Home/Blog
My research field of satellite passive microwave remote sensing took off like a rocket (pun intended) when the first Special Sensor Microwave/Imager (SSM/I, built by Hughes Aircraft) was launched in mid-1987 on the DoD series of weather satellites (DMSP).
We SO anticipated that first instrument…good calibration, and high frequency channels to estimate precipitation over land. The previous NASA instruments (ESMR-5, -6, and SMMR) were a good start, but had limited channel selection and less than optimal calibration strategies.
The SSM/I instrument series was later redesigned to incorporate the temperature sounding channels (SSMIS, built by Aerojet). (By the way, we don’t use these in our UAH global temperature monitoring work, since we receive very little money to produce the UAH datasets and incorporating an entirely new series of instruments would be a major effort).
But the real benefit of the SSM/I series of satellite sensors was the production of the “ocean suite” of products: integrated water vapor, surface wind speed, integrated cloud water, and rain rate. These continue to be produced by several investigators, and I use those produced by Remote Sensing Systems (RSS).
To help interpret the SSM/I measurements, let’s start with the HadSST3 sea surface temperatures (SSTs) measured since July, 1987, which is when SSM/I data first became available. (All of the following time series are monthly global anomalies since July, 1987; some have trailing 6-month averages plotted as well). It shows the well-known warming up until the 1997/98 El Nino, then roughly level temperatures since then.
Fig. 1. Monthly global oceanic HadSST3 anomalies from July 1987 thru Feb. 2014.
The first SSM/I field to address is total vertically-integrated water vapor, which closely follows the SST variations:
Fig. 2. Monthly global oceanic anomalies in SSM/I total integrated water vapor.
The water vapor variations lag the SST variations by an average of one month. A regression relationship reveals an average 10.2% increase in vapor per deg. C increase in SST. This is larger than the theoretically-expected value of 6.5% to 7% increase, a discrepancy which can be interpreted in different ways (more evaporative cooling of the ocean stabilizing the climate, or more water vapor feedback destabilizing the climate — take your pick).
Next, let’s examine the surface wind speed variations from SSM/I. These have been compared to literally millions of buoy wind measurements, and are quite accurate. In fact, I would wager these are by far the best estimate of changes in global ocean wind speed we have:
Fig. 3. Monthly global oceanic anomalies in surface wind speed from SSM/I.
We see there was a slight (1-2%) increase in ocean wind speed from before the 1997/98 El Nino to after, which at least qualitatively might be supportive of Trenberth’s claim of increase ocean heat storage and surface cooling temporarily cancelling out anthropogenic global surface warming. I have not looked into whether a 1-2% change in wind speed could have such an effect, so feel free to comment on this. Note also that the last year or so hints at a reversal of this increase back to pre-1998 wind speeds. If wind speeds remain at the lower level, it will be interesting to see if surface warming resumes. I’m making no predictions on this.
The SSM/I rain rate variations are always quite noisy. Warm conditions tend to show more rainfall, but the strong 1997/98 El Nino curiously shows little effect, and there is a hint of increasing ocean rainfall in recent years:
Fig. 4. Monthly global oceanic anomalies in rainfall from SSM/I.
Finally, let’s look at what I think is the most interesting SSM/I variable from a climate change standpoint, total integrated cloud liquid water (CLW):
Fig. 5. Monthly global oceanic anomalies in integrated cloud water from SSM/I.
The variations in cloud water show some interesting low-frequency behavior. I have previously discussed the fact that these cloud water variations are correlated with CERES-measured net radiative flux, and so provide a proxy measurement for the net radiative imbalance over the ocean which suggest some portion of recent warming was simple due to a natural decrease in cloud cover.
The updated regression relationship I get is 0.24 W/m2 loss in Net (solar plus IR) radiative energy for each percent increase in SSM/I cloud water, a scale factor we can then apply to the cloud water graph to get a Net radiative flux graph:
Fig. 6. Monthly global oceanic anomalies in Net radiative flux estimated from SSM/I cloud water variations, using a CERES-based scale factor of 0.24 W/m2 per percent cloud water change.
