This article follows:
- Part One – which explained a few basics in energy received and absorbed, and gave a few useful “numbers” to remember
- Part Two – which explained energy balance a little more
- Part Three – which explained how the earth radiated away energy and how more “greenhouse” gases might change that
What is albedo? Albedo, in the context of the earth, is the ratio of reflected solar radiation to incident solar radiation. Generally the approximate value of 30% is given. This means that 0.3 or 30% of solar radiation is reflected and therefore 0.7 or 70% is absorbed.
Until the first satellites started measuring reflected solar radiation in the late 1970’s, albedo could only be estimated. Now we have real measurements, but reflected solar radiation is one of the more challenging measurements that satellites make. The main reason for this is reflected solar radiation takes place over all angles, making it much harder for satellites to measure compared with, say, the outgoing longwave radiation.
Reflected solar radiation is one of the major elements in the earth’s radiation budget.
Over the 20th century, global temperatures increased by around 0.7°C. Increases in CO2, methane and other “greenhouse” gases have a demonstrable “radiative forcing”, but changes in planetary albedo cannot be ruled out as also having a significant effect on global temperatures. For example, if the albedo had reduced from 31% to 30% this would produce an increase in radiative forcing (prior to any feedbacks) of 3.4W/m2 – of similar magnitude to the calculated (pre-feedback) effects from “greenhouse” gases.
Average global variation in albedo (top) and reflected solar radiation (bottom)
(click on the image for a larger picture)
(click on the image for a larger picture)
The first measurements of albedo were from Nimbus-7 in 1979, and the best quality measurements were from ERBE from November 1984 to February 1990. There is a dataset of measurements from 1979 to 1993 but not from the same instruments, and then significant gaps in the 1990s until more accurate instruments (e.g. CERES) began measurements. Satellite data of reflected solar radiation from latitudes above 70° is often not available. And comparisons between different ERB datasets show differences of comparable magnitude to the radiative forcing from changes in “greenhouse” gases.
Therefore, to obtain averages or time series over more than a decade requires some kind of calculation. Most of the data in this article is from Hatzianastassiou et al (2004) – currently available here.
The mean monthly shortwave (SW) radiation budget at the top of atmosphere (TOA) was computed on 2.5 longitude-latitude resolution for the 14-year period from 1984 to 1997, using a radiative transfer model with long-term climatological data from the International Satellite Cloud Climatology Project (ISCCP-D2)..
The model was checked against the best data:
The model radiative fluxes at TOA were validated against Earth Radiation Budget Experiment (ERBE) S4 scanner satellite data (1985–1989).
The results were within 1% of ERBE data, which is within the error estimates of the instrument. (See “Model Comparison” at the end of the article).
It is important to understand that using a model doesn’t mean that a GCM produced (predicted) this data. Instead all available data was used to calculate the reflected solar radiation from known properties of clouds, aerosols and so on. However, it also means that the results aren’t perfect, just an improvement on a mixture of incomplete datasets.
Here is the latitudinal variation of incident solar radiation – note that the long-term annual global average is around 342 W/m2 – followed by “outgoing” or reflected solar radiation, then albedo:
The causes of reflected solar radiation are clouds, certain types of aerosols in the atmosphere and different surface types.
The high albedo near the poles is of course due to snow and ice. Lower albedo nearer the equator is in part due to the low reflectivity of the ocean, especially when the sun is high in the sky.
Typical values of albedo for different surfaces (from Linacre & Geerts, 1997)
- Snow 80%
- Dry sand in the desert 40%
- Water, sun at 10° 38% (sun close to horizon)
- Grassland 22%
- Rainforest 13%
- Wet soil 10%
- Water, sun at 25° 9%
- Water, sun at 45° 6%
- Water, sun at 90° 3.5% (sun directly overhead)
Here is the data on reflected solar radiation and albedo as a time-series for the whole planet:
(click on the image for a larger picture)
Over the time period in question:
The 14-year (1984–1997) model results, indicate that Earth reflects back to space 101.2Wm-2 out of the received 341.5Wm-2, involving a long-term planetary albedo equal to 29.6%.
The incident solar radiation has a wider range for the southern hemisphere – this is because the earth is closer to the sun (perihelion) in Dec/Jan, which is the southern hemisphere summer.
And notice the fascinating point that the calculations show the albedo reducing over this period:
The decrease of OSR [outgoing solar radiation] by 2.3Wm-2 over the 14-year period 1984–1997, is very important and needs to be further examined in detail. The decreasing trend in global OSR can be also seen in Fig. 5c, where the mean global planetary albedo, Rp, is found to have decreased by 0.6% from January 1984 through December 1997.
The main cause identified was a decrease in cloudiness in tropical and sub-tropical areas.
Model Comparison
For those interested, some ERBE data vs model:
(click on the image for a larger picture)
When calculating the planetary average albedo, it’s important to calculate it as the ratio of the global reflected to the global incoming radiation. If one calculates the albedo by latitude first and then averages the albedo after correcting for area, you can get the wrong answer because total incoming radiation also varies by latitude. The sum of the products may not be equal to the product of the sums.
That does seem to be a problem for the short term climate sensitivity, though. The UAH global average anomaly linear trend for 1984-1997 produces a temperature change of only 0.2 C. That’s a short term climate sensitivity of less than 0.1 C/W, ignoring all the other possible forcings. Mauna Loa CO2 was 344 ppmv in 1984 and 363 ppmv in 1997. That’s ~0.2 W/m2 forcing change, so it is indeed small compared to the albedo forcing. A 3 C for doubling CO2 is a climate sensitivity of about 0.6 C/W. Feedbacks don’t care about the source of the forcing to a first approximation.
