Guest post by Dr. Pat Michaels – reposted (with permission) from World Climate Report
A new paper just published in Geophysical Research Letters by Roger Davies and Mathew Molloy of the University of Auckland finds that over the past decade the global average effective cloud height has declined and that “If sustained, such a decrease would indicate a significant measure of negative cloud feedback to global warming.”
Davies and Molloy are quick to point out that part of the decline from 2000 to 2010 in cloud height is due to the timing and variability of El Niño/La Niña events over the same period, however, there still seems to be evidence that at least part of the decline may remain even when El Niño/La Niña variability is accounted for.
Figure 1 (below) shows the history of the effective cloud height, as determined by Davies and Molloy from satellite observations, from March 2000 through February 2010.
Figure 1. Deseasonalized anomalies of global effective cloud-top height from the 10-year mean. Solid line: 12-month running mean of 10-day anomalies. Dotted line: linear regression. Gray error bars indicate the sampling error (±8 m) in the annual average (source: Davies and Molloy, 2012).
The dotted line is the linear trend through the data as determined by Davies and Molloy and has a value of -44 meters per decade (+/- 22m). However, clearly the trend is influenced by the large negative departure centered around the beginning of 2008 that was related to a moderate La Niña event in the Pacific Ocean. To avoid the influence of the this event, Davies and Molloy calculate the difference between the cloud heights during the first and last years of their record and still find a decline of 31 m/dec (+/- 11m). Although this latter technique doesn’t fully account for the El Niño/La Niña signal in the record, it does at least give some indication of the influence of the large negative departures in the latter half of the record, and indicates that the overall decline is not simply an artifact of a single event.
The average global cloud height is linked to the average global temperature—generally, the higher the average cloud height, the higher the average surface temperature, and vice versa. The tie-in is related to the height in the atmosphere from which clouds radiate long-wave radiation to space. The higher up they are, the cooler they are, and thus the less radiation they lose to space, which means the surface stays warmer.
Davies and Molloy calculate that on a decadal basis, the radiative forcing from increasing greenhouse gases is the same as that caused by either a decrease in the total global cloud amount of ~0.3% (which would allow more short wave radiation from the sun to hit the earth’s surface) or an increase in the global average cloud height of ~19 meters (about 62 feet). All to say, that clouds play a major role in the earth’s climate and that small changes in cloud characteristics can add to (via positive feedbacks) or offset (via negative feedbacks) the warming pressure put on the climate from increasing greenhouse gases. A point well-recognized by Davies and Molloy when they write “Changes in cloud properties in response to rising surface temperatures represent some of the strongest, yet least understood, feedback processes in the climate system. “
Davies and Molloy hoped to better our understanding of cloud behavior by quantifying changes in cloud heights as determined from data obtained from the Multiangle Imaging SpectroRadiometer (MISR) carried aboard the Terra satellite. The MISR data provides stereo imaging that can be used to determine the heights of clouds. The MISR data is not perfect, as it misses very thin clouds (like high level cirrus) and very homogeneous clouds (like some cirrus from thunderstorm anvils), but perhaps its biggest shortcoming is that the period of available data is still pretty short (i.e., only begins February 2000). Nevertheless, an investigation of what data is available from the MISR instrument can provide some insight as to the variability of cloud heights and their relationship to the earth’s climate.
Which was the main purpose of the work of Davies and Molloy.
In full recognition of the limitations of the data, here is how Davies and Molloy conclude their paper, in their own words:
Finally, we note that the climate data record of [effective cloud height] anomalies may ultimately indicate a measure of long-term cloud feedback that may be quite separate from the correlations discussed above [i.e., correlations with El Niña/La Niña]. Ten years is unfortunately too short a span for any definitive conclusion, as the linear trend in global cloud height of -44 +/- 22 m over the last decade is partly influenced by the La Niña event, and may prove ephemeral. The difference between the first and last year of the decade, not directly affected by the La Niña event, is -31 +/- 11 m. If sustained, such a decrease would indicate a significant measure of negative cloud feedback to global warming, as lower cloud heights reduce the effective altitude of emission of radiation to space with a corresponding cooling effect on equilibrium surface temperature. Given the precision of the MISR measurements, we look forward to the extension of this climate data record with great interest.
