Climate models fail to ‘predict’ US droughts

Parody of ExxonMobil logo designed by St Leger...
Parody of ExxonMobil logo designed by St Leger used in the US (Photo credit: Wikipedia)

Story submitted by WUWT reader “Clyde”

Simulations identify past megadroughts, but at wrong times.

“This would be a fine country if it only had water,” observes a settler looking at the barren west Texas plains. “So would Hell,” replies a despairing farmer.

That old Texas joke probably originated in the 1950s, when the state was baked by its most relentless drought in recorded history.

Reliable forecasts of future ‘megadroughts’ would be a boon to farmers and water managers. But results presented last week at the annual assembly of the European Geosciences Union in Vienna suggest that such forecasts are still beyond the reach of current climate models.

The results were puzzling. Although the simulation produced a number of pronounced droughts lasting several decades each, these did not match the timing of known megadroughts. In fact, drought occurrences were no more in agreement when the model was fed realistic values for variables that influence rainfall than when it ran control simulations in which the values were unrealistically held constant.

“The model seems to miss some of the dynamics that drive large droughts,” says study participant Jason Smerdon, a researcher at Lamont-Doherty who studies historical climate patterns.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
May 9, 2013 3:18 am

But they don’t give up: ‘“Climate models are not perfect, but they do the big things really well,” says Tierney.’
I’d say ENSO is a Really Big Thing. And the models don’t do it well at all.

Bloke down the pub
May 9, 2013 3:38 am

When their models are as accurate as a piece of seaweed and a pine cone, then maybe people will start to listen.

May 9, 2013 3:48 am

GIGO again. When models work within reality they might work but until then we have to suffer this c–p.
If the models keep failing then the theory is wrong. Change the theory not reality.

May 9, 2013 3:52 am

In IT circles, it is known as GIGO, Garbage in, garbage out.
I would think that they would know that by now.
They must be slow learners.

May 9, 2013 3:55 am

I suppose a symposium topic on model failures is of great use to the modeling community, but it doesn’t translate to the warmists’ predictions of eminent doom.

Dr. Paul Mackey
May 9, 2013 4:27 am

“The results were puzzling”.
No they are just plain wrong.

son of mulder
May 9, 2013 4:31 am

How many times? One can’t predict chaotic systems in any detail like when and where, very far into the future.

May 9, 2013 4:33 am

Speculative but a possible scenario:
Last summer California current was cooler than normal, hence less evaporation. There are number of reasons why that could happen, I would suspect the Kuroshio-Oyashio and Alaskan currents system. NOAA itself states:
“The Kuroshio Current (red arrow) runs off the coast of Japan. It is a strong and very large warm surface oceanic current traveling 40 to 121 km/day at 1.6 to 4.8 km/h. At latitude of. about 35N it divides to form an eastern branch flowing nearly to the Hawaiian Islands and a northern branch that skirts the coast of Asia and merges with the waters of the cold Oyashio Current to form the North Pacific Current.
The Oyashio Current (green arrow) is a cold subarctic current that flows south and circulates counter-clockwise along the western North Pacific Ocean. The two currents collide near eastern shores of Japan forming the North Pacific Current.
Any change in the balance of these currents of the coast of Japan will reach Canada in just over a year and California few months later.
It is likely that the current systems was temporarily disturbed by tectonic movements during earthquake magnitude 9 of Japanese island of Honshu in March of 2011.
.Wikipedia lists seven strong earthquakes of M8+ in the area since 1920s (I added coincidental major US droughts, occurring a year or so later) :
01 September 1, 1923 M8.3
March 2, 1933 M8.4 – Major drought 1934
December 20, 1946 M8.1
March 4, 1952 M8.1 – Major drought 1953-4
May 16, 1968 M8.2
September 25, 2003 M8.3
March 11, 2011 M9.0 – Major drought 2012
Any March (spring equinox time) Japan’s major earthquake ( 3 out of 3) has a high probability of causing major drought in the USA 16 months latter, the time required for cold currents to reach California coast. This could be just a coincidence, but NOAA could do well to imitate a further research, since the US could get a year + warning, for the farming communities of possible impending drought.

Bruce Cobb
May 9, 2013 4:38 am

“But the uncertainties don’t change the larger picture, scientists say. “Climate models are not perfect, but they do the big things really well,” says Tierney. “We can be pretty confident that the southwest will warm and that water will become scarcer.””
They are “pretty confident”, are they? Based on what? Oh, right. Their precious, failed C02-centric models. Doom-casting is easy. The only difference between them and the crazy man on the street corner holding up a sign is the white lab coat.
Of course, the idea of water scarcity is based on many factors, not least of which has to do with demand, which is a separate issue. Climatists just love to conflate things though.

