Tuesday, August 15, 2017

Robert Gould Shaw and the Massachusetts 54th Regiment Memorial



The Massachusetts 54th Volunteer Infantry was the first African American regiment formed in the US Civil War.  Lead by Robert Gould Shaw the regiment endured heavy casualties and fought with valor. Shaw was killed in the attack on Fort Wagner South Carolina in 1863.  Sgt William Carney was the first African American to earn the Congressional Medal of Honor for his bravery in that fight.

Dedicated in 1897, the monument was created by August St. Gaudens and paid for by private donations.  The relief shows Shaw and the troopers marching off to a war that would free their countrymen from chattel slavery.  It sits opposite the State House on Beacon Street in Boston.

At the dedication
The military units present began to march past the Memorial, led by 65 veterans of the 54th Regiment. Some of the officers wore their Civil War uniforms, but most of the enlisted men were in their best frock coats. Black veterans from the 55th Massachusetts and the 5th Cavalry were also present. Among the men of the 54th, Sergeant Carney carried the American Flag. The sight of him elicited cheers from the onlookers who knew of his exploits. The 54th veterans laid a large wreath of Lilies of the Valley before the monument. All of this deeply moved Saint-Gaudens:

"Many of them were bent and crippled, many with white heads, some with bouquets... The impression of those old soldiers, passing the very spot where they left for the war so many years before, thrills me even as I write these words. They faced and saluted the relief, with the music playing 'John Brown's Body'…. They seemed as if returning from the war, the troops of bronze marching in the opposite direction, the direction in which they had left for the front, and the young men there represented now showing these veterans the vigor and hope of youth. It was a consecration."


Monday, August 07, 2017

Making the Elephant Dance as Performed by Ned Nikolov and Karl Zeller

As John von Neuman put it
With four parameters I can fit an elephant, and with five I can make him wiggle his trunk.
and, as fate would happen, along comes Ned Nikolov and Karl Zellner with "New Insights on the Physical Nature of the Atmospheric Greenhouse Effect Deduced from an Empirical Planetary Temperature Model".  It's basically 22 pages of word salad and Eli may later return to pointing out some of the more amusing light fingered moves, but here the Bunny will only provide a small amuse bouche with the "interesting" exercise in fitting five numbers with four free parameters, two unphysical constants and a free choice of fitting form.

So briefly, what goes on is to fit the five average surface temperatures of five plants or moons (Venus, Earth, the Moon, Mars, Titan and Triton.  Wait you say, that's six, not five, but they leave Titan out of the mix because  (Eli told you this would be yummy) to an arbitrary functional form

y = a exp(bx) + c exp(dx)

Wait you say again, ok, that is four parameters and the functional form plucked out of thin air, but what is x and y.  That's kind of interesting and more than a bit light fingered but you have to watch the moving cup.  The independent variable is a ratio of pressures (Ps/Pr).  Ps is the pressure at the surface, Pr, well that's interesting, Pr starts out as the "minimum air pressure required for the existence of a liquid solvent at the surface, hereto called a reference pressure (Pr)" but about a page further on it morphs into 
For a reference pressure, we used the gas-liquid-solid triple point of water, i.e., Pr=611.73Pa [38] defining a baric threshold, below which water can only exists in a solid/vapor phase and not in a liquid form. The results of our analysis are not sensitive to the particular choice of a reference pressure value; hence, the selection of Pr is a matter of convention.
The alert out there have noticed that the minimum air pressure required for the existence of a liquid solvent at the surface kind of depends on the temperature of the surface, and would vary widely from planet to planet. Of course worry bunnies like Eli might ask:  What liquid?  Water exists as water on the surface of the Earth, if there was any as steam at Venus and as ice at all the others if it exists there at all.  For Venus maybe CO2, but at the surface of Venus CO2 is a supercritical fluid and you can't tell the difference between liquid and gas.  At Titan, there are oceans, but oceans of methane, so any useful Pr is going to be wildly different for all of these bodies and, in the case of the Mars, and Triton some pretty fancy liquids are going to be needed.

Selection of water as the solvent of choice is then both arbitrary and unphysical.  But why do Nikolov and Zeller insist on using it? Turns out their elephantine trunk waving depends on using dimensionless variables, but restricting Pr to an inappropriate value independent of the planet is equivalent to stripping the units off of the surface pressure Ps.

How about y.  y is defined as the ratio (Ts/Tr) with some really serious trickery buried in TrTr is defined as a reference temperature,
the planet's mean surface temperature in the absence of an atmosphere or an atmospheric greenhouse effect.
At this point no bunny should be surprised to find that that ain't quite that.  Whoa.  Where that come from.  Old timers may remember Eli's old friends Gerlich and Tscheuschner who were also in the business of trying to falsify the greenhouse effect, by as was pointed out, not understanding what the greenhouse effect was.  As Science of Doom put it
 Gerlich & Tscheuschner have written an amazing paper which had the appearance of physics yet failed to address any real climate science.
Eli and several distinguished bunnies had a run at G&T, but, of course, as such things go, the majicians never give up, and one may anticipate a visit from Nikolov and Zeller too.  Good times to be had.

Anyhow, one of the results was a nice arXiv article by Arthur Smith explaining how the surface temperature of a rotating planet varies with the rotational period and the heat content of the surface, which for the earth is basically that of water.

The parameter λ for the Earth is 0.04 and describes the ratio of the energy absorbed from the sun in a day to the heat capacity of the surface.  The effective temperature is the temperature determined by emission from  the surface on a non-rotating body needed to maintain thermal equilibrium.  Depending on your model Arthur showed, as was well known, that the average temperature of the surface of a rotating planet without greenhouse gases has to be less than the effective temperature.

