I was re-reading an old post of mine entitled “How Thunderstorms Beat The Heat“. I say “re-reading” because I couldn’t remember writing some parts of it. Yes, it was only from two years ago … but during those two years, I’ve researched and written 83 other scientific posts here at WUWT, plus another 152 political and other posts at my own blog … so things can get lost in the flood.
In any case, I got to thinking about the following graphic from that post. It shows how much evaporative cooling occurs as a result of evaporating the observed amounts of rainfall.
Figure 1. Scatterplot of evaporative cooling versus sea surface temperature. Evaporative cooling is calculated from TRMM satellite-observed rainfall data. Each dot is a 1° latitude x 1° longitude ocean surface gridcell. The conversion factor from rainfall to evaporation is that 80 watts per square metre (W/m2) of solar radiation applied for the period of one year will evaporate one cubic metre of seawater. Thus, one metre of annual rainfall is equivalent to 80 W/m2 of evaporative cooling.
Now, this graph shows the amount of rainfall evaporative cooling, or alternatively the amount of rainfall. I realized that there is an oddity … wherever ocean temperatures average above about 26°C or so, you will get rain. Might get more rain, might get less, but the area in the lower right has very, very few gridcells—not much ocean surface is both hot and rain-free.
I decided to take a closer look at just the tropical data in Figure1, and to show it in terms of the underlying rainfall data rather than evaporative cooling as in Figure 1 above. Figure 2 below shows that result:
Figure 2. Scatterplot of tropical rainfall over the ocean versus tropical sea surface temperature (SST). Each dot is a 1° latitude x 1° longitude tropical ocean surface gridcell.
The oddity is the clearly defined minimum rainfall at the lower right in the graph. For example, the graph says if the average SST of a given gridcell is 28°C, that gridcell will get at least a half metre of rain. Might get more, might get a lot more, but other than a few outliers that is the minimum rainfall you’ll get at an average SST of 28°C.
What does this look like in the real world? Well, here’s the map of the average annual rainfall.
Figure 3. Rainfall data from the TRMM satellite. The satellite only covers the area from 40°N latitude to 40°S latitude. Units are metres of rainfall per year.
You can see the clear evidence of the Inter-Tropical Conversion Zone (ITCZ) thunderstorms and associated tropical downpours just above the Equator in the Atlantic and the Pacific.
Next, Figure 4 shows a detailed temperature map of the warmest areas of the ocean …
Figure 4. Temperatures of the areas of the ocean that average 27°C (81°F) or warmer. Red area above Australia is the “Pacific Warm Pool”
When you compare the two figures, you can see the close relationship between average temperature and average rainfall.
To return to the question at hand, to me the oddity shown in Figure 2 is that there is a minimum rainfall for a given temperature. (There is also a maximum rainfall for a given temperature, following a similar curve, but it is not as well defined).
Not only is there a minimum rainfall for a given temperature, but the slope of the minimum rainfall continues to increase as the average temperature increases. Some experimentation yielded the following heuristic hyperbola (blue line) delineating the minimum temperature.
Figure 5. Same as Figure 2, with the addition of the hyperbola (blue line) which approximates the limit of the minimum values.
This lets us quantify the rate of increase of the minimum rainfall as temperatures warm. It’s shown in the second column in Figure 5 above, headed “Rain Slope”. For example, by the time the sea surface temperature gets up to 27°C, minimum rainfall is increasing at the rate of an additional 146 mm of rainfall per degree C of surface warming.
Finally, let me return to where I started. This was a discussion of evaporative cooling as measured by rainfall. As I mentioned above, a metre of rain per year requires 80 W/m2 over the year to evaporate that amount of precipitation. So we can interconvert between amounts of rain and the equivalent amount of evaporation needed to provide the water for that amount of rain.
As a result of this interconversion ability, Figure 5 also lets us look at how fast evaporative cooling is increasing with each additional degree C of sea surface temperature. This is shown in the third column in Figure 5, headed “Evap Slope”. At 27°C, for example, the cooling is going up by 12 W/m2 per degree C …
One of the results of this relates to the oceanic temperature maximum. It has been noted for some years that in the open ocean, almost nowhere is the average temperature 30°C or greater. You can see this in Figures 4 and 5 above, and I discuss this temperature maximum in a post linked in the endnotes below. Only 1.5% of the individual tropical ocean temperature gridcells shown in Figure 5 are above 29.5°C, and only 0.15% of the gridcells are above 30°C.
To explain the existence of this oceanic temperature maximum, let me suggest that there is plenty of cooling inherent in the minimum rainfall data shown in Figure 5 to put a solid upper limit on ocean temperatures. By the time you get up to 28°C or 29°C, the evaporative cooling is increasing at a remarkable rate. In practice, this means that at ocean temperatures up near 30°C, any extra incoming solar energy merely increases evaporation, with only a minimal increase in the sea surface temperature. This keeps the average sea surface temperature under 30°C everywhere in the open ocean.
