700 meters
Deep ocean = 700-2,000 meters
Abyss = > 2,000 meters
Hansen, J., Mki. Sato, P. Kharecha, and K. von Schuckmann, 2011: Earth’s energy imbalance and implications. Atmos. Chem. Phys., 11, 13421–13449, doi:10.5194/acp–11–13421–2011.
Llovel, W. J. K. Willis, F. W. Landerer, I. Fukumori. 2014. Deep-ocean contribution to sea level and energy budget not detectable over the past decade. Nature Climate Change, 5 October.
The thermal expansion coefficient of water is very temperature dependent; warm water expands a lot more than cold for a given heat input, so this is very worrying and a double whammy so to speak. It's interesting that global sea level rise has been nearly constant at c 3mm per year for the past few decades. I worry that there will be a rapid acceleration sometime soon
Are the relative volumes of the above depth categories know?
I’ve recently completed a pretty serious blog post dealing with climate change, with reference especially to the new NASA report on the lack of deep sea warming discussed by Laden, which as I see it, could make a huge difference to the debate. And no, I’m not a “denier,” but a card carrying lifelong Democrat, liberal to the gills. I’d appreciate feedback from anyone reading here in the form of comments, positive or negative. http://amoleintheground.blogspot.com/2014/10/common-sense-on-climate-change.html
victorag - The uncertainty range in the Llovel paper is many times larger than their estimate, but from a physically-based perspective one has to wonder why, if the abyssal cooling trend is correct, geothermal warming in the abyssal ocean appears to stopped stone-cold dead. We would expect geothermal activity to be still be going on down in the very deepest parts of the ocean, as it's a component of the thermohaline circulation.
As for continued global warming (aka the pause):
And the remaining 1% which is warming the atmosphere:
The RSS satellite data appears to be the odd one out. We'll need more data to say for sure, but it looks like the atmosphere is still warming too, albeit at a slower rate than the previous two decades. SkS has a three-tiered rebuttal to this myth in the works.
That the climate models are remarkably close to the observed temperature trend over the recent decade, taking into consideration all relevant factors, is an explicit demonstration that we can get this temporary surface temperature slowdown even when the Earth's climate sensitivity is around 3°C per doubling of CO2.
As for climate policy, one needs to consider the ecological and agricultural impacts of warming and ocean acidification if one expects to be taken seriously.
Smith@2
From the Llovel paper:
I have some questions about the numbers that Greg Laden gave. He wrote [with my modifications for ascii text]:
"One of the key numbers is the energy imbalance where the ocean absorbs extra AGW produced heat. Energy imbalance is measured in terms of Watts per m^2. The present study yields a value of 0.72. A previous study reported 0.54. Other estimates have varied in this range. Llovel et al point out, however, that these differences may be due to differences in the ocean depth considered in each study and the time periods covered."
Before I ask these questions, I would like to quote what one named "BBD" said in the comments toward the end of the comments section at October 8, 2014 at 6:58 PM
http://blog.hotwhopper.com/2014/10/a-lot-more-heat-is-found-in-ocean.html?showComment=1412755125544#c8146988397082631758
for the article "A lot more heat is found in the ocean" posted Wed Oct 8 20154 at the site HotWhopper:
http://blog.hotwhopper.com/2014/10/a-lot-more-heat-is-found-in-ocean.html
Here is the quote from BBD's comment [with my modifications for ascii text]:
"...Nobody has claimed all of the extra heat has gone there although some of it obviously has. I hope you take the point, sloppy language here is encouraging sloppy thinking.
From Llovel14:
Therefore, we estimate the heat uptake by the upper 2,000m of the global ocean to be 0.72 +- 0.1 W m^2. Our estimate is slightly larger than the recently reported estimate of 0.54 +- 0.1 W m^2 for the upper 1,500 m layer computed over 2005-2010 and the estimate of 0.56 W m^2 for the 0-1,800 m layer over 2004-2011
The planetary energy imbalance is ~0.6 - 0.7W/m^2, so the OHU estimate in L14 would seem to account for all of it."
The first paragraph above seems to be a direct quote from the Llovel paper itself. The context of this comment above by BBD seems to be on Trenberth's "missing heat" that those who reject mainstream climate science say is still missing. But BBD seems to say that via the given numbers, seemingly directly from the Llovel paper itself, this paper more than closes the gap and so therefore there is no more "missing heat" problem.
My questions are mainly to those who have access to the paper itself:
Are Laden and BBD using the term "energy balance" differently or applied to different things? BBD says, "The planetary energy imbalance is ~0.6 - 0.7W/m^2" while Laden says, "Energy imbalance is measured in terms of Watts per m^2. The present study yields a value of 0.72." But the Llovel paper according to the above quote from the paper uses the term "heat uptake" for the 0.72 +- 0.1 W m^2 measure.