Why use an SSM/I estimate of CERES Net radiative flux, instead of CERES directly? Mostly because CERES is available only since 2000, whereas SSM/I is available since 1987. But also, the CERES measurements are very difficult, with the reflected solar flux (which dominates the CERES-SSM/I relationship) having a strong angular dependence. The SSM/I measurements are instead thermally-based (microwave emission) and have no such angular dependence. Finally, radiative fluxes are so important (e.g. being the basis for global warming theory) that any independent means of estimating them are worth looking into.
Be careful in interpreting the estimated radiative fluxes in Fig. 6 because they could have an offset. Since the anomalies I compute (by definition) sum to zero over the entire time series, that means the total time-integrated radiative energy flux also sums to zero. So, while the graph in Fig. 6 suggests energy loss by the global oceans over the last 5 years, it could be the whole curve needs to be shifted upward. There is no way to know. The CERES fluxes have already been adjusted to match the increase in oceanic heat content, which was a logical thing for the CERES Team to do since the absolute accuracy of CERES is ~10 W/m2, whereas the increase in ocean heat content in recent years (IF you believe the warming estimates) correspond to only a few tenths of a W/m2 imbalance. The main value in the graph is to identify possible changes over time.
Others might see some relationships in the above plots that I haven’t noticed; I’ve made the Excel spreadsheet available for those who want to play with the data.
(Note that a possible El Nino this year would temporarily dominate everything else, as would any La Nina afterward. I’m instead talking about the longer term evolution of the cloud cover of the global oceans and what it might mean for global temperatures on decadal time scales.)
It is interesting to me that post sub-solar activity which became firmly established post 2005 that the oceans show a corresponding increase trend in overall energy loss.
I see a correlation.
Roy,
You have two major problems with your technology:
1. Calibrated thermometer is the ONLY instrument used by all fields of experimental sciences to measure temperature and you have no right to re-define what the real temperature is all about
2. All your ‘calibration’ is referencing something that does not exist – global temperature
So the only ‘proof’ of your ‘invention’ is referencing non-existent number, which is contrary to basic principles of experimental sciences.
I hope this helps,
Dr Darko Butina
PS. I would love to receive your comments on my paper that deals with daily Tmax/Tmin data using state of art pattern recognition tools and also why are you ignoring the paper by Essex et al 2007 which asks ‘Does the Global temperature exist?’ with categorical answer ‘No’.
Dr. Darko global temperature does exist it is just hard to measure.
Dr. Spencer is rightly trying to show trends, which is a big part of what climate science is all about.
Salvatore,
First of all, I was asking Roy, properly trained scientist, to explain what is wrong with using the ONLY instrument that is designed to measure the ‘temperature’ and not you for your personal opinion. Secondly, anyone who knows anything about physical sciences knows that if you can’t measure this ‘global temperature’ by a instrument that it is designed to measure temperature then the ‘global temperature’ does not exist. For your education, and to remind Roy about the facts that he should still remember, the mechanism by which thermometer works is that it reflects thermal equilibrium between the molecules in thermometer (like mercury in old days) and molecules that surround it. In other words, it is measure of kinetic energy of the molecules that surround thermometer. So if thermometer cannot detect either increase or decrease of kinetic energy of the system, than that system is neither warming or cooling. And finally, the climate sciences should be about the same thing as any science should be – about the scientific truth and not looking for ‘trends’ in non-existing theoretical space.
In few words, as an old professor of physic said to his students: “boys keep in mind that a thermometer measures always the temperature of itself”
I fully agree.
Have a nice day Dr. Butina.
Massimo
If energy loss keeps increasing sea surface temperatures should follow.
I wish they would come up with a product through satellite data that would show the total monthly percentage of global albedo/cloud coverage.
Dr. Spencer that would be a great product to come up with.
So Dr. Darko, what do you suggest when it comes to finding out about global temperatures/ trends?
I CAN’T WAIT TO HEAR THE ANSWER.
Dr. Darko just throw out all the data past and present.
That is your answer the blind leading the blind.
I think the evidence of distant past and more recent past temperature changes is quite good and only getting better.