Here, we’re in a two-week spell of no rain. It looks as though after a very wet last winter, this winter may be very dry. There’s a technical name for that kind of thing. They call it “weather” …
That’s the downside of that dang weather stuff. I say abolish the environment, it takes up too much room …
The upside of the weather is that it is a lovely calm sunny day today, about 75°F (24°C), with the tiny bit of visible ocean glittering and winking in the far distance. The cat outside the front door gives the peaceful sunshine two thumbs up. Or it would if it had thumbs …
Best of this wonderful life to everyone,
w.
AS ALWAYS: When you comment, please QUOTE THE EXACT WORDS THAT YOU ARE DISCUSSING. This will allow all of us to be clear on exactly what you are talking about.
Data:
TRMM Data is here, see the bottom of the page for the NetCDF file.
CERES Data is here, I used the EBAF Edition 4.0 dataset.
Further Information on the Ocean Temperature Maximum
Argo and the Ocean Temperature Maximum 2012-02-12
It has been known for some time that the “Pacific Warm Pool”, the area just northeast of Australia, has a maximum temperature. It never gets much warmer than around 30 – 31°C. This has been borne out by the Argo floats. I discussed this in passing in “Jason and the…
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
There is a very interesting paper by Robert Homes published in Earth Sciences in December and posted on Notrickszone yesterday. It basically states that the temperature of the earth can be calculated with no need of greenhouse gases. Therefore all these calculations everyone is doing is really only looking at how energy moves around on the earth and not the cause. I am only an engineer and will therefore refrain from making any comments on this matter.
Al
30 degrees C must be the magic temperature where any additional energy instantly bypasses the atmosphere and surface waters and instead warms the deep ocean where it hides. /s
Wills-my greatest fear is a cat with opposable thumbs…
Had a Maine Coon that would’ve loved it..
Best cat I ever had…
but thumbs -no.
@ur momisugly Willis Eschenbach
An interesting and informative commentary denoting the correlation between annual average rainfall and annual average sea-surface temperatures as denoted on your “Figure 2. Scatterplot” of the area of the equatorial Pacific that is bounded between 23.5°N latitude and 23.5°S latitude.
Now I am not a fan of, or a believer in the importance of, …… anything that has been calculated to have an “annual average, …… especially the often touted global or regional annual average near-surface air temperatures. Said “calculated averages”, IMHO, are only useful as reference data,
Anyway, my curiosity begs the question of, …… does the 18°C axis on your Figure 2 Scatterplot graph equate to a combined Solar position of the Sun being directly overhead at the Tropic of Cancer (northern solstice) and the Tropic of Capricorn (southern solstice), …….. whereas the 29°C axis on your Figure 2 Scatterplot graph equates to a Solar position of the Sun being directly overhead at the Equator (equinox)?
Great stuff as always. I believe you are writing future PhD theses for the Philistines. These graphs relate evaporative cooling to temperature. Is the cold rain that falls and, presumably cools the surface some more included in this relationship?
Gary Pearse February 6, 2018 at 10:31 am
Yes, but …
The amount of energy to evaporate seawater at 30°C is on the order of 2400 J/g … hang on … OK, with salinity at 35 PSU it takes 2427 J/g.
The amount of energy to heat seawater is on the order of 4 J/g / K-1. Now, the water will have to warm from say 5°C to 30°C, which brings it up to about 100 J/g.
But that is still a fairly trivial (~4%) component of the total.
w.
I’m getting a very different result from the JRA-55 atlas.
http://virakkraft.com/Evap_SST_JRA-55.pdf
Thanks, LGL. That shows NET evaporation (evaporation less rainfall).
w.
No, it shows evaporation.
It seems to me that, for lack of any other explanation, the strong negative feedback from increased evaporation, (including the resultant increase of albedo from more clouds that would have to have formed from it) likely explains why earth’s temperature seems to have had a hard limit of about ~23 degrees (globally so a few degrees higher in tropical waters or just above the knee in WIllis’ scatter plot) over the last 600 million years regardless of major CO2 fluctuation and even plate tectonics.
Disaster averted! Not only will earth not “burn up” if we keep warming, the area of rain forest is just going to keep growing in size (like it already has).
From Flim Flam Flannery of the Climate Council: “Flannery Forecasts Perpetual Drought
Over the past 50 years southern Australia has lost about 20 per cent of its rainfall, and one cause is almost certainly global warming.”
Flannery should read Willis’ paper.
Hi Willis,
Your methods to display global measurements in bulk are tremendously valuable. They allow fast comprehension, they lead one on to further exploration of the theme. IMO, you should be generously funded to develop these even more, with the objective of creating a new national teaching source, among other aims.
