Guest post by Willis Eschenbach
There’s a new study out from NOAA called “Probable maximum precipitation (PMP) and climate change”, paywalled of course, which claims that global warming will lead to a 20%-30% increase in “probable maximum precipitation”. The abstract says:
Probable Maximum Precipitation (PMP) is the greatest accumulation of precipitation for a given duration meteorologically possible for an area. Climate change effects on PMP are analyzed, in particular, maximization of moisture and persistent upward motion, using both climate model simulations and conceptual models of relevant meteorological systems. Climate model simulations indicate a substantial future increase in mean and maximum water vapor concentrations. For the RCP8.5 scenario, the changes in maximum values for the continental United States are approximately 20–30% by 2071–2100. The magnitudes of the maximum water vapor changes follow temperature changes with an approximate Clausius-Clapeyron relationship. Model-simulated changes in maximum vertical and horizontal winds are too small to offset water vapor changes. Thus, our conclusion is that the most scientifically sound projection is that PMP values will increase in the future due to higher levels of atmospheric moisture content and consequent higher levels of moisture transport into storms.
When I heard that number, a 20%-30% increase in maximum rainfalls, my urban legend detector starting ringing like crazy.
Figure 1. The authors’ guess at how much more rain will be falling by the end of the century.
So … why did my urban legend detector go off from this claim? It has to do with energy.
The press release quotes the authors as saying:
“We have high confidence that the most extreme rainfalls will become even more intense, as it is virtually certain that the atmosphere will provide more water to fuel these events,” said Kenneth Kunkel, Ph.D., senior research professor at CICS-NC and lead author of the paper.
Now, the increase in maximum rainfall is said by the authors to be due to the increase in water vapor in the air. It’s unclear if the 30% increase in maximum rainfall will be matched by a corresponding overall increase in rainfall. However, it is highly unlikely that an increase in water vapor will only increase maximum rainfall events. The authors themselves say that their projections show “a substantial future increase in mean and maximum water vapor concentrations”.
So to be conservative, let’s cut the 30% increase in maximum water vapor down to a 20% increase in mean water vapor, and see what that looks like.
I want to determine how much energy we’re talking about here. Suppose the rainfall were to go up (on average) by about 20% globally. Right now, the globally averaged rainfall is on the order of a metre of rain over the entire surface per year, a bit more or less depending on who is measuring. Twenty percent of that is 200 mm. So we need to evaporate an additional 200 mm over every square metre of surface to produce the stated increase in rain.
It takes 2260 joules of energy to evaporate a gram of water. For each square metre we need to evaporate 200 mm, or 200 kg of water. To evaporate that much water takes 4.52e+8 (452,000,000) joules of energy.
Now, a joule is a watt-second. We need 4.52e+8 joules of energy every year to evaporate the additional water, which is 4.52e+8 watt-seconds per year. Dividing that by the number of seconds in a year (3.16e+7) gives us the change in constant 24/7 watts needed to evaporate that much water. Remember, this is an increase in the constant watts of energy striking every square metre of the planet.
And that number, dear friends, the amount of additional energy needed to increase global evaporation and thus rainfall) by 20%, turns out to be 14.3 W/m2. That’s about the amount of energy increase from three doublings of CO2. Yes, CO2 would have to go from the current ~400 ppmv to about 3,200 ppmv to provide that much extra forcing …
So my urban legend detector is still working fine. There’s nowhere near enough energy available to power that claimed jump in rainfall.
Now, I could leave it there, since the energy necessary to make their claims possible doesn’t exist. But in order to confirm that finding, my plan of further inquiry was to see whether either the intensity of rainfall events or the mean rainfall has changed over the last century. People are always claiming that we don’t have any controls for our experiments when we study nature. But nature provides its own experiments. To start with, we have the warming since 1900. On land, according the Berkeley Earth Surface Temperature data, the temperature has gone up about a degree over that time … but did the rainfall go up as well?
Figure 2. Global precipitation over the land, in mm/day. Data Source 1901-2009: CRU TS 3.10.01 (land)
OK … no increase at all in global rainfall, neither in the monthly means nor in the maximums. So no support for their claims there.
