Guest Post by Willis Eschenbach
One of my great pleasures is to come across a new dataset. Turn me loose on new observations of this magical world, and there’s no telling where I’ll end up. Thanks to a recent article here on WUWT I got to thinking about water vapor. Some research found the RSS 1° gridded “total precipitable water” (TPW) dataset. Total precipitable water (TPW) is the mass (or sometimes the depth) of water in a 1 metre by 1 metre column from the surface to the top of the atmosphere, if it all fell as rain. The RSS dataset has the TPW (for the ice-free ocean areas only) since 1988. Figure 1 shows the average values, in kilograms of water per square metre. Note that the RSS dataset only covers the ice-free oceans.
Figure 1. Total Precipitable Water.
Now, there are a few interesting things about Figure 1. First, you can see why they call it the “wet tropics”. There’s lots of water in the air.
Next, the horizontal red band just above the equator delineates the effect of the band of thunderstorms perpetually boiling along the length of the inter-tropical convergence zone (ITCZ).
You can also see why CO2 is called a “well-mixed” greenhouse gas, and water vapor is not. The amount of water in the air varies from the poles to the tropics by more than an order of magnitude.
Seeing Figure 1 made me think that I could estimate the change in the poorly-named “greenhouse effect” due to a given change in water vapor. Ramanathan proposed that the magnitude of the clear-sky atmospheric greenhouse effect could be measured as the amount of upwelling longwave radiation (ULR) from the surface that is absorbed by the atmosphere. Ramanathan also observed that the variation in the strength of the clear-sky greenhouse effect was an effect of the variations in water vapor.
To show the close relationship between variations in the atmospheric absorption of the surface radiation, and the total water vapor seen in Figure 1, Figure 2 shows the atmospheric absorption as revealed by the CERES data:
Figure 2. Average atmospheric absorption of upwelling surface longwave radiation, clear-sky CERES data. Calculated as the amount of longwave (infrared) emitted by the surface minus the amount observed at the top of the atmosphere.
Seeing those two figures gave me the idea that I could actually measure the amount of change in downwelling radiation from a given change in precipitable water vapor. So here is a scatterplot graph relating the two:
Figure 3. Scatterplot of Total Precipitable Water (logarithmic, horizontal scale) versus Atmospheric Absorption (vertical scale). Dashed vertical line shows global average value. Dotted lines show the range of the global average value over the period.
This is quite an impressively tight result, particularly given that the two variables (absorption and TPW) are from totally different datasets. I note that this is experimental validation of the IPCC’s statement about the underlying physics, viz:
The radiative effect of absorption by water vapour is roughly proportional to the logarithm of its concentration, so it is the fractional change in water vapour concentration, not the absolute change, that governs its strength as a feedback mechanism. IPCC AR5 WGI Box 8.1
More than just validating the IPCC claim of a generalized logarithmic relationship, however, this has allowed us to actually quantify the relation between the two. It also allows us to differentiate that relationship in order to determine the slope of the atmospheric absorption as a function of water vapor. That slope turns out to be 62.8 / TPW. At the average TPW value in Figure 3 of 29 kg/m^2, this gives us a slope of 62.8 / 29.0 = 2.2 W/m2 increase in absorption per kg/m2 change in TPW.
That is to say, we get a bit over two watts per square metre of increased absorption for every additional kilo of atmospheric water per square metre.
Now, that is an interesting finding which we can combine with the following look at the change in global average total precipitable water since 1988:
Figure 4. Decomposition of the total precipitable water data (upper panel) into the seasonal (middle panel) and residual (bottom panel) components.
Some things of interest. First, in the bottom panel you can see the effect on TPW of the El Nino episodes in 1997/98, 2010/11, and 2015/16. You can also see that we haven’t quite recovered from the most recent episode.
Next, there is a clear trend in the TPW data. The total change over the period is ~ 1.5 kg/m^2, centered around the long-term mean of 28.7 kg/m^2.
And utilizing the relationship between water content and atmospheric absorption derived above, this indicates an increase in downwelling radiation of 3.3 W/m2 over the period.
Now, please note that this 3.3 W/m2 increased forcing from the long-term increase in water vapor since 1988 is in addition to the IPCC-claimed 2.3 W/m2 increase since 1750 in all other forcings (see Figure SPM-5, IPCC AR5 SPM). The IPCC counts as forcings the long-term changes in the following: CO2, CH4, Halocarbons, N2O, CO, NMVOC, NOx, mineral dust, SO2, NH3, organic carbon, black carbon, land use, and changes in solar irradiance … but not the long-term changes in water vapor.
This leads us to a curious position where we have had a larger change in forcing from water vapor since 1988 than from all the other IPCC-listed forcings since 1750 … so where is the corresponding warming?
Sunny today, I’m going for a walk …
w.