Even if we say all the energy from the decrease in albedo went into the ocean there’s a problem. The change in ocean heat content (OHC) has been used to prove there’s a net radiation deficit of 0.85 W/m2 (heat in the pipeline). But if this wasn’t corrected for the increase in albedo forcing, that deficit changes to a surplus when albedo forcing is included and the planet has been radiating ~1.5 W/m2 more than expected.
Another way of looking at it is that clouds have near zero feedback effect. The increase in absorbed radiation at lower cloud coverage is nearly exactly matched by the increase in the fraction of emitted radiation from the surface that reaches space. That’s still a problem for the GCM’s, though, as, IIRC, many of them have clouds as a positive feedback. But it’s also a problem for those who claim that clouds, especially low tropical clouds, are a strong negative feedback. That includes the cosmic ray/cloud hypothesis.
I was just wondering, why is there no albedo value for clouds? is it because we believe they are a net zero in the equation or just an oversight in this list. personally, and I’m no scientist, it seems like cloud cover would cause a net cooling effect. ie. if the planet were under 100% cloud cover, I imagine it would be quite chilly after a period of time without sunlight. like-wise I can imagine the whole planet a lot warmer with 0% cloud cover. I’m sure there’s a good deal of discussion and not a strong “concensus” on the issue of the effect of clouds in the (or the lack of clouds) climate models. It seems to me that this would be a critical part of any valid model, and should be something that climate science can not ignore. Having said that, I don’t find a lot of climate science articles about the subject of clouds and their effect on the energy budget.
everyman:
The list is of surface properties (a clarification I will put in the article).
Clouds have a very high albedo, and you can see more about their overall effect in Clouds and Water Vapor – Part One
oops, just found your article “clouds and water vapor part 1”.
haven’t read it entirely yet, but skimmed it and it looks very interesting.
I’ve read several articles that claim the current IPCC climate models do not include cloud cover as a variable. is that true? and that the formation of clouds is not fully understood. if so, how can we presume to model the climate?The last time I heard someone say they believed something was true but could not explain how it came to pass was when I was in church. I really hate to see any field of scientific study relying on faith to defend it’s claims. Is that where we end up when discussing clouds and their effect on climate trends?
De Witt, you say: “Another way of looking at it is that clouds have near zero feedback effect. ” I thought over all clouds had a negative feedback or radiative forcing;
Click to access 7_Dupont.pdf
Cohenite
The document you linked (good one, BTW) does not mention feedback – only net radiative forcing.
Maybe you are mixing up both concepts?
Alexandre, this is my understanding of cloud forcing:
FSW (cloud) = FSW (clear) – FSW (all-sky)
FLW (cloud) = FLW (clear) – FLW (all-sky
Fnet (cloud) = FSW (cloud) + FLW (cloud)
Fnet (cloud) => Clouds have a cooling effect on Climate if Clouds have a warming effect on Climate if > 0
I would be very happy for someone to explain how that differs from cloud feedback.
SoD, thanks for another post–I always learn a lot here.
Regarding the graphs in the post–why does the southern hemisphere have larger variation than the northern hemisphere in the albedo and incoming/outgoing flux? Is it the difference in the land/water area?
And now, OT but in the spirit of diverting the albedo thread to talk about clouds 🙂
I finished reading Archer’s “Understanding the Forecast”, and regarding clouds he said that higher clouds have an overall warming effect and lower ones have an overall cooling effect. I believe I understood why, but I had 2 questions:
The explanation that he gave for higher clouds reflecting OLR was that even though they are thin clouds with less water available they are still “opaque” enough to block those wavelengths. Is the transition point for becoming “opaque” to OLR just the point when the water vapor condenses to droplets?
He also mentioned that ice crystals reflect SR significantly better than water droplets, but didn’t give much detail beyond that. I assume that ice crystals would form more readily in the higher clouds, so that would increase the amount of SR that was blocked, but how well do ice crystals block OLR?
Thanks.
Bill Stoltzfus:
Thanks for the kind comments.
The earth is closer to the sun (perihelion) in Dec/Jan, which is the southern hemisphere summer. And the highest surface albedo is from ice/snow at the south pole.
So during the southern hemisphere summer the solar insolation is higher and the proportion of surface reflected radiation is a maximum. Contrast that with the northern hemisphere – when the solar insolation is a maximum (Dec/Jan) the northern latitudes can not reflect very much.
On clouds, I didn’t understand the points from David Archer’s book. “..higher clouds reflecting OLR..”
Clouds don’t reflect OLR. Longwave is absorbed and then re-radiated. Can you check exactly what he wrote?
Cohenite
FSW (cloud) = FSW (clear) – FSW (all-sky)
FLW (cloud) = FLW (clear) – FLW (all-sky
Fnet (cloud) = FSW (cloud) + FLW (cloud)
Fnet (cloud) => Clouds have a cooling effect on Climate if Clouds have a warming effect on Climate if > 0
Yes, I guess that matches what I understand about cloud radiative forcing too. And AFAIK the net effect is a cooling one (Fnet>0). We seem to agree here.
I would be very happy for someone to explain how that differs from cloud feedback.
One thing is what effect you have once you add clouds to an atmosphere: they will reflect a lot of shortwave back up, and trap some LW in the lower atmosphere.
Another thing is how clouds would respond once you have warmed the climate for some other reason (say GHE or the sun).
If you get more clouds, then the feedback will be negative, as it will dampen the original forcing.