Here are some suggestions along those lines. You have probably thought of them already, so pardon me if I wrongly come across as teaching about sucking eggs.
Several of your past graphs have made me go again and again to the Ted talk of Hans Rosling and other clips of his. (Sadly, Rosling died about a year ago.) https://www.ted.com/talks/hans_rosling_shows_the_best_stats_you_ve_ever_seen
The Rosling methods are excellent to display a third variable on a conventional XY axis graph. I would love to see your above Figure 1 Scatterplot of evaporative cooling versus sea surface temperature running on a time basis, be it year by year or season by season or whatever third variable is chosen.
The movements on some Rosling graphs remind me of the group flight patterns of birds. Videos at
http://www.dailymail.co.uk/news/article-2913471/Watch-flocks-birds-create-beautiful-patterns-sky-Netherlands.html
Some simple mathematics are at http://www.audubon.org/magazine/march-april-2009/how-flock-birds-can-fly-and-move-together
They note “It turns out that only three simple rules suffice to form tightly cohesive groups. Each animal needs to avoid colliding with its immediate neighbors, to be generally attracted to others of its kind, and to move in the same direction as the rest of the group. Plug those three characteristics into a computer model, and you can create “virtual swarms” of any sorts of creatures you like. They change density, alter their shape, and turn on a dime—just as real-world birds do”. I am still thinking if this has any practical outcome.
Coming back to the essence of your essay here, maybe you are close to answering a query from an earlier post about what regulated the feedback in the hypothesis of cloud-controlled thermostats for climate. I asked about the “set point” or reference value that is part of many feedback mechanisms. Now I am starting to feel that it might be tied to primarily the coefficient of thermal conductivity of sea water, regulating the rate at which ocean water heat can get to the surface to be evaporated until that temperature-limited behaviour is seen, as on your Figure 1. But as you note in a comment about outcomes of gravity changes, it is a very complicated matter. Geoff.
Geoff, thanks for your support.
Roslings methods are indeed interesting. If you go to gapminder.org you can play with his animations. I’ve even experimented with uploading my own datasets to see what they can do.
When I need a third variable I generally use color … and yes, seeing how it moves and changes would be interesting and not that hard to do. Just what I need … another project. However, it would be a worthwhile one. I use a function that I wrote called “sketch” to do Figures 1, 2, & 5. It uses semi-transparent dots, and it takes the two datasets as input variables.
It wouldn’t be hard to wrap that into a function called “sketchmovie” … let me think about that …
w.
@Willis Do you have any clue how to quantify the effect of “rain chill”? I am asking because it is vital point in determining net cloud forcing. Rain falls from clouds in will lower temperatures in correlation with cloudiness, but that is not due to CF of course.
To illustrate the issue, these are data from tropical (US) stations – temperature vs. low cloud condition (up to 12.000ft)
http://i736.photobucket.com/albums/xx10/Oliver25/cloudsvstemp%20tropical.png
And here is the amount of rain according to (low) cloud condition..
http://i736.photobucket.com/albums/xx10/Oliver25/rangliste%20prec.png
If you wonder why there is rain from clear skies, again these are only clear up to 12.000ft..
Anyhow, to perfection the subject (which obviously shows a net positive cloud forcing!!!), it would be great to quantify that rain chill factor.
Thanks, Leitwolf. Not sure exactly what you mean by “rain chill”. Thunderstorms cause cooling in a wide variety of ways, including but not limited to:
Reflection of sunlight
Increased evaporation due to storm winds at the base
Downwelling “rain wind”, the entrained wind that comes down from the clouds with the rain
Virga
Increase in sea surface roughness, which increases albedo
Breaking waves with spray, which increases evaporative surface area.
“Whitecaps” and blown foam (spume) which increases albedo
I see no way to disentangle these from say the effect of a cold front without rain …
w.
Right, there are a lot of factors, but the most basic idea I had in mind is rain itself. As it originates in higher altitudes where it is much colder than at the surface, the rain itself will be cold. Rain (or snow) will warm up, as it falls through the warmer, lower atmosphere, but still its temperature will lag far behind. Especially when it is generally warm, one can fairly well see how a little shower quickly drops temperatures.
Overall there are a couple of factors which are affiliated with cloudiness, which will lower temperatures, but are not cloud forcing. So it would be great to single them out, but as I told, I see no way to fairly quantify them.
All I know is, that their impact will be overproportionately strong with OVC skies, which then would look much warmer if we could sort it out. In fact we might have a very straight linear trend, with temperatures being the warmer, the more clouds there are. Which is amazingly funny of course, as we are talking about low clouds in the tropics. That should be the clouds with the biggest negative CF of all.
Ooops .. I meant to reply to the previous thread