May 9, 2013 5:08 am

Being from Texas, I was amused by the opening story. The joke may be older than we think – from simply observing the record, the great Texas droughts appear to occur on a 60 – 70 year cycle. Interesting given the oscillation of the PDO, makes me wonder if there’s some connection.
Of course Texas is undergoing a multi-year drought currently; the greatest recorded drought was in 1950 – 1957, and the book “Droughts” from the Texas State Historical Ass’n includes this statement: “One of the worst droughts in Texas history occurred in 1884–86,” Those are the statewide droughts, but that book also notes that: “There has been at least one serious drought in some part of the state every decade of the twentieth century.”
So, “drought” in Texas really should be thought of as normal. This is a semi-arid region, after all, it’s heavy rain that is the abnormality around here.
Although I’ve been out in West Texas and seen deep channels cut by what must have been long lasting, heavy rainfall – makes me ponder what it was like here during the last Ice Age, when the Mammoths roamed.

May 9, 2013 5:15 am

It must be a problem with the recorded dates of the droughts. The records need a revision to bring them in line. Hey, if it works for temperatures it should work for recorded rainfall.

Alec aka Daffy Duck
May 9, 2013 5:22 am

AMO positive and PDO negitive predicted the drought

May 9, 2013 6:13 am

What I took from this article was that even when they’re wrong they’re still right. I think that was the point of it.

Frank K.
May 9, 2013 6:24 am

The “scientists” know the numerical solutions to the models are wildly uncertain. Yet, when their fellow extremists post “parody” Exxon-Mobil signs which were design to incite the public (and based largely on lies and distortions) the “scientists” sit idly by and twiddle their thumbs, hoping secretly that their next big money grant will be approved by the DOE or NSF.

May 9, 2013 6:44 am

OK, the “Theory” (TM) is definitively wrong.
In my scientific education I was taught that if a theory is wrong it needs to be modified to take account of the new observations. Also a theory is only a theory if it fits observations in reproducible fashion. Until it does so, it is only a hypothesis.
The hypothesis of CAGW and the models which underpin it do not fit observations, therefore the hypothesis is wrong and needs to be modified to take account of observations.
Seems pretty clear-cut to me.

May 9, 2013 7:10 am

Locusts and hydrologists have known for years what climate scientists fail to notice. Drought cycles in the US and Africa follow the same cycle, and the locust breeding cycle is timed to match.
The farmer’s almanac does a much better job of predicting drought than the climate models. But of course the Almanac is Astrology, while the climate models are Science.
To this date there is no computer simulation that can accurately predict the twice daily ocean tides. Yet we try and model climate using just such a simulation.
The only reliable prediction of the tides comes from Astrology – the orbit of the sun, moon and planets is matched to observations and this gives us a prediction based on cyclical patterns. Early humans learned to predict the seasons in just the same way, long before they understood the cause.
The simplest explanation is that you cannot model complex systems such as climate and ocean tides by computer simulation. They are too complex. Small inaccuracies quickly add up to form large errors that render the simulation meaningless. Instead you must forecast these complex systems based on historical observations of cyclical behavior.

May 9, 2013 7:14 am

We need a mega pipeline project to bring water to Texas. Last year the water table dropped by an average 3.5 feet, with a median decline of 1.8 feet. Deep wells levels can jump up and down depending on rain fall so we need watch this carefully. We can’t run our thermal power plants if the farmers use up too much water. Here is a great site to monitor the Texas water table levels.
Back during the great droughts of the 1950’s you could find water at around 70-f00t depths. Now pumpers hit water at 150 feet.

May 9, 2013 7:19 am

ferdberple says:
May 9, 2013 at 7:10 am
” Instead you must forecast these complex systems based on historical observations of cyclical behavior.”
That would work much better, but it leaves no room for CO2 as the control knob of temperatures. So they don’t do that.
It would be a great idea if the EU postponed further squandering of resources in CO2 climate science and renewable energy subsidies until the EU has solved its insolvency problem.
But that’s not how Berlin operates, oops I meant Brussels. I only wanted to *think* “Berlin”. Honest.

May 9, 2013 7:33 am

“”The model seems to miss some of the dynamics that drive large droughts,” says study participant Jason Smerdon, a researcher at Lamont-Doherty who studies historical climate patterns.”
Good understatement Jason. Obviously Jason Smerdon needs to do more research on historical climate patterns and he wants more research money. On a less sarcastic note, if we could predict droughts it would definitely be a good thing. But I doubt the researchers are prepared to scrap their current hypotheses and models. Therefore they will continue to waste money. Unlike research funded by private investors who require real results for their money, tax payer funded research does not require real results. It is dependent only on meeting political agendas.

