Part 2: How Sea Ice Modulates Ventilation of Stored Ocean Heat and Warms the Arctic

Jim Steele

Examines the three major factors affecting sea ice extent, ventilation of subsurface heat and Arctic temperatures and explains why warming air due to a cooling ventilating Arctic ocean should never be averaged with other temperatures to create a global temperature statistic intended to indicate earth’s heat storage.

Transcript of video available at:

Jim Steele is Director emeritus of San Francisco State University’s Sierra Nevada Field Campus, authored Landscapes and Cycles: An Environmentalist’s Journey to Climate Skepticism, and proud member of the CO2 Coalition.

4.8 9 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
October 20, 2021 10:21 pm

Unfinished sentence below the 1993 Nature paper chart:

The effect of sea ice on Arctic temperature is more dramatically exhibited by the 

Reply to  Sunsettommy
October 20, 2021 11:15 pm


? 😉

Jim Steele
Reply to  Sunsettommy
October 20, 2021 11:29 pm

Thanks sunseet,

it should read and does now “exhibited by the Dansgaard-Oeschger events” the transcriber butchered the spelling so was deleted but I was distracted and neglected to replace it with the correct spelling.

Bryan A
Reply to  Jim Steele
October 21, 2021 1:15 pm

I liked PAUSE better.

Joel O'Bryan
October 20, 2021 10:22 pm

Just to be a skeptic party pooper, anomalies must be averaged. Not actual temperatures. No one in the climate scam averages thousands of actual temperatures to then find the anomalies. Even they know that makes zero physical sense to average temperatures from different surface/masses.

But even averaging anomalies allows for a titanic size iceberg hole for mischief in the “homogenization” and infilling between dwindling rural stations to craft a narrative on temps.
Missing a few rural stations??? No Problem, just interpolate between tow urban stations. Cooling problem solved.

What the homogenizers and infilling liars can’t obliterate though is the recorded data that the 1930’s were the highest recorded station temps across the NH by far.

Reply to  Joel O'Bryan
October 20, 2021 11:17 pm

An average global temperature or average global anomaly doesn’t make sense

Temperatures are local on a daily, weekly, seasonal basis not global

Stephen Lindsay-Yule
Reply to  Joel O'Bryan
October 21, 2021 12:18 am

Climate change is about the energy balance of the earth. Temperature is proportional to the amount of light hitting a square meter per second. Averaging temperature over different surfaces tells us how much of that light is being absorbed, Therefore knowing if the balance is positive or negative. Your second paragraph speaks why anomalies can distort earths temperature. As past data can be altered however scientifically accepted it is by those changing the data.

Jim Gorman
Reply to  Stephen Lindsay-Yule
October 21, 2021 4:32 am

Temperature is not proportional due to different specific heats and masses of the various components. As an example use H2O whose latent heat removes energy from the system and hides it thereby not raising temperature.

Reply to  Jim Gorman
October 21, 2021 5:13 am


Also, a common meaning of ‘proportional’ is:

having the same or a constant ratio

Suppose we have a flat piece of metal in orbit. It’s at an angle to the sun such that the equilibrium between its absorbed and radiated energy produces a surface temperature of 100k.

Now let it rotate such that it receives twice the energy. Because the energy it radiates is proportional to T^4, its surface temperature will be 119k not 200k.

So, for a couple of reasons, surface temperature is not “proportional to the amount of light hitting a square meter per second.” In fact, in the arctic in the summer, where the melting ice is a powerful phase change, and there’s a huge influx of energy from the atmosphere and ocean, the temperature has only a small relationship with incoming solar radiation.

Crispin Pemberton-Pigott
Reply to  Stephen Lindsay-Yule
October 21, 2021 6:12 am

Stephen, unfortunately your explanation is too simplified to communicate the concepts.

Climate change is about the enthalpy of the system (total energy content). Temperature is an indicator, but not of the enthalpy. Temperature times thermal mass IS an indicator of energy content.

If the temperature drops and the heat was transferred to a “heavier” medium, the enthalpy may not have changed at all. So it is fair to state that temperature is an “indicator” but not a “quantifier”. Enthalpy is all about quantities. Temperature is a quality.