According to the calculations of Davies and Molloy, the negative climate forcing from a decrease in the average global cloud amount during the past 10 years has more than offset the positive forcing from an increase in greenhouse gases from human activities. It is little wonder that the rate of global temperature rise during this period has been so paltry!
Davies and Molloy write that they “look forward to the extension of this climate data record with great interest.” We want to be the first to second that sentiment.
Reference:
Davies, R., and M. Molloy, 2012. Global cloud height fluctuations measured by MISR on Terra from 2000 to 2010. Geophysical Research Letters, 39, L03701, doi:10.1029/2011GL050506.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Peak Fog is on the way.
Another great cloud paper from 2012 is
http://meteora.ucsd.edu/~jnorris/reprints/Loeb_et_al_ISSI_Surv_Geophys_2012.pdf
It has an amazing amount of detail. On cloud top height they find a similar result but also show that the amount of high cloud decreases in warmer conditions which would generally neutralize the affect of higher clouds.
ShrNfr says:
February 9, 2012 at 3:20 pm
I would hate to have to tease out the AMO and ENSO effects along with the effects of lower solar activity.
The paper above has a first go at that. The good thing from that paper is the sense they give that the instruments in space have the potential to get to the bottom of the problem of clouds.
Alan S. Blue says:
February 9, 2012 at 12:21 pm
“Could we have a plot of the ‘ENSO Meter’ for the same time period?
Just switching back and forth from the page in the sidebar … not sure how well they correlate, really.”
Dunno. But it would look a bit like this.
http://i40.tinypic.com/e0o4fo.png
“the negative climate forcing from a decrease in the average global cloud amount during the past 10 years has more than offset the positive forcing from an increase in greenhouse gases from human activities. ”
Something wrong here. Why are the clouds declining? Because it is cooling! In the polar regions cummulus are about 600 feet; at the equator its 6000 feet or more. You cant have global warming going on with the clouds declining in height, so you can’t say increased forcings from GHG are being neutralized by negative feedbacks from declining cloud height. Have I got this right?
GeoLurking: Just curious what software you used to digitize the graphic data (of Fig 1). It looks like you’re using DPlot (great program); are you using one of their recommended program?
Robert Brown says:
February 9, 2012 at 3:18 pm
“O2 and N2 lose heat by transferring it to the CO_2 and H_2O which can then radiate the heat away, the only way any part of the upper atmosphere can lose heat.”
Not the only way. I’d be tempted to give it a pass if you had wedged “significant” in there somewhere. O2 and N2 both radiate and absorb, too, though at much lower rates. But, NO and O3 are produced in the upper atmosphere from these constituents. Production of NO causes airglow, and O3 is a GHG at relevant wavelengths.
You are also neglecting another very significant GHG: CH4, which exists in the stratosphere at significant concentrations.
Robert Brown says:
February 9, 2012 at 3:18 pm
“O2 and N2 lose heat by transferring it to the CO_2 and H_2O which can then radiate the heat away,”
I used to think that too, as indeed old text books indicate oxygen and nitrogen don’t radiate at atmospheric temperatures. Well, they don’t as a result of quantum energy level jumps. But the downward radiation is full spectrum blackbody radiation and it now seems apparent that oxygen and nitrogen do in fact radiate as a result of electron acceleration during collisions – nothing to do with quantum energy steps.
“If the CO_2 is cold when it radiates the heat away, it dissipates the heat slowly and overall temperatures underneath rise”
The atmosphere may well retain extra thermal energy, but temperatures underneath can only rise if there is an actual transfer of thermal energy downwards. No amount of downward radiation will be converted to thermal energy in cooler layers, as per Johnson’s Computational Blackbody Radiation.”