May 9, 2013 7:52 am

“Being from Texas, I was amused by the opening story. The joke may be older than we think – from simply observing the record, the great Texas droughts appear to occur on a 60 – 70 year cycle. Interesting given the oscillation of the PDO, makes me wonder if there’s some connection.”
Several years ago the JPL did a study on the Great Plains Megadrought of the late 1920s and 1930s. They concluded that a combination of cold North Pacific SSTs ( PDO) coupled with persistantly warm SSTs in the Central Atlantic (AMO), caused the the southerly fetch from the Gulf of Mexico to be diverted southwestwards into old Mexico. While they didn’t give precise metrics, they did conclude that droughts are more likely when the North Pacific was cooling while the Central Atlantic maintained warmer than normal SSTs. Instead of southerly low level wind flow across Texas, thewinds become easterly (ie drier)

May 9, 2013 8:13 am

As far as Texas is concerned, we always get plentiful rain during El Ninos, when Global temps spike. So in Texas, droughts correlate with the cooler episodes and abundant rain correlates with warmer episodes, in a global sense. These irrefutable facts of climate stand the alarmists like Hansen on their heads.
The hard fact of global warming is that precipitation increases with temperature, as at the end of the last ice age, when it doubled, as shown by the ice cores.

May 9, 2013 8:44 am

johnmarshall says:
May 9, 2013 at 3:48 am
If the models keep failing then the theory is wrong. Change the theory not reality.
But that is not the Marxist way.
BTW, not GIGO. GIGO is when a skilled chef is given rotten meat. He will produce a meal that is less then ideal. This is is different. It is giving USDA Prime to a trail wearied cub scout with a roaring camp fire and a tin mess kit.

May 9, 2013 8:48 am

Texas and much of the country West of the Mississippi goes through periodic droughts and wet periods. Some of these places in fact are predictably cyclical. Over 40 years ago and a number of times since I have spent time in Texas and recognized that because of these periodic droughts it would be possible to use this opportunity to buy a small ranch property there. Texas calls me but for a number of reasons I never choose to follow up on this dream. But the point is anyone can observe and see that these droughts are cyclical. So if you have a dream of a 20 acre gentleman’s ranch in Texas hill country now is the time. I can guarantee you someone not yet schooled in texas drought cycles has a property that no longer can produce an income for them to pay the mortgage and their loss is your gain. Just be aware that 20 years down the road another period of drought will occur and you will need to survive them to once again fully enjoy your property once the rains come back. By the same token you can expect periodic floods and winters from hell.

James at 48
May 9, 2013 5:15 pm

No doubt. Predicting PDO would be a bear. Predicting anything with longer time frames than PDO would be darn near impossible.

Janice Moore
May 9, 2013 10:27 pm

“But that’s not how Berlin operates, oops I meant Brussels. I only wanted to *think* ‘Berlin’. Honest.” [Dirk H, 5/9/13 at 7:19AM]
Herr Dirk, you prove your honesty by such a post. You go, Dirk!
I’m not German, but I, too, would prefer it to be NOT Brussels but: Berlin and Antwerp and Paris and London and … .
The only service the EU can do for the world is to die quickly and quietly and SOON.
Ditto for the U.N.

Goode 'nuff
May 9, 2013 10:36 pm

Texans could run a pipeline to Lotawata, Oklahoma.

Steve Keohane
May 10, 2013 8:12 am

vukcevic says:May 9, 2013 at 4:33 am Not to discount your interesting correlations, but in Colorado we have drought with La Niña and rain otherwise. mpainter says:
May 9, 2013 at 8:13 am
Alludes to the same for Texas.
It is my understanding that the migration to the SW America, also known as the Desert Southwest, during the last three decades of the 20th Century, occurred during unusually wet times. Now people who live there are surprised by the lack of water.

May 10, 2013 9:20 am
David Cage
May 12, 2013 1:39 am

The standard the models should be judged by is that from just studying history. The US gets drought the years the UK gets very cold and wet. That event can be picked up really easily from the UK rainfall data. It interestingly also appears to match the years the north west passage is deemed right for crossing attempts both successful and otherwise.
Could the Arctic ice melt be the trigger, not the result of climate change, is a question I really would love to see investigated by a real climate scientist with an open mind. Historically it would appear that at least from 1600 literature would seem to back this idea. I know it is not very scientific but it is better quality evidence than much of the so called science actually used.

%d bloggers like this: