Guest post by J Storrs Hall
There are several ways to predict what the temperature trends of the next century will be like. The standard method of prediction in science is to create a theory which embodies a model, test the model experimentally, and then run it into the future for the prediction. There is another way, however, which is simpler in some ways although more complex in others. That’s simply to remember what’s happened before, and assume it will happen again.
Here’s a record of what’s happened before, which most WUWT readers will be familiar with. It’s the GISP2 Greenland ice core record, shown for the Holocene:
I have shamelessly spliced on the instrumental record in red (by setting the temps in 1850 equal); it is the HadSST record.
When I first started looking at GISP2 it seemed to me that there were several places in the record that looked very much like the sharp spike in temperature we’re experiencing now. The obvious thing thing to do seems to be to overlay them for an easy comparison:
Here I’ve plotted the 400 years following each minimum in the record that leads to a sustained sharp rise. There were 10 of them; the first five are plotted in cyan and the more recent 5 in blue. You can see that in the latter part of the Holocene the traces settle down from the wilder swings of the earlier period. Even so, every curve, even the early ones, seems to have an inflection — at least a change in slope — somewhere between 200 and 250 years after the minimum.
The hatched black line is the average of the 5 recent (blue) spikes. The red dots are the uptick at the end of GISP2 and HadSST, spliced at 1850. Note that ALL the minima dates are from GISP2.
Prediction of the 21st century is left to the reader as an exercise.
Read ’em and weep.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
theduke says:
July 1, 2011 at 7:18 pm
“Ancient astronomers knew of many patterns in the movements of planets, but they could explain diddly and predict diddly.”
Ancient astronomers could predict many things. The position of the sun and moon (Stonehenge comes to mind). The could also predict the position of the planets in the various constellations. Predicting something using historical data is a valid scientific method, even if the mechanism is unknown. Climate science is the same way. Weather forecasting is the same way. We do not understand the whole mechanism, but we can extrapolate. Models are a valid means to test our predictions, but they are only that, models not laws not hypothesis. They are a tool and must be used with caution. Ask any good weather man on the accuracy of his short term forecasts. Personnaly I like the graphics of the “forecast weather”, but I’m still very skeptical about their accuracy..
NikFromNYC says:
July 2, 2011 at 11:43 am
If you merely circle wagons and don’t allow a factual debate ’bout Greenland, then you in this hour, don’t deserve a shower. Self-selection takes no prisoners. Thinkers fled from this debate, already. It’s Summer outside. No BANDWIDTH LIMITED issues exist, outside. Real people, they exist, out there. They want to meet you. Nothing would make them happier. Go out. Report back. Negative self-selection prevents my message from having any effect, here, in a especially contrived junior science backwater.
You hours do not affect.
I have read many hundreds of thousands of comments, and have myself written many tens of thousands of comments and replies in various places while on-line over the past years.
I must confess, however, that the above completely confounds me. Please explain just what the heck you are trying to say. But avoid the adult beverage (or recreational drug) this time. 8<)
In a red (by highlighting the graph from Figure 1 in a new Figure 3), or some other such method, please show which “upticks” you are using to create the second figure.
Dear Mr Hall,
Congratulations on a very interesting and useful observation. I would like to incorporate these charts in a future presentation, with attribution. To that end, if possible please email me the Excel files the graphs come from so I can re-plot them for Powerpoint. [trimmed, by request. Robt]
Thanking you in advance.
Just a quibble, but the GISP2 measurements were made in Greenland so it would be better to use the DMI data than HADSST when “shamelessly” splicing on the instrumental data post 1850:
http://diggingintheclay.files.wordpress.com/2010/12/coastal-average.png
John Q. Galt says:
July 1, 2011 at 5:57 pm
“Nightly lows in upstate New York are getting ridiculous. I had to close the windows last night as it was absolutely frigid.”
Yeah, my mom (western NY state; southern tier) has been complaining every time I speak to her that the furnace has been running every night and some days too. And it’s raining almost every single day. Meanwhile with me here in south central Texas we have an exceptional (beyond extreme) drought and temps are running well above average with 100F+ days rolling along like boxcars on a long train.
Be a good chap and send a little of that rain and cold down my way!
re; to those whining about Greenland not being representative of the whole planet
So Mann can use trees from just one region to reconstruct past global temperature, foraminifera from just one location to reconstruct past global sea level, ice cores from antarctica can be used to reconstruct past global CO2 level, but somehow Greenland ice cores are only applicable to Greenland?
