Guest Post by Bob Tisdale
The KNMI Climate Explorer has added a number of datasets to their Monthly observations webpage, where users select desired data based on global coordinates. (Many thanks to Dr. Geert Jan van Oldenborgh of KNMI.) The new datasets include, under the heading of Ocean mean temperature, the National Oceanographic Data Center (NODC) Vertically Averaged Temperature Anomaly data of the global oceans. The data are supported by the Levitus et al. (2009) paper World Ocean Heat Content and Thermosteric Sea Level change (0-2000 m),1955-2010. Basically, the NODC Vertically Averaged Temperature data are the temperature component of their Ocean Heat Content data. KNMI has added the vertically averaged temperature anomaly data for the depth ranges of 0-100 meters (1955 to present), 0-700 meters (1955 to present) and 0-2000 meters (2005 to present).
Note: KNMI has also added to their Climate Explorer the Japanese Meteorological Agency (JMA) Ocean Heat Content data for the depths of 0-700 meters. We’ll compare them to the NODC’s data in a future post.
In this post we’ll take a quick look at the vertically averaged temperature anomaly data for 0-2000 meters. That’s the period when subsurface temperature measurements are dominated by ARGO floats, providing a reasonably complete view of ocean temperatures to depths of 0-2000 meters. We’ll examine the other depth ranges in future posts.
PRELIMINARY NOTE
The NODC present their data on a quarterly basis. KNMI has converted it to monthly data by assigning the quarterly values to each of the respective months. As a result, the monthly data take on a step-like appearance, with lots of little plateaus. We could smooth the data with 3-month filters to eliminate those tiny steps, but in this post it would serve little value because we’re looking primarily at warming (and cooling) rates.
GLOBAL AND HEMISPHERIC WARMING RATES
Figure 1 presents the NODC vertically averaged temperatures anomalies, for the depths of 0-2000 meters, for the 1st quarter of 2005 through the 2nd quarter of 2014. Since the 1st quarter of 2005, the oceans of the Southern Hemisphere for the depths of 0-2000 meters have warmed at a rate of about +0.05 deg C/decade, based on the linear trend. On the other hand, the linear trend for the Northern Hemisphere oceans is considerably less, warming at a rate that’s about 10% of the Southern Hemisphere rate. That’s quite a difference in a world where manmade greenhouse gases are supposedly responsible for the warming and where manmade greenhouse gases are said to be well mixed.
Figure 1
OCEAN BASIN COMPARISONS
I’ve divided the ocean basins into subsets using the same coordinates I use for the sea surface temperature updates:
- Arctic (65N-90N)
- North Atlantic (0-70N, 80W-0)
- North Pacific (0-65N, 100E-90W)
- Indian (60S-30N, 20E-120E)
- South Atlantic (60S-0, 70W-20E)
- South Pacific (60S-0, 120E-70W)
- Southern (90S-60S)
The cooling rate of the Arctic Ocean is so out of proportion to the other ocean basins that I’ve shown it separately in Figure 2. One would want to attribute the spike in 2007 to the seasonal sea ice loss that year. Curiously, similar spikes do not appear during 2012 and 2013 when there were comparable or greater seasonal losses.
Figure 2
I’ve excluded the Arctic Ocean from Figure 3. As shown, there is a wide range of warming rates for the rest of the ocean basins to depths of 2000 meters. The Indian Ocean has had the greatest warming during the ARGO era, more than twice the warming rates of the South Atlantic and South Pacific, and about 3 times higher than the rate of the North Atlantic. Yet, for the depths of 0-2000 meters, the North Pacific and the Southern Ocean surrounding Antarctica show basically no warming over the past decade.
Figure 3
WHICH LATITUDES WARMED AND WHICH ONES COOLED
Figure 4 presents the warming and cooling rates (deg C/decade) of the global oceans on a latitudinal-average (zonal-means) basis. The vertical axis (y-axis) is scaled in deg C/decade. The horizontal axis (x-axis) is scaled in degrees latitude, with the South Pole on the left at -90 (or 90S), with the North Pole on right at +90 (or 90N) and with the equator at the center at zero degrees latitude.