My Usual Request: We can minimize misunderstandings by being specific. If you disagree with me or anyone, please quote the exact words you disagree with, so we can all understand the exact nature of your objections. I can defend my own words. I cannot defend someone else’s interpretation of some unidentified words of mine.
My Other Request: If you believe that e.g. I’m using the wrong method or the wrong dataset, please educate me and others by demonstrating the proper use of the right method or identifying the right dataset. Simply claiming I’m wrong about methods or data doesn’t advance the discussion unless you can point us to the right way to do it.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I suspect some of our alarmist friends will point to this and say ‘there’s the positive feedback that we said would happen’.
But it’s not positive feedback, as it has been happening during the “pause” in warming …
And in either case my question remains—where is the projected warming claimed to result from the 3.3 W/m2 increase in water vapor forcing?
w.
“where is the projected warming claimed to result from the 3.3 W/m2 increase in water vapor forcing?”
Willis, I suppose you yourself think that the rise in temperature is nihilized by the upwards energy transport by the extra H2O, inclusive a possible rise in the upward velocity of the rising water vapour / condensed air.
Got to agree with Willis on this one. There is a rise in water vapour, but that rise was not ‘forced’ by an increase in temperature driven by an increase in CO2. Quite the opposite. With no rise in temperature at all, particularly in the tropics where most of the water is, the total water mass has increased. Whodunnit?
Whatever the mechanism, it appears that an increase in water vapour concentration has caused an increase in CO2 concentration. Mankind spews a very large mass of water vapour into the air each year – a huge amount. Is this the first identifiable fingerprint of man-made global whatever-it-is?
It is obvious that an increase in the total water mass is going to create more clouds sooner or later. So Willis, is there a link between this water concentration data and your cloudiness data, say, in the tropical zone where it matters the most?
If there is 3.3 W/m^2 additional forcing from water vapour perhaps there is at least a corresponding loss due to shading by the clouds formed earlier each day because of the additional water vapour. That also seems obvious. We might soon have a good understanding of the temperature governor mechanism and the reason(s) for its 8 month delay.
As I see the graph, I think warming and TPW are pretty well aligned. Both goes up until about 2001, then have a pause until 2014 and then goes up again.
Why not do an analysis of this?
You may for instance make a scatterplot of warming vs water vapor
/Jan
Water vapour increases when temp increases, Temp increases when cloud decreases or during el nino.
The lack of warming is caused by convective cooling which increases exponentially with temp.
simple.
There’s another effect, where the incremental latent heat removed by increasing evaporation offsets the incremental solar input, keeping ocean temperatures from rising much above 300K. You can see this in a scatter plot of surface temperature vs. water content, where water content becomes nearly asymptotic to a temperature around 300K and which is most pronounced over the oceans.
All this evaporation is the driver of the band of tropical thunderstorms, which like a Hurricane, leaves cooler water in their wake demonstrating that weather has a net negative feedback like influence. I say ‘feedback like’ because the IPCC’s understanding of feedback systems is horribly broken.
Outside my door. It is currently 98F with a dew point of 76F.
There isn’t a 3.3 W/m^2 increase in WV forcing because WV is not well mixed in the atmosphere. Increasing precipitation where there had already been a lot of precipitation does almost nothing to increase the forcing, maybe it even decreases it due to the increase of latent heat transport?
The energy from the forcing is reflected back to space by increased clouds?
Increased clouds would count as increased precipitatable water, right?
Like Lindzen’s iris effect?
A negative feedback, stabilizing our wonderful earth.
They say it went and hid in the ocean, I think.
Cispin,
“It is obvious that an increase in the total water mass is going to create more clouds sooner or later.”
You would think so, but water increases monotonically with temperature, while cloud coverage increases monotonically up to about 0C and then starts to decrease with temperature until it starts to increase again near the equator. The reason is because the fraction of the surface covered by clouds is Lindzen’s iris that adapts to the needs of the system in order to achieve radiant balance. The reason for the change at 0C is because the characteristics of the system changed, where below 0C, ice reflects the same as clouds, thus the only effect of incremental clouds is to warm the surface. Above 0C, clouds now reflect more than the surface and incremental clouds decrease the average temperature. An interesting point is that while the cloud area decreases with increasing temperature, cloud volume increases linearly with water vapor which increases monotonically with temperature. There are many scatter plots showing the sensitivity of almost every climate variable to every other climate variable as measured using the ISCCP weather satellite data set.
http://www.palisad.com/co2/sens/index.html
Humidification lowers air temperature. The Arizona (or other areas of the globe) monsoon is all the evidence you need. That or a psychrometric chart. Might not be the whole story, but has to be part of it…
Willis
Thanks for fascinating analyses.
With the increase in precipitable water, there appears to be a corresponding increase in precipitation.