Nikolov and Zeller reproduced Smith's results in another paper with one very strange twist.  In their model  they insist that every planet without an atmosphere will have the same surface as the moon.  (Basically λ =20 in the figure above.)  Using the bare moon Tr now Tna, is, again arbitrary, but let's go ahead and look at the fit which is all John von Neuman told you it would be

Further hand waving ensues.  Nikolov and Zellner will soon be here to entertain you.  Eli warns the bunnies they are indefatigable and will tell you to read the paper.  Eli's advice is if you want some laughs go ahead.  Scott Denning has been trapped into the endless circle, so be sure to take some survival rations for him.



Friday, August 04, 2017

The face of a champion




Context (link here if the video doesn't work, just scroll down):

Wednesday, August 02, 2017

Tyler Cowen and defensive innovation: reasonable idea, problematic execution

Tyler Cowen has an interesting point about "defensive innovation" meant to prevent future problems, wherein he describes climate change as a future problem and Tesla as a defensive innovator that solves the future problem without necessarily improving human lives over baseline conditions.

It's a reasonable point viewed in isolation, that we may overestimate how technology improves lives over the baseline condition when it just solves problems we anticipate. The problem is overplaying the point. One example is Cowen referencing a cure to Alzheimer's as a defensive innovation - yes it's true that Alzheimer's will likely become far worse of an affliction as our population ages in the next few decades. OTOH, it's already a tremendous problem today - as an economist, Cowen should appreciate the tremendous, current economic cost in treatment and decreased earnings from Alzheimer's today, let alone the negative utility from human suffering (and to be fair, he doesn't completely ignore this).

Climate change is a comparable example - for one thing, it's already harming us. For another, environmental impacts are rarely unique - the pollution that causes climate change has many other negative impacts on health and the environment.

A thought experiment:  compare a modestly optimistic future for our Earth in the year 2050 relative to Earth 2 in another universe, exactly like ours except some quirk in physics or technology keeps greenhouse gases from building up in their atmosphere. Earth 2 will be still be using coal and petroleum in 2050, with resulting air quality health impacts and devastated environment from mining and spills. They'll talk about switching to wind and solar, but those technologies will be expensive because Earth 2 hadn't spent many decades subsidizing research in those fields. Earth 2 will also be poorer than us because fossil fuel energy there will be more expensive than cheap renewables here.

Other than the climate impacts, Earth 2 will be worse than Earth. Climate mitigation is making Earth a better place.

One final point - from a policy perspective, it doesn't matter whether an innovation is defensive or an improvement over current conditions. If it fixes a problem in the future at an acceptable cost, then it doesn't matter whether that problem exists in the present or just in the future - you should still fix it. Climate change exists in the present, but regardless, the future catastrophe is well worth preventing.

Friday, July 28, 2017

Eli Responds to Gerlich and Tscheuschner


Sometimes Eli gets the feeling he was created for Twitter.  In answer to the old question of how a cooler atmosphere with greenhouse gases can warm the surface


Extended remarks from John Tyndall in 1859, who put it better
[T]he atmosphere admits of the entrance of the solar heat, but checks its exit; and the result is a tendency to accumulate heat at the surface of the planet.

Original Climateball

Andy Dessler set off a bit of a tweetstorm asking where the definition of climate sensitivity as 2xCO2 came from.

Turns out the answer is Arrhenius in 1896
but, to meander to the point of this post for another reason Eli was guided to an old favorite, Through the Looking Glass, and the Original Climateball dialog between Alice and Humpty Dumpty.  To just pull a few lines out of the mouth of the egg:
'You don't know what you're talking about!' cried Humpty Dumpty. 'How many days are there in a year?'

'Three hundred and sixty-five,' said Alice.

'And how many birthdays have you?'

'One.'

'And if you take one from three hundred and sixty-five what remains?'

'Three hundred and sixty-four, of course.'

Humpty Dumpty looked doubtful. 'I'd rather see that done on paper,' he said.

Monday, July 24, 2017

Issues with bat/wind turbine study, worse reporting, and awful Op-Editing

(Maybe worth emphasizing this is Brian writing, not Eli or John.)

A new study gives some reason to believe that wind turbines have secondary effects on bat mortality compared to other anthropogenic factors like intentional killing, accidents, and the imported fungal disease called White Nose Syndrome. You'd have trouble knowing that though after reading the bad reporting in Scientific American and worse Op-Editing by a Manhattan Institute 'scholar'. To be fair to the secondary reports, the study itself could've done a better job discussing its results.

Disclaimer time! I'm not a scientist, let a alone a bat specialist, maybe I'm off. Regardless, said specialists probably will get something useful from the study, but it's broader public effect isn't so great.

So, problems with the study:

1. It is not a study of bat mortality. It's a study of Mass Mortality Events (ten bodies or more). If you don't know the relationship between MMEs and overall mortality (and apparently we don't), then you don't know the importance of MME causes to bat conservation. Aside from one throwaway sentence (that many bat species are gregarious and therefore likely to have MMEs) this issue isn't discussed. There's also no discussion of habitat destruction except when the destruction creates MMEs, when habitat destruction is likely far more important than any other factor.

2. Actually, it's not a study of MMEs, it's a study of reported MMEs. In other words, nobody went out and did transects in places with bats to survey for MMEs - this study just looks at reports of MMEs however they came to be, creating a significant bias because what gets reported is not even an attempt at a random survey of MMEs. This is not well-acknowledged in the study, with an important exception saying wind turbine reports are biased higher because of mandatory reporting requirements that don't exist for other causes. Subsequent reporting on this study by others omits this disclaimer.

3. Worst of all IMHO is they included qualitative reports of MMEs (i.e. reports of "many" dead bats) and they did not adjust measured mortality quantitatively for the number of deaths in each MME. So a MME of 10 bats counts equally in their study with one killing 10,000 bats (and they acknowledge some MMEs at that level and higher). I think this is worst of all because it seems like something they could actually fix, while the first two problems are limitations they couldn't fix but could have acknowledged more readily.

Related to this last flaw of what MMEs were considered is that they excluded MME reports of food markets and of bat imports. My cynical take is they excluded those categories because they'd overwhelm the others and highlight the problem of selection bias for reported MMEs. They say they excluded food market reports because it's been studied elsewhere, an explanation that doesn't make sense when surveying relative causes of mortality.