If you get less clouds as a response to that first forcing, then the overall cloud feedback would be positive, as it will somehow reinforce it by diminishing the cloud cooling effect.
cohenite,
The instantaneous effect of clouds measured at the surface appears to be negative. The observed longer term global effect, however, based on the data from the paper cited in the original post, appears to be very close to zero. Maybe the loss in cloud cover was mostly cumulus clouds whose effect, according to the presentation you cite, is close to zero.
scienceofdoom:
I didn’t explain it well, AND I got my terms confused–of course you’re right that clouds absorb and re-radiate LR. I will have to go check Archer out from the library again.
everyman. Clouds are included in models. For some more detail, see faq on climate models
For hooks into the literature, best to start with IPCC WG1, Chapter 8, Atmospheric processes.
Trenberth’s current Albedo estimate is 29.83%. Clouds are at least half of this value.
Anyone run across the estimated total Earth Albedo during the Ice Ages and especially the last glacial maximum?
——————–
The greenhouse effect estimate of 33C ignores the fact that at least half of the Albedo is the result of clouds (which would also be contributing to the 33C greenhouse effect though the impact of water vapour that the Albedo values are based on).
So, let’s exclude that portion and what we have is:
– CO2/GHG only Greenhouse Effect = 288K – [1366 W/m2 * (1-0.15) / 4 / 5.67e-8]^.25 = 20C
Now the global warming formula have to rewritten because the CO2/GHG greenhouse effect is only 20C.
—————-
Earth’s Albedo has varied throughout the last 650 million years between 25% to 50% (29.83% today) depending on the continental arrangements and the amount of ice that builds up and spreads out from the poles. That is a big range and the values are capable of explaining the majority of the temperature changes over the period.
Bill Illis:
Can you explain this statement a little further?
The earth absorbs 1367 * (100%-30%) /4 = 239 W/m^2 averaged over the surface of the earth.
This implies an effective radiating temperature of 255K. But we measure an approx 288K average at the surface of the earth.
If there were less clouds because of less water vapor I’m sure things would be different. But they aren’t different..
That’s why I can’t work out your point?
The 33C greenhouse effect is calculated based on an Albedo value of 30%. But (a little more than) half of that Albedo value is clouds (or water vapour).
We shouldn’t count the negative impact of water vapour (through the Albedo impact) and exclude the positive impact of that very same water vapour. So, let’s exclude the water vapour impact on Albedo entirely as one way to solve the dilemma.
Put 15% into the formula instead = 1367 * (100%-15%)/4 = 290 W/M^2.
The effective radiating temperature is now 268K or 20K less than the current surface temperature of 288K.
One could say water vapour has both a positive and a negative impact on the climate. Overall, one would have to conclude the negative impact of clouds is greater than the positive impact of water vapour since the actual effective radiating temperature (255K) is lower than no-water vapour radiating temperature (268K) [15% * 1367 W/m^2/4 = -51 W/m^2 = -13K is a big number].
This result is total counter-intuitive but it does indicate that water vapour is a negative feedback. The Albedo impact of clouds is greater than the positive greenhouse effect of water vapour. At this point the math really breaks down unless one also realizes that the Stephan Boltzmann equations are also logarithmic. The temperature impact per Watt declines as the Wattage goes up.
We start with 268K effective radiating temperature, we subtract -51 W/m^2 out of the climate (clouds reflecting sunlight) – radiating temperature is now 255K, We then add-back 75 W/m^2 of CO2/GHG forcing and add-back 75 W/m^2 of positive water vapour forcing and the math now works.
We get 288K at the surface and 20K of net GHG/water vapour greenhouse effect (and water vapour can have 75 W/m2 of positive greenhouse effect and -51 W/m2 of negative Albedo effect but the negative can still out-weigh the positive since Stephan Boltzmann is logarithmic).
I think someone will have had to play around with the formulae and the greenhouse effect math a lot to be able to see this.
Let’s take a real-world example as well.
Let’s look at the Sahara temperatures in Sudan (15N, 30E) versus Equatorial temperatures in Zaire (0N, 30E). Let’s use April since this is about the right time to take into account the slight lag of temperatures versus the Equinox.
So both areas are getting about the same solar energy in at the TOA. Sudan has almost no water vapour and Zaire has a large amount of water vapour.
Temperatures in Sudan in April at 15N, 30E average 32C (hot days cooler nights, temp change is 23C from day to night) and Zaire averages 22C (day and night difference is 9C).
So water vapour and cloud albedo reduces the temperature average by 10C.
I guess I should also add the no-cloud surface Albedo of the desert in Sudan is also much higher at about 40% than the Equatorial jungle of Zaire at about 25%.
So the temperature differential is even greater since the surface Albedo (with no cloudiness in both locations) should make Zaire about 15C warmer than Sudan (rather than the 10C cooler as the current water vapour levels and clouds dictate).
Interesting point Bill; you may be interested in these papers:
http://www.scienceonline.org/cgi/content/abstract/320/5873/195
http://www.agu.org/pubs/crossref/2010/2009JD012050.shtml
I’m not sure if I follow your calculations, but you raise a crucial point; which is how much does CO2 contribute to the 33C greenhouse temp?
cohenite:
You can see the answer here – CO2 – An Insignificant Trace Gas – Part Five
So of the 33C greenhouse temperature you say ~ 13C is from CO2?
I’m sorry, my maths is terrible and State of Origin is on; the CO2 contribution would be ~ 9.5C?
cohenite:
Well, roughly speaking. As there is 4th power relationship between temperature and energy it may be better to say 27% of the total “retained” longwave energy is due to CO2.