Figure 4
During the ARGO era, the greatest warming to depths of 2000 meters occurred at the lower mid-latitudes of the Northern Hemisphere and the mid-latitudes of the Southern Hemisphere. The greatest cooling occurred in the mid-to-high latitudes of the Northern Hemisphere, with a comparatively slight cooling in the lower latitudes of the Northern Hemisphere and in the Southern Ocean surrounding Antarctica. Note also that the cooling in the Northern Hemisphere encompasses more than the Arctic.
Regional ocean cooling to depths of 0-2000 meters is difficult to reconcile in a world where greenhouse gases are supposed to dominate climate. If natural factors can cause the oceans to cool, they can also contribute to the warming.
CLOSING COMMENT
For a couple of years, we’ve been illustrating and discussing how horrendously the latest and greatest climate models simulate the surface temperatures of the global oceans. See the most recent posts:
- Maybe the IPCC’s Modelers Should Try to Simulate Earth’s Oceans, and
- IPCC Still Delusional about Carbon Dioxide
Some readers might suspect that if the models cannot simulate the surface temperatures of the oceans, they will prove no better when attempting to simulate the warming and cooling to depths of 2000 meters. Unfortunately, the climate model outputs for subsurface ocean temperatures are not presented in formats so that they can be easily added to the KNMI Climate Explorer. And it’s unlikely that the climate science community will present those model failings.
More averaging of unrelated temperatures. Junk.
How are the temperatures unrelated, Jeff Alberts? For the most part, they’re using the same measurement technology for a single metric.
Bob, there is no single temperature for the Earth or all the oceans. The measurement technology is irrelevant. You can’t take a measurement in one place, and average it with other measurements taken thousands of miles away and end up with anything meaningful.
Using a single measurement technology consistently is a good thing. Now if only the measurement methodology was robust enough to provide value. I am sure we have instruments that can measure in hundredths of a degree, I am skeptical we have a methodology for getting the temperature of all regions of all the oceans of all the world to a hundredths of a degree. This to me is not feasible at this time. Put another way, what is the margin of error?
Numbers are infatuating, they are infatuating in and off themselves, endless facinating possibilities and relationships. So there is a ton of numerical data from tons of temperature data points. Having a lot of data points is easy, manipulating those data points is fun, validating the efficacy of those data points is much more difficult.
Alx, the point is not whether technology exists to take temperatures of all points on the planet at once, the point is that averaging (or taking the mean, median, whatever) all of them together is physically meaningless. Each point reading is an entity unto itself. Averaging it with another point reading on the other side of the planet is not physically valid. Intensive properties.
Alberts,
Averaging the surface air temperatures across the globe into one single value in attempting to somehow track the energy inside the system, is indeed a pointless endeavour, because you would have to take into account the wildly fluctuating heat capacity of air. Raising the temperature of dry, cold Arctic air by one degree requires a lot less energy than doing the same to humid, warm tropical air. In other words, all you need to do, theoretically, to raise such an averaged global temperature value for the Earth is to move more of the tropical energy to the extratropics than before. No need to increase the total amount of energy at all.
When it comes to OHC, however, it’s a bit different. Since the heat capacity of liquid water doesn’t change appreciably within Earth’s temperature range, averaging temperature measurements across the globe to account for the total energy content does make sense – in theory at least. That is, if the measurements and the averaging process were flawless.
Jeff, you are quite wrong. The temps with latitude pattern is highly informative of poleward heat transport which as some have commented could be a real concern.
All climate is basically poleward heat transport. Its implications can be quite big.
That last chart strongly suggests that the Sun is still warming the middle of the Planet, but that warmth can no longer keep up with the cooling taking place near the poles.
That does not bode well for world.
The Red & Blue areas look approximately equal to me
Except that you need to squeeze the x-axis (or something) near the poles to remind yourself that the surface area per deg latitude decreases toward the poles. However what really counts is the volume of water at each latitude, I suppose, and that would be even more difficult to envision from these graphs.
Not sure? Zero to 2000 meters is a huge area. If all the warming is in the first 200 meters, then yes the recent Sun may have been a factor, however, if the warming is manifesting at greater depths then it could be circulating warmth from centuries ago. Or the near surface waters could be warming from depth.
And of course these numbers need error bars.
That’s the way it strikes me, too. I’d better keep scrolling down, and catch up with comment. Brett Keane
Great post Bob. Thanks!
Who would of thunk it? The oceans have warm spots and cold spots in them just like the lakes I swam in as a kid!