The MSM misrepresentation most in evidence is the claim that the temperature tells us specifically about the heat contained by the climate system. It does not. However it is very easy to deliberately miscommunicate what is happening.

You will never see the word “enthalpy” in an MSM piece, even though that is the term needed. The enthalpy of a system is partly stored in the chemistry (chemically stored energy), temperature, and vibrational energy considering the mass being vibrated. The oceans contain by far most of the energy in the climate system. A change in ocean chemistry (with no change in temperature) changes the total energy. Similarly a change in temperature with no change in chemistry. To speak of “averaging anomalies” v.s. “temperatures” is a distraction. The issue is the total energy.

We don’t know the mass, chemistry and temperature of many elements of the system, so no proclamations about cooling or warming are meaningfully accurate. That has not stopped all sorts of people, scientists included, from making sweeping and loose claims as to what the system enthalpy was, is, and where it is heading.

It is important from a skeptical point of view to be clear that as soon as it is established that the variations – warming an cooling – of the system enthalpy are rooted in the oceans, the whole “back-radiation” falls away as it is not possible to warm an ocean with IR so varying GHG’s won’t make a meaningful difference.

And further, it is important to understand why Mach objected so specifically to Arrhenius on the matter of CO2 and temperature enhancement. Find and read Arrhenius’ 1895 paper, and then read a translation of Mach’s response pointing out that the radiation bands Arrhenius proposed as potentially causing higher temperatures were already saturated, or nearly so. The potential for warming, Mach wrote, is thus a small fraction of the number in the conceptual analysis first proposed. Arrhenius published a couple of small papers admitting there were issues and by 1915 it was generally evident that Mach’s analysis was superior. Still is. That is why “warmists” always refer to the 1895 paper. It sounds scarier.

After more than a century, the basics of the matter have not been effectively communicated to the public, with thousands of charlatans (most of whom know better) prying funds out of gullible governments to, “address the climate crisis”. The “solutions” proffered usually have very poor (low quality/unbelievable) claims about the energy output of their devices, ignoring the energy input needed by a society to create them.

For example, about 3/4 of the energy output of a wind turbine would be required to create another one to replace it after 7-12 years. Only 1/4 of the energy would be available to do something else such as cook lunch or charge a car. To create a sustainable society we must analyse the energy inputs to manage the whole technology. I think for solar panels the output is net-negative when mining and recycling are included. To date they are playthings of the elites.

Steve Z
Reply to  Crispin Pemberton-Pigott
October 21, 2021 1:16 pm

Crispin makes an excellent point about enthalpy, which is a measure of heat content. Enthalpy relative to a fixed reference state, for a substance of a given phase, is usually calculated as

H = mCp (T – Tr)

where H = enthalpy
m = mass
Cp = specific heat at constant pressure
T = actual temperature of substance
Tr = reference temperature

On a mass basis, the specific heat of air is roughly 0.24 kcal/kg-C, while that of water is about 1.00 kcal/kg-C, meaning that on a mass basis, if 1 kg of air was cooled by 1 C, it would warm 1 kg of water by only 0.24 C.

The dominance of water in the enthalpy balance is even greater on a volume basis. Air at sea level pressure and 15 C has a density of about 1.2 kg/m3, but that of water is about 998 kg/m3. If 1 cubic meter of air is cooled by 1 C, the enthalpy lost is 1 * 1.2 * 0.24 * 1 = 28.8 kcal.

If this heat is transferred to 1 cubic meter (998 kg) of water, the temperature rise of the water is only 28.8 / (998 * 1.00) = 0.029 C. Put another way, if 1 m3 of water was cooled by 1 C, the heat lost could warm 1 m3 of air by 34.7 C.

The equation above only applies to substance in a given phase (gas for air, liquid for water). Once a phase change is involved (such as melting ice), the change in enthalpy is much larger. The heat required to melt ice is about 80 kcal/kg.

When global warm-mongers try to project that a slight increase in the average temperature of the atmosphere would cause the Greenland and Antarctic ice caps to melt, causing massive sea-level rise and flooding of coastal areas, they overlook this heat balance.