A bit offtopic maybe, but didn’t we establish already that effective temperature and average temperature are not directly comparabale. Meaning the atmosphereless blackbody earth (emission wise) with the same albedo, would have a lower average temperature than 255K (which would imply a higher greenhouse effect than 33 K).
I remember this part about the effective temperature being different than the average one (hey, this is looking more like a study group than a blog 🙂
But I did not understand your point, here. Is the present global effetive temperature 289K, too?
Not much point really, I just don’t like the 33 C value that keeps popping up now and then. I’d prefer if those numbers weren’t used very close to each other – 255 K, which is the effective blackbody temperature for the amount of energy the surface receives, and 288K, which is the measured average surface temperature.
A few questions:
1) Is the asymmetry of the cloud shapes important in the albedo calculations? If the clouds ‘lean’ in one direction as a result of large scale wind pattern, the optical properties of the clouds could be asymmetric (think window blinds).
The H2004 paper uses something called “cloud asymmetry parameter” (p2685), but if I understand correctly (read, if the results of 5 minutes of googling can be trusted 😉 this seems to concern the optical properties of the cloud particles, not the bulk shape of the clouds.
2) In calculating the mean monthly incoming solar flux (page 2677/78), the H2004 paper uses a parameter mu_n, identified as “the cosine of the mean daily solar zenith angle”. Shouldn’t it be “the mean cosine of the daily solar zenith angle”? The formula given there seems to me to be just that.
3) On page 2681, the authors compute the ocean reflectivity, using an incidence angle phi defined as cos^{-1} mu. Is mu the same as mu_n? I suppose it isn’t; averaging of the reflectivity formula over the incident angles could give a different value than the formula applied to the arccos of the mean cosine.
4) Is there some 1 sigma errorbar given for the global 0.296 albedo?
So in about 15 years from 1984 to 1998, the increase in forcing due to albedo changes from tropical clouds was twice that of the total CO2 forcing increase over the previous 60-years. In 1998, we get a huge 150+year record El Nino.
Is this the scale of natural variation Spencer is talking about?
Is the albedo change this a huge positive feedback?
What albedo change do you mean? According to what I understood from the graph, the albedo change in the 84-98 period was in the order of 0.05% – and least part of this is atributtable to polar ice extent loss as a response to warming.
The figures mentioned at the beginning of the text refer to an hypothetical 1% change.
It’s data, see Figure 5 [Hatzianastassiou (2004)] above. 0.6% Albedo decline in 15-years and the earth is sucking up 2.3W/m2 more forcing. SOD claims this was tropical/sub-tropical cloud changes.
Also, 84-98 was not a warming period unless you cherry-pick the endpoints… it’s bumpy with a flat trend.
Could you please walk me through here?
The way I understood it, the red line of Fig. 5 is defined as that linear function at the top right. I assumed ‘x’ is years. That makes 0.0096W/m2 in 15 years. I calculated the albedo decline (0.05%, not 0.6%) pretty much the same way. How did you get your results?
About the 84-98 period: putting aside the doubtful importance of surface temperature and significance of a 15-year period in climatic terms, I have just plotted the period in Excel with a linear trend (NCDC data), and it’s clearly positive.
x is in months. 0.014 W/m^2 * 14 yr * 12 months/yr = 2.35 W/m^2 change over 14 years.
Thanks.
SOD: Hatzianastassiou determined from a linear fit (Fig 5b) that OSR dropped at an average rate of 0.014 W/m^2/month or 1.68 W/m^2/decade between 1984 and 1997. Isn’t it true that we shouldn’t draw any conclusions from a trend until we establish that an appropriate confidence interval doesn’t include zero change? Since the data are strongly correlated from month to month, proper derivation of a confidence interval requires correction for autocorrelation. Is it normal for climate scientists to draw conclusions (and use the word “significantly” in the abstract) without calculating a confidence interval?
Hatzianastassiou and your post also tend to confuse observed OSR (as measured by ERBE over 5 years) with MODELED OSR (calculated from observations of the earth’s weather over 14 years). Doesn’t the uncertainty in the Hatzianastassiou’s trend for OSR include both the variability of modeled OSR (from natural variability in the weather) plus the uncertainty in the raw data used by the model? For example, if our ability to detect and categorize clouds and aerosols changed over these 14 years, then those biases will flow through into the modeled OSR.
The abstract claims that changes in tropical clouds are responsible for the modeled trend in OSR, but the paper provides no regional data to support this conclusion. Before accepting this conclusion, don’t you expect to see a robust 14-year trend in tropical clouds of appropriate magnitude to explain the observed trend in modeled OSR?
What does Hatzianastassiou’s work suggest about our ability to project OSR with GCM’s? Hatzianastassiou’s model reproduces ERBE data with a RMS deviation of 11.5 W/m^2, giving a 95% confidence interval of about 20 W/m^2 – a reasonable 7% of the overall OSR range. However, to understand how reliably GCM’s predict global OSR (and therefore climate), we need to assess uncertainty on a global, rather than grid cell, basis.
According to Table 3, ERBE observed 0.28 W/m^2 more OSR being emitted from the Northern Hemisphere than the Southern during 1985-1989. Hatzianastassiou’s model predicts 1.06 W/m^2 less OSR emitted from the Northern Hemisphere. Based on hemispherical differences (an n of 1), one might anticipate an uncertainty of about 1.3 W/m^2 (This estimate presumably could be refined by bootstrapping.) It needs to be supplemented with the roughly 1 W/m^2 overestimate for overall global OSR (101.83 vs 100.90 W/m^2), with the complication that ERBE measurement are equally uncertainty. Therefore, even when we can observe cloud cover from space, measure the optical properties of aerosols, and “cheat” by incorporating albedo data for land from ERBE (see top of p 2682); the best we can currently do produces uncertainty in global OSR that is the same magnitude as the radiative forcing from anthropogenic GHG’s. Furthermore, unlike Hatzianastassiou’s model, GCM’s rely on projected, not observed, inputs such as clouds and aerosols.