Well, thee are warming spots, and cooling spots, depending on the error bars??
What ever happened to polar amplification?
If the trends by latitude were the opposite of what is shown, it would undoubtedly be heralded as the “fingerprint” of AGW.
In 30 years, it seems plausible to me that Figure 4 could be flipped on the x axis. Which is what it may have been 1980 to 2004. Thus the fingerprint of man’s CO2 that got the whole AGW alarmism going in the first place.
What are the error ranges for these temperature measurements? I’m having a great deal of trouble accepting that the ARGO floats measure temperature changes to within .03C/decade.
Exactly. This looks suspiciously like noise.
My thought exactly. What do the one and two sigma bars look like? 0.001 degrees resolution? I have an Agilent multiplexer that can read that, but for a whole ocean?
With data that varies so much combined with a low trend line (slope), a person should also do the confidence limits of the slope, i.e,, the degrees C / decade change.
There are only 3000 argo floats in use. That’s not enough data to tell the temperature of the English channel, let alone the world’s oceans! I’ve said it before but if argo floats were a fish they’d be an endangered species.
Forget error bars; we don’t have any data!
Even if the ARGO’s can measure temp changes to 0.03C / decade world wide, breaking the world up into 34 five degree latitude bands must increase the uncertainty within each band by a factor of about 6 (sqrt(34)).
I have another objection to Figure 4: The spacing of the X-Axis is equi-latitude. But this is an Area chart, so visually one is looking at the areas.
As a better approximation , the X-Axis should be linear with cos(latitude).
The interval 65-80 N does not cover as much of the sphere (3.93%) as 0-5 N (4.36%) When you take into account the proportion of each band covered by water and not land, 65-80N is smaller yet. and very different than 65-80 S.
I present data that are available, as they are available. If you’d like to further analyze them, follow the links.
I could not figure out how to get the data, so as an approximation I estimated the data from the first figure for the northern hemisphere which has a slope of 0.005 C / decade. My resesults for the best fit were close at 0.00459 C / decade. However the correlation coefficient of the data to the fit equation was only, 0.0129, the standard deviation of the difference of the data and the estimated value from the slope is 0.988 C / decade and the 90% confidence interval for the slope was very large from -0.096 to +0.105 C / decade.
The equation turned out to be -9.212 + 0.0004591Y. (Y = year).
This indicates that data with so much variation or noise makes it hard to infer very much. It seems like the data for the other hemispheres have a similar problem.
Is it possible to track the energy flux (numerically) from the Tropics and the Subtropics to the Northern Latitudes?
I suspect you’d want the data in absolute form for that. Unfortunately, the NODC only presents their data as anomalies.
I found some errors in my estimates with the small numbers from graphs in the first figure. I will redo later, The standard error and the range are much less than I showed. I will have to redo. Sorry.
I corrected the data I used in a previous reply from the Northern Hemisphere from the first Figure in nthis paper. I used one value for the “Plateau” which represents 3 months of constant data with a single value. The data I used is shown below for years starting in 2005 in 0.25 year increments.
Using multi regression software I get the best fit slope of 0.0009 C / decade with a 90% confidence interval range of -0.0091 to +0.011. If the rest of the regions have about the same amount of variability, this means we can use their estimated slope +- 0.01 C / decade for an approximate 90% confidence interval.
It is not clear why I get a much lower value than the author did of about 0.005 C / decade.
The data I used is for years starting in 2005 in 0.25 year increments:
-0.0050 0.0070 -0.0150 -0.0060 0 0.0090 0.0030 0.0090
0.0100 0 0.0200 -0.0100 0.0110 0.0070 0 -0.0050
-0.0190 -0.0100 -0.0080 -0.0030 0 0.0100 -0.0060 0.0090
0 -0.0160 0 -0.0095 -0.0085 -0.0090 -0.0140 0
0.0100 0.0050 0.0200 0.0180 0 -0.0030
Bob, the last chart is very interesting: does it make any difference if you calculate Jan-Mar 05 to Jan-Mar 14? In other words, are there any residual seasonal effects to be concerned with?
RERT, the data are presented as anomalies. One would think there would be very little seasonal impact. But there is a minor difference in the trends if I delete the last quarter. The curves are very similar but the cooling rates are impacted. I’m not sure if it’s a seasonal component or the fact that we’re looking at such a short term.