See: Global Dimming and Brightening Wild 2015. Inset graph ~ slide 25?
A quick search brought up:
Improved Historical Analysis of Oceanic Total Precipitable Water Thomas M. Smith, Phillip A. Arkin
DOI: http://dx.doi.org/10.1175/JCLI-D-14-00601.1
Wang, X, Zhang, K, Wu, S, Fan, S and Cheng, Y 2016, ‘Water-vapor-weighted mean temperature and its impact in the determination of precipitable water vapor and climate change analyses’, Journal of Geophysical Research, pp. 1-20. (Paywalled but looks interesting.)
Happy hunting.
Willis,
“But it’s not positive feedback, as it has been happening during the “pause” in warming …”
I second what Jan said. It looks to me like you have two peaks associated with El Ninos and two plateaus, one before the earlier El Nino and one at a higher level between the two. I think that is a lot like the temperature record. Might be worth plotting the water vapor forcing vs. temperature and compare the slope to the IPCC feedback.
“And in either case my question remains—where is the projected warming claimed to result from the 3.3 W/m2 increase in water vapor forcing?”
But you can not tell the difference between warming due to water vapor and water vapor due to warming. You can only see if the slope is correct. The IPCC mean is 1.6 W/m^2/K (Table 9.5), so that would take 2K warming, which has not happened. The question then is: What are your error bars like?
where is the projected warming claimed to result from the 3.3 W/m2 increase in water vapor forcing?
LWIR emmissions are ‘isotropic’, that is they act in all directions equally. In other words, the IR photons shoot off in every direction, not just down. Thus, if down-welling LWIR is increasing then so must up-welling LWIR. The latter is lost to space. The absorption and radiative properties of water vapor and CO2 are the same in that increasing either gas in the atmosphere increases the amount of LWIR lost to space which logically must cool the atmosphere. The basic premise of looking for a temperature increase from CO2 or water vapor forcing would seem to be backwards
Water vapor is different than CO2 however in that its radiative effect is dwarfed by its phase change effect. A cloud containing water droplets and water vapor will resist temperature change – either upwards or downwards due to evaporation or condensation absorbing or releasing latent heat. To understand the temperature impact of water vapor in the atmosphere, you also have to know how much water in the form of water droplets is present. I don’t think you have that data.
“projected warming claimed” what do you define as warming? A rise in the energy content of air near the ground or the temperature of the air near the ground? This is my biggest issue, CAGW likes to mix two different things depending on what gives them their desired answer. I know you are not intending to do that here, but it is an important distinction, the higher humidity has more heat content but not necessarily higher temperatures as you already know. So which would one are you referring too.
NASA 2015:
“Increased Rainfall in Tropics Caused by More Frequent Big Storms
A new study based in part on NASA satellite data has shown that an increase in large, well-organized thunderstorms is behind increased rainfall in the wettest regions of the tropics.
(….)
“The observations showed the increase in rainfall is directly caused by the change in the character of rain events in the tropics rather than a change in the total number of rain events,” said lead author Jackson Tan, who conducted this research while at Australia’s Monash University but now works at NASA’s Wallops Flight Facility, Wallops Island, Virginia. “What we are seeing is more big and organized storms and fewer small and disorganized rain events.”
(….)
The study helps chip away at one of the big questions facing climate change science: To what degree will a warmer world accelerate the water cycle and patterns of rainfall and drought? In particular, this study revolves around what scientists call organized deep convection – in short, large thunderstorms. These storms make up about five percent of the weather systems in the tropics but are responsible for about 50 percent of tropical rainfall.
While this study does not delve into what’s causing the increase in large storms, it does reveal a tight correlation between this trend and increasing rainfall.
Analyzing rainfall and cloud data from 1983 to 2009, Tan found that these large storms are still producing similar or even less amounts of rain on average, but they are happening more often.
(….)
Co-author George Tselioudis, a researcher at the NASA Goddard Institute for Space Studies in New York, authored a paper in 2010 that defined trends in organized deep convection in the tropics and sparked Tan’s curiosity on what was driving those trends.
“This work changes our perception of why tropical precipitation is increasing,” Tselioudis said. “We thought it was because the warmer atmosphere holds more moisture and, therefore, when storms occur they rain more. But that doesn’t seem to be the case. Instead, the warmer tropical atmosphere becomes better organized to produce large storms more frequently.”
The researchers analyzed rainfall data from NASA’s Tropical Rainfall Measuring Mission (from 1998-2009) and from the Global Precipitation Climatology Project (from 1983-2009). These rainfall data were compared to cloud data from the International Satellite Cloud Climatology Project for the period from 1983 to 2009
While climate models have long predicted that a warmer world would see an acceleration of the water cycle and changes in rainfall patterns, models have not as of yet produced this change in the frequency of organized deep convection.
