Can longwave radiation heat the oceans?

[Drakkith]

An increase in CO2 atmospheric concentrations will increase IR radiation in the atmosphere and as a result will increase downward IR radiation from the atmosphere to the Earth's surface. Where this "extra" CO2 strikes land it will warm the surface, increase convection, conduction and radiation increasing surface atmospheric temperatures.
As we've already touched on - where the "extra" downward IR strikes the ocean the effect is much less clear. Does the "extra" energy mainly increase evaporation rates or does it lead to increased temperatures in the water column?

As far as I know the increased evaporation rates are due to the increase in temperature, so you can't have one without the other.

 

[Bystander]

"Total Solar Irradiance," ref. 17 and 18 from Wiki http://en.wikipedia.org/wiki/Sunlight , 1360.5 -1363 to 1365.5-1367 W/m²; a 1.5 - 2.5 W/m² variation over long-term solar cycles, and 0.3% difference between total irradiance as measured in two series of measurements covering ~ 70's to present day. Forty to fifty years, ± 0.3 %, is decent calorimetry. Earth's orbit, aphelion and perihelion, other parameters are known to far smaller uncertainty, and TSI as a function of day of the year can be expressed to same uncertainty as that of the measurement series. So we know incoming irradiance to ± 0.3 % year-round. Need to check Wiki's citations, but they look decent (no J. Irreproducible Results type stuff).

Since Δq ~ 4σ_S.-B.T³ΔT the uncertainty in solar irradiance shouldn't affect black body numbers significantly, and we can take this much as well known.

Next step is to identify uncertainties in the 1050 and 1120 W/m², for direct surface insolation and directly transmitted plus scattered surface insolation figures from Newport, ref. 3 in http://en.wikipedia.org/wiki/Sunlight . Haven't got this one handy at the moment. Use of the word "about" in the Wiki article may not mean anything, but, "one step at a time."

 

[Loudzoo]

As far as I know the increased evaporation rates are due to the increase in temperature, so you can't have one without the other.

That's exactly my understanding. The downward longwave radiation works to increase the temperature of the ocean skin which increases evaporation, ceteris paribus. The question we hope to answer is whether this process can ever heat the mixed layer or ocean bulk. If changes to DLR are quickly offset by changes to the evaporation rate (in a linear fashion) then "long term" ocean temperatures are unlikely to be sensitive to changes in DLR.

 

[my2cts]

Short answer: of course it can and it will, all other effects being equal. Any radiation will. Its a no brainer.
Somewhat longer:
A large fraction will be absorbed by the water, some is reflected. Depending on the wavelength this happens in the top 100 microns (6.3 um) , still 1 million or so molecules thick, or in the top few meters (10 um). This process delivers heat to the ocean. This does not mean the ocean temperature rises since so many other mechanisms are involved.

 

[my2cts]

but atmospheric radiation only heats the top few molecules./QUOTE]
The top mmm or meters . That is a lot more than the thickness of a water molecule.

 

[Loudzoo]

Re: my2cts
Thanks for the clarification. Yes - my "top few molecules" comment for representing one side of the argument is I suspect an exaggeration. Its not my understanding but what others have said.

This does not mean the ocean temperature rises since so many other mechanisms are involved.

Its this question that I really want to drill down on and get a handle on.

 

[my2cts]

You need to do really good measurements, easier said than done, or run a really good model.
Maybe the latter is your best option.

 

[PeterDonis]

Without greenhouse gases in the atmosphere the Earth's atmosphere would be some 33C cooler on average.

This factoid is often quoted, but it leaves out a key factor, namely latent heat transport via the hydrologic cycle: water evaporates from the surface (mainly the ocean), rises into the upper atmosphere, and condenses there into clouds and precipitation. The net effect is to transport heat (as latent heat carried by the water vapor) from the surface to the upper atmosphere, where it can more easily escape to space.

The reason this is important can be illustrated by considering the following series of scenarios:

(0) A hypothetical Earth with an atmosphere completely transparent to both incoming and outgoing radiation. The average surface temperature of such an Earth would be approximately 255 K, based on thermal equilibrium with incoming Solar radiation covering the cross-sectional area that the Earth presents to the Sun.