Thanks Bob, looks as good as the same to me, which is comforting! R.
To my simple mind, Figure 4 demonstrates how the the great Laurentide ice sheets get cold NH summers they would need to accumulate year-after-year, decade-after-decade, century-after-century.
It is not scientifically sound to believe that we can really measure the average temperature of the seas to an accuracy of 100ths of a degree Kelvin or derive meaningful trends from a ten year data set.
While graphing such a data set may be interesting, basing any conclusions, predictions or projections on the data is premature. One might rightfully find some areas of interest to do research on from this data, but that is all.
Nor can any sensible deductions be made when each ARGO float is effectively measuring such an enormous volume of water. Some time ago I calculated that each float is responsible for measuring the equivalent of, IIRC, some 4 BILLION Olympic-sized swimming pools. To argue that an accuracy of even 1 degree is laughable.
If Argo floats float, then what keeps them evenly distributed? In addition to your good point about the vast volume each float represents, doesn’t random longitudinal or latitudinal travel introduce additional ambiguity?
Good post Bob. The global average surface temperature is dominated by 2 hots spots. The Antarctic Peninsula and the Arctic.
http://landscapesandcycles.net/image/95498578.png
There are 2 competing explantions.
Alarmists suggest this is Arctic Amplification as CO2 melts ice and heat accumulates as the dark oceans absorb more heat. Natural climate cycles suggest changing winds from the Arctic and Antarctic Oscillations removed insulating ice cover in both regions and the higher temperatures resulted because more heat ventilated from the oceans. Winds blowing across open waters that dominate along Eurasia, extend that warming into northern Eurasia.
The rapid rate of ocean cooling at both those hot spots would is strong evidence the Arctic Amplification is not happening, and the natural climate change explanation ventilating is far more consistent with reality.
Steele,
There is no actual Antarctic Peninsula hotspot. That map is purely an artefact of the start and end dates chosen. There has been no warming on or around the Antarctic Peninsula over the last 30-35 years. That’s what the data tells us. Which means that, whatever warming there was, it ended more than a climatological period ago. So much for the continued and strengthening effect of the steady increase of atmospheric CO2.
http://okulaer.wordpress.com/2014/09/13/a-case-to-prove-a-point-the-claims-of-major-ongoing-antarctic-peninsula-warming/
Much of the Antarctica surface data is in-filled by a fellow named Gavin. RSS, the troposphere satellite data does not reflect the in-filled areas, and , in general, the surface T is reflected faithfully in the troposphere, except where GISS in-fills. http://stevengoddard.wordpress.com/2014/10/15/polar-amplification-6/ Check out the center of Africa as well; no data for the central region, in-filled from stations all around, yet RSS shows cool.
Hey Bob, what do you think of this Unisys plot of the oceans? I have been looking at this on a daily basis for many years and this is the most cooling I have seen in the northern oceans…. Your thought?
http://weather.unisys.com/surface/sst_anom_new.gif
Their color scaling is very confusing, because they use the same colors for different values. Yellow appears 4 times on that scale. That’s why I no longer look at UNISYS.
I’ve been looking at the SST anomaly maps from CMC for a couple of months. Not as confusing.
http://weather.gc.ca/saisons/image_e.html?img=2014101700_054_G6_global_I_SEASON_tm@lg@sd_000&title=daily&bc=sea
Animations are here:
http://weather.gc.ca/saisons/animateweb_e.html?imagetype=images_loop&imagename=……..00_054_G6_global_I_SEASON_tm@lg@sd_000.png&nbimages=1&clf=1&bc=sea
Cheers
Oops. Here’s a hyperlink to the CMC SSTa animation webpage.
Hi, Bob, I have to disagree about the colours being confusing. In either a negative or a positive direction there is no reuse of colours until 10degC or so, which like hurricanes in Hertford or Hampshire hardly ever happen. And the eye easily picks up which areas are cool and which are warm.
Mind you I prefer their older product http://weather.unisys.com/surface/sst_anom.gif .
Like Dennis, to me it seems to me UNPRECEDENTED (I already showed my family).
Rich.
See – owe to Rich, thanks for pointing that out. Here’s the other problem. Referring to the color scale on the map above, it is impossible to tell what color represents a zero temperature anomaly. Do the blues and greens extend into positive values, as they had in the past?