Disclaimer time again! I didn't read the supplements which might give some defense against my criticism, but they weren't attached to the Google Scholar link. I supposed I could've been more industrious and contacted the authors, but I also think these flaws should've been addressed in the study itself.

Anyway, some props to the study for disclosing its limitations even if they could've highlighted them and done things differently. Below are the key category results IMO - they disclose and then ignore the figures in parentheses, we will do the opposite (and note the study includes other categories that aren't relevant to this post):

Category                          Total MMEs (and order of magnitude for maximum number carcasses in single events per category)

Intentional killing           205 (10x5)
Accidental                         66 (10x4)
Wind turbines                  281 (10x2)
White Nose Syndrome   266 (10x4)

What I take from this is that of these four categories, wind turbines kill the fewest bats - by two or more orders of magnitude. Contrast this with the study abstract that says

 Collisions with wind turbines and white-nose syndrome are now the leading causes of reported MMEs in bats.

Scientific American takes that to mean

wind turbines are, by far, the largest cause of mass bat mortality around the world

And Robert Bryce of the Manhattan Institute:

wind turbines are now the largest cause of mass bat mortality. 

People who don't read the original study (Bryce doesn't even link to it) are unlikely to catch the importance of the qualifier "mass," or that they're measuring events and not the numbers of bats killed in mass events. Even with all that, Scientific American's "by far" is completely wrong. Bryce gets many other things wrong or exaggerated in his anti-wind jeremiad.

One wrinkle to this is that the study shows a large change in MMEs since 2000, with far fewer of other categories while nearly-identical, large numbers of MMEs occur from wind turbines and White Nose Syndrome (maybe 37% of events from turbines and 36% from WNS). That still doesn't change the fact that WNS kills bats by two orders of magnitude more, nor that there's a reporting bias to show more wind turbine MMEs.

I'm not rejecting that turbines killing bats are a concern (my idea btw is to put high-frequency, very short distance sonar warning noisemakers on problematic turbines), just how it's being discussed. The study does mention climate change as a future and present-day impact on bats. Bryce somehow omits that.

What this all needs is perspective.

Sunday, July 23, 2017

How is A not B?


As part of another post not finished, Ethon took a look over at the nameless one's twitter feed where he has posted a talk.  Now, not to be snarky, but it appears that the nameless one has really put his hand on the problem with discussing climate change, and, it is summarized in one slide, to which Eli asks how is A not B


This may not be the point that the nameless one was trying to make, prescribing more interchange amongst the sides, but Eli asks who blocks more honest bunnies than the nameless one?

And why was he not struck by the contrast?  Also for Fernbach et al.

Friday, July 21, 2017

The joy of meh over California climate law

I've been following but not talking about the runup to California's recently re-enacted climate legislation, authorizing more action from now to 2030 and removing the legal uncertainty that existed in the previous law over cap-and-trade.

I didn't know what to say about the bill supported by the mainstream environmental groups as part of a deal with industry and Western States Petroleum Association, versus alternatives supported by harder-line enviros. People who knew more than me about what was going on seemed torn (although they eventually lined up one way or another) while people who knew less were very confident.

The good thing about it is that the choice in California was between striking a compromise that still ended up as probably the strongest legislation of any state, versus taking a risk on something stronger that might fall apart. That's not the choice on tap in most other states.

Everyone has biases. The mainstream environmental bias is to make deals, and the hardline bias is to reject deals. I'm pretty amazed that WSPA took the deal - it's going to be hard for them to argue in other blue states in the West that they should accept nothing after having accepted this. That to me is an important gain.

Regardless, this is what we've got, and now executing on it is the important thing.


UPDATE:  Steve makes a good point in comments - prior law requires 40% reductions in GHGs by 2040, so WSPA had little leverage as opposed to facing potentially-crippling regulation. I had known about that, but stupidly overlooked it when thinking about tradeoff choices for this post. Still, California has backed off on regulatory approaches in the past (e.g. Zero-Emission Vehicles), and the precedent for other states still applies.

My other beef about this is I haven't seen reported what the new cap and trade price floor and ceilings are going to be, other than they're low. I even quickly skimmed the law's text and couldn't figure it out.



Wednesday, July 19, 2017

Keeping Up With the Times Part I


So some pro-tohedgehog (hate to meet the amateur type) has a post on why people don't trust science, with a nice little Socratic dialog.  Only problem is that pro has not yet emerged from hibernation and things have changed or never were.  Let Eli playfully fisk this with instructions for the whiners

Imagine this hypothetical, but potentially very real, conversation with a non-academic:
1: “This research paper has been published, and therefore is scientifically valid.”
Well usually scientists say something like there is a huge number of papers out there on this point, and if they think that a publication is chancy will let you know but whatever
2: “But it’s paywalled, I can’t access it. How do I know it’s valid?”
There are a number of things a bunny could do.  In Climateball Speak do your due diligence.  First search for it on Google Scholar.  You can search under the name of the author or the title of the paper or whatever.  Then look to the far left hand corner.  Often there is a link to an on line open version


If that does not work, why then you click at the bottom on All xyz versions.  That usually only pops up links only to abstracts, but it sometimes brings up a copy of the paper.  For the example above there are two other links to pdf versions. That's another win.

If not send a nice Email to the corresponding author (usually shown by a superscript, something or other in the on line journal abstract which you find by clicking on the title of the paper) who will, in IPCC speak, very likely send you a copy.  If you are really old fashioned send a re-print card.   It will amuse them if they are as old as Eli, it will confuse them otherwise.  Win-win.

Don't start your Email by accusing the corresponding author of being in the pay of whomever you are venting on that day.  The text of the reprint card is not a bad place to start.