The new finding also raises an intriguing next question: Why are these weather systems organizing in the atmosphere more frequently? That’s not yet known, Tan and Tselioudis said
(….)
(….) There has to be more emphasis on the dynamics. If you look at the grand challenges of figuring out the water cycle – this is where the emphasis is moving to.”
Source: http://www.nasa.gov/press/2015/goddard/march/increased-rainfall-in-tropics-caused-by-more-frequent-big-storms-0
First off – thanks Willis for this. An interesting and new perspective on water vapor. That is always a good thing.
Do US weather balloon data give us enough information to get TPW for north America? I would like to see this same analysis done there.
Bernie
Realtime is being modeled… Will take,some time to see where the source data are.
https://earth.nullschool.net/#current/wind/surface/level/overlay=total_precipitable_water/orthographic=-101.82,25.47,403
Yes that is nice but will need the source data for sure. Thanks.
Bernie
Bernie,
I was going to ask a similar question. Where is the corresponding data for the land? A question that was raised in the article that ‘precipitated’ this article, was what, if any, anthropogenic effects have there been on water vapor over land?
You noted the correspondence to the ENSO.
Also note that data starts from 1988 and 1988 was the biggest La Nina on record. So the starting point (more like several months into the starting point considering there is a lag) is close the lowest TPW we should see while 2015-16 was one of the biggest El Ninos on record so the ending point is close to the highest TPW we should see.
Text versions of the data 60N-60S and 20N-20S available here.
http://data.remss.com/vapor/monthly_1deg/tpw_v07r01_198801_201606.time_series.txt
I compare the RSS water vapor dataset to the NCEP Renalysis dataset, and the ENSO and the IPCC AR5 climate model forecasts in this chart going back to 1948.
Bill Illis
That is cool, but RPC 6.0 is not a believable scenario. Is there something more believable that you can plot? If it was 2.1 or some such, it might start to provide ‘guidance’ if not prophecy.
Bill, and while you are at it, could you reposition the upper left key.. It overlays the peak of the 1983 ENSO value. Thx
Thanks for this chart. It suggests the RSS derived upward change in TPW and so WVF is partly an artifact of the time period. So the derived forcing is overestimated.
Crispin in Waterloo
3.0C per doubling is based on a 7.0% increase in water vapor per 1.0C increase. Almost every climate model uses this value (a few are at 6.0%).
So 3.0C later, all the of the RCP scenarios have a lift of 22.0% in water vapor.
Bill, Judith Curry and I had quite a go on this. 7% per 1C is canonical Clausius Clapeyron. Models do use that at the surface. They don’t at altitude. But AR4 black box 8.1 was adament that they still behave as if this was so was supported by observation, when it actually isn’t. See comment below.
7% is the value that Wentz et al reported measuring in their July 13 2007 SCIENCE paper: “How much more Rain will Global Warming bring ? ” They said it was the one result that the GCMs agreed with.
I believe this value is also derived from the Clausius-Clapeyron equation.
G
Thanks Bill, George and Ristvan.
And to think that it is this week that I am helping someone with calculating condensation on plastic-wrapped heavy objects in descending aircraft. Dew you have a formula ;~) ?
Fortunately WUWT’s Nutty Professor is backing me up.
Great article Willis
You have demonstrated and quantified the positive feedback mechanism of water vapor here.
Some warming produce more water vapor, which produce some additional warming, and so on. Not enough to get a runaway process, but enough to give a positive feedback, which confirm one fundamental element in the IPCC models.
/Jan
NO ! More water vapor equals more clouds, and less sunlight reaching the surface.
So it is a definite negative feedback. It is the principal regulator of earth’s Temperatures.
G
In fact it actually shows this.
The bulk of the water is at the equator and the bulk of the solar radiation enters there. This is how the TPW remains in the tropical region. The difference between CO2 and TPW in the atmosphere poor mixing.
The fact that the 3.3W has not boiled the world is in itself fantastic proof.
No, that is not how the thermodynamics works.
There are no doubt that water vapor is a greenhouse gas, and warm air can hold more water vapor than cold air before it condensate.
Clouds are liquid droplets or small ice crystals. These droplets and ice crystals forms when water vapor condensate. It takes more water vapor to condensate in warm air than in colder air.
That means that if the temperature could increase and the water content was held constant, the clouds would disappear.
If the temperature was constant and the water content increased, the amounts of clouds would indeed increase.
However, what is happening is that both air temperature and water vapor are increasing, and that will not necessary give more clouds.
/Jan
More clouds would indicate more radiation released at altitude – clouds can’t form unless the vapor they form from loses some energy. Less than half that energy will return to warm the lower atmosphere below the clouds. Indeed, the the formation and dissipation of virga below the clouds indicates that additional energy is being captured in the lower atmosphere and carried aloft. The system of cloud formation must be a net cooling process even with “downwelling radiation from CO2 and water vapor.