I think Ice Core data ,and the technology we have to evaluate it today ,along with present satellite temperature technology gives us a reliable way of knowing what has taken place on the globe temperature wise in the past and what is currently happening in the present.
In contrast to Dr. Butina’s conclusions.
I don’t believe in AGW, and the question to question AGW theory is NOT to take the approach the temperature data is no good, but rather come up with alternative reasons for AGW, and why the temperature has changed.
The approach that temperature data is simply wrong will never fly and is the wrong way to go about trying to prove AGW theory (which is wrong )is wrong.
Data is all there is and it has been pretty consistent, until AGW theory came about, and manipulation of temperature data started taken place. Which is an entirely different matter.
However , prior to the current global warming fiasco(1980 or so -present) temperature data conclusions from many various sources past and present were pretty darn consistent.
It is this manipulation which is the reason why it is hard to measure global temperature let along global temperature trends.
I think the satellite temperature data will put a rest to this. Thanks Dr. Spencer for this great source of temperature data.
http://stevengoddard.wordpress.com/2014/04/13/climategate-scientists-getting-rid-of-the-1940s-temperature-spike-in-the-arctic/
Thanks, Dr. Spencer. Very good article.
In SSMI-cloud-water-thru-Mar14.png it can be seen how from mid-2009 Cloud Water shoots up, not to return to its previous status, but to take a step up and remain elevated.
In http://www.drroyspencer.com/wp-content/uploads/SSMI-Net-rad-flux-thru-Mar14.png from mid-2010 Net Radiative Flux takes a dive and a step down and remains depressed.
I flipped one vertically to compare to the other and yes, they visually correlate somehow.
Well, Roy, with all that water vapour data for various locations one would think someone, somewhere would have tried to confirm that water vapour is a greenhouse gas causing warmer surface temperatures. Where’s the study?
My study showing water vapour cools is not hard to replicate. To prove me wrong you would have to produce a similar study proving water vapour warms by about 10 degrees for each 1%, as is in effect claimed by GH advocates.
The Ranque-Hilsch vortex tube provides evidence of the gravito-thermal effect. You would need to provide contrary empirical evidence.
You would also need to produce a valid (but different) explanation as to how the necessary thermal energy gets into the Venus surface in order to raise its temperature by 5 degrees during its sunlit hours.
BigWaveDave considers the gravito-thermal effect (seen in the vortex tube) worth your time thinking about …
“Because the import of the consequence of the radial temperature gradient created by pressurizing a spherical body of gas by gravity, from the inside only, is that it obviates the need for concern over GHG’s. And, because this is based on long established fundamental principles that were apparently forgotten or never learned by many PhD’s, it is not something that can be left as an acceptable disagreement.”
The greenhouse conjecture would violate the laws of physics. It is totally wrong.
The difference between you and me, Roy, is that I understand the physics which explains why water vapour cools, and why the gravito-thermal effect is a reality.
You don’t understand because you don’t understand entropy and you confuse it with enthalpy, as is common in climatology circles when they excuse decreases in entropy by claiming there’s no problem with enthalpy.
Consequently, Roy, you misunderstand why your “Misunderstood … ” article is wrong about isothermal conditions, even though the evidence that a force field induces a thermal gradient is staring you in the face in a Ranque-Hilsch vortex tube.
And because that “33 degrees of warming” is fully explained by the gravito-thermal effect (which you cannot disprove) the whole greenhouse conjecture is proved false.
Joe Postma (PSI author) still thinks that the surface temperature of the oceans (where the thin surface layer is almost completely transparent and solar radiation passes down into the thermocline) is determined by that solar radiation, even though it mostly passes through the thin surface layer of water.
It is not.
The temperature of the thin surface layer of the oceans (which plays a primary role in determining mean surface temperatures as measured) is itself determined primarily by non-radiative processes, not by radiation.
It’s so obvious that Postma is wrong about this!
Roy
Increase in wind speed will not increase ocean heat storage as Trenberth claims. The effect of higher wind speed is to lower the static pressure and increase evaporation. It cools the sea surface. This slows down convective heat transfer to deep ocean as the temperature gradient decrease. Turbulent mixing by wind occurs only near sea surface.