So how about local maximum rainfall events? Are those going up?
For this, we can turn to the temperature and precipitation records of England. For the Central England region, we have daily temperatures and daily precipitation records since 1931. Since 1931, the average Central England Temperature (CET) record has gone up by just under one full degree. So we should see any thermal effect on the maximum rainfall. With that 1°C temperature rise as the backdrop, here’s the maximum central England daily rainfalls, month by month, for the last eighty years.
Figure 3. Maximum daily rainfall, 1931-2012, Central England. Data Source Photo Source
Here, we find the same thing. There is no evidence of any increase in maximum rainfall events, despite a 1° temperature rise.
Hmmm …
The part I really don’t like in all of this is that once again, all of their claims are built on computer models. But what I don’t find is any serious testing of their whiz-bang models against things like the global or the CET temperature and rainfall records. In fact, I don’t see any indication in any venue that any computer models are worth a bucket of warm spit when it comes to rainfall. Computer models are known to perform horribly at hindcasting rainfall, they do no better than chance.
So once again, we’re back in the land of Models All The Way Down. I gotta confess, this kind of thing is getting old. NOAA and NASA appear to be falling further and further behind reality, still churning out useless studies based on useless models.
Just one more waste of taxpayers money.
w.
Elegant and entertaining as usual, Willis.
My years spent working with ‘old blokes who had observed much and said little’ on farms made me aware of cycles in weather and someone uphill mentioned 11 years for a precipitation cycle, which sounds about right to me. But what would I know, I’m not a climate pscientist!
atarsinc says: April 6, 2013 at 8:14 pm
Mike McMillan says:April 6, 2013 at 6:54 pm. …
Mike, that seems right, but you’re still dumping the heat into the lower troposphere. You know, that part of the atmosphere where weather happens. JP
We’re dumping heat into the upper troposphere, too. Thunderstorms can get up ten miles on occasion.
Aside from a few Sherpas, Peruvians, and Yetis, we live at the bottom of the troposphere. What’s important is getting the heat off the surface where we are. Whatever is going on in the rest of the troposphere doesn’t mean much to us unless it affects the surface (air travel excepted).
We care a lot more about the PDO, AMO, and ENSO than we would otherwise because they screw up the weather. We care about the UAH and RSS satellite LT not because it affects us, but because it keeps a reality check on GISS and East Anglia’s fiddling with the surface temperature records, which give EPA and governments yet another excuse to control our lives. /mindless rant
Ian W says: April 6, 2013 at 7:30 pm
… You can watch this heat being released in real time just go to
http://www.ssd.noaa.gov/goes/east/natl/flash-rb.html
There you will see the outgoing infrared as seen by the GOES East Satellite. Notice how the weather systems show up.
[to] Mike McMillan …
So not a net-net process as we can see energy departing from the frontal and storm systems
Spiffy video.
However, it doesn’t show amount of energy departing.
If you tick the ‘IR Temp’ box at the top, it turns on the temperature scale above the rainbow legend at the bottom. Quite the opposite of what you’d expect. Red is very cold, blue is warm.
Clouds are pretty opaque to IR, so the loop is looking at cloud top temperatures, the higher the colder. Intense storms are very high, thus very cold tops, so the color is a measure of what’s going on down below, not outward radiation.
Willis, All of the energy needed to raise that much humidity would not have to come from the CO2 forcing, some would come from the water vapor itself, since it is a positive feedback. It is a shame that the models under represent the rest (negative feedback part) of the water cycle. Wentz in Science (2007) found that precipitation increased in proportion to the humidity in the observations, while NONE of the models represented more than half of this.
However, the authors were shameless in representing the scientific soundness of their projection as if it were conservative when the scenario they use is well acknowledged to be aggressive. For example:
“The RCP8.5 emissions scenario projects greenhouse gas emissions with a continuous rise in radiative forcing to about 8.5 Watts per square meter in 2100 (19) and therefore represents an aggressive warming scenario.”