Bob,
I am using http://www.tropicaltidbits.com/analysis/cdas-sflux_ssta7diff_global_1.png ,the colour anomoly temp gradations are finer (0.2°).
I think the color scheme would be just fine if you chopped it off at plus 6 and minus 6. The rest is totally superfluous for anomalies.
The Ice around Antarctica looks unusually warm.
Bob – a few months ago there appeared on Unisys to be unusually warm SSTs. You commented on this as if it was a real phenomenon. Is it only cold SSTs that are confusing?
In the north Pacific and Atlantic, I meant to add.
While it is often written “0-2000 meters”, I feel it is important to note that this is the upper 2000 meters of the oceans. The “deep ocean” temperatures are not, as far as I understand, being measured anywhere near the extent of the ARGO system.
Our study is motivated by the desire to analyze one deep current long enough to detect slow changes with time of the deep ocean circulation. This includes interannual time scales and shorter termvariab ility down to semidiurnal frequencies. The short-termvariab ility on tidal frequencies is presented to provide data for model verification. The best candidates for finding significant mean flows with current meters are generally felt to be either along the Western boundary of the deep oceans (Warren, 1981), or in deep gaps between abyssal basins (Whitehead, 1998). A total of 20 long-time measurements now exist below 2000m (Hogg, 2001).
From – Variability of Antarctic bottomwater flow into the North Atlantic – Richard Limeburner, J.A. Whitehead, Claudia Cenedese 2005
Finding –
We conclude that this current is an extremely steady long-term flow. For climate changes over decades, deep ocean flows through other gaps might thus show small but very important trends if small changes in their speed can be measured over long times with high precision. Conventional instruments meet the requirements. The most important future task is to locate the pertinent deep gaps.
Is it possible to get vertically averaged temperature anomalies for 700-2000 meters?
Not in an easy-to-use form. You could weight the two datasets, though, and subtract one from the other. See example here:
http://bobtisdale.wordpress.com/2013/07/04/rough-estimate-of-the-annual-changes-in-ocean-temperatures-from-700-to-2000-meters-based-on-nodc-data/
Excellent post more than answers my question. Thanks!
nutso fasst wrote, “Is it possible to get vertically averaged temperature anomalies for 700-2000 meters?”
I second that request.
Sea surface temperatures are very seasonal. Below 100 meters the seasonal temperature fluctuations are gone. Vertically averaging the seasonal top 100 meters with the 1900 meters below that doesn’t really tell us much about deep ocean temperature trends.
The IPCC says nearly all sea-level rise (SLR) due to thermal expansion occurs in the top 700 meters. That’s probably right. More specifically, out of 1.6 ± 0.5 mm/yr SLR that the IPCC attributes to thermal expansion over 1993–2003, they say all but 0.1 mm/yr was in the upper 700 meters. In fact, most of that is surely in the top 100 meters.
The thing is, thermal expansion of the upper ocean does not affect sea-level elsewhere. Rather, when the density of the water in the upper layer of the ocean changes, the water rises or falls in place, because gravity balances mass, not volume. This is most dramatically visible in the case of ice, but it’s just as true for liquid water.
http://www.sealevel.info/iceberg_1521870c_annotated.jpg
Thermal expansion increases water volume by a percentage of depth. So if you warmed up the top 10 meters of water by enough to get a 1% increase in volume (maybe by 5 °C, depending on temperature), the surface would rise 10 cm; or if you only warmed the top one meter of water that much the surface would rise by one cm.
But that type of “sea-level rise” is a local effect. It would be detected by satellite altimetry over the open ocean (if those measurements were reliable, which is questionable), but sea-level rise due to thermal expansion in the upper layer of the ocean does not cause lateral flows of water or significant coastal sea-level rise.
Contrast that to sea-level rise from melting grounded ice (glaciers, mountain snowpack, etc.). When grounded ice melts, and the meltwater finds its way to the sea, sea-level is affected everywhere. So that’s the only type of sea-level rise that matters for coastal planning.
Of course, ice and snow are always melting, and water is always freezing, and snow is always accumulating. Those processes are nearly balanced: in an average year, the Earth has a net loss of somewhere between 100 and 200 cubic miles of grounded ice (probably closer to 100 than 200), which finds its way into the oceans as meltwater and rain.