Cloud cover can have big impact on climate. I estimate an 85% cloud cover equals -6.6 W/m^2. At no feedback scenario equals 2 C cooling. This is a little ice age. Global temperature in the last ice age was about 12 C or just over 2 C cooler than present. But present cloud cover is below 70%
One moment you talk about the Little Ice Age, then you talk about the “last ice age.” We are still in an ice age. Do you mean the last glacial period? The periodicity of glacial periods is far more likely to be regulated by changes in the eccentricity of Earth’s orbit which occur in a cycle of about 100,000 years and are caused by Jupiter’s gravity. These changes affect the annual mean distance of the Earth from the Sun, and hence the insolation level.
Refer: http://en.wikipedia.org/wiki/Quaternary_glaciation
Dr. S,
It can heat the ocean, see some of Bob Tisdale’s description of neutral and la nina processes.
Clear skies and wind pool heat in the indo-pac and draw cool deep ocean water in the eastern pac surface. Wind moves this water to the west as the clear skies warm it.
The winds cause the warm water to pool in the indo-pac, pushing water down to depth. I think it is plausible that this pool also may interact with the deep ocean in several ways. There may be some mixing. There may be deep ocean rivers that touch this pool at times. There are many plausible mechanisms to transfer heat from the warm pool, and not just to the deep ocean. Rivers I hypothesised could just as well move water to another surface region like the tropics or arctic.
Well I dont think water vapor can be a positive feedback as during the 1998 El Nino, water vapor and temperature cooled just as rapidly as it warmed when the event was over
This raises an interesting point. I believe it’s possible that there are positive feedback processes which occur initially which strengthen a disturbance, which then switch to negative feedback as the disturbance dissipates. Also, it probably doesn’t makes sense to talk about individual feedbacks, since they interact with each other. It’s a complicated issue.
Roy: Why don’t you just do a study, as I have, comparing long term mean precipitation and temperature data in the warmest month(s) for inland tropical cities, the temperatures being adjusted for altitude? I have provided my data and results that show negative correlation, from which we have evidence that water vapour cools: it does not warm by 10 degrees for every 1% in the atmosphere, as the GH advocates claim. The reason it cools is because it lowers the thermal gradient, so the thermal plot rotates and becomes lower at the surface end, resulting in lower supported temperatures in the early pre-dawn hours. It’s not hard to understand, Roy.
Here’s the methodology, Roy. Rather than waste time with the aliens, come down to Earth and study what’s really happening here ….
Appendix – Temperature-rainfall correlation
It is a fundamental requirement for there to be a radiative greenhouse effect that water vapour and suspended water droplets in the atmosphere should have a warming effect, because these are by far the most prolific greenhouse gases. This warming effect is supposed to account for most of the “additional 33 C degrees” in surface temperatures, increasing the thermal gradient from an assumed initial isothermal (level gradient) state to one in which the surface temperature is about 30°C warmer. Then carbon dioxide and other radiating molecules are supposed to raise the temperature a little more up to a total of 33 degrees above the level gradient value. Furthermore, if carbon dioxide levels increase, it is assumed that the level of water vapour would increase as a result, and so more warming is expected, multiplying the effect of carbon dioxide with this extra positive feedback.
However, it is well known and acknowledged that water vapour leads to a lower thermal gradient, otherwise known as the “wet” or “moist” adiabatic lapse rate. Rather than the dry rate (calculated from the -g/Cp quotient to be -9.8C/Km) high levels of water vapour are known to reduce the gradient to about -7C/Km and even down to -6.5C/Km in the very humid Equatorial regions. The main argument in this book would thus suggest that, because water vapour makes the thermal gradient less steep, we should expect a lower surface temperature when the new radiative equilibrium is established. Thus it appears that water vapour should have a net cooling effect.
It seems remarkable that this apparent contradiction does not appear to have been investigated with what could be a relatively low cost study, compared with the funds that have been spent on other climate research. Because of this, the author spent just a few hours analysing temperature and rainfall data for 15 cities, in order to give an indication of how a more comprehensive study could be conducted.
It was considered most appropriate to select towns and cities within the tropics, which extend between the Tropic of Cancer (at about 23.5° North) to the Tropic of Capricorn (at about 23.5° South) because the Sun will be directly overhead any particular city twice a year. By selecting data for the hottest month this will usually correspond to the month in which the Sun passed through its Zenith, or the following month. As other variables may have affected the Northern Hemisphere, it was decided to limit the study to the Southern Hemisphere and to select the hottest month out of January, February or March, though nearly all turned out to be January. Such a selection avoids the need to make compensations for the angle of the Sun at latitudes outside the tropics.
It is noted that flat islands such as Singapore have very regular maximum and minimum daily temperatures, and this is almost certainly due to diffusion, convection and wind from the air just above the ocean surface, where the air temperature is governed by the water temperature. A similar effect occurs to a lesser extent with coastal cities, as well as with some cities that are close to large inland bodies of water. Hence it was decided not to include cities that were less than 100Km from the coast or such bodies of water.
It was also considered that there would be a need to adjust temperatures to what would be expected at a common altitude, and 600m was selected. Cities with altitudes outside the range 0 to 1200m were then excluded so that uncertainties relating to assumed temperature gradients (lapse rates) would be unlikely to exceed about half a degree at the most. It was decided to use a temperature gradient of -7C/Km for the third with the greatest rainfall, -8C/Km for the third with the least rainfall and -7.5C/Km for the middle third of the cities in the sample.
The above exclusions tend to rule out Indonesia, Papua New Guinea and other Equatorial island regions such as are found to the north of Australia. As the study was restricted to the Southern Hemisphere, it was decided to limit it to latitudes from 16.0 to 24.0 degrees south as this would include Alice Springs in Australia (latitude 23°40’S) which was considered close enough to the Tropic of Capricorn, as well as most tropical regions in Australia (AU) except those close to the Northern coastline. It also of course included a slice of both Africa (AF) and South America (SA) and, from these three continents, a total of 15 cities were selected, there being six in Australia, five in Africa and only four in South America where several were ruled out by altitude.
Cities which were within one degree of either the latitude or longitude of a previously selected city were not considered. However, once it was determined that a city met the requirements for altitude and coordinates, it was included in the study before referring to any temperature or rainfall data, so none were excluded for any “exceptional” reasons relating to such data, except for Emerald in Queensland Australia for which the source of data [3] had no rainfall information.
It is appreciated that rainfall may not be an accurate indicator of the thermal gradient, but neither would relative humidity be any better, because suspended water droplets also play a part in reducing the gradient, as does the release of latent heat when it rains.
Means of Adjusted Daily Maximum and Daily Minimum Temperatures
Wet (01-05): 30.8°C 20.1°C
Medium (06-10): 33.0°C 21.2°C
Dry (11-15): 35.7°C 21.9°C
Doug,
I’ve said this to you before, but the reason why water vapor cools is because it’s interconnected with the formation of cloud cover. On global average, cloud cover cools by about 20 W/m^2, which is a lot given clouds are far better IR absorbers than even water vapor. In addition, the removal of water vapor from the surface as latent heat has a strong cooling effect on the surface. It’s the combination of these two interconnected effects that results in net cooling.
Or it’s because the combination of cloud caused (from solar reflection) and evaporative caused cooling is stronger than the increased IR opacity from increased water vapor — that higher water vapor concentration shows a net cooling effect.
See my paper that shows evaporation at 6%/C increase cause surfsace cooling and latent heat transfer to the atmosphere from cloud formation thsat also moves more heat outward causes more negative feedback than IPCC’s positive feedback. -0.74 C / C vs +0.562 C / C feedback back factors.
http://wattsupwiththat.com/2014/04/15/major-errors-apparent-in-climate-model-evaporation-estimates/
or for a pdf email me at [email protected].
Also, increased water vapor absorbs more incomming solar radiation. Negative feedback factor of about -0.15 C / C for 6% more water vapor.
What is the average climatology of precipitable water vapor over oceans between 60S to 60N. I’ve seen a value of 28.5 kg/m2? It impacts the percentage change of course.
I plot the RSS data against NCEP reanalysis and they are very, very similar. NCEP reanalysis only provides for 4.5%/C versus the lower troposphere temperatures and slightly lower against Hadcrut4. Should the comparison be made to Ocean SSTs?
I note the latest value of TCV from RSS for February 2014 is -0.106 kg/m2. Did SSTs drop in January or February that much? Lower troposphere and surface temperatures did.
Would we not expect an increase in wind speed associated with a larger area of warm water increasing the convective nature of the tropical pacific?
Thanks for the interesting data, none the less.
Doug Cotton, do you ever notice your behaviour is not one that anyone really wants to be exposed to? Maybe a change in approach would do you good!
http://youtu.be/GDwLPMOzHLY
Really would!
This is why I like this page: there are always new perspectives to learn from. I am trying to put this into the general concept that high clouds warm and low clouds cool. Thanks for the good food for thought.
You can’t prove us wrong, Roy. Many now realise the gravito-thermal effect is a reality and you can’t prove it doesn’t exist in a Ranque-Hilsch vortex tube. There’s nothing quite like this empirical evidence which thus proves any greenhouse warming effect is pure fiction.
Wow – DC has now morhped into an “us”. Is this why he posts his nonsense using so many different aliases?
Roy W Spencer Ph.D. says:
“Also, it probably doesn’t make sense to talk about individual feedbacks, since they interact with each other.”
It definitely does not make sense. Compare what W. Ross Ashby says in “Introduction to Cybernetics”, 1957:
“Such complex systems cannot be treated as an interlaced set of more or less independent feedback circuits, but only as a whole.”
The reason for these futile discussions in separate-feedback terms, is a confusion between science as the study of practical action and as the study of nature “as it comes upon us”. Feedbacks were INVENTED by engineers and simply indicate deliberate physical connections backwards in machines. If a feedback in a machine does not “work” as expected it is simply torn out. To talk about feedbacks in Nature is merely to make a hopeful analogy to human design. As soon as it is obvious that it is a complex system – stop with the analogy!
I think the use of feedbacks is an excellent tool in estimating global warming, but I question if the climate models do it correctly. The important thing is to relate them to a dimensionless number: the change surface temperature caused by the feedback of interest for a 1 C prior change in temperature, but before the feedback “feedbacks on itself”. We call this F, the feedback factor. For a single feedback this causes an infinite set of diminishing feedbacks (assuming the value of F is less than +1; if not we get thermal runaway). Standard theory of convergence of infinite series handles this. The feedback multiplier after the feedback process is complete is M = 1/(1-F).
If F = +0.5, we get a series from an initial change of 1C, additional changes of .5 + .25 + .125 + .065 . . . which converges to 1 plus the original change of 1 for a total of 2. Using the equation for M means we don’t have to do these series. Now it is important that if there are a number of feedbacks that the feedbacks F values are combined and then a single M value is calculated. If the reverse is done positive feedbacks are incorrectly weighted higher that negative ones. A simple example shows this. Assume we have two feedbacks, +0.5, and -0.5. These would cancel and there would be no net feedbacks. However, if separate M multipliers were calculated, we get M1 = 2 and M2 = 0.667, the product of which is a final feedback multiplier of 1.33. I have seen a number of people make this mistake. They hear that water vapor feedback can double what CO2 does and they start from this value. On the other hand the effect of two or more positive feedbacks are understated if separate M values are combined.
One rare concern: Negative feedbacks that sum more negative than -1 can cause oscillation if there is enough delay in the system.
There may be special problems where feedbacks interact with out the common connection of a surface temperature change. One example is evaporation and cloud feedbacks. More evaporation will produce more clouds, but more clouds will decrease evaporation especially over the ocean where clear days can cause about 5 times the evaporation of cloudy days even though average temperature and humidities are the same. From this we conclude that cloud variation is a strong negative feedback on itself: More clouds, fewer clear days, less evaporation, fewer clouds, more clear days and evaporation, etc. If these delays were long enough oscillation could come about or show up an as intermediate type that looks like a damped, declining oscillation or “ringing”. Anyway this cast doubts on the IPCC values of high positive cloud feedback. By the way, it was not based on “warm weather makes less clouds”, but it was devised as a way to explain the temperature rises the models estimated that could not be predicted by the other feedback values See Soden and Held (2006). They call it a “residual”. It looks like “fudging” in order to make simple energy balance models agree with the climate models.
From a chemical engineers perspective and increase of around 10% for a 1 C change in SST is exactly what I would expect. Vapor/liquid equilibria is a surface phenomena whereby the sea surface is at equilibrium with the adjacent air layer not necessarily the bulk atmosphere. The removal and replacement of this layer by wind etc determines the evaporation rate. The water vapour in the bulk atmosphere is determined by mass transfer from the surface . This is determined by wind and mixing effects. Assuming a constant mass transfer coefficient the difference between the sea surface air layer and the bulk atmosphere will be constant. Say this difference is 0.0066.
SST Water conc at surface Bulk atmos.
15 0.0168 0.0102
16 0.0179 0.0113
The perc increase in the surface layer is 6.5%. The increase in the bulk atmosphere is 10.8%
A significant paper I just discovered by Dai (2006) may explain which evaporation increases close to 10% / C.
http://www.cgd.ucar.edu/cas/adai/papers/Dai-cmep-paper.pdf
It shows relative humidity (RH) drops a little with temperature and evaporation is very sensitive to RH. His data from 15,000 weather stations from 1975 to 2005 show global specific humidity rises 4.9 % / C and over the oceans at 5.7 % / C. The ocean value indicates a slight drop in RH about of 0.5% and indicates an evaporation increase about 8% / C. A drop of RH of only 1% gives evaporation up to 10% / C. The IPCC and the modelers would call both of these “consistent with constant RH”. Reduced humidity means less positive water vapor feedback and increased negative feedback from the transfer of latent heat through evaporation from the surface to the atmosphere. On the latter, see my paper that shows at 6% / C evaporation increase more than offsets present estimates of positive water vapor feedback
http://wattsupwiththat.com/2014/04/15/major-errors-apparent-in-climate-model-evaporation-estimates/
or for a pdf email me at [email protected].
Equations are below on effect of RH variations.
For constant wind speed and wave action, evaporation is proportional to saturated vapor pressure x (1 – RH), where RH is expressed as a fraction rather than a percentage.
For a typical ocean surface temp of 17 C and 70% RH, using Bolton’s empirically verified equation (accurate within + – 0.3%) which is more accurate at sea level than the C-C version,
D. Bolton 1980, The Computation of Equivalent Potential Temperature
MONTHLY WEATHER REVIEW, 108, 1046-1053
P = 6.112 exp((17.67 t) / (t + 243.5))
P at 17 C = 19.3634 mbar
P at 18 C = 20.6258 mbar
For constant RH:
20.6258 / 19.3634 = 1.0652
Or a 6.52 % / C increase in both specific humidity and evaporation for constant RH at 17C.
To get a 5.7 % increase in specific humidity in going from 17 to 18 C, RH at 18 C must drop from 70% to 69.46%.
Ratio of specific humidities:
(20.6258 * 0.6946) / (19.3634 * 0.7) = 1.0570
Ratio of evaporation rates:
(20.6258 * (1 – 0.6946)) / (19.3634 * (1 – 0.7)) = 1.0844
At a 69% RH at 18 C we get:
Ratio of specific humidities:
(20.6258 * 0.69) / (19.3634 * 0.7) = 1.05
Ratio of evaporation rates:
(20.6258 * (1 – 0.69)) / (19.3634 * (1 – 0.7)) = 1.10
It’s surprising how sensitive evaporation change rate is to changes in RH. The sensitivity drops with the original RH values. At 40%, a 1% drop in RH increases evaporation by 8.3% / C rather than about 10%. On the other hand, specific humidity reduction drops to 3.86% / C, so its sensitivity increases with reduced original RH.
I have a plot of specific humidity and evaporation changes vs. the new RH, but don’t know how to paste it here.
My email address shown above is incorrect. It should be: [email protected]. It had an extra j.
Hello very good internet site! Person. Stunning. Amazing. I’ll bookmark your web site in addition to consider the for in addition? We are pleased to find many helpful info the following inside the publish, you want figure out a lot more procedures about this respect, many thanks giving.