From the supplementary material of:
http://www.pnas.org/content/early/2013/03/28/1216006110
I’m always amused watching people try to wrap their heads around the most simple concepts.
Temperature and water vapor concentrations are intimately connected. If one changes, the other compensates. It’s simple. Convection explains all thermostatic function for atmospheric temperature.
Example: temperature increases, more water evaporates. Water vapor forms clouds that reduce input to the system, and rises to whatever altitude required to radiate energy, then returns… whether it’s slow or fast (storm). Temperature decreases.
Example: temperature decreases, less water evaporates, fewer clouds blocking solar input, less energy radiated out to space from airborne water vapor. Temperature increases.
Missing the OBVIOUS steps in between might lead a researcher to conclude that water vapor could increase if temperature increases, but apparently they need to consider what happens to that water vapor once it is in the atmosphere. They missed that. It doesn’t just sit there and rain more. It goes away, quickly, and takes energy with it.
At MOST, even a significant change in CO2 levels will make transient temperature changes while the system auto-adapts. Again, the concept of long-term climate change from altering a trace gas is ludicrous. One day everyone will laugh at ever thinking it was even possible, let alone happening.
200 mm, or 200 kg of water
Willis, need a volume to convert to weight.
Code Tech is right.
Willis’ thermostat idea is fairly simple. The sun comes up over the ocean, sunlight hits the water, slows down, and gets absorbed, raising the temperature. Evaporation increases, convection lifts bubbles of humid air up to the condensation level where they form puffy clouds and lose their oomph. Heating continues throughout the morning, more and more clouds form. Eventually, the rising bubbles start drawing moist air in from the sides, and with a steady supply of moist air, we tip over into thunderstorm mode, hauling tremendous amounts of heat and water from the surface to altitude..
With extra insolation or a generally warming climate, the thunderstorm heat engine just starts up sooner in the day and runs longer. Simple.
As a side note, a cloud need only rise above 18,000 feet, and half of the earth’s greenhouse gasses that would slow spaceward long wave radiation are beneath it.
Thanks Willis for the post and David for mentioning my work.
The very notion of probable maximum precipitation (PMP) is not scientific and signifies a failure of the meteorological and hydrological communities. Generally I avoid using that categorical language, but I am fully convinced that it’s a tragic failure; it is a shame that it is still in use. I explain the reasons in http://itia.ntua.gr/758/ . In brief, the idea of an upper bound in precipitation and flood serves political aims, not scientific ones. It is appreciated by decision makers who want to fool people saying that a certain construction, if designed by PMP, entails no risk. Risk cannot be eliminated, it can only be reduced at an acceptable level, and of course decrease of risk is associated with increase of cost. Evidently, the new “climate-change” setting of the PMP idea serves additional political aims.
In http://itia.ntua.gr/23/ I have shown that a PMP value, when estimated using the statistical method by Hershfield, has a probability to be exceeded at any year equal to 1/60000. Hershfield’s method relies on rainfall data only and if modified by assigning a probability of exceedence becomes a statistically consistent method. The other method, which as I understand was used in this study (and, in addition to rainfall, uses atmospheric moisture data, the Clausius-Clapeyron equation etc.) still is a statistical method. But it is a very bad statistical method and I doubt if it can be remedied in any way (see http://itia.ntua.gr/701/).
Stephen Richards says: April 7, 2013 at 1:36 am
200 mm, or 200 kg of water
Willis, need a volume to convert to weight.
1 cubic centimeter of water = 1 gram by definition
1 liter, 1000 cc = 1 kg
1 cubic meter = 1 metric tonne, 1,000,000 cc
In answer to an earlier comment, I believe there is an agenda.
Google agenda 21 for dummies.
Global Warming can be seen as a pretext for profligate govts to grab extra taxes “to save the planet”, & for banksters to make millions on carbon cap & trade etc, but it can also be seen as establishing a global problem to ease forward in the public mind the need for a global govt.
This is the UN Agenda 21.
Unfortunately Agenda 21 also calls for the “removal” of 6 of every 7 persons on the planet.
This is what I believe is the driver for the present Middle East wars & African conflicts.