SOD: Do you disagree with this pessimistic assessment of our current ability to model OSR?
Not too much can be made of trends without an understanding of their statistical significance. Sometimes its best to involve theoretical statisticians in the review of scientific work if conclusions are being made based on analysis of data and making correlations, cause & effects, etc. The statistics can get tricky quickly so its best to consult a theoretical statistician no matter what the scientific discipline.
Of course I’m talking about having statisticians review scientific papers prior to publication, not blog postings. I think the main purpose of scientific blogs is to stimulate thinking and discussions from people of diverse disciplines. This is the best climate science blog I’ve seen.
Frank
That’s a very good point. You can see some results from climate model CCSM3 under the heading Reflected Solar Radiation in Models, On – and Off – the Catwalk – Part Two.
These are compared against ERBE 1985-1989, widely considered to be the most accurate dataset.
You can see the systematic errors in prediction even when averaged over the 5 year period.
SOD (reply to July 8, 2010 at 8:59 pm):
Thank you for directing me to your “Models, On – and Off – the Catwalk”, which I hadn’t seen earlier. I have some questions about your “common misconceptions” that I will post in Part I – where it belongs (but where it will be hard to get and find a response.)
Science of Doom,
Fig. 5b shows a 2.35W/m2 change in forcing due to albedo changes in the period (according to our understanding). It’s really a lot.
The AR4 shows a graph based on models where even the strongest radiative forcing change (GHG) barely exceeds 2W/m2 in 150 years. (references included) Albedo changes in this graph are way smaller.
I understand they don’t show something directly comparable. The IPCC graph does not have one single line for total albedo. Even so, the difference is too large.
How do we reconcile these two apparently conflicting bits of information? Is one of them plain wrong, or maybe obsolete? Maybe uncertainty ranges?
Science of Doom,
I just noticed this little exchange with cohenite:
Cohenite:
So of the 33C greenhouse temperature you say ~ 13C is from CO2?
SOD,
Well, roughly speaking. As there is 4th power relationship between temperature and energy it may be better to say 27% of the total “retained” longwave energy is due to CO2.
I might believe you if the Earth’s atmosphere were free of water vapour and clouds. The water complicates things. For example, the measured adiabatic lapse rate would closely conform to simple one dimensional theory if there were no vapours around.
I read all your stuff about “An Insignificant Trace Gas” but am unconvinced owing to inadequate accounting for the effects of water vapour.
gallopingcamel:
It’s not easy to picture all the different elements and how they fit together.
The inappropriately-named “greenhouse” effect is the “trapping” of longwave radiation from the surface to the top of atmosphere.
So if 390W/m^2 leaves the surface and 240W/m^2 leave the top of atmosphere, the “greenhouse” effect is 150W/m^2.
There are different ways to view it of course, but this is the conventional measure.
If we agree that this is what we are measuring, then it becomes quite “straightforward” to work out the effects of the various different gases.
“Straightforward” doesn’t mean easy. But there isn’t much room for doubt. What most people struggle with is that they can’t do the calculation (solving the radiative transfer equations) for themselves and there is a lot more water vapor than CO2.
It’s hard to accept someone else’s calculation when you don’t know them well.
If you’re “unconvinced” I’m fine with that, but I do like to try and explain things better, so if you had to point to a particular area, which of these would it be:
a) the “greenhouse” effect = the difference between longwave radiation leaving the surface and the top of atmosphere = 150W/m^2
(We can measure this – obviously it varies across the globe and across time, so the value is the global annual average)
b) the theoretical calculation of this effect is done using the radiative transfer equations which calculate absorption and emission at each layer in the atmosphere, using the concentration of each gas, the temperature at that level and the measured absorption at each wavelength
c) if we compare the theoretical calculations with/without CO2 and with/without water vapor we can compare the relative effect and see that water vapor is around 2.5 times the effect of CO2.
Regarding the effect of CO2: I recall a ‘realclimate” post giving an effect for CO2 of between 26% and 9%. My impression of that was that there is plenty of interference and overlap in the effects of water vapor, CO2, clouds, etc.
I get the impression from the “realclimate” article that with NO water vapor, clouds, or other greenhouse gases, CO2 would have a
150 watts*26%= 39 watt effective increase.
Just take away CO2, and water vapor, clouds, methane, etc would have a
150*91% = 136.5 wattt effect, leaving a reduction of only 13.5 watts resulting from the removal of all CO2 greenhouse effects.
Here’s the link I referred to earlier:
http://www.realclimate.org/index.php/archives/2005/04/water-vapour-feedback-or-forcing/
scienceofdoom,
Thanks for your comments which make perfect sense. The main difficulty lies in our limited understanding of how feedbacks affect the initial forcing from the radiative transfer processes.
Water in its many forms (vapour, liquid and solid) participates in some of feedbacks, both positive and negative. When one tries to put it all together the uncertainties are too great for any meaningful predictions of future climate to be made.
gallopingcamel:
Excellent to know, thanks.
So perhaps the bit I didn’t explain well is that the calculation is not a forecast/prediction/projection of what might happen in the future.
It’s just a calculation of what proportion of last year’s / last decade’s “greenhouse” effect was from CO2. Ie, out of the 150W/m^2 what value can be attributed to CO2?