The total mass of air over 1 square meter of the Earth’s surface at sea level is about 10,330 kg. If the temperature of this entire column of air was warmed by 1 C, the total heat required would be 10,330 * 0.24 * 1 = 2,480 kcal. This would be enough heat to melt 2,480 / 80 = 31 kg of ice, so that the resulting volume of water would be about 0.031 m3, or enough to cover the 1 m2 of ice surface by a depth of 0.031 m = 3.1 cm of water.

The total area of land-based ice caps (mostly Antarctica and Greenland) is about 15 million km2 = 1.5(10^13) m2, so that raising the temperature of the entire atmosphere over the ice caps by 1 C would produce a volume of water of 0.031 m * 1.5(10^13) m2 = 4.65*10^11 m3.

If this additional water was distributed in the world’s oceans, whose total area is about 510 million km2 = 5.1(10^14) m2, the net sea level rise would be

4.65(10^11) m3 / 5.1(10^14) m2 = 0.00091 m = 0.91 millimeter.

For the warm-mongers who are worried about a 2 C rise in the temperature of the atmosphere over 50 years or so, the total sea level rise due to atmospheric heating would be about 1.82 millimeters, or 0.072 inch. Basically a drop in the huge “bucket” of the world’s oceans.

Sea levels have been rising over the past century at a rate of 2 to 3 mm/year (depending on where it is measured), so an extra 1.8 mm or so over 50 years due to extra CO2 in the atmosphere is a very minor contributor.

The heat balance of the earth is dominated by water in its three phases–ice, liquid water, and water vapor (which results in clouds and rain), and the circulation of ocean currents. Any marginal heating of the atmosphere by absorption of infrared radiation by CO2 has a negligible effect, similar to the effect on the total weight if a fly lands on an elephant.

Reply to  Steve Z
October 21, 2021 9:07 pm

But, if the temperatures above the poles really went up by a degree, very little if any would melt as the temperature is almost always well below zero C. For ice to go through the phase change at the ice/water boundary the ice needs to be warmed up to 0°C first and then add in the latent heat amount.

Nicholas McGinley
Reply to  Stephen Lindsay-Yule
October 21, 2021 6:36 am

As several have already pointed out, there are numerous problems with averaging temperature, and also with assuming that average temperature of some number of surface locations conveys physical information about the heat content and changes thereof in the atmosphere.
Books could be written about this subject, in fact one could fill a library with books about this.
But one can use a single example of a very simple case to illustrate:
Consider the average temp of two surface locations taken at just two points in time.
If one location is a place at sea level, and the other is a place at 5000′ of elevation, and on Day 1 the sea level place is at 30° C, and the place at 5000′ is at 0.0° C, and on day 2 the sea level place is at 0.0° C and the place at 5000′ is at 30°C, then the average temperature of the two locations is the same on day 1 as it is on day 2, but the amount of energy is not at all the same.
The air at 5000′ averages about 15% less dense than the air at sea level.

When one considers that the amount of moisture can be and very often is vastly different at any two disparate locations on different days, and even the same location on different days, it is immediately obvious that what is being measured is not heat content of the air, or even changes in heat content.

Simply put, there are numerous independent variables that prevent any such assertion of equivalence and direct comparison from being valid.

John Dowser
Reply to  Joel O'Bryan
October 21, 2021 12:21 am

If you repeat the kitchen experiment from part 1 and measure only anomalies on the kitchen floor and on the stove you still will measure some average increased value for your kitchen while no heat was added to the initial pan. The point is that any negative anomaly of the energy flux of the pan does not translate in any measurement. Not without introducing other ways to quantify the pan with water you just heated.

Reply to  Joel O'Bryan
October 21, 2021 3:20 am

Averaging opens the door for all kinds of manipulation. My doctor uses a non contact gadget to look in my ear and writes down a temperature. How about nominating Mauna Kea to represent the earth temperature wise? Yes there will be all kinds of variations around the mean but it will be much harder to claim manipulation.

Reply to  Joel O'Bryan
October 21, 2021 4:20 am

What you aren’t getting is that statistical models, including “global average temperatures” are based upon an assumption that test data are being drawn from a population of like members. If the members are not alike, then the notion of averaging them no longer makes sense from a statistical model perspective.

It’s like saying that the average weight of sea lions and woodpeckers is 400 pounds. That may be true, but it is still meaningless. A finding that the average weight of sea lions is 800 pounds (or whatever it might be) is meaningful when describing a given species of sea lion. Ditto for a finding that the average weight of woodpeckers is 8 ounces, or whatever it might be. That’s meaningful too.

Jim Steele’s position is that Arctic temperatures, air or water, are not like temperatures, air or water, anywhere else on the planet, so a statistical average including Arctic temps is meaningless.

Jim Gorman
Reply to  Joel O'Bryan
October 21, 2021 4:30 am

If you had a herd of Clydesdale horses and Shetland ponies would you “average” height measurements and expect an answer that is worth anything? Would you make harnesses for the average, would you feed each based on the average. You need to ask if averaging anomalies in the NH and the SH provide an answer to anything.

Lastly, converting to anomalies hides the variance among different locations. That makes no sense at all. Besides that, when drawing graphs it makes anomaly growth look fantastic when 1/100th of a degree is really only a very small percent of the actual total measurement.

Lastly, to be useful, the math behind anomalies need to be such that you can convert temps back to the real measurement. But that wouldn’t allow for good propaganda would it? If you want a really useful number, find a global baseline and then convert individual stations to anomalies. At least, that wouldn’t hide as much of the variance in real temperatures as is going on today.

Reply to  Jim Gorman
October 21, 2021 4:39 am

would you feed each based on the average

This actually does make sense. I mean you can have an average horse for calculations. Of course you don’t feed every horse the average.

Lastly, converting to anomalies hides the variance among different locations

It’s hilarious how unable you are to understand these simple, basic things. Anomalies are just the thing minus a fixed quantity. That doesn’t change the variance. Averaging does, but that’s a different animal.

Reply to  nyolci
October 21, 2021 9:15 am

Nope. Anomalies are not the temperature minus a fixed quantity, they are the temperature minus an average for EACH location. The average is not a fixed quantity, it’s a calculated variable. That requires many measurements to form many different averages, one for each location. In actuality, it’s worse than that – there are no fixed temperature stations in the ocean, there are a bunch of moving bouys, so to get the average at a location, temperatures must be interpolated from nearby locations, but the density of bouys isn’t uniform, the bouys tend to congregate in areas where the winds blow them. But it’s even worse than that, anomalies require seasonal correction, and the correction is not a constant factor either.

You haven’t thought this thing through, nyolci.

Reply to  Meab
October 21, 2021 9:37 am

Well said. It’s hilarious how Nyolci is unable to understand these simple, basic things.

Reply to  Graemethecat
October 21, 2021 11:37 am

Well said. It’s hilarious how Nyolci is unable to understand these simple, basic things.

Try to be smart? Without success 🙂

Reply to  Meab
October 21, 2021 11:37 am

they are the temperature minus an average for EACH location

[redacted] this is exactly what I claimed. There’s a fixed quantity known for each location. This is calculated once for each base period (and for each location). From that point on it is just a constant subtracted from the value when calculating the anomaly of the location. Furthermore, this does not change the sd of the anomaly. Actually, this was the original point.

anomalies require seasonal correction

This makes the thing a bit more complicated but again, this doesn’t affect the sd.

[watch the language, please. mod]

Reply to  nyolci
October 22, 2021 2:15 pm

Mod, please don’t force your religious fundamentalism on us.

Nicholas McGinley
Reply to  nyolci
October 22, 2021 2:20 pm

If you want to define terms being used in a conversation, you need to avoid sloppy language and saying things that are not the case.
Now you are whining while doubling down on being wrong.
You must be used to talking to stupid people.

Reply to  Nicholas McGinley
October 23, 2021 1:02 pm

you need to avoid sloppy language

It was not sloppy, this is that simple.

Reply to  Joel O'Bryan
October 21, 2021 8:28 am

Averaging anomalies introduces a new bias exaggerating high latitude measurements (which vary greatly daily, seasonally and annually) vs. meridional records ( which vary much less). For example, Clive Best provides this animation of recent monthly temperature anomalies which demonstrates how most variability in anomalies occur over northern continents.
comment image

For a discussion of the bias resulting from this see:

Wim Röst
October 21, 2021 12:07 am

Well-informed, well analysed, well explained. Thanks!

Wim Röst
October 21, 2021 1:59 am

Melted ice enhances the quantity of water vapor in the Arctic air. Water vapor is the main greenhouse gas, not being present in large quantities under cold Arctic conditions like with covering sea ice. The huge rise in water vapor over now ice-free areas has a huge greenhouse warming effect, but the warming of air noticed is also telling that the ocean below is cooling. Arctic warming is most time misunderstood.
By water vapor, Arctic warming is enhanced in a second way. The buoyancy of Arctic air rises strongly, both by warmer air and by the higher quantity of light water vapor molecules now present in the air over open water. Low-pressure areas will develop by the rising air and when they develop very near to the North Pole, they can only ‘suck in’ surface heat from the South. That air always is warmer and more humid than the Arctic air present before. Plus, the northward movement of air will take with it extra low-pressure areas from the Mid Latitudes. Northward atmospheric heat transport (the heat being mostly derived from warmer ocean waters from lower latitudes) is completing the ‘Arctic Amplification’ of what started as a small change somewhere that resulted in some Arctic ice melt. And what finally will end up in a colder Arctic ocean.
Usually ‘cooling’ takes more time than ‘warming’. Warming is quick, cooling slow. The reason could be that ‘warming the surface and the atmosphere above’ is easier than ‘cooling the deep’. When a large inflow of warmer than normal subsurface waters warm the Arctic Ocean (think about tens of thousands of cubic kilometers), it takes a lot of time to remove all of the extra heat content of that warm subsurface inflow. Periodic removal of the ice cover simply is necessary to restore the old situation.

The bigger picture is, that when the Sun shifts to the equator (as happens in this phase of the Earth’s orbit) the poles are cooling steadily, but very slowly. More ice and snow develop than thousands of years ago (during the Holocene Optimum) and the coverage of more sea ice keeps more Arctic heat below. Any ice cover of the Arctic prevents an unhindered cooling of the Arctic ocean. The more ice, the higher the disturbance of weather patterns when sea ice starts melting and weather patterns shift to a warming modus. Therefore the Dansgaard-Oescher events have higher temperature effects during a Glacial than in our warmer Holocene. But now, when because of orbital reasons the North is cooling, variability is rising. The Little Ice Age was the first clear sign, our following rapid Modern Warming a second one.

Warm is stable, [general] cooling means: more unstable, with larger temperature fluctuations, up and down. The Big Picture is that we (bit by bit) are entering more glacial circumstances, the actual higher variation in temperatures over decades and over centuries being part of that general cooling down.

Jim Gorman
Reply to  Wim Röst
October 21, 2021 4:35 am

“The huge rise in water vapor over now ice-free areas has a huge greenhouse warming effect, “

Exactly how does latent heat cause warming?

Joel O'Bryan
Reply to  Jim Gorman
October 21, 2021 9:05 am

It does in model land: The presumed amplified wv condensation produces the mid-tropospher hotspot in the tropical latitude bands. The failure to observe what was/is model predicted to be observable (by satellite MSU readings and radiosonde balloons) is the clearest indication that the climate models are grossly wrong in their outputs about future warming.

Reply to  Jim Gorman
October 21, 2021 9:23 am

It doesn’t over time it causes cooling.

Wim Röst
Reply to  Jim Gorman
October 21, 2021 3:44 pm

Jim Gorman: “Exactly how does latent heat cause warming?”

WR: It is water vapor that causes warming just above the surface. Surface radiation is absorbed by water vapor (and by clouds) and this absorbed energy is not directly lost to space. More energy remains near the surface: more water vapor results in less heat loss.
Above sea ice temperatures easily fall to minus 30 Celsius. No sea ice means a temperature at the surface around 0 degrees Celsius, 30 degrees warmer. Much more radiation is lost by the warmer surfaces and trapped just above the surface by absorbing water vapor. When less energy is lost, the atmosphere near the surface remains warmer.

Who looks at the huge spikes in temperature sees when low-pressure areas with more water vapor (and more advected energy) did enter the Arctic:

Matthew Sykes
October 21, 2021 2:01 am

Yep, arctic sea ice insulates the ocean, it keeps heat in.

Ben Wouters
October 21, 2021 2:44 am

Agree with the mechanism for oceanic heat ventilation.
The ventilation of ALL deep oceanic heat can only occur at the (high) latitudes of the Arctic and Antarctic regions since solar maintains an insulating layer for much of the oceans.
In those regions ALL accumulated geothermal heat must be ventilated, giving the mechanism for increasing or decreasing deep ocean temperatures.
Ventilation > geothermal input gives cooling.
Ventilation < geothermal input gives warming of the deep oceans.
Realize that the ~100 mW/m^2 geothermal heat is sufficient to warm the average oceanic column 1K every ~5000 year.
Heating int he Arctic must be much faster since those waters are less deep and geothermal heating is considerably higher due to the high underwater volcanic activity on top of the “standard” geothermal flux.  

Nicholas McGinley
Reply to  Ben Wouters
October 22, 2021 2:34 pm

Not so.
Upwelling occurs in places far removed from the Arctic.
The poles add cold bottom water to the oceans, on a net basis.
Upwelling can occur by mechanisms such as geographical shape of coastlines combining with offshore or long shore winds that push away surface water; intense surface storms and wave action that cause deep mixing; advection of polar air into midlatitude regions resulting in rapid surface cooling and subsequent flipping between surface and water at depth; and by a gradual process as new cold bottom water is slowly but constantly fed into the bottom water zone, forcing water to rise by being replaced.
All of these can be enhanced by warming caused by subsurface volcanic and tectonic activity, and this type of warming of subsurface/bottom water can operate independent of other mechanism.
Just off the top of my head.

It is usually wrong to speak in absolutes, and using all caps to make such assertions is unseemly and jarring.
Both of which simply draw more attention when saying something that is just plain wrong.

October 21, 2021 3:26 am

The antarctic is much colder than the arctic, so I’d say landed ice does a much better job of modulating heat release.

John Tillman
Reply to  pochas94
October 21, 2021 6:37 am

The Antarctic ice sheets are at high elevations. Arctic sea ice lies at sea level. Also, the floes move around Antarctic glaciers flow, but not in the same ways as sea ice.

Captain climate
October 21, 2021 11:15 am

I love these videos. They explain ocean currents and the climate in no bullshit ways. And they point out the obvious that heat transport and not CO2 are responsible for arctic warming.

October 21, 2021 11:36 am

Excellent and much needed article.
Short comment – atmospheric CO2 boosts downwelling of oxygen and ventilation of the deep ocean:

Atmospheric CO2 is good for the deep ocean – Odyssey (

Don’t let them try to tell you that rising CO2 depletes ocean CO2.
As usual, the opposite is true.

Reply to  Hatter Eggburn
October 21, 2021 2:16 pm

“Don’t let them try to tell you that rising CO2 depletes ocean oxygen

Ulric Lyons
October 24, 2021 6:29 am

Rigor 2002 is completely backwards. Arctic warming is negative NAO driven not positive AO driven. Directly by a meridional jet stream increasing warm humidity events and cyclones into the Arctic, and indirectly by negative NAO driving a warmer AMO. There was no rapid warming or sea ice loss in or from 1989, it began in a deep negative NAO episode in 1993, and continued through a negative NAO regime 1995-1999. The 1970-80’s were more positive NAO than negative NAO, with the positive NAO episodes exactly in the years when the AMO and Arctic were colder.
During centennial solar minima, negative NAO conditions typically increase, that brings colder weather to northwest Europe, and it drives a warmer AMO and Arctic.
Like during the Dalton Minimum when British naval ships noted a large loss of Arctic sea ice in 1815-1817.
It’s also a mistake to look at only the winter season AO or NAO, as many of the larger negative NAO anomalies affecting the sea ice have occurred in summer when they have had more impact. Like with the Arctic cyclones in summers 2007 and 2012, during months of deeper negative NAO conditions.

comment image

%d bloggers like this:
Verified by MonsterInsights