When I see American putting the produce of millions of acres of its land into its gas tanks, while food riots, not a desire for democracy, causes “The Arab Spring”, then I know the Main Stream Media are bought & paid for, & something is far far wrong.
I was discussing this problem with a young lady recently, & she said to me “good, there’re too many people on the planet”
Word is obviously not getting through to folk that there is no need for a chicken little panic: that world population growth is slowing & is expected to peak & decline as prosperity grows; that we have reached peak land & do not need to put further acres under the plow; that CO2 is plant food & is greening the deserts; that CO2 is a coolant, (if I have that correct).
I reckon this site could provide a right service if it could link articles from Matt Ridley, Allan Savory, Willis & whoever else under a heading such as Bright Future or something like that, but I must admit I have no idea how big an ask this is.
Thank you Anthony, mods & commenters for a most enjoyable & informative site.
Please keep up the great work.
You may publish all or nothing or whatever bits you choose.
Regards,
JD.
I must agree with Nick Stokes, WernerBrozek and others here. The energy input to produce the initial 20% increase in water vapour would only be required once, not every year.
Energy used to supply the latent heat of evaporation is not lost. It is returned when the water vapour condenses. It is returned to the atmosphere at that point which in turn radiates back to earth.
This energy could only be lost to the Earth system by radiation to space. The amount radiated to space is not going to change. The amount radiated to space will always be such as to balance the incoming solar energy (in the long term).
Come on. Whether it condenses inside a thunderstorm or not makes no difference. Energy cannot be created or destroyed by doing mechanical work. This is the first law of thermodynamics.
So the energy is only needed once.
Nice lack of correlation between temperature and rainfall in the CET though.
Nick writes “The only way it can leave the planet is as OLR. But that requires a warmer atmosphere.”
Arguments of additional energy stored in the oceans for example dont translate to additional energy being lost in events. CO2 doesn’t buy more variability at a fundamental “cant argue with the physics” level. We can only watch and see what effect it has in the atmosphere and so far predictions are wrong (ie atmospheric hotspot).
MikeB says: April 7, 2013 at 3:17 am “It is returned to the atmosphere at that point which in turn radiates back to earth.”
Very sad, you believe “Back Radiation” can return 100% of the heat released to the atmosphere back to the earth. That is a new one even for those that believe in BR in the first place.
What I find really sad is that this passes for Science, it is bringing science and scientists in to disrepute and to think it comes from the NOAA.
Seriously, MikeB, you have just demonstrated a profound lacking of logic.
The energy IS lost by radiation to space. Are you saying the atmosphere is a closed system? Really???
MikeB says:
April 7, 2013 at 3:17 am
I must agree with Nick Stokes, WernerBrozek and others here. The energy input to produce the initial 20% increase in water vapour would only be required once, not every year.
Energy used to supply the latent heat of evaporation is not lost. It is returned when the water vapour condenses. It is returned to the atmosphere at that point which in turn radiates back to earth.
/////////////////////////////////////////////////////////////
Whilst I need to think more about it, I do consider that there is merit in what Nick Stokes moots, in that the energy gained will be locked into the system and not lost from the system (unless radiated away into space to the extent that the system is not a fully closed system).
However, whilst that suggests that all or at any rate the bulk of the energy is only required once, it begs the question to what part of the system is the energy returned and when so returned, is that energy in a position to perform the required work in our to once again evaporate water.
You suggest that the energy is returned to the atmosphere, but this is where I have some difficulties and upon what I need to think more. I am not convinced that 100% of the energy is returned to the atmosphere. Second, it may be returned at height in the atmosphere where it is less useful. Of course, all of this may equalize over the course of time so that one simply reaches a new equilibrium point at which the one off energy is locked in the system and the system has a different rainfall profile as a consequence of the different energy equilibrium point.
But is not Willis’ point that presently, DWLWIR is about 320 W m^-2 which contributes towards the overall energy budget and with that budget, we presently have about 1m of rain, and to get 20% more rain (ie., 1.2metres) one would need a revised energy budget at which DWLWIR is about 334 W m^-2 (which of course, would give a different temperature equilibrium) and an energy budget equalising around that figure. The increase in DWLWIR would gradually grow as GHGs were increased.