The same concept applies for predictions of more CO2 – the calculation of the radiative effect “prior to feedbacks” is quite solid. After “feedbacks” the uncertainties are much much greater.
But the uncertainties aren’t in the radiative effect of CO2, they are in what happens to water vapor, clouds etc.
Though those feedback uncertainties are heavily constrained by physical limits, past records and experimental limits on effects possible in an earth-like atmosphere.
The additional problem is that uncertainties are not helpful.
It’s as likely to be a doubling or less (therefore 2C per doubling of CO2 because of feedbacks) as to be more than a fourfold increase (4.5C per doubling). Additionally, it can’t be less than 1C per doubling because the feedbacks have to be positive. All we can say about the upper end is that it can’t be infinity. Or beyond.
And at 1C per doubling BAU still leaves us boned (2C warmer will leave us 7m higher sealevel, because it did last time it was that warm and we haven’t built up the continents out any further…). But just as likely is 5C per doubling which means we’re already boned because if we drop to zero CO2 production, we’re still going to see 2C warming and we won’t get cooler enough to rebuild the land ice for centuries after.
So the uncertainties go from
We’re boned under BAU by 2100
to
We’re boned now (and we HAD BAU, so this is BAU too).
So the only way out is to get off BAU and hope that the feedbacks are at the bottom end.
Mark,
Your assumption that the net feedbacks have to be positive is not based on any supportable fact. Almost all feedbacks from almost all other natural processes are negative unless a system is balanced at an unstable location.
The theory is based on CO2/GHGs controlling the absolute level of water vapour almost one-for-one. No CO2/GHGs and there is almost no water vapour as well.
The 3.0C per doubling formula (at extremely low levels CO2 levels) indicates that almost the entire 33C greenhouse effect is accounted for by CO2/GHGs. [Now we not sure whether the logarithmic nature stays intact at very low levels, but it shows that the nature of the 3.0C per doubling proposition results in nearly complete influence over water vapour levels as well].
So, it doesn’t make much sense to separate the greenhouse effect into water vapour and CO2/GHG components unless you also allow water vapour to vary independently of the CO2/GHG temperature-induced level.
I think the log relationship doesn’t hold for CO2 below 20ppm, where it becomes a linear rather than log relationship.
From memory…
Mark,
I went as far as to list Modtran’s results ppm by ppm until some 50ppm.
It looks like the part below 20ppm is not only nonlinear, but its relatonship to radiative forcing is “more than logarithmic” (everytime you double concentration the effect is less than the previous amount).
Alex, I think that’s because MODTRAN is doing more than you think.
Think of it this way: Look at it like a photon looking from the bottom to outer space. Each absorber is a black hole, swallowing you up if you fall into it.
If there are very few CO2 molecules then when one absorbs it won’t be blocking flux toward other CO2 molecules. Adding another CO2 molecule would therefore add more absorbers without interference.
Then when they start to become significantly obscuring, the chance that one CO2’s “black hole” will fall over the black hole of another CO2 molecule, you’ll get a less than linear response until there’s nothing but black holes.
This is the “saturated gas” argument.
Why it falls down is because the size of those black holes depends on what energy photon you are. Further away from a peak absorption line the black hole is smaller, falling down as the log of the frequency difference (at least far enough in the wings anyway). Therefore, though the peak IR is blocked, the output of the earth is not monochromatic, so more photons that would have fallen between the gaps of the black holes will find themselves caught.
There will be a transition between linear and logarithmic and that will start as linear, become less than linear but more than logarithmic, then become logarithmic (or as close as we can be bothered with).
MODTRAN could be including pressure broadening and scale height influcences which aren’t considered here, but for small changes, there should be a log dependent and a linear dependent part with a variant mix of both joining them as CO2 proportion goes up.
I have a question on Albedo. . .
I have seen charts of the recent ice ages showing temp vs time and these show a saw tooth pattern with a temp change of the order of 6 to 8 C.
I have seen an analysis of the temp change which goes along the following lines:-
– CO2 doubled which gives a forcing of 4W / m2 which will raise the temp by 2C
– given that actual temp is rised by 3 to 4 times this amount this must be do to positive feedbacks?
– hence doubling manmade C02 will cause a 2C rise plus 3 to 4 times this amount in positive feedback?
Guess if this is true we are in deep do do!
Now to my question. From your article on Albedo I believe you indicate that a 1% change in Albedo (from ~30%) would have a similar forcing to a doubling of C02 i.e. ~4W/m2). Now I am guessing during these previous ice ages much more of the planet would be covered in ice and hence the Albedo would be vastly different from today? Can you estimate the change in Albedo between these previous ice age states and the forcing this would create? How much of the 6 to 8 C change is the change in Albedo likely to account for?
I suppose I should try and do the calculation myself (I will give it a go when I get chance) as it seems to be relatively simply geometry?
SOD – the site I come to when I want a geniune unbiased view!
Just attempted a quick calc:-
– assuming current ice cap is at 70 Deg Lat
– Ice Age 40 Deg Lat
– Considering Northern Hemisphere only I make the changed ~ 10% of the area of a 2D circle.
– As this is mostly land mass I have assumed a change in Albedo from 0.4 to 0.8
– this gives a ~ 4.5 percentage points change in Albedo 30 to 34.5 ?
If (big if) this is correct does this not explain the whole of the ice age temp change?
Cheers
Neil:
It will be an interesting calculation to try. Probably many people have already done it, so there might be papers on it.
The much bigger question – and plenty of papers attempting to propose a solution – what caused the earth to go into the ice age?