Also, BBD seems to say that since 0.72 is greater than the range ~0.6-0.7, Trenberth's "missing heat" is more than covered by the 0.72 +- 0.1 W m^2 heat uptake (as the paper puts it). Is BBD saying this and if so, is BBD right - does the Llovel paper do this and say this in this quote above from that paper?
Please, would someone with the requisite knowledge and access to the paper answer these questions and clear all this up? (And please feel free to include mathematically oriented information.)
[PS] Fixed links
Okay, the "deep" ocean and the "abyssal" ocean are the same thing--but is the "deeep" ocean even deeeper than that? :)
Rob Painting @ 4
Looking forward to the three-tiered rebuttal.
#1 StBarnabas:
Good point!
I used this water density calculator to create the graph below showing how one kilogram of sea water expands with temperature. Note that the expansion per °C of warming doubles from 0°C to 5°C and nearly doubles again from 5°C to 15°C.
I wonder if the nearly constant sea level rise the last two decades despite increased melting of land ice can partly be explained by this difference in thermal expansion. If some of the ocean heat uptake during the last 20 years has shifted from the shallow and warm parts to the deeper and colder parts this would reduce the total thermal expansion even if the total heat flux into the oceans remained the same. (not all Joules of OHC is equal!) This reduced thermal expansion may have offset the increased sea level rise from melting ice sheets, but that situation won’t last. Either changes in the ocean circulation will lead to more upper ocean warming again, or increased melting of land ice will overwhelm the reduced thermal expansion and accelerate the sea level rise.
I disagree with the analytical logic of the statement in the posting "...with heat that might otherwise add to the atmosphere seemingly entering the ocean on a regular basis...". In this disagreement I'm assuming that the statement relates to GMST fluctutations being due to the oceans absorbing heat (else my disagreement might be void). It is not that heat enters the oceans which could reduce an increase in GMST, it is that colder deeper water must rise to the surface if warmer surface waters are being pushed downwards. It is this colder deeper water reaching the surface waters that restrains an increase in GMST to less than that naively expected from solar or atmosperic changes just as it is warmer water rising to the surface (such as, obviously, El Nino) that boosts an increase in GMST to more than that naively expected from solar or atmosperic changes. It is not that "heat that might otherwise add to the atmosphere" gets added instead to the oceans. Rather It is that the temperature of the oceans varies between -0.2 degrees and +5 degrees (apart from a trivial ~3% in the tropics and sub-tropics sitting on top in a tiny pool) but GMST is +14.6 degrees. To make this important point (this isn't semantics) another way, it requires only 12 zettajoules to raise GMST by 1 full degree including land to a depth of 6m and ocean to a depth of 1m (which I'm arbitrarily considering the "surface") so it is not sensible to state or imply that 150 zettajoules of heat added to oceans the last 10 years (for example) was "instead of" GMST going up an additional 0.3 degrees or some such over that 10 years. It's quite simply additional cold water from that vast reservoir of coldness getting to the surface. Obviously, a corollary to that is that the additional cold water from depth getting to the surface will cause it to warm by SWR, increasing OHC.
grindupBaker @9, I am not sure I understand your point. The slow down in warming in GMST has resulted in a trend about half of the expected, and hence an increase in temperature over the purported 15 years of the slowdown of 0.15 C. By your estimates, that represents 1.8 Zettajoules. That represents 1.2% of the approximately 150 Zettajoule rise in ocean heat content over the same time. Note that there was also a 150 Zettajoule rise in ocean heat content in the 15 years prior to that when GMST rose as expected. The point then, is that the difference in the rate if rise in GMST between the two periods is:
1) A result of additional warming of the oceans; but that
2) That additional warming is so small a component of the total warming as to be negligible.
In short, they are saying the additional heat is going into the ocean, not that all of the increase in OHC is due to surface heat going into the ocean instead.
Further, and with respect to the cold water, if the cold water at the surface cools the surface temperatures, then it must do so (at least in part) by being warmed by the surface. More directly, the colder the surface water, the harder it is for heat to escape from that water given a warm atmosphere and hence more of the incoming heat is retained in the ocean. Consequently, I do not see how your description differs from that which you are criticizing.
Smith@2 & Andy Skuce@5 NOAA has 4,267 metres average ocean depth and Woods Hole Oceonographic has 3,682 metres. I don't know why the huge range (Woods Hole inclusion / NOAA exclusion of some shallow seas maybe ?). If "average ocean depth" is based on its volume and the area covered at the surface including continental shelves then if it's 4,267 metres then above 700 m and above 2,000 m cannot possibly in any practical way be more than 16.4% and 46.9% respectively (they can be less). If it's 3,682 metres then above 700 m and above 2,000 m cannot possibly in any practical way be more than 19.0% and 54.3% respectively. I infer that the "20%" and possibly the "50%" appear to be approximations not within 1% of actual.