My take on Willis’ article is that to achieve that new equalibrium point one would need much more than a doubling of CO2.
However, I consider this to be an overall simplification for a point which I believe that Nick raised earlier, namely that any momemt in time, there are many molecules of water vapour on the cusp of condensating and only a little extra energy is required to tilt these over the balance from ome state to another.
Is one of the problems in all of this the assumption that “It takes 2260 joules of energy to evaporate a gram of water.”(which needs slight correction for temperature)? This is of course, the position at sea level, but not at altitude. If the increased precipitation is to come from water vapour at altitude then less energy is required.
I have a number of comments to make, and I think the water vapor issue is the most important one in the climate debate (given half the expected warming comes from increased water vapor) so bear with me.
The Classius Clapyeron relation says that water vapor increases by 7.0% per 1.0C increase in temperatures. So, by 2100, when temperatures are up 3.25C, water vapor is supposed to be 22% higher than average.
Indeed, all the climate models have this type of increase built in. I downloaded the multi-model mean for AR5 from the IPCC data center awhile ago. It has now been reorganized and I can’t find it anymore (registration was required).
This is the RCP 6.0 scenario water vapor numbers which do rise 22% by 2100 versus the actuals to date. [Note the NOAA study is the RCP 8.5 scenario which means GHG forcing will get to +8.5 W/m2 by 2100 (and 12.5 W/m2 by 2250) which is a very high estimate – CO2 doubling is only +3.7 W/m2 so 8.5 W/m2 is an over-the-top scenario and I don’t know why it is used so much. RCP 6.0 is much closer to the track we are now].
The IPCC AR5 water vapor forecasts are already far off.
http://s14.postimg.org/nnhgk7ku9/Water_Vap_IPCC_AR5_2100.png
Last month, water vapor was about 2.0% higher than the long-term mean (versus the climate models which have it at 6.0% higher). Last month was close to ENSO-neutral which is important since the ENSO affects global and tropical water vapor numbers by a huge amount. It is the main driver.
Water vapopr cycles through the atmosphere each 9 days. There is 40 times more rainfall (1000 mms/year), than a typical area has in it atmosphere (24.5 mms or 1 inch).
The climate models have a slightly lower precipitation increase (about 4.0% per 1.0C), versus the water vapor increase (7.0% per 1.0C). To get more water vapor in the air given it cycles through so fast, thee must be less precipitation than there is evaporation. (I’m not sure where this really comes from, but that is the assumption).
Precipitation forecast from the AR5 multi-model mean for the RCP 8.5 scenario (from the Climate Explorer). An increase of 14% by 2100 (just be nitpicky here, the values in this multi-model run are about 40% too high, but who knows what data source they were forced to use).
http://s23.postimg.org/5npwc9svf/icmip5_pr_Amon_modmean_rcp85_180_180_E_90_90_N_n.png
I would like to see more analysis of whether this is physically possible.
Regarding the 4.52e+8 joules of extra energy per year required to evaporate that much more water, it is actually a lot less than the GHG forcing is which is about 9.08e+13 joules/year right now. [2.86 W/m2*3.18e+13 joules/W/m2/year].
Willis
Interesting article.
Just a quick question. You appear to me to be assuming that all this extra rainfall is going to come from the oceans (average temperature say 15degC – although the bulk will no doubt come from the tropical or sub-tropical oceans which are somewhat warmer).
However, is it not the position that we already have a lot of water vapour in the atmosphere, far more than that which precipitates? This [surplus] water vapour has already been the beneficiary of such energy that was required to break the bonds in the liquid state and to change it from liquid to vapour.
Is it not likely that a large percentage of the additional rainfall will come from water vapour already in the atmosphere (possibly already reasonably high up in the atmosphere) such that less energy is required than you are proposing?
To put it a different way, you state that “It takes 2260 joules of energy to evaporate a gram of water” (depending upon the temperature of the water). But should you be considering how much energy is required to evaporate water not in its liquid form but in its vapour form.