The rate at which thermal energy leaves the surface is not affected by what happens up in the atmosphere. The adiabatic lapse rate is affected by the relative humidity, not the amount of carbon dioxide.
Now, the relative humidity has been falling this last decade as we have seen in another WUWT article. This increases the adiabatic lapse rate. You might think that this would make the clouds colder, but they have been falling too we now learn. So perhaps they are still about the same temperature.
The only transfer of thermal energy to the surface is in precipitation. Thus, if the clouds are staying at about the same temperature this last decade, we should see no significant change in temperatures – which is the case – QED
Sorry – correction: No amount of downward radiation will be converted to thermal energy in warmer layers, as per Johnson’s Computational Blackbody Radiation.”
Svensmark’s theory suggests low level cloud formation during times of cosmic ray flux.
http://physicsworld.com/cws/article/news/45982
I am not convinced about that trend line in Fig1 from 2003 onwards, it looks fairly flat.
Doug Cotton says:
February 9, 2012 at 6:39 pm
“The atmosphere may well retain extra thermal energy, but temperatures underneath can only rise if there is an actual transfer of thermal energy downwards.”
There is a continual transfer of thermal energy downwards. It comes from the Sun. When it is impeded from leaving, it pools up until equilibrium between incoming and outgoing can be reestablished. That pooling up leads to a rise in temperature.
It is not unlike a dam across a river. Putting a dam across a river, which has a constant source pouring water into it, will cause the water to pool up until it overflows the top of the dam, and equilibrium is reestablished. The water level behind the dam is higher, even though the dam itself is not a source of water.
Robert Brown says.
Sorry, I call BS!
Surely, if O2 & N2 can absorb energy by conduction, they can radiate in their respective absorptive/emissive frequencies.
RGB is an appropriate initial btw as I do consider you as a white knight.
Bart says:
February 9, 2012 at 7:04 pm
Doug Cotton says:
February 9, 2012 at 6:39 pm
Since you’ve broadcast your crackpottery here for all to see, it is only fair that I take it as an opportunity to broadcast mine . 😉
This is where the analogy gets interesting. Suppose the dam is infinitely high, but it has two rows of floodgates in it. One, at low level, we will call the CO2 floodgates. One, at higher level, we will call the CH4 floodgates.
The water rises until it starts flowing out of the CO2 floodgate. But, the outflow isn’t enough to establish equilibrium before this row is saturated, so the water keeps on rising. Eventually, it reaches the level of the CH4 floodgates. Here, the water has enough of an outlet that the level stabilizes.
Now, we add more CO2 outlets. To make the analogy fit, let’s assume that we had to raise the level of the CO2 floodgates a bit to fit more in. If the CO2 floodgates had previously been sufficient to allow an equilibrium level to be established, that equilibrium level would be pushed higher.
BUT, since the level previously rose to the CH4 gates, and we put more CO2 gates in below that level which are now able to remove higher volumes of water, the equilibrium level will go down.
I hope maybe this analogy will get a read from some people who did not understand what I was talking about on an earlier thread.
Adding CO2 does not necessarily raise surface temperature, because there are other radiative emitters in the atmosphere which have interacted to create the equilibrium temperature.
Good,honest research.
Maybe the clouds are getting shorter because there is less heat to go around?
Or, there are more clouds nearer the ground?
Clouds look different in the IR at say 15 micron than they do in the visible at say 0.5 micron. Also, transparency is wavelength dependent. To save a literature search, does anyone know the wavelength used to estimate the height of clouds as reported here?
This is why computer models have no prospect of working. The behavior of clouds has a huge impact on the amount of solar irradiance received at the surface, Changes in cloud pattern alter both the albedo and the amount of solar irradiance absorbed by the atmosphere
The variations are all but infinite since variation occurs depending upon the size of the cloud both 2 dimensionally and 3 dimensionally,the composition of the cloud, the altitude at which the cloud forms, the location where the cloud forms, the time of day the cloud forms and dissipates, the day & month when the cloud forms.
This is infinitely more complex since a cloud forming say over the snow covered Himalayas will have a significantly different impact on albedo than a similar sized cloud forming over a tropical rain forest So one has to look at the ground surface albedo that is being interrupted by the cloud to get a proper appreciation of albedo change. Of course, surface albedo is often seasonal and therefore the changes in albedo change also with the season.
The latitude of cloud formation and the date and time of day is very material to azimuth issues, quite obviously a cloud forming at say 11 am and dissipating at say 4pm in the tropics has a significantly different effect to a similar cloud forming say at 2pm and dissipating at 5pm over mid Canada.
There is no such thing as average cloudiness and our estimate of cloudiness conditions need be out by just 1 or 2% to fully explain the 20th century warming.
Since the process by which clouds form, dissipate and how and why they take the composition they take is not sufficiently well understood to have a reasonable stab at modelling. . .
gyptis444 says:
February 9, 2012 at 2:23 pm
Thanks much.
John M says:
February 9, 2012 at 6:05 pm
“GeoLurking: Just curious what software you used to digitize the graphic data (of Fig 1). It looks like you’re using DPlot”
Yes, I set the params of the plot after measuring the extent of the the graphic and setting it at as the background. A little tedious, but it works.
I’ve used it for about two years now… great program. What I liked about it was it’s integration to Excel. (OT: I even managed to get the coordinator of Involcan (El Hierro) pissed off at me for my interpretation of GPS data (quadratic sheet) for his “volcanic crisis.” Had to yank a few vids off of my Youtube channel… most of the vids were done using Dplot and a video capture program)
My head spins. Is it CO2 that drives the climate? Is it sun spots? Is it total solar irradiance? Is it the pacific decadal oscillation? Is it Cosmic Rays? And on and on and on and….. Now cloud height? WUWT?
Seriously, if the climate is this complicated (or MIGHT be this complicated), what makes us think we have an idea whatsoever that we have the faintest idea of where it’s going?
I would like to embarrass gyptis444 by seconding the recommendation to watch:
http://sites.agu.org/fallmeeting/scientific-program/lectures/bowie-and-named-lectures/6dec/
g444 says “I strongly recommend this as it is extremely enlightening as to the specific details of cloud physics which have not been adequately characterised.”
I strongly recommend this for that reason, and ALSO because it implicitly indicates how much has been achieved already. In addition, it indicates how climate scientists really talk to each other, and what one of the key roles of simulation models is in the actual practice of climate science.
Indeed, the availability of the principal AGU lectures online is a wonderful thing. In future I’ll attend the less prominent sessions and watch most of the big name lectures after the fact.
re: richard verney says:
February 9, 2012 at 8:11 pm
The problem with clouds is compounded by natures trick of camouflaging a cloud’s complexity, behind such a simple outward appearance. We have been underestimating them since modern science began. As everyone knows, almost everything can influence cloud formation, and cloud formation steers climate. Smoke, cosmic rays, dust, pressure waves, electrical charge, chemistry, etc. etc. all dance within them. We have been chasing our tails for a century, trying to manipulate one factor or the other, in order to manipulate clouds (usually to induce precipitation).
As this cloud altitude observation demonstrates, it is still confounding us all today. As usual, quantification and assigning direction of effect, prove difficult. Still, the prize is worth the effort. GK
Our little watery blue ball, held captive by old Sol, and perturbed by a captive moon, and various other divers influences. Fights back to reset the thermostat for the passengers. Reset takes a while and the rampant sun experienced for a few cycles has reset our world to cold, thus the change in clouds. The rather sudden holiday old Sol decided to take will mean an over shoot of temperature toward cold. If old Sol does not wake up. there will be no reset signal to warm, and our thermostat will be set at cold. I do not like cold.
DesertYote says:
February 9, 2012 at 3:24 pm
“10 years is not enough time to indicate anything at all.”
I quite agree. You are a sound voice of caution.
But surely it’s long enough to hint at where to look?
Reblogged this on The Blogspaper.