For clarity, my response above was to George E. Smith in his saying that
This is not correct. More water vapor do not necessary mean more clouds as long as the air temperature increases. Hotter air can contain more water without forming clouds. This will increase the greenhouse effect, just as the IPCC models says it does.
Figure 2 in Willis article, which is actually a direct measurement of the greenhouse effect, confirm this.
/Jan
JKA – more water vapor does mean more clouds only at a higher elevation. (The ALR is what it is.) and more clouds leads to more cooling, eventually.
Also, the greening of the earth could be, in part, due to slightly increased pptn not just CO2.
Willis – yet another thought provoking article – thanks!
BACullen writes:
Do you have any link to articles, which support this?
/Jan
Jan, which has more clouds, the Sahara or the Amazon rainforest?
Well Jan, you can have it any way you want it.
More water vapor means MORE EVAPORATION. That is the common name for the process that produces water vapor in the atmosphere.
More evaporation means MORE PRECIPITATION. That is the common name for the process that removes water vapor from the atmosphere.
In most places on earth, it is a tradition to have clouds before you can get to precipitation.
Overall, precipitation and evaporation must balance over the globe. That is to ensure that the oceans remain below us, rather than above us.
Ergo, more water vapor means m more clouds; which are a net cooling effect.
But if where you live you don’t have clouds before precipitation then that would be an exception to the general rule.
G
George says:
Sorry George, but it is not clear what you actually mean with this.
More water where?
If you by “More water” mean “more humidity in the lower atmosphere”, it is clearly not correct. More humidity in the air means less evaporation.
You cannot by pure logic deduce that a larger amount of water in the atmosphere leads to more or less flux of water to or from the atmosphere. It can be correct, but any such claim has to be based on empirical evidence. So where are the scientific articles that says that the combination of higher temperature and more water vapor in the atmosphere necessarily give more clouds?
Ha, ha, I perceive it as just funny, let’s keep it funny not nasty.
/Jan
No, that would be implying that anthropogenic CO2 is the reason for the increased water vapor, not ENSO cycles. (see Bill Illis’ post above)
NASA make the question of cloud forcing seem very simple in their intro to the subject for the general public ( such as me) :
-“The study of clouds, where they occur, and their characteristics, play a key role in the understanding of climate change. Low, thick clouds primarily reflect solar radiation and cool the surface of the Earth. High, thin clouds primarily transmit incoming solar radiation; at the same time, they trap some of the outgoing infrared radiation emitted by the Earth and radiate it back downward, thereby warming the surface of the Earth. Whether a given cloud will heat or cool the surface depends on several factors, including the cloud’s altitude, its size, and the make-up of the particles that form the cloud. The balance between the cooling and warming actions of clouds is very close although, overall, averaging the effects of all the clouds around the globe, cooling predominates.”-
http://earthobservatory.nasa.gov/Features/Clouds/
Jan Kjetil,
but enough to give a positive feedback, which confirm one fundamental element in the IPCC models.
:
How come – when it’s already been reassured that GCM’s don’t care about clouds. How could they with grid sice 100’s mile of size.
Johann writes:
Because this is about water vapor not about clouds.
Clouds are water droplets, not water vapor.
/Jan
All those “forcings” are not additive like resistances in series. Think parallel resistances with water vapor and clouds being the big ones.l
But hey, it is -clear sky- atmospheric absorption. TPW makes atmosphere unclear.
And clouds are not part of TPW… Separate metric.
Well Steve when was the last occasion that you got precipitated on by TPW without a cloud in the sky.
Clouds are the part of TPW that is already on its way to precipitating, which is an essential step to precipitation.
NO clouds = NO precipitation. (of H2O)
G
Sorry if I phrased it ambiguously. TPW and condensed water are separate measured values, on different sides of the phase transition. The cloud part of the water is no longer TPW.
By George! Steve Fraser is right, in a very specific way…
http://www.remss.com/measurements
There are separate measures for TPW (vapor only) & CLWC (cloud liquid water content).& “Rain Rate” (RR) which includes liquid + solid water (snow, ice).
Deliberately side-stepping the obvious link between high TPW & CLWC is a bit cheesy. Playing as Devli’s advocate though, seems the “best” estimate of total water in all forms is RR – CLCW + TPW.
Bah! RR is mm/hr while TPW & CLWC are mm. Adding TPW + CLWC is a better estimate of total water, but misses frozen water.
But they correlate.
Doesn’t it also matter where in the vertical column the water vapor is found? Isn’t water vapor at higher altitudes supposed to have a larger effect and isn’t it true that water vapor has been decreasing?
That is, if you have more evapotranspiration due to increased downwelling IR you get more low level water vapor but this leads to stronger convection, more clouds and more condensation which lowers water vapor at high altitudes.