I agree. I guess we can rule out CO2 as according to the charts I have seen it was at relatively low levels at the “tipping point” and did not start to rise for some time after? I have also heard that the Mak cycles don’t seem to have enough Umph?
Another equally interesting puzzle given the obvious positive feedback from ice / albedo – what stopped the rise in temperatures? Given that retreating ice wall amplifyer the effect (though less as more ice disappeared) and that by this time the Co2 effecting was “kicking in”?
If (always a big if) CO2 did not cause the Ice Age and assume did not stop it – was there any significant effect or was it simply a “by product” that has a marginal effect? However, this does not preclude AGW being an issue now.
Another topic on a similiar theme. It is believed that at some point in the past we had “snowball” earth and a possible cause for this ceasing was CO2. (Volcanos pumped CO2 into the atmosphere which could not get disolved into the oceans – as they were covered in ice so concentrations kept increasing until they caused the ice to mely etc). This sounds pausible. But what concentrations of CO2 would be needed to counter the Albedo effects of Snowball earth? e.g. doubling CO2 ~4W which is similiar to 1% change in Albedo, but wouldn’t Snowball earth change Albedo by several tens of 10% in Albedo?
Best Regards
“I guess we can rule out CO2 as according to the charts I have seen it was at relatively low levels at the “tipping point” and did not start to rise for some time after?”
The *start* was milankovich cycles.
But after ~800 years, the warming was from the CO2 and its feedbacks (like, for example, polar ice removal).
The insolation changes are not enough to explain the temperature differences and if the ice albedo reduction were enough of a feedback, we’d be in SERIOUS trouble right now, since this feedback doesn’t care WHY it’s getting hotter, merely that it does, and so the 1C warming from a doubling of CO2 would be increased by that same feedback to 4, 5, 6 or more degres C per doubling. That 1 C per doubling is from extremely complex and complete (and immutable without being EXTREMELY obvious in other unrelated fields like semiconductor lasers in your BluRay player) models of radiative transfer and quantum mechanics.
And if there are countering negative feedbacks reducing CO2’s effect, these also don’t care where the warming is coming from, therefore they’d counter this albedo change feedback too.
The paleoclimate work done by Mann et al go into this sort of thing in the paleoclimate, and there’s an entire chapter in the WG 1 part of the IPCC report:
http://www.ipcc.ch
“Volcanos pumped CO2 into the atmosphere which could not get disolved into the oceans – as they were covered in ice so concentrations kept increasing until they caused the ice to mely etc”
Actually it was because the long term sink for carbon is in weathering of rocks. If the rocks are covered with ice, they can’t weather and take up carbon.
Again, the paleo work by Mann and others (referenced in WG1) will lead you to the papers of interest in this.
Not sure who the rocks would “weather” in snowball earth as they would be under 1km of ice – unless I missed something?
Cheers
They didn’t.
That is why CO2 built up because rocks WEREN’T weathering.
Mark,
Firstly apologises for not responding sooner – vacation.
Second, apologises for not taking the time to read your reply more throughly (I was in the airport waiting for my plane). On doing so I believe we are effectively saying the same thing. i.e. in Snowball earth there is no where for the CO2 to “go” so it keeps building up etc.
However, since the albedo change in snowball earth would be “huge” I wonder how mcuh CO2 would be needed to overcome this forcing? I suspect a considerable amount?
Lastly, I understand that the weathering of rocks is simply a theory to explain how the CO2 is removed from the atmosphere? At first glance (with admitedly no further analysis) this does not seem particularly plausible? This does not mean it is incorrect (CO2 causing global warmer falls into a similar catagory but basic physics as admirable demonstrated on this site show this to be true). Would not the return of vegatation be a more plausible cause? or the long term effects of plants and animals dying and the subsequent calsification of the remains be a better model?
Food for thought.
Regards
“- Considering Northern Hemisphere only I make the changed ~ 10% of the area of a 2D circle.”
PS the land of the midnight sun doesn’t get much sunlight on the ground since the ground is at a slant. This significantly reduces the effect.
When the axial tilt is greater, then the effect is greater and this is one of the three main cycles of the Milankovich cycle that explains past ice ages where no other explicit force (like supervolcanoes, asteroid strike or coal burning humans) was extant.
What helped the antarctic was that large landmass that moved to slowly cover the south pole. That stopped a lot of not-frozen water from getting to the pole and moderating the deep winter cold like it can in the North.
You might want to consider the “2D circle” and re-think your comments about ground slant? It is basic geometry. I re-did the calcalutions with these assumptions:-
– Ice Age Min – 75 degs lat (many this is too extreme but this end has far less affect)
– Ice Age Max – 40 deg (Madrid mid Spain
– Northern hemisphere only
= This equates to about 10% of the area of the “2D Circle”
– I then assumed the change in Albedo was 0.4 if this moved from current to ice.
– I assumed this is all land which it is not. However, effect would be more pronounced if it is sea, the this would probably not freeze so far south?
Result = 4.4% change in Albedo?
Happy for someone to point out errors in my calc?
Cheers
The earth isn’t a 2D circle, however.
It’s why you divide the solar flux by 4 to change from the 2D disk the flux from the sun makes at 1AU to the 3D sphere that the earth is recieving that radiation upon.
I.e. the change at 40N to 42N is much greater than the change 70N to 72N. Even if you use a “per square meter” figure, since that square meter will present only Cos(lat) square meters to the sun to receive or reflect solar radiation.
Additional reinforcement comes from reflection being dependent on the angle of incidence.
Therefore the feedbacks would be large early on in the exit, but less later on.
That is why CO2 comes into the picture.
“Happy for someone to point out errors in my calc?”
