Sunday, July 15, 2018

On Records


A distinguishing mark of a new record in a time series is that it exceeds all previous values another is that the first value in a time series is always a record. 

Given a stationary situation with nothing except chance, aka natural variation, the number of new records should decline to zero, or pretty close, as the series extends in time.

If governed by natural variation the rate at which it declines should be a marker of the nature of the wings, e.g. the distribution characterizing the climate from which the weather is sampled.

It’s pretty hot out there, with all sorts of new record highs being recorded. They are not declining to zero. This is strong and convincing evidence of increasing global temperatures.




That, and it being pretty hot out everywhere

Wednesday, July 11, 2018

The Betts Test

Having engaged himself in a worthy round of hole digging while trying to explain why he never used what he himself acknowledged as the excellent Skeptical Science’s explanations about climate issues, Richard Betts fell back on defending Ethon’s favorite, Roger Pielke Jr., against the Skepticals listing said Roger as a climate misinformer.

This too did not go so well until Richard came up with the Betts test, that there were nowhere to be found on said list those who were really out of their mind worried about the dangers of climate change (the name of Peter Wadhams came up in this regard and indeed Dana Nucetelli has a few words on this)

However Eli finds the Betts test a good one and would point to an interesting example previously discussed here on Rabett Run, specifically Roger Jr.’s Convenient Truthiness. Now at the time, Roger’s Prometheus had not yet been pecked to death by Ethon, and Rabett Run was but a small vanity blog, but Eli had asked over there why Roger reserved all his fire for Mike Mann, Al Gore and the IPCC. The answer came

Question: Why don’t I write about glaciers, solar variability, Fred Singer, or Pat Michaels?  
Answer: I don’t know anything special about glaciers, solar variability, or the issues which are often discussed by Fred Singer or Pat Michaels. By contrast, I do know something about disasters and climate change. In fact, I know a lot, perhaps as much as only a few dozen people.
Eli was not particularly impressed
Other than the fact that the Google turns up a mess of Singer / Michaels pronuncimientos about climate change and disasters, we here at the Rabettorium were under the impression that the good Prof. Pielke runs a SCIENCE POLICY INSTITUTE and is always telling us that he loves science policy and we don't.

S. Fred and Pat have been playing in the science policy patch like forever, and maybe before. I am morally certain that RPJr has never, ever read a single word of that stuff. On the other hand, I am a bunny, and you know about the morals of hares. In the words of Dorothy Parker about a particularly childish children's book, Tonstant Weader Fwowed up.
UPDATE: Well, it looks like Roger knows S. Fred well enough to invite him to lecture his classes. FWIW, Trenberth, Sawitz et al also came. Roger also knows Pat Michaels work well enough to cite it in his publications on hurricane damage and elsewhere, as well as the fact that his father and Michaels are co-authors and long term collaborators. In short, the deniability here is not even plausible.
You want the links bunnies, well go to the original

Then, not a single comment from the audience now, well, ATTP has gathered a few but he always does a better job of that than Eli, but none at the time.

As to Roger, the Betts test is a wonderful razor to tear apart the envelop of his latest and Richard might consider taking a Betts test hisself.  Eli will be back with the results.

Tuesday, July 03, 2018

Hansen 1988 Retrouvé

Celebrating the 30th anniversary of the 1988 publication of Global climate changes as forecast by Goddard Institute for Space Studies three-dimensional model Hansen, J., I. Fung, A. Lacis, D. Rind, S. Lebedeff, R. Ruedy, G. Russell, and P. Stone Eli along with others (like Gavin Schmidt, Nick Stokes and others) Eli went back and took a look at what had been discussed in these parts as well as those.  There are a lot of very good people who contributed to that paper, so it henceforth hereabouts shall be known as HFLRLRRS.

One of the most important things is that the current discussion focuses on the predictions of forcing as well as outcomes.  The first is more or less ecomonics, the second physics.

That being the case, HFLRLRRS were pretty good on both.  Back in 1988 Hansen described the choice of scenarios
For the future, it is difficult to predict reliably how trace gases will continue to change. In face, it would be useful to know the climatic consequences of althernative scenerios. So we have considered three scenarios for future trace gas growth, shown on the next viewgraph. 
Scenerio A assumes the CO2 emissions will grow 1.5 percent per year and that CFC emissions will grow 3 percent per year. Scenerio B assumes constant future emissions. If populations increase, Scenerio B requires emissions per capita to decrease. 
Scenerio C has drastic cuts in emissions by the year 2000, with CFC emissions eliminated entirely and other trace gas emissions reduced to a level where they just balance their sinks. 
These scenarios are designed specifically to cover a very broad range of cases. If I were forced to choose one of these as most plausible, I would say Scenario B. My guess is that the world is now probably following a course that will take it somewhere between A and B
In 2006, Eli pointed out that the CO2 prediction was eerily accurate


It shows that many details in emission scenarios are unimportant, or rather that false estimates in one direction are most likely going to be cancelled by false estimates in the other for a different forcing agent, and that on average the scenarios should be useful for larger periods.
Which is what happened using the illustration from Real Climate


A neat thing in HFLRLRRS which has not been commented on much were the color coded maps of temperature anomalies by decade in the future which can be compared to measurements. Eli finds this much more interesting and informative about the early GISS model than the time one dimensional variation of averate global temperature change.  Comparing the result for the 2010s with measurements between 2014-2016 show that HFLRLRRS captured the broad picture but was not perfect


Arctic amplification is clear in the model and 30 years later in the observation.  Otoh, the Antarctic in the model is clearly too warm as is the Arctic south of Greenland. The third pole (the Tibetan Highland is too warm also. It would probably be better to map out more of the current decade wrt the El Nino - Southern Oscillation, 2016 being a monster El Nino, but the period also includes about a year of La Nina.
Another, not remarked much upon thing is that  HFLRLRRS calculated the temperature change compared to the interannual variabilty of their 100 year control run.  The change is everywhere (with the exception of the Indian Ocean off Perth) positive compared to the null of the control.  What would be interesting (to Eli) to compute would be the model interannual variability for the 2010s compared to that of the control run.  Indeed were Eli in charge, CIMP runs would include such a metric.