Go to the authors' (all of them, sequentially) web pages.  Authors often list their publications with links to open copies held locally, or to preprints of same.  Only takes a moment

Let's say this doesn't work.  Well you could go to a local university library and try and find the paper.  If it is more than a cup of coffee away, you should check the catalog to see if they have the journal and what you need to get access.  If you are nice they will IPCC level very likely let you in, you may have to show ID, and depending on the circumstance let you use their on line services as well as look in the stacks.  These days with smart phones you don't even have to buy a copy card.  There are, of course, local rules.  Eli has been using this method for years where he lives because his place did not have subscriptions to and he has a nice little deck of copy card.  Here is the policy at the University of Maryland College Park
Catalog Visitors can search the University Libraries catalog from on or off campus, regardless of one's affiliation with the university.

Databases On campus, anyone can access the databases without restriction. Off campus, only currently registered UMD students and currently employed faculty and staff can access the site-licensed databases.

Photocopying and Printing in the Libraries

Photocopying and printing are available for a fee. There are no coin-operated photocopiers or printers in the University Libraries, so visitors will need to purchase a Photocopy Card in order to copy/print. Ask at any library Information & Reference Service Desks for prices and information on obtaining a card. Library computers

Visitors are invited to use public library computers, but first must obtain a guest account. Please note that guest accounts are not compatible with Mac computers that boot only into Mac OS. Apply at any campus library Circulation Desk.

Photo ID is required. Acceptable forms of identification include driver's license, state-issued ID card, passport, military ID, school ID, or other institutional ID with photo and unique identifying number. Library computers are available to users on a first-come, first-served basis. 
Pay attention to local rules, by experience, UK university libraries are much more difficult to get access to but the British Library has an on demand service which delivers electronic copies for £5.35 each. 

You could go to your local town, city, state library and ask for an interlibrary loan or a photocopy, you could even pay the charge to rent or buy the paper (horrors).

But let us say that none of that works for you.  In a pinch, of course there is always sci-hub but as with Kodi add ons there are issues oh my there are issues and more issues.  In this sort of thing Dr. Ruth has good advice.

UPDATE:  Read the comments after reading the post.  The Ever Helpful Bunnies (You know who you are) have added a number of additional ways to get what you want. 

Finally a word about publication policies.  Granting agencies the world around have in the past decade required that publications their work sponsors be openly available, often after a six month to a year period.  Publishers have responded by charging different amounts for publications that are immediately open as opposed to those that are open after embargo.

Some publishers (even reputable ones) have gone to a completely open publication model with costs covered by either the authors or by the granting agencies or their institutions

So yes Virgina, if you can't get a copy of a published paper you are not trying very hard.

Monday, July 17, 2017

A Rumination on Energy Costs


So Eli returns from vacation with a report from Ethon who has taken up Twitter on the economics of energy sources.  To put it simply, nuclear and hydro are very long term investments, efficiency, wind and solar are investments and fossil fuels are an addiction.

The first thing to understand is that each of these is subsidized.  The second is that proponents of each of these thinks all the others are subsidized and their favorite is not.  That's another post.

The cost of fossil fuels is pretty much the cost of the coal, oil and gas, although, of course, there are infrastructure costs, but a reasonable estimate (and Eli is the most reasonable bunny you could ever meet, as a colleague just wrote, reasonably insane perhaps, but reasonable nonetheless) .  To be more exact the cost of the fuel is about 70% of the total cost and a portion of the capital cost is the infrastructure to move the fossil fuel to its final resting place before it goes up in smoke (e.g. pipelines, railroads, ships, etc).

The competition, efficiency, nuclear, hydro, solar and wind are capital intensive.  For practical purposes, efficiency is 100% capital, an upfront cost.  At least in theory people weigh the cost of money vs. the money they save, and that depends on the interest rate.  That's theory.  As a practical matter people and businesses are very reluctant to invest even with payback periods of a year or two.

That explains the role of regulations and subsidies, to get people to do what they rationally should do but irrationally won't.  Good examples of this are, for example, power companies paying or subsidizing compact fluorescent bulbs so they didn't have to build more power plants, or building code insulation requirements, or fleet mileage, etc.  Each of these can be played but each of these has a rational effect.

Nuclear, hydro, wind solar are the opposite of fossil fuels with about 70% up front capital cost and 30% operating costs (close enough).

Nuclear to start, comes in large lumps and has a long time between when you issue the bonds to build, spend the money to build and the plant comes on line and starts trading electrons for cash to pay the bonds.  This can only be done by governments, or with guarantees from the government.  The most successful example is France, which took a political decision in the 1960s/70s to go nuclear for electrical generation and provided the resources to do so to EDF which is 85% government owned.

Big hydro is pretty much the same story with the add on that the lake behind the dam covers a lot of ground which requires eminent domain seizures.

So it is pretty clear that nuclear/hydro build out is best suited to places with strong, stable (gotta last more than a decade, let's not talk about the proliferation risk) and well funded central governments, China, France, Russia, maybe India.  The US could do it, but the  free market folk and the NIMBYs would never allow it. (Caveat:  Folk have been talking about small nukes for almost as long as fusion.  Eli is a show me bunny)

Wind and solar are distributed.  The generating facilities are small and inexpensive, Eli could even affords some rooftop solar, and even industrial strength wind and solar are cheap as compared to hydro and nuclear, well within the reach of your local source of electrons.  But, of course, the wind don't always blow and the sun is on a fixed schedule.

However, a big enough network can bring power from where the wind is blowing, the sun shining and there are work arounds like thermal/hydro storage.  Still, as the Bunny agreed in 2006

In a Science Policy Forum article entitled "A Road Map to US Decarbonization", (available in part in the Energy Bulletin) Reuel Shinnar and Francesco Citro point out that while nuclear is well suited to support baseload electricity generation, solar is ideal for handling peak demand, being most available, when most needed, during the hot days.
Moreover, we still have a decade or two where baseload could be handles by gas turbines which have spin up times under 30 minutes, and for those times when there is excess wind/solar, why free markets were made for bunnies who know an opportunity when they sniff it

oh yeah, Russell has a special on offer over at Real Climate



Sunday, July 09, 2017

Trump caught by his defamation of women he assaulted

Last fall I suggested both Trump and his campaign could and should be sued by the women he called "liars" when they accused him of assault. Trump also claimed he'd sue for defamation, that was just another lie of his.