I did, but you didn’t like the 3D sphere bit.
Marc,
Apologises again but I do not follow you 2D to 3D argument? The divide by 4 is simply to convert energy impinging on the earth which is 2D to the whole of the earths surface which is 3D. You consider Albedo after this – check of Earth Eng Budget P1?
Here’s a theory (just a theory nothing more):-
– Mak cycles cause earth to heat up maginally when in Ice Age.
– Feedback mechanism caused by reduction of ice heats plant up ~5deg and we lose the ice sheets until this effect “runs of stream” due to geometry factors.
– either during or at the end CO2 released by addition temp increases temp ~1deg.
– when CO2 levels return to normal due to:-
o rocks
o plants
o something else
o any of the above
this starts to cool plant by 1deg
– feedback mechanism caused by increasing ice do to cooling caused by CO2 loss causes plant to cool by 5 deg
– Natural limits of feedback reached due to physical properties stop at normal ice age limits (N.B. This last one is real flakly!)
Now this is only a theory and it is relatively simple but, at least on first passing could explain what happened? Does this mean this did happen – NO. But would not this be a good starting point? i.e. think of a simply model and try and find ways to disprove it?
Even if the above it true this in no way means that we have not got an issue with CO2 now! Current levels are way beyond these previous cycles and we are into a completely different scenario. For me all I would take from the past ices ages are that small changes can have profound impacts on the earth an we should be careful with the “experiment” we are currently conducting by pouring huge amounts of CO2 into the atmosphere.
I look forward to our continued dialogue.
P.S. Went to California for vacation – had a great time!
Perhaps you’ve addressed this; we all get to the same place at different times.
The radiative forcing for the albedo change 2000-2004 (+0.9 W/m2) is about 0.37C*. A different NASA article I just read noted a change of about +2.4W/m2 from 1977 to 1997. All this seems enough to account for the 0.7-0.9C* increase since 1977. Mr. Watts et al work indicates that perhaps 0.3-0.4 C* of the Hansen/NASA/GISS/NOAA increase in global temperatures since then is incorrect temperature corrections. This leaves NOTHING for CO2. Even saying each calculation is optimistic in its own effects, it leaves very little for pCO2.
Am I missing something? Is the change in Earth’s albedo since 1977 as measured by satellites, and some modest reduction due to confirmation bias in the data adjustments, equal to the noted temperature increases in the global system?
The IPCC did not find albedo changes sufficient, if by that I take Hansen’s paper on the subject to reflect IPCC thought. The number 0.9 or 2.4 W/m2 seems to contradict that position that solar/albedo changes are important.
Doug Proctor:
Do you have a reference for the NASA article?
SOD,
From my reading of your articles I have concluded that the radiative forcing of doubling CO2 is around 3.7 and this would correspond to a 1.2C rise in temp. What happens if we keep adding more CO2? Does this doubling relationship still hold? By my caculations if we start with 0.02% and keep doubling after 13 times we would be at 80%+ of the atmosphere would be CO2? Would this equate to a forcing of 3.7*13 and a temp rise of 1.2*13 (15.5 deg)? Or am I completely off base here? As a curiousity – what would be the radiative forcing and Temp rise if the atmosphere was 100% CO2?
Still trying to conceptualise CO2 causing “snowball” earth to melt? By my rough cut calculations the reduction radiation being absorbed by the earth due to Albedo changes would be far greater that this (double? i.e 30C).
I wonder if it would snow in Snowball earth? i.e. if it does not snow that would the ash from the volcanos eventually “discolor” the snow and reduce albedo?
Many thanks
Neil:
The log relationship for radiative forcing from CO2 is an empirical fit based on models that have been run up to about 1000ppm – as far as I know.
It’s a good rule of thumb in the absence of any other information but important to remember that it isn’t some fundamental law.
As for the ancient past and snowball earth, these are difficult questions to figure out – all of the turning points are challenging.
See the lightweight Ghosts of Climates Past that one day may get something useful added.
SoD
Many thanks – in the process of downloading AR4. Figure it is about time I took a look! Off to Russia next week so should have plenty of “airtime” to digess.
Regards
Sod
FYI – Interesting article on “Snowball Earth” if you have not seen this http://www.space.com/scienceastronomy/snowball-earth-ice-age-winters-101104.html
Regards
Does anyone have a link to an up to date graph of earth albedo? There is a good one at this nasa link but the data only goes up to 2005. Is there anything more up to date?
I am particularly interested in the apparent increase in albedo since 1998. Has it been sustained in the last few years?
RobB:
You can access the CERES data yourself – at http://ceres.larc.nasa.gov/order_data.php.
It takes a bit of trial and error to find the right product. I have previously spent a bit of time finding what I wanted – for example, in Understanding Atmospheric Radiation and the “Greenhouse” Effect – Part One.
[…] For example, the % of reflected solar radiation is now known to be quite close to 30%. That equates to around 103 W/m² of solar radiation (see note 2) that is not absorbed by the climate system. Compared with the emission of radiation from the earth’s climate system into space – 239 W/m² – this is significant. So we might ask – how much does this reflected % change? How much has it changed in the past? See The Earth’s Energy Budget – Part Four – Albedo. […]
[…] For example, if the albedo had reduced from 31% to 30% this would produce an increase in radiative forcing (prior to any feedbacks) of 3.4W/m2 – of similar magnitude to the calculated (pre-feedback) effects from “greenhouse” gases. …. Science of Doom […]
[…] Link January 18, 2014 sarahaworth Leave a comment […]
Wow! At last I got a weblog from where I can genuinely obtain helpful data concerning my study and knowledge.