Cake:having:eating
Greenland:global temperature
canary:coal mine
theduke says:
July 1, 2011 at 7:18 pm
“Ancient astronomers knew of many patterns in the movements of planets, but they could explain diddly and predict diddly.”
Nonsense. They could predict that the sun would rise in the east and set in the west. They didn’t know why. All they knew was it happened with reliability in the past and knew of no reason why it would NOT continue that way into the future.
Climatology is the analysis of past patterns and extension into the future. Is it reliable? To a certain extent, yes. Is it as reliable as having a theory of climate? No.
Back in the days when I was (very successfully) playing the stock market company quarterly reports which happened to be glowingly good would always have a caveat: Past performance is no guarantee of future performance. That by no means that past performance has no predictive value. Often it’s the best and only predictor you have to work with.
Derek Sorensen says:
July 2, 2011 at 1:27 pm
“(read: working out potential and actual customer’s credit-worthiness). The golden rule in that area is “The future is like the past,” and it seems to work out more often than not.”
True, but caution is advised and it’s always best to take all the data you have, however meager into account. The classic example is shoe size. If I were hiring people for positions where high intellect was essential and I didn’t have much time or data to work with I could screen candidates by their shoe size because, statistically, the larger the shoe the more intelligent the person wearing it. However, if we have an additional bit of information like the candidate’s age, we can do better screening due to the fact that shoe size:intelligence connection is radically skewed by baby shoes. Adult shoe size has very little relation to intellectual capacity although it would still be a good screening tool for NBA recruiters.
Nice contribution! According to GISP2 there were 7 longer periods that had an increase in temperature of more than the 0.6° C that we experienced during the 20th century. When you look at the glacial part of the GISP2 record ( ~50.000 – 10.000 BP), one counts 29 periods that exceed the 20th century increase. See: http://www.klimaatgek.nl/cms/?De_dogma%27s:Unieke_temperatuurstijging Translate button in the right upper corner of the (Dutch) site.
NikFromNYC says:
July 2, 2011 at 11:43 am
“If you merely circle wagons and don’t allow a factual debate ’bout Greenland, then you in this hour, don’t deserve a shower. Self-selection takes no prisoners. Thinkers fled from this debate, already. It’s Summer outside. No BANDWIDTH LIMITED issues exist, outside. Real people, they exist, out there. They want to meet you. Nothing would make them happier. Go out. Report back. Negative self-selection prevents my message from having any effect, here, in a especially contrived junior science backwater. You hours do not affect.”
Are you smoking crack?
With an as yet undetermined appendage R. Gates writes on July 2, 2011 at 11:35 am:
“Uh, this particular post has nothing to do with climate science…if you haven’t noticed.”
Bzzzzzzzt! Wrong!
The OP is pure unadulterated climate “science”.
http://en.wikipedia.org/wiki/Climate_science
My emphasis.
LazyTeenager says:
July 1, 2011 at 5:11 pm
“Let’s look for example at cereals like rice and wheat. As suitable temperature zones for these crops move towards the poles, what happens to the amount of land available?”
It doesn’t decrease as there’s no dearth of continental crust in higher latitudes. What actually happens is total amount of arable land increases as the permafrost zone shrinks. Just as beneficial, growing seasons get longer. In a few decades they might (again, as the Vikings did) have a long enough growing season in southern Greenland to grow apple trees and silage for livestock. In more temperate higher latitude climes they’ll be able to squeeze in two growing seasons instead of just one.
The bottom line is a warmer planet is a more productive planet insofar as life is concerned. Barren rocks and ice aren’t conducive to plant growth. Plants are the primary producers in the food chain. Plant productivity determines how much animal productivity can be supported.
Well, folks, looks like I’d been scooped on this one — some decades ago, which is as it should be, if there were real climate scientists out there:
http://stevengoddard.wordpress.com/2011/05/26/1979-before-the-hockey-team-destroyed-climate-science/
I would bet that they were looking at just the same kind of peaks in the record that I was.
Observational vs. controlled studies
The OP is an observational study which is pretty much the only kind of study possible in climate “science”. I use the scare quotes because climatology is more actuary than it is scientific.