This could be the reason for the lack of warming and also the reason for no hot spot. I think you really need to know the changes at various altitudes to compute a true feedback.
The different hemispheres’ monthly pattern can be seen (1997-2004 data) in Figure 11 of Wang, et al. (2006) ” A near-global, 2-hourly data set of atmospheric precipitable water from ground-based GPS measurements”; free full text = wiley.com/wol1/doi/10.1029/2006JD007529/full
Willis you have fallen into the trap of using the data before it has been suitably adjusted.
The stratosphere is cooling, the surface is warming (due to reduced clouds) the net result is an increased lapse rate and increased convective cooling that overwhelms other feedbacks. That is why OLR in increasing contrary to AGW.
To the question of where is the warming, there are three partial answers which collectively might explain the whole.
1. As Bill Illis points out, Enso versus RSS time period suggests the change in WVF is likely overstated.
2. Your many previous posts on the thermoregulatory effect of tropical thunderstorms via albedo and humidity washout. Higher TPW suggests more such activity would occur.
3. All specific humidity is not created equal in terms of forcing. What matters most for GHE is the upper troposphere where ULR can finally escape to space. Upper troposphere water vapor gets there via thunderstorm convection cells. Lindzen’s adaptive infrared iris suggested higher temp/humidity, bigger more violent thinderstorms, so less anvil moisture detrainment. He focused on negative consequences for cirrus cloud formation (since icy cirrus warm because ice is transparent to incoming sunlight but opaque to ULR), but cirrus is also a proxy for upper troposphere specific humidity. Higher TPW also suggests a dryer upper troposphere through these mechanisms. And indeed, there are both satellite and corrected radiosonde reading which suggest this drying over the relevant time period. See, for example, Paltridge 2009 for sondes and John 2011 for satellite, both discussed in the climate chapter of The Arts of Truth with several footnotes.
ristvan, my response below is much narrower. In the larger scheme of things, if there is more TPW and a higher rate of transfer from surface water to TPW to cloud condensation to precipitation, there is a good chance that the entire nominal increase in energy of DWLWIR has been used up in increasing the rate of the hydrological cycle. I think this is like your number 2.
In a dynamic system with spatial and temporal heterogeneity, there are many possibilities.
A fascinating analysis and correlation. Much more relevant than the number of pirates. Any idea if this idea has been explored in scientific literature?
But, but, but… I like the Pirate correlation. 🙁
Why is there more TPW?
Because there are fewer clouds.
Why fewer clouds?
1) Fewer plankton-sourced cloud condensation nuclei
and/or
2) Fewer galactic cosmic ray-source cloud condensation nuclei
If it is the first, we could be in part to blame. Plankton get their micronutrients from airborne dust. We increase ground cover through irrigation, CO2 fertilization, and tree planting so there is less dust going into the oceans.
If it is the second, it is a natural cycle and we have to decide if trying to control it is a good idea.
Either way, if it is a problem, it can be fixed through ocean iron fertilization: http://russgeorge.net/
2) Fewer galactic cosmic ray-source cloud condensation nuclei
Except that the cosmic ray flux has been increasing due to lower solar activity, so it is hardly your number 2)
Please help me out here Leif. If the cosmic ray flux increases, wouldn´t that create more cloud condensation nuclei, or did I not understand Svensmarks theori correct ?
vboring said:
2) Fewer galactic cosmic ray-source cloud condensation nuclei
could be the reason. I pointed out that solar activity has been decreasing and therefore cosmic rays have been increasing, contrary to vboring’s ‘explanation’.
What about the reduction of Sulphur because of the clean air acts?
Willis and mulder. Suggest exploring: Wild’s presentation with links to papers:
Global Dimming and Brightening: Decadal changes in sunlight at the Earth’s surface Martin Wild Dec 2015 AGU Press Conf.
Wild shows trends of declining sulfur and aerosols with corresponding “brightening” since ~ 1985.
Note the different SOx and brightening trends between northern and southern hemispheres.
If you find similar differences in TPW that could further evidence for that cause.
vboring,
Something else to consider: http://www.hardydiagnostics.com/articles/Ice-Forming-Bacteria.pdf
Bacteria may well be more important than salt crystals, dust, and plankton for providing condensation nuclei. Yet, most people are probably unaware of the role that they play. They can also affect the temperature at which water freezes.
Have humans had any influence on airborne bacteria? I’d say that there is a good chance that we have because of the widespread and common use of antibiotics, not only for ourselves, but for livestock.
the decrease in plankton could be due to increase in uv light reaching the oceans . uv light is a plankton killer.
The analysis was done to correlate TPW to change in outgoing radiation over clear sky, which doesn’t really tell us anything about the big picture does it? For instance, the increase in TPW might very likely effect how often skies are clear and how much heat latent heat transfer there is.
Hello Willis. I shall presume to give you a home work assignment.
The U.S. Climate Reference Network (USCRN) has 10 years of data that can be put into a graph. I believe that Mr. Watts has a list of a few still functioning, well sited stations that have been in operation for over 100 years. So if a graph of those 100+ years were in the last 10 years to match the USCRN graph or not match, that would be interesting.
Hausfalter posted something relevant on this over at Judith Curry’s on 2/9/16. IMO weak, but still relevant.
Precipitable Water? We should be able to see a large red dot over the British Isles…
I offer this recent image from the world of aviation weather services. Notice the prevalence of hail at high altitude in the hot, humid summertime. This illustrates the effect of higher water vapor content on CAPE and updraft velocity. Stronger updrafts >> more hail at altitude >> stronger upward heat delivery by liquid-to-solid-to-liquid-again phase change as hail forms, falls, and re-melts. I realize this image is for land, not oceans, but still the concept applies. In the global warming alarmist movement, in my opinion, the fixation on upwelling/downwelling radiative heat transfer distracts from the obvious power of phase change and vertical movement far above the surface. By the way, anyone can see this frequently updated aviation weather product without an account by going to www[dot]duats[dot]com.
?dl=0
Even just the average change in entropy is significant ~9kJ/kg every night as temps drop from max to min.
I found that it was TPW that I was measuring with my IR thermometer pointed straight up, and where we live we see large swings in humidity, which also shows up with the ir thermo (they just need a calibration from one of the gov IR stations).
Here’s the impact of dew point/humidity on my local temps.
The 97 El Nino altered the land based surface stations daily minimum temp, like a change in TPW would, and my climate sensitivity calculation show only one extra-tropic band with an increase in CS right when temps stepped up
It might be possible to look at global TPW prior to this step, and after this step to see if the routing of this band of Atm PW changed it’s path altering the land based stations.
The two possible effects are an increase in water vapor, or a reduction in clouds in this band that coincided with that El Nino.
interesting one Mike, how do we distinguish cause from effect there? Which is driving which?
What it appears to me is, first the oceans drive surface weather patterns, they both influence the path of weather, they also are the source of all of the water vapor that carries a lot of energy from the tropics to the poles, warming the surface while cooling along the way.
The “step” in temps after the 97 El Nino, was a recorded increase in min temps, but only for the stations in the 20-30 N latitude band. I’m not sure (yet) whether this was from a change in dew point (the path TPW took changed), or a change in cloud patterns. This change in min temps were then propagated into the global temperature average.
Why doesn’t anyone tell us that all of that temperature increase came from a change in the recorded min temp at stations in one region?
Now one question I haven’t really answered is did these changes record an actual increase in the planets temperature, or whether there really was no warming, just the land change was measured, and the reciprocal cooling where this area of increased min temps moved from was not.
I found a place to look at TPW, http://images.remss.com/cdr/climate_data_record_browse.html
it’s easy to step a month at a time and watch the generation of water vapor, 98 does have a very strong band of high TPW at the equator, but it fades, and would not be the source of a long term change in minimum temps.
If atmospheric humidity has declined how has Total Precipitable Water increased?
http://www.c3headlines.com/greehouse-gases-atmosphereco2methanewater-vapor/
http://c3headlines.typepad.com/.a/6a010536b58035970c017ee7c6621a970d-350wi
Steve, I was wondering the same thing. Can someone provide a succinct explanation of the difference?
Thanks Willis. Not sure if I should have been…but I was startled by the incredibly tight correlation between the two. I mean, I understand the theory, but to see it so…starkly…was a surprise. Thanks again.
rip
Thermalization of terrestrial radiation explains why CO2 has no significant effect on climate. See ‘EPA mistake’ at http://globalclimatedrivers.blogspot.com
The analysis at that link in no way rules out that increased water vapor could be contributing to warming. Water vapor has increased as a result of primarily, increased irrigation, cooling towers at electricity generating facilities, and increased burning of hydrogen rich fossil fuels. More water vapor in the atmosphere means more warming and possible acceleration of the hydrologic cycle.
Adding a water vapor factor will certainly help explain the growing separation between measured average global temperatures and the expected temperature decline accompanying other factors as shown at Figure 17 at the link.
The remaining issue is whether or not the water vapor driven warming will be significant enough to mitigate the otherwise expected cooling.
And utilizing the relationship between water content and atmospheric absorption derived above, this indicates an increase in downwelling radiation of 3.3 W/m2 over the period.
This leads us to a curious position where we have had a larger change in forcing from water vapor since 1988 than from all the other IPCC-listed forcings since 1750 … so where is the corresponding warming?
That increase in DWLWIR occurs at the water surface, because that is where that water vapor came from.
This is related to a question that I have often asked here, at ClimateEtc, and at RealClimate: What would actually happen at the water surface if the DWLWIR increased a little bit: some warming, some warming with increased evaporation; increased evaporation without warming? I think that the answer is not known. However, about 550 times as much energy is required to evaporate a gram of water as to warm a gram of water by 1 C, so even if there is a tiny increase in the evaporation rate, there can only be a tiny increase in temperature from an increase in 3.3 W/m^2.
Thank you again for your essay.
What happened to the Log (TPW) as shown in your Figure 3? You used linear(TPW) in your average gradient but the linear(TPW) gradient will reduce as TPW increases. What will that do to your calculations?
To Willis: Thanks for your usual great mathematical analysis. However, unless I have misread your result, you assumed linearity between water vapor concentration and absorbance. This is roughly true for small absorbances (up to about 10%), but linearity fails badly above 50% absorbance. And most of the water vapor absorption lines in the bond-bending vibration band from 1200-2200 cm^-1 are almost completely saturated (near 100% absorption), so much so that the literature doesn’t even show spectra extending above 1500 or 1600 cm^-1 (for example, the MODTRAN spectrum available at https://en.wikipedia.org/wiki/Radiative_forcing , which closely models an actual spectrum at Guam, available at http://climateaudit.org/?p=2572 ). So increasing water vapor by 7% does not increase absorbance by 7% (I estimate this is a factor of 3 too high, if we include the partially saturated water vapor lines from 200-600 cm^-1 which correspond to changes in rotational states only in the ground vibrational state). Below 200 cm^-1, the MODTRAN curve shows a match with a 220 K Planck black body emission curve, since the average translational kinetic energy at 220 K is about the same as the energy of a 200 cm^-1 photon (so constant complete absorption is followed by complete emission, producing a Planck black body emission curve). This won’t change much with a small 4% change in water vapor. From 200-600 cm^-1, the actual spectrum departs from the 220 K black body curve, indicating that absorption is not 100% complete in the entire path length from the Earth’s surface to 10 km (where T = 220 K), so the Signal measured by the satellite looking downward is due to modified Beer-Lambert absorption (modified by a small emission term, following the Schwarzschild Equation), In another Comment re water vapor, I showed that even the 7% figure is too high: climate sensitivity on doubling CO2 must include the smaller TOA emission from the 62% of the Earth’s surface that is clouded (the MODTRAN curve was calculated for a cloudless surface; you can see that the TOA emission at 300 ppmv CO2 is 260 Wm^2, fully 20 W/m^2 higher than the mean value of 240 W/m^2 needed for energy balance. The TOA emission above the clouds must therefore be 228 W/m^2 at energy balance. From this, including the extra emission from the stratosphere on doubling CO2 (which means the Earth’s surface emission doesn’t have to increase as much for energy balance), climate sensitivity on doubling CO2 (not including water vapor and cloud feedbacks) is about 0.6 degrees, not 1 degree. This decreases the Clausius-Clapeyron increase in water vapor on doubling CO2 from 7% to 4%. And as others have pointed out, increasing water vapor is likely to increase cloud cover, which produces a net negative feedback. I estimate that this negative cloud feedback is about -0.2 degrees, which cancels most of the +0.27 degree positive water vapor feedback. The net result is that the 1 + 2 = 3 degree climate sensitivity estimate is about a factor of 5 too high. It is AT LEAST a factor of 2 too high, as the logarithmic relation between CO2 and warming means that an increase in CO2 from 280 to 400 ppmv is an increase by a factor of 2^0.5146 , which means 0.5146 doublings. If one doubling really produces 3 degrees warming, then 0.5146 doublings ought to produce 0.5146(3) = 1.54 degrees, which is almost a factor of 2 higher than the 0.8 +/- 0.1 degree in the historical record from 1850 to 2015. This is way outside the error bars of 0.8 +/- 0.1 degrees. This explains why all the fancy computer models based on a climate sensitivity of 3 degrees do not match the historic record, and why they are doomed to make even worse predictions of future warming, as saturation (diminishing returns) means even less warming on increasing CO2 linearly. For example, just look at the green and blue absorption curves in the MODTRAN computed spectra (which I assume are accurate): they are almost identical, except for a slight increase in areas at absorptions due to sidebands centered at 618 and 721 cm^-1. The stated radiative forcing of 3.39 W/m^2 is only 8.9% of the total CO2 absorption ditch, which corresponds to about 38 W/m^2. This is nowhere near the 100% increase in area expected on doubling CO2, if absorbance were linear, and not severely restricted by saturation effects.
If I have misinterpreted your calculations, I’m sorry, but I do not normally spend much time reading WUWT articles carefully (most of the Comments raise valid points of disagreement).
rogertaguchi, good comment.