That sounds like a lot of water, but compared to the volume of the oceans, it’s not much. You’d have to add the meltwater from ~95 cubic miles of grounded ice (= 362 Gt = 395 km^3) into ~87 cubic miles of fresh water and add it to the oceans to raise globally averaged sea-level by just 1 mm. Ref:
http://sealevel.info/conversion_factors.html
daveburton,
Thanks for this explanation. You should add it to your sealevel.info page, which I think is excellent. I send it to everyone who hyperventilates about sea level rising 700,000 miles by 2100 AD and flooding western civilization.
The climate4you page has an excellent writeup on why the polar regions have enhanced greenhouse gas effects and the warming there is the “fingerprint” of what the added CO2 is doing. Yet this data doesn’t show much warming of the oceans at the poles at all, in fact a cooling. It seems contradictory that the N. Atlantic and N. Pacific data in Figure 3 show warming and no change, respectively, while the average for the northern latitudes in Figure 4 is obviously an overall cooling. Is that because the most volume of water in the northern hemisphere is in the areas where warming is occuring?
If net water were warming they’d get warmer in the middle and colder at the poles where they melt ice from the bottom. New ice can still grow in the cold air on top.
These results are really concerning. They seem to imply that overall transport of heat in the oceans is slowing. Either the currents themselves are slowing, or, the currents have less efficacy to transport heat for some unknown reason.
Paleoclimate data does show that when solar flux was reduced during the Little Ice Age, transport of warm water northward measured through the Florida Straits was reduced by 10%. This was linked to lower solar irradiance, the spinning down of the subtropical gyre and southward shift of ITCZ.
Read Lund et al (2006) Gulf Stream density structure and transport during the past millennium
They wrote:
“We also estimate that Little Ice Age volume transport was ten per cent weaker than today’s. The timing of
reduced flow is consistent with temperature minima in several paleoclimate records, implying that diminished oceanic heat transport may have contributed to Little Ice Age cooling in the North Atlantic. The interval of low flow also coincides with anomalously high Gulf Stream surface salinity, suggesting a tight linkage between the Atlantic Ocean circulation and hydrologic cycle during the past millennium.”
RE:
A C Osborn
October 17, 2014 at 7:37 am
That last chart strongly suggests that the Sun is still warming the middle of the Planet, but that warmth can no longer keep up with the cooling taking place near the poles.
That does not bode well for world.
=====================
I agree. These data are highly concerning.
Yes James it’s the Sun and the sun has gone quiet, so look for cooling in the near future.
Sun Spot gives an unsupported answer. We can calculate the amount of solar insolation being received into the oceans in the middle of the Planet for both clear sky and cloudy sky conditions. Thus we can also calculate the change based on cyclic TSI solar parameters under clear sky and cloudy sky conditions. We can then calculate the change in ocean T to the depth of absorption related to the various solar irradiance spectrums. Result? Because the middle band of ocean is such a vast entity, the change in T is below the capacity of ocean temperature sensors to measure accurately, consistently, and repeatedly to the level of this variation. The Sun is a less likely source of temperature trend. Our own much more highly variable planet is the more likely source of such trends.
I dont actually think that current solar trends can affect poleward heat transport which is by ocean currents with century/millenial timescales.
Bob, please correct me if I am wrong, but when we are talking about numbers so small, 0.03 degrees Centigrade per decade (top graph), aren’t we really talking about immaterial changes? If they are immaterial, why are we continuing to spend so much time and resources measuring them?
Because even small increases in the ocean temps take massive amounts of energy. Energy that will (if the science is right) eventually be reflected in the atmospheric temperature.
“eventually” > 1000 years? Lots of time to use up fossil fuels and perfect fusion power.
Energy that will (if the science is right) eventually be reflected in the atmospheric temperature.
Except for those Laws of Thermodynamics –
The first law of thermodynamics basically says that the energy that goes into a system cannot be lost along the way, but has to be used to do something … in this case, either change internal energy or perform work.
How much of the ‘Massive Amounts of Energy’ is used in generating the flow and friction losses of the ocean currents?
Second Law of Thermodynamics:
It is impossible for a process to have as its sole result the transfer of heat from a cooler body to a hotter one.
The supposed amount of energy equates to 0.06 C. Explain how the temperature gets hotter than that.