So he has been sued, and is trying to wriggle out of it. We'll see what'll happen - the idea that defamation is legally-protected "hyperbole" won't go far, but the problem of suing in state court is a little more serious. The argument is that federal courts are supervised by the executive branch's co-equal, the Supreme Court, to prevent shenanigans while state courts are not. My understanding though is that it's usually not hard to find some reason to file a state law claim in federal court, so this is at most a delaying tactic.

Another issue is that only one woman has sued Trump so far out of the dozen or so he defamed (obviously suing a vindictive millionaire president is not an easy thing to do). A case would essentially come down to credibility - Trump has zero and could be torn apart in court, but you still need a significantly-more credible plaintiff. Playing the numbers game would help, but there's still time for others to bring their own suits.

I've got mixed feelings about the plaintiff lawyer being Gloria Allred. I consider it a strike against a lawyer to be one who hogs the limelight as opposed to lawyers who put the client in front while the lawyer concentrates on winning the case. OTOH, among Trump's many lies is that he doesn't settle suits - he settles them all the time. Allred as a great publicist could help make the downside for Trump sufficient that he settles. Again, we'll see what'll happen. 


Monday, July 03, 2017

Hoisted by my own petard

This picture of nothing is what used to be my favorite and most-convenient gas station, just a week or so ago. I doubt it'll be replaced by a gas station.

I stand by an argument I made four years ago that EVs will do more than become more convenient as they gain market share - they will create a virtuous cycle of making gas vehicles less convenient because the market supporting the gas engine infrastructure, like gas stations, will shrink.

In my case, there's another gas station almost as convenient as this one, but it only takes debit cards that aren't convenient for me. Others are a little further away, but the point remains that the convenience is decreasing. Gas stations are disappearing around the country - maybe EVs have only played a marginal role in that so far, but I expect they'll play a bigger role in the future, and still the balance between gas and EVs keep shifting.

I think in many or most two-car households, having at least one car be an EV is more convenient than two gas vehicles - you choose the EV for as many trips as possible and charge it up when you park or at home, and waste less time going to gas stations. As 200-mile EV range and fast-charging become standard, and as gas stations keep disappearing, the relative convenience will keep moving towards EVs and the virtuous cycle will accelerate.


UPDATE:  some good comments, as usual.
  •  I agree that increased fuel economy has reduced gas demand and gas station numbers, and that in turn reduces the convenience of gas engines. By itself though it's not a virtuous cycle, except to the extent that gas mileage improves further. Absent further improvement, there should be a stabilized point where the number of stations balance with the new, lower demand.

  • I also agree that factors making land more valuable for uses other than gas stations are the primary motivators so far in reducing station numbers, particularly in dense urban cores. While extraneous, this also reduces gas engine convenience. Like increased gas mileage, it only reduces convenience up to a point as opposed to being a virtuous cycle.

  • Fernando's right that gas stations will (and have) reacted to find non-gas ways to boost sales. As they go further in that direction though, they'll have fewer pumps or no pumps, and the inconvenience will still increase. 

  • Gas engine repair and maintenance will also become less convenient - EVs need less maintenance and different equipment, so fewer mechanics will train on gas engines. Those repair bays will get replaced with expanded coffee shops and (possibly) electric charging stations.

  • While EVs have only a small impact so far on gas demand, they can affect what's happening right now based on people's expectations of the future. Take for example a family business that owns a dozen gas stations, with the parents nearing retirement and kids deciding whether to take over versus having a very different career. Those kids may well be concerned about what EVs will do to the business in 10-15 years (and should be) and tell the parents to sell instead. Anyone else thinking of a 10-year investment knows there's a risk that EVs will significantly hurt the resale value. These EV effects on gas stations are happening now.

  • Not an expert on this, but I'm guessing that ultimately there's not much long-term future for stand-alone EV charging stations replacing gas pumps, except on interstate highways. As EV range gets further and further above 200 miles, and as fast chargers become ubiquitous at work, shopping, and home, there just won't be a need except on highways where lots of people are traveling long distances.

  • And finally, the above is mostly predictive rather than policy-related, but if the virtuous cycle is real then it does have policy implications. Aggressive long-term EV targets are achievable and should be pursued because there's a virtuous cycle effect we have barely experienced yet that will make them work. Outright bans on gas engine sales like those proposed in the future for Norway and elsewhere will be politically feasible because the writing will be on the wall concerning EV superiority.

Friday, June 30, 2017

Team B, Red Teaming and Steve Koonin

The Trump administration has been pushing the idea of a "Red Team" to re-examine the EPAs endangerment finding.  As Eli has pointed out, this is not a Red Team effort but a Team B attempt.  Red teams bring in experts in the field who have not been involved in a specific report to critically evaluate it.  Team B was a group run by Richard Pipes to exaggerate Soviet capabilities in the mid-1970s in order to justify an arms build up by tearing down CIA analysis.  Team B brings in ideologues to write a report that conforms to the ideology of those who commission their work. 

Joshua Rovner, in his book, Facixing the Facts, National Security and the Politics of Intelligence"points out where the Pipes Team B exercise went

The actual intelligence picture was irrelevant.  Team B simply assumed that Moscow was actively seeking any technology that would allow it to gain a decisive strategic advantage
 And Team B's imagination was quite fertile 
The Team B exercise corrupted the estimative process in ways that were wholly predictable.  The theoretical benefits of competition were lost because the composition of Team B was lopsided, because the panel spent as much time criticizing the intelligence community as it did evaluating the Soviet threat, and because the outside group relied on open sources.  The administration was warned of these problems in advance but did not intervene to insulate the NIE process from political bias.  On the contrary, it allowed the exercise to proceed in order to satisfy domestic political imperatives.
 There is much more at the link detailing the disastrous errors in the Team B report but the worst outcome was Star Wars, as the following Reagan administration used it as justification for the Star Wars build out. 

So let the bunnies count the ways that this administration's EPA administrator will build out his Team B
  • The membership of Team B Climate will be lopsided
  • Team B will spend as much time criticizing the IPCC and National Academy reports as evaluating the threat from climate change
  • Team B will rely on open sources, lord help us, like Watts Up With That, Curry's Climate Etc.
Any bunny thinking not, well Eli has a few carrots to wager on each proposition.

History Commons has a long discussion of Team B's fantasies including
Lack of Facts Merely Proof of Soviets' Success - One example that comes up during the debate is B’s assertion that the USSR has a top-secret nonacoustic antisubmarine system. While the CIA analysts struggle to point out that absolutely no evidence of this system exists, B members conclude that not only does the USSR have such a system, it has probably “deployed some operation nonacoustic systems and will deploy more in the next few years.” The absence of evidence merely proves how secretive the Soviets are, they argue. 
Climateball players have seen this before, and indeed, the run up to the Iraq War featured exactly the same playbook (see History Commons).

Brad Plumer points to Joseph Majkut at the Niskanen Center wondering what could be wrong with such an exercise.   Now Brad is a reasonable guy and the Niskanen Center is reasonable as real conservatives can be, but when Eli points out that the pawn is poison Majkut replies
Koonin, of course, is the apparatchik who tried to hijack the APS's drafting of their statement on climate change which required, amongst other things, that wiser heads on the drafting committee step in and Koonin huffing off in full regalia.  Eli has written several brilliant posts on the entire farrago but there was one thing that he missed coming from early on in the process, February 2013, which shows what Koonin was up to
The type of statement APS should make – simple & declarative or one that incorporates many details – needs consideration. The Intergovernmental Panel on Climate Change (IPCC) is also due to report on climate change in 2013; using their review as a trigger for an in-depth look at the APS statement is appropriate.

Commentary: J. Trebes agreed that using the IPCC review as a trigger is appropriate. Using it as a scientific basis for our statement will mitigate scientific argument within the APS. S. Koonin cautioned that APS should create its own statement and make its own judgment, separate from the IPCC report.
And he tried, oh my how he tried.


Friday, June 23, 2017

Russia, China, and cyber acts of war

I've been dumbfounded by the willingness/denialism of Republicans and of parties in other democracies to be the manipulated tools of the hostile Russian government. Apparently, rabid nationalism can be turned on and off as circumstances warrant.

OTOH, until recently I've been moderately less-exercised about what the Russians had actually done. Hostile countries steal secrets, so that level of espionage is standard activity. They also selectively release stolen secrets, leavened with lies, to weaken their opponents, so there's nothing out of line in that Russian behavior. What's out of line is that it actually contributed to their preferred candidate's success in the US, with little political blowback during the election (or afterwards).

Maybe this specific kind of election interference should be considered worse than it is normally considered as espionage, but the limited reprisal Obama has appeared to authorize seemed appropriate.

One thing that is different is the more recent news that Russian tried to do more than steal information and spread lies, but appear to have made a serious effort to hack the election systems. What I've read is that was more of a recon than an actual attempt to change the results, but it is different from everything else they did.

Everything else is normal espionage that requires a normal level of retaliation. Hacking an election to stop the elected candidate from getting office is overthrowing the American government, and it's an act of war. I think it's equivalent to an assassination attempt. Maybe it's not overtly violent, but unless you're a pacifist, there are things that are equivalent or worse than violence, and overthrowing democracy is one of them.

An act of war doesn't require a declaration of war in response, it just requires a proportionate response. Hacking the medical records of senior Russian officials and changing their medical prescriptions strikes me as a proportionate response.

I don't know if that needs to be done now (and won't anyway given that the Russian candidate now runs our country) but should be the guide for the future, and communicated to Russians for purposes of deterrence.

The one other factor that I haven't seen discussed is how Russian and Chinese behavior seem so different. China is our real, long-term rival. Putin's incompetence has used up half of the time Russia has to transform itself before oil becomes useless, and there's zero likelihood he'll now start a transition. China, on the other hand, is not engaged in these kinds of political attacks on the US, and that's interesting from a foreign policy perspective.

Seems like China is treating us a potential future enemy - its massive hacking of our systems are designed to crash those systems if it needs to in the future. Russia is treating us as a current enemy. Different tactics, requiring different responses from us.

Wednesday, June 14, 2017

Advice From teh Communicators

An evergreen across science communication is that scientists don't know how to communicate science.  Eli has confronted this issue before at the cost of ticking (permanently) off a bunch of communicators

. . .  a whole lot of other people appear to think that scientists are lousy communicators, and indeed, a whole lot of scientists agree and there are workshops, meetings and even, shudder, blogs, devoted to self improvement, or not. This goes into the file under missing the point.

It's not that scientists are or are not lousy communicators (say that and Eli will lock you in a room with Richard Alley for example), but that journalists are lousy communicators. It's their fucking (emphasis added) job and they are screwing it up to a fare-thee-well. It ain't just climate either. What journalists produce often makes the average cut and paste student paper blush with modesty
Well that, of course points to the communicators, who are not just journalists, and indeed some journalists are doing a good job communicating science, others, of course, not so much.  The not so much camp is dominated by the opinion communicators like Bret Stephens, like Matt King Coal Ridley, like James Didn't Read the Literature Delingpole and others.  The perversity of this is the New York Times, which hired at the same time Bret Stephens and Brad Palmer Plumer and now Lisa Friedman in addition to the esteemed Justin Gillis.  Of course what happens is the trio of reporters best stories get stepped on by the Opinion (don't have anything to do with us boss) Section's know nothings, the public hears cacophony, rolls eyes, decides nothing is settled, climate change is just a side show and moves on.