(1) A hypothetical Earth with an atmosphere that had the same greenhouse effect as the actual one, but was still completely transparent to incoming radiation, and which had no other heat transport processes (i.e., no convection, no hydrologic cycle, no weather, etc.). The average surface temperature of such an Earth would be (IIRC--it's been a while since I looked at the details of these calculations) about 333 K, which is 60 C or 140 F--i.e., hotter than the hottest surface temperature ever recorded on the actual Earth (134 F in Death Valley).

(2) The actual Earth, with both the greenhouse effect and the hydrologic cycle. The average surface temperature is about 288 K.

Notice that the 33 C (or 33 K) temperature difference quoted is between (0) and (2), but that difference is not just due to the greenhouse effect; it's due to a balance between the greenhouse effect and the hydrologic cycle. It's also worth noting that the average surface temperature is actually not the best thing to look at to understand what's going on: you need to look at the profile of temperature vs. altitude in the atmosphere.

 

[PeterDonis]

You need to do really good measurements, easier said than done, or run a really good model.

A model is only going to be "really good" if it is repeatedly improved based on comparison between the model outputs and really good measurements. So no matter what, you need really good measurements. One of the key issues with our current (relatively poor) understanding of the Earth's climate is that we don't have really good measurements, and without them, there's no way to spot where our models are going wrong.

 

[PeterDonis]

As a human being I'd love to think CAGW is bunk because otherwise we're in serious danger.

It's worth pointing out here that this does not follow; even if "CAGW is bunk" (which I personally think it is, at least the CO2 alarmism part of it), we could still be in serious danger, because we don't understand very well how the climate works. We could be in for another ice age soon; we don't know whether, or when, this might happen. Various other human activities besides CO2 emissions could be significantly affecting the climate (my personal candidate is land use--we have drastically changed the land surface of the planet); we don't know which ones, or how big their effect is. So the common "skeptic" argument that, since CAGW is bunk, there's nothing to worry about, is not a valid argument.

 

[Bystander]

As far as I know the increased evaporation rates are due to the increase in temperature,

Vapor pressure is a function of temperature. Evaporation rate is a function of heat transport to the phase interface and of material transport from the interface.

many other mechanisms are involved

That's what we're trying to pin down.

good measurements

These are what we're trying to identify.

good measurements.

(ditto)

other human activities besides CO2 emissions could be significantly affecting the climate (my personal candidate is land use--we have drastically changed the land surface of the planet); we don't know which ones, or how big their effect is. So the common "skeptic" argument that, since CAGW is bunk, there's nothing to worry about, is not a valid argument.

"... could be significantly affecting the climate ..." Yes. At the moment, we're trying to pin down what's been measured, and how well, rather than what probably needs to be examined in detail.

"Jemand anderes?"

 

[Loudzoo]

Thanks Peter and Bystander - very valid points in my humble opinion.

 

[Bystander]

Seems to have settled back down to a "dull roar." Apparently no concerns regarding insolation rate at the upper boundary, 1363 W/m² ± 0.3 %, + 3 % at perihelion and - 3 % at aphelion, the uncertainty of orbital parameters being insignificant, with an eleven year solar cycle variation of ± 0.05 - 0.08 %, peaking at 1979, 1990, 2001, and 2012. Stet?

Area of the Earth disc intercepting solar radiation is ~ 1.25 x 10⁸ (km)², with an atmospheric halo area at 100 km altitude of 4 x 10⁶ (km)², ~ 3 % of total intercepted radiation. "Total" over the disc, or "mean" rates per unit area (W/m²)? "Total" is a bit perilous given lack of data on a fine enough scale; "mean" is likewise perilous with its implication of integration of data on a meaningfully fine scale. "Mean" and "total" also include the possibility of remote (satellite or lunar instruments) radiometry. The object at this point is to examine the uncertainties in the 1050 and 1120 W/m² values for directly transmitted surface insolation, and directly transmitted plus scattered surface insolation cited (Newport http://www.newport.com/Introduction-to-Solar-Radiation/411919/1033/content.aspx ) in the Wiki "Sunlight" article. There are no citations in the Newport article and that pretty much leaves us in the "air" regarding partitioning of reflected, transmitted, and absorbed radiation fluxes. Anyone aware of measurements of total reflection, or transmission over a broad band, rather than just visible or other limited bands, by geosynchronous or lunar or other suitably remote instruments?

Energy/power balance may not be necessary at this stage of the argument, but I'll feel a lot more comfortable examining the IR details if there is a total balance with which they must be consistent.

Okay, now everybody jump in, and try to stick to balancing total insolation with total of reflected, transmitted, and absorbed radiation. Please do not use "albedo" without specifying a definition/band width.

 

[Bystander]

So, it's off to the library for J. Gee. Whiz, Geochim. et Cosmochim., Phys. Rev., and what else? Any suggestions? I've found nothing regarding the uncertainties that bother me in two days on the net.

 

[FactChecker]

Are you only interested in the direct warming by IR? It seems like warming the surface must indirectly warm the lower levels either by heat transfer down or by reducing heat transfer up.

 

[Bystander]

Are you only interested in the direct warming by IR?

The "interest" is in seeing/finding/determining the "state of the art" regarding energy balances and uncertainties in the various radiation energy exchange/transfer mechanisms among surface, atmosphere, and oceans.

 

[FactChecker]

Bystander - I found the following passage on a CAGW skeptic's website (http://climaterealists.com/index.php?id=4245)
"However the effect of downwelling infrared is always to use up all the infrared in increasing the temperature of the ocean surface molecules whilst leaving nothing in reserve to provide the extra energy required (the latent heat of evaporation) when the change of state occurs from water to vapour. That extra energy requirement is taken from the medium (water or air) in which it is most readily available. If the water is warmer then most will come from the water. If the air is warmer then most will come from the air. However over the Earth as a whole the water is nearly always warmer than the air (due to solar input) so inevitably the average global energy flow is from oceans to air via that latent heat of evaporation in the air and the energy needed is taken from the water. This leads to a thin (1mm deep) layer of cooler water over the oceans worldwide and below the evaporative region that is some 0.3C cooler than the ocean bulk below."
The last sentence does seem to be validated with this paper: http://www.nature.com/nature/journal/v358/n6389/abs/358738a0.html
But is the rest fair?

I think this analysis must be close to the answer. The latent heat of evaporation must be a big player in this. If the surface temperature consistently remains below the lower temperature, that implies that light frequencies which can penetrate to lower levels are enough to maintain a steady heat transfer up to the surface. But clearly the IR warming the surface must reduce the rate of heat transfer. So IR indirectly allows the lower layers to stay warmer than otherwise.

 

[Bystander]

must be close to the answer

Possibly. It would be a more compelling argument with some actual numbers. As it stands, it's difficult to distinguish from a variety of perpetual motion schemes that run the energy around in circles with qualitative explanations to the point that no energy balance is possible.

 

[FactChecker]

Possibly. It would be a more compelling argument with some actual numbers. As it stands, it's difficult to distinguish from a variety of perpetual motion schemes that run the energy around in circles with qualitative explanations to the point that no energy balance is possible.

"perpetual motion schemes"? What are you referring to?

 

[Bystander]

"perpetual motion schemes"?

IR evaporates surface film, warming atmosphere, which emits IR, which evaporates surface film ... ? Bothers me.

 

[Loudzoo]

Bystander - many thanks for your time on this. I too have been hunting for some actual data but am yet to find much. What little I have found (such as this: http://journals.ametsoc.org/doi/abs/10.1175/JCLI-D-11-00262.1) suggests to me that the additional 1 watt / m^2 downward longwave radiation at sea level that doubling CO2 atmospheric concentrations would generate cannot be easily traced within the context of +/- 10 watts / m^2 errors in measurements. Furthermore, an additional 1 watt / m^2 in the context of an average of 350 watts / m^2 is unlikely to affect long term warming rates of the ocean even if you could measure to that level of accuracy . . .

 

[Bystander]

Comparing total insolation (1363 ± 0.05 to ± 0.08% variation over 11 year solar cycle) to GMT isn't getting me anywhere either. Should be a 10 -15 mK "scallop" showing up in GMT, GISS other temperature records with ~ 11 year period, and if it's there, it's not obvious. May be too far below the noise level of the original measurements.

 

[PeterDonis]

the additional 1 watt / m^2 downward longwave radiation at sea level that doubling CO2 atmospheric concentrations would generate cannot be easily traced within the context of +/- 10 watts / m^2 errors in measurements.

This is an issue with many key measurements of the climate: the effect that is supposedly there is smaller (sometimes much smaller) than the error in the measurements.

 

[Bystander]

the effect that is supposedly there is smaller (sometimes much smaller) than the error in the measurements.

At which point we've about exhausted this discussion? And should give up and lock it?