Even Bart & Homer Simpson obey the Laws of Thermodynamics.
Actually El Nino tells us the ocean temp has a very real and quick impact on Atmosphere. Take a look at 1998
Simon says, “Because even small increases in the ocean temps take massive amounts of energy.”
That’s very correct. But looking at it the other way, those “massive amounts of energy” have only raised the ocean temperatures a few hundredths of a deg C, which simply increases the” background state” by that same amount.
Simon says, “Energy that will (if the science is right) eventually be reflected in the atmospheric temperature.”
What scientific papers are you referring to? The heat absorbed by the oceans cannot be released to the atmosphere unless it rises to the surface. Then again, the oceans have only warmed by a few hundredths of degree.
Bob
“Then again, the oceans have only warmed by a few hundredths of degree.”
According to NOAA’s national climatic Data Center, the global ocean surface temperature for August 2014 was the warmest on record…. 1.17 degrees F (0.65 degree C) above the 20th century average. And the combined land and sea was also the warmest on record for the month of August at .75 C above the 20th century average.
Yes, but energy dissipated in the oceans, and released at the surface through a .30 degree increase over a century IF the trend continues for ten decades, and IF the measurements are accurate, still only translates into a relative atmospheric increase of .30 degrees. Nothing to worry about, and certainly beneficial.
CAGW cannot happen if the energy is going into the oceans.
Simon, RSS and UAH do not show the atmosphere nearly as warm as the GISS surface readings, which are departing from the satellites at a record rate.
0.03C/decade is 0.3C per century – about half the warming caused in the last century so it’s not insignificant. However, the question is how accurate are these ocean warming numbers.
Is this raw or “corrected” Argo data?
It’s from the NODC, Robert. It’s been corrected.
Then I don’t trust it. But I appreciate your work with it.
I too was curious about how the data had been corrected. A 2012 paper by Xue et al A Comparative Analysis of Upper-Ocean Heat Content Variability from an Ensemble of Operational Ocean Reanalyses had come to a consensus that the upper 300 meters have not warmed or cooled since 2003. The Wunsch and Heimbach 2014 paper had the upper 700 meters warming until 2009 and then a slight cooling from the peak, However your data presentation Bob suggests the upper oceans are still warming, except in the polar regions. Your thoughts?
Bob, are these anomalies based on raw data or adjusted data.
There have been a few posts over the last year or two suggesting that adjustments were being made to the ARGO data or that very cold readings were removed but not the very warm readings.
Secondly how much of the claimed warming is in surface waters compared to say 700 to 2000m ?
Hi Peter Foster. They’re the adjusted data. I provided a breakdown of the temperature rises at 0-700 meters and at 700-2000 meters in the following post:
http://bobtisdale.wordpress.com/2013/07/04/rough-estimate-of-the-annual-changes-in-ocean-temperatures-from-700-to-2000-meters-based-on-nodc-data/
Thanks Bob, the oceans appear to be quite stable (not surprisingly). I don’t see error bars but I would imagine they are in range of the fluctuations being observed when speaking of hundredths of a degree variation.
Thanks Bob, your posts are always informative.
I have been thinking recently about the ocean heat content data and how it relates to atmospheric temperature. Shouldn’t the data for these two always add up to give us a relatively best guess for Earth’s total heat content? I know that there are changes in the rate of energy that the Earth receives due to orbital variations and clouds, but there shouldn’t be any acute rates of change in total heat content without an obvious reason such as volcanic eruption, right?
In other words, if there is supposedly a large change in total ocean heat content, e.g. 2012.5-2013, then shouldn’t we see an instant change in in atmospheric temperature that coincides with it? I think it would be very interesting to see the two data compared to see how well they agree. If the ocean heat content and atmospheric heat content show a strong negative correlation then that would suggest the data are good, but if they didn’t wouldn’t that mean there are errors somewhere in the data? I think you would be the man for the job. Thoughts?
Robert W Turner: “In other words, if there is supposedly a large change in total ocean heat content, e.g. 2012.5-2013, then shouldn’t we see an instant change in in atmospheric temperature that coincides with it?”
Not necessarily. The rise in atmospheric temperature may have resulted from a release of heat from the oceans, which would have cooled the ocean basin where the release took place.
True. Yet the mean difference between the SST and the atmosphere float between a very small range, likely variable , as you say, depending on disparate ENSO phenomena. (This answers Simon’s concern on how the SST can have a relatively short term larger affect on the surface record.)
However, the mean ocean T will, in the long term, moderate whatever the surface is doing. Therefore energy going into an average ocean T is, in affect, dissipating the potential warming or cooling.
Thanks Bob, interesting, as usual.
Bob
Excellent article as usual. It crossed my mind that deploying ARGO in the Arctic probably presents no few practical problems, including the fact that the sea surface they are supposed to pop up to is quite likely to be solid and not very penetrable . Satellite coverage might also be an issue as not all satellites have high enough inclination orbits to see very high latitudes. A little Googling leads me to believe that there is an Argo like program for the Arctic (POPS?), but not when (or if) the units were deployed or how many there are.
Anyway, I was wondering if the somewhat anomalous Arctic Ocean results might not be a result of poor coverage. I thought I’d ask if you know.
Aren’t the northern latitudes cooling data due to the heat lost from the melting of the ice?
If so, then a better average would be derived from excluding the northern latitude data.
Then the higher average warming of the balance would be closer to what would be expected??
There is not increasing ice loss.
Bob Tisdale,
I learn so much from your posts. And I am becoming such an expert in my know-nothing friends’ eyes. 😉
“KNMI has added the vertically averaged temperature anomaly data for the depth ranges of 0-100 meters (1955 to present), 0-700 meters (1955 to present) and 0-2000 meters (2005 to present).”
I can see it coming like a freight train down the track. A new study, peer view, with a spaghetti graph, claiming “We have found the missing ocean heat“.
Sometime before its discovery in 1999, the Gakkel Ridge – a spreading center on the floor of the eastern Arctic Ocean about 1800 km long running from the NE corner of Greenland almost to the continental shelf off the northern edge of Siberia – released enough lava to blanket at least 10 sq km of the floor of the Arctic Ocean.
That much lava represents a substantial input of heat into the Arctic Given the limited exchange of deep waters between the Arctic Ocean and any other large body of water, it should still be cooling down from that event. This may contribute to the Arctic cooling rate depicted in Figure 2.
I have not yet seen a quantitative or even semi-quantitative accounting for heat introduced to the deep oceans by the vulcanism which seems to be occurring there continually in a variety of places and intermittently in many more. http://en.wikipedia.org/wiki/Supercritical_fluid#Hydrothermal_circulation
Bob Tisdale:
Your figure 2 looks like a step change. It has an upward spike as did the 1998 El Nino temperature spike.
The zonal data is consistent with an increase in polar deep water formation with an increase in near equatorial upwelling promoting more low cloud formation. (Especially if this data has been adjusted warmer by say 0.05 deg. C.)
Bruce, would you not need to know where, at what depth, the cooling and warming is happening? (By the way, over ten years these floats must move a lot. Has the mean latitude changed? Has the mean depth changed, IE, more or less Argo floats in deeper, or shallower then 2000 meter water?
David, yes, ideally I would love to see zonal data for only the top 200-300 meters where solar energy input is greatest- but I will take whatever I can get!
Bruce, curiously look at figure 8 of Tim Ball’s post, “A Simple Truth; Computer Climate Models Cannot Work”. The precipitation anomaly, appears to be the opposite of the ocean changes. Do not know what it means or over what period that is from. (As always, error bars?)
Bob has some posts that give some of this, broken out to different depths.
David I’m sorry, I’m not sure what he is trying to show?
If you check the errors in the measurement of ocean temperatures, you will find there is nothing to talk about; all changes are within experimental error.
I think it would be interesting to correlate at least some of the readings (down to about 200 meters or so) with major high pressure systems that sustain themselves over a period of time. Details left to the scientists, of course.
Actually, IS there any data on where the major highs have positioned themselves around the planet over the past few decades?
Still haven’t seen, as AC Osborne wrote, why cooling both ends doesn’t mean a deficit of energy?
This thee-hundredths of a degree of warming of the ocean’s depths was accomplished (not measured) only after the adjustments to Argo data because of alleged poor calibration by manufacturers of Argo instruments. It was tough to get Argo to cough up what they wanted to hear, but if the data doesn’t say what you want, torture it until it does.
Reblogged this on Globalcooler's Weblog and commented:
Always a pleasure to see Bob Tisdale doing what he does best.