Of course, there are not just reporters, there are communications experts, the various deficit modelers and the cultural cognition folk and more.  Most of these are simply trying to cut themselves a piece of the pie.  RPJr when he was in the business was a great one for pie slicing.

ATTP has a recent comment on this based on a talk Doug McNeil gave.  And sums it up as
The environment can be difficult and challenging; we should try to say interesting things but also be careful of what we say; it should be relevant but not too complex; we should know the audience, and we should repeat the message.
As fate would have the June 2017 copy of APSNews came across Eli's mailbox (the Heartland Institute never set the Bunny so much as a cross word) and on the back page was an essay by Bill Foster, a member of the US House of Representatives and a PhD physicist who sums up science communication with this gem of advice
On the campaign trail, I learned that there is a long list of neurons that you have to deaden to convert a scientist's brain in to a politician's.  When you speak with voters, you must lead with conclusions rather than complex analysis of underlying evidence -- something that is very unnatural to a scientist.  You also have to repeat your main campaign message over and over again, since you will be lucky if a typical voter will hear you speak for a few seconds -- and those few seconds have to include your campaign message.

Wednesday, June 07, 2017

0.04% Is a Lot of Molecules

An evergreen in the denial crowd is that CO2 is only a very small part of the atmosphere so how could it make a difference.

ADDED: In the comments Mark B points out that

The silliness is that it is precisely because CO2 is a very small part of atmosphere that humans are able to meaningfully change it's concentration. For example we are depleting O2 at the same rate as we are adding CO2, but the change is a negligible percentage of the normal content so only the most pedantic would dwell on it. That is, we've changed the CO2 concentration by about 45% and the O2 concentration by about 0.06%.
The short answer is that the atmosphere is very big.   Eli has a nice BOE (back of the envelope, not quite a Fermi problem but Eli would be quite pleased if others thought it in the neighborhood of same) answer

"... the estimation of rough but quantitative answers to unexpected questions about many aspects of the natural world. The method was the common and frequently amusing practice of Enrico Fermi, perhaps the most widely creative physicist of our times. Fermi delighted to think up and at once to discuss and to answer questions which drew upon deep understanding of the world, upon everyday experience, and upon the ability to make rough approximations, inspired guesses, and statistical estimates from very little data." 
 It starts by estimating the number of molecules in a m3 of air.  Well a Bunny who knew Loschmidt's number 2.7 x 1019 cm-3 or 2.7 x 1025 m-3  (which is the same thing since 1 m3 is 106 cm3 ) could start there or you could rearrange the ideal gas law
pV = nRT to n/V = p/RT
Since 1 atm is 101, 325 N/m², the gas constant R is  8.314 J K-1mol-1 and 0 C is 273 K
n/V = 101 325 N/m² /(8.314 J K-1mol-1 x 273.15 K) = 44.64 mol m-3 
which is a little surprising, since the average weight of a molecule of air is ~ 29 g or 0.029 kg so a cubic meter would weigh 1.3 kg but that is another direction.  In any case since there are 6.02 x 1023 molecules per mole that gives us Loschmidt's number again, in case a bunny has forgotten it or 2.69 x 1025 m-3.

If 400 ppm or 0.04% of that is CO2 there are  1.07 x 10 21 CO2 molecule in a cubic meter.  A useful estimate of the average distance between CO2 molecules is the inverse of the cube root. of the number density.  That is 4.5 x 10-8 m.

So how does that compare to the wavelength of light at which CO2 absorbs light in the IR.  Hmm, that's about 14 microns.  A micron is a millionth of a meter, So how many CO2 molecules are there along one wavelength of IR light where it is capable of absorbing.

About 300.

That's enough

Monday, May 29, 2017

Hybrid renewable systems

Kind of new to me, but obvious enough:  wind blows some of the time when the sun doesn't shine, so put wind generation and solar on the same location and reduce some infrastructure cost while getting less-intermittent power. Obviously it won't work everywhere, but it helps. I read somewhere (and sadly can't find the link now) that night winds are very reliable in India during monsoon season, and India's the big challenge now that China is all-in on renewables, so this could be huge.

Alternatives include renewables with large hydro and with power storage. And my personal favorite, floating solar panels.

Tangential thought: we would live in the energy world that denialists think we live in if it weren't for solar and wind (and soon, battery storage). I mean that denialists argue we can't maintain a modern lifestyle without fossil fuels. How that translates within their minds into climate change not happening is unclear, but regardless, that view of the energy picture has been wrong for a decade. And now even the denialists have to add a throwaway statement that "I support solar and wind too" before defending massive pollution of our environment.

Is it just luck that wind and solar and hopefully storage are taking off in terms of cost savings just in time to save us from ourselves? Certainly it's also a function of years of government-funded research, but other fields like wave power, instream hydro, and biofuels have had the same research with limited results. Maybe I'm just looking at the gift horse in the mouth, but if the technology for solar and wind were 20 years behind where they are today, then we'd be in a hell of a mess on climate. I'm curious why it's worked out relatively well.

Some thanks to Jimmy Carter perhaps, starting something that Reagan couldn't quite totally bollox?

Thursday, May 25, 2017

Government Regulations and the Law of Chocolate Chip Cookies

Ms. Rabett when still employed worked in a nest of libertarian types who complained about how complicated government regulations were, and how it was stifling business and most importantly how they could make more money without them.  The same folk would wail about how bad government was when something bad happened and there was no law or regulation to stop it. There ought to be a law they screamed.

Now Ms. Rabett is more than somewhat of Eli's disposition or as she would say visa versa and she realized that governments do not always write crazy regulations to pass the time of day, but more often because some outstanding libertarian tries something crazy and goes nah nah there ain't no law agin it.  There are so many regulations because there are so many libertarians out there trying to game the system.

Many years ago, Angry Bear explained the Law of Chocolate Chip Cookies

So the process continues… Eventually, the Army has a spec that indicates even situations that a rational person would say – “This makes no sense. Everyone knows that.” But the rational person wouldn’t realize that when the Army specifies that no sawdust is to be used in making flour, or that no more than X parts of per million of rat droppings will be in the cookie, that the Army has a damn good reason for having that in there, namely that some upstanding leader of the community who waves a flag and is a member of the local Kiwanis actually tried to pass such things off on American military personnel. And of course, that upstanding leader of the community who waves a flag and is a member of the local Kiwanis is happy to lecture one and all about how much more efficient the private sector is than the public sector – exhibit A being the Army’s specs on making a chocolate chip cookie.

CO2 Atmospheric Absorption Is NOT Saturated

It is certainly an evergreen claim by the climate change disbelievings crew that the absorption of CO2 in the atmosphere is saturated.  What does saturated mean to them is a useful question to ask. A useful answer would be that the atmosphere is optically thick at the greenhouse effect relevant frequencies/wavelengths where CO2  absorbs, between about 620 and 840 cm-1.

It would also be useful to describe what is meant by optically thick and optically thin.  To do that we first need to define optical depth.  Optical depth is the fraction of light blocked in passing through a medium.  The transmission is the percentage of light that gets through.  Something is optically thick at a particular wavelength if no light can get through it, It is optically thin if most or all of the light can get through.  If an absorption is not optically thick, it can't be saturated

If the disbelievers are right at current concentrations CO2 is optically thick over the entire region.

We can check on that using Spectral Calc, a program that allows us to calculate the spectrum based on precision and verified measurements.  Let us imagine that the atmosphere is a tube with 400 ppm CO2 at 296K.  How much of the light is absorbed in a 1 m tube


At this point those interested in only the bottom line can skip down to the bottom of the post and pick up the figure the bunnies need for their tweet.

ORIENTATION

Most of the spectrum is due to transitions from the CO2 ground vibrational level to the first excited vibrational level  The sharp peak in the center is called the Q branch composed of lines that are very close together and corresponds to transitions where the rotation(al quantum number) of the molecule does not change.  The band to the left is the P-branch for transitions where the rotational quantum number decreases by 1.  The band to the right is the R-branch where the rotational quantum number increases by 1.

The two little sharp peaks to the right and left of the main bands are Q-branch transitions between excited vibrational levels.  Even at room temperature a small percentage of the molecules are vibrationally excited by collision.  Of course, they can also lose energy by collisions but there is an equilibrium between excitation and de-excitation by collisions with nitrogen and oxygen molecules (mostly) and a thermally driven equilibrium population in each vibrational level.  If a bunny squints really hard she can see the corresponding P and R-branches. These are called hot bands. Why the excited vibrational levels are split and even what excited levels they connect is complicated.  Google books provides an explanation.

If the distance is increased to 10 meters the lines of the 0-1 band are optically thick but there is still space between them, however, the lines do have wings and the wings overlap so even over a 10 m path, there is a noticeable underlying continuum mostly caused by collisional broadening.  The hot bands on either side of the Q branch are now easy to see.  The Q branch 0-1 band is optically thick
At 100 m or 0.1 km the 0-1 transition is almost optically thick and the 1-2 bands are very clear.  Using the squintosope, Q branches for higher lying hot bands can be seen at the edges
For a 1 km path length, most of the 0-1 transition is optically thick (saturated in the disbelieving sense) but light from the surface would still be seen in the wings, where the hot bands are.  
BOTTOM LINE
Finally at 10 km, while the center of the CO2 absorption is optically thick, there are still regions of the spectrum where light from the surface will get through the atmosphere.
Of course, increasing the amount of CO2 in the atmosphere will decrease the transmission in the wings of the bands.  At 560 ppm
and returning to 280 ppm
There are a few things that Eli has not considered in this post but they all would DECREASE the calculated optical thickness. Temperature and pressure decrease with altitude.  This post assumes both are constant. Their effects will be considered in detail in follow on posts,  Simply put the optical depth is directly proportional to density and path length, thus decreasing density with altitude, decreases the average optical depth and increases transmission across the spectrum.  Second at lower temperature there is less population in the excited vibrational levels and the hot bands at the edges of the spectrum are weaker, decreasing the optical depth in the wings, and increasing it in the center 0-1 band.  Since the 0-1 band IS optically thick at very small path lengths anyhow, this increases transmission.  Third, each of the lines is substantially broadened at atmospheric pressure.  A narrower comb of lines is optically thinner.  This would substantially decrease the continuum absorption between the lines.

Bottom line, the 667 cm-1 CO2 vibrational absorption is not optically thick across the entire region of absorption. It is not saturated.

Wednesday, May 24, 2017

What Rich Lowry said about Erdogan thugs attacking protestors in America

He got this one right, at least. Go read.

You don't often get a chance to read a National Review piece and agree with every word.

Monday, May 08, 2017

Probably Not The Place For This


Eli has been watching the reports on today's Senate hearing which features the Ted Cruz - Sally Yates death match.  The general feeling is that Yates did to Cruz what Macron did to Le Pen.  However, rather than getting too deeply into the legal parts of their interchange, the Rabett would like to point out that Yates READ most of her initial answer to Cruz (starting at 1:49 in the video below).

She was clearly prepared for the question.




You might ask what little birdy whispered in her ear, well, let's go to the video from three months ago




Somebunny was paying attention.

Friday, May 05, 2017

The New York Times approach to climate change


Behold the CrapWaffler, the writer that the New York Times thinks is a contribution to the climate change debate. It's what happens when you hire a climate denialist with the implied condition of employment that they can't completely lie about climate change, but merely smear uncertainty and misdirection about undertaking reasonable action (and Stephens still managed to get important things wrong).

The New York Times thinks it has added to the breadth of discussion on climate by getting as close to wrong as possible while not saying much of anything.



Stephens is shocked, shocked, that people would accuse him of "closet climate denialism". The term denier fit Stephens perfectly in 2015 when he wrote that temperatures would be about the same in 100 years, unless he was lying at the time about what he believed. It would be helpful if he now