I was just reading a SciAm article on data mining in the medical field comparing results of observational studies and controlled studies. One successful example of observational study is linking increased risk of lung cancer with cigarette smoking. One unsuccessful example is estrogen replacement therapy for older women with declining estrogen levels. An observational study found that women with declining estrogen were at increased risk for stroke and cardiovascular disease. A controlled study found that estrogen replacement therapy did nothing to reduce risk of stroke and cardiovascular disease and only increased the risk of breast cancer. The reason for the error is that women with declining estrogen were seeing a doctor more often, which is how the lowered estrogen was detected in the first place. The actual relationship between estrogen and cardiovascular health was that the women with low estrogen were seeing a doctor more often and the doctor was, at the same time, assessing cardiovascular health and through earlier detection warding off cadiovascular disease.
racookpe1978: I suspect Mark V Shaney … http://en.wikipedia.org/wiki/Mark_V_Shaney
as for the points I picked, http://i52.tinypic.com/20qb4eb.png
Dave says:
July 2, 2011 at 8:07 am
Bill Illis: Hi! Where do your two graphs come from please?
———————————–
Antarctica Epica Dome C 800,000 year temperature history
ftp://ftp.ncdc.noaa.gov/pub/data/paleo/icecore/antarctica/epica_domec/edc3deuttemp2007.txt
Greenland NGrip ice cores extended back to 123,000 years ago.
[Use this formula to convert d18O% to Temp C anomaly = (d18O% + 13.4) / 0.9 + 24.78 ]
http://www.iceandclimate.nbi.ku.dk/data/2010-11-19_GICC05modelext_for_NGRIP.xls/
Henry@DaveSpringer
I just wanted to throw another piece of wood (log) on the (cooling) fire:
http://www.letterdash.com/HenryP/ooops-global-cooling-is-coming
Actual recent observation. What do you think of this?
When a state gains power, endorsed by other states, and claims that the climate is a concern of national security, and also claims that other state’s national security, and by extrapolation soveriegnty, are thretened by the climate and that state seems to be on the edge of offering a solution if you just “do as i say”, the problem is beyond temperature graphs and who is right or wrong.
The problem becomes political, phylosophical and idealogical survival.
With Germany gaining UN chairmanship and making the noises it’s making, HHhhmm, just old, crusty and suspicous am I.
Looking at the time stamp of the last reply, I suspect nobody will read mine, but here it is anyway.
First off, my math says 1850 to 2011 is 161 years, yet graph two puts us at ~230 years on red. If all the other lines are correct on the x-axis then we (actually you and our children, I won’t see it) might be lucky enough for 40-90 years of general rise in temperature, based on past patterns.
Second, on graph one, the red line seems over extend. Please double check your ‘splicing’.
So now let’s look at the trends. Draw a straight line starting from the Minoan warm peak, through the Roman, MWP and Current WP; this is the systems topside resistance, for whatever reason, to temperature rise in the last 4,000 years. All peaks are roughly evenly spaced showing a non-random pattern. IF the temperature does not break through to the topside as it did after the cold period ~5280 BP, then it must go down.
Now, let’s look at the support the system displays on the low end: draw a line connecting two major bottoms; One at ~6260 BP and the other through ~1580 BP. The 2 lines cross ~1100 years from now which, based on the time period of peaks, would be roughly where the next PEAK is AND that peak is at or below the low of the LIA!
So what does this pattern imply?
1. We are probably at or near the peak temperature of the Current WP with a significant decline coming in the very near future.
2. That decline, based on the extent of previous declines will break through the support line we just drew through 6260 and 1580 BP. That’s very bad because that means their is no other system bottom support displayed anyplace in the in the Holocene pattern except during the Younger Dryas and we know how far down that went.
3. We know that looking at the typical past interglacial periods of the ice age we are in, the Holocene warm period has pretty much played itself out. A large temperature drop, 80K years of advancing ice sheets and a massive decline in the Human population (down ~5 billion from present) is on the way soon. Hopefully beyond all our lifetimes but it is going to happen as sure as a free falling hammer will hit the ground.
So, as I see it, the rational thing to do is plan how to survive as a civilization and as individuals / groups in the coming deep freeze.
David Loufek says: July 3, 2011 at 2:48 pm
So, as I see it, the rational thing to do is plan how to survive as a civilization and as individuals / groups in the coming deep freeze.
What we would learn on the way to preparing would be much more beneficial than scrapping the world economy and science over an unsubstantiated theory of warmth.
MM IT TOOK A WHILE FOR THE TROLL TO COME OUT, MUST BE THE COLD!
Bill – thanks for the data – Dave
HenryP says:
July 3, 2011 at 12:31 pm
“All we need is two things.
Free enterprise and oil.
The rest takes care of itself.”
We are screwed if this next election doesn’t bring us some “change”…