Your comments would be appreciated.
Okay, I think it’s clear that the evaporation and condensation involved in rain moves heat upward about a half mile. I guess the question I’ve got is, what then? I have this (possibly naive) intuitive notion that the higher up the heat accumulates, the more it will radiate into space as OLR. I freely admit that this is nothing but an intuitive notion. Anyone care to either help me put this idea on more solid footing or smash it if it’s incorrect?
My idea runs along these lines. Layers in the atmosphere, absorbing and reemitting. Some of the energy gets emitted upwards, some downwards, random distribution. It seems to me that the higher up the layer is that we’re looking at, the better the chances for any given amount of energy to make it out to space as OLR. ~shrug~
Tear this down for me gents. 🙂
Sorry, I screwed that calculation up at the end. I was using km2 in a calculation.
Should read.
Regarding the 4.52e+8 joules of extra energy per year required to evaporate that much more water, it is actually a lot MORE than the GHG forcing is which is about 9.08e+7 joules/year right now. [2.86 W/m2*3.18e+7 joules/W/m2/year].
In fact it will be twice as much extra energy than is provided by GHGs in 2100 under the RCP 8.5 scenario which is.
8.5 W/m2 * 3.18e+7 joules/W/m2/year = 2.70e+8 joules/year
Interesting. They haven’t gone through the right energy models or the Classius Clapyeron has been inappropriately used.
MikeB writes “Energy used to supply the latent heat of evaporation is not lost. It is returned when the water vapour condenses. It is returned to the atmosphere at that point which in turn radiates back to earth.”
That latent heat is transported a couple of kms up into the atmosphere where half of it radiates further up and away from the surface and the other half radiates towards the earth but is recaptured into the atmosphere well before it gets there. Well before it gets involved in further evaporation for example.
You dont get anything for free, if the energy is being transported high into the atmosphere (where half continues away) then its not available for heating at the surface.
Same lack of change in precipitation can be seen for CONUS at NOAA NCDC.
And, at least some of the latent heat released by condensation of precipitation will be cloud top, so radiate away as OLR, requiring additional new input heat the next year. On balance, Willis back of the envelope calculation of the additional heat is more right than wrong. Plus, it is plain illogical to think the extra evaporation gets somehow stored up for a very few extraordinary precipitation events. That is not how weather works.
The paper is just more unscientific unsound conclusions, opposite to what the abstract says.
richard verney
A falling raindrop has kinetic energy , when it hits a surface it makes a noise. All this requires energy which will be lost from your cycle, plus any water vapour from the atmosphere that falls in precipitation will have to ultimately be replaced by the transport of water from the surface. increase the rainfall rate and you have to increase the energy input rate .
I think I see what has been done by these hucksters. Using a psychrometric chart for 1000 M (est. cloud level), 5.3 C (adiabatic lapse rate from 14.5 C), and 100% humidity one gets 0.0062 kg H2O/kg air. In order to increase the carrying capacity of air by 30% one has to increase it’s temp by nearly 4 C to 9.1 C. Actually, it’s considerably more complicated than this but as a zeroth estimate this gives us an idea of what has been done, i.e. dramatically increased temperatures.
Just another example of climatology GIGO to generate funding. Models, models, models, but no understanding of basic physical concepts.
Thanks yet again, Willis!
Nick Stokes and Werner Brozek suggest that the energy only needs to be added once, since it remains within the climate system (otherwise there would be more energy leaving the Earth than arriving).
Whilst convincing, it leads to a strange situation. If energy to evaporate the water is only need once, say one years additional forcing leads to this extra water vapour, then what happens to next years forcing? Clearly, this extra forcing is not needed to evaporate the extra water, so it must lead to a further increase in evaporation. Ie, 400mm. As this energy is only needed once, rather than every year, then the year after that, this same extra forcing will lead to an even higher level of evaporation.
The logic therefore leads to an absurdity – that a fixed level of additional forcing will lead to more and more evaporation with respect to time. Therefore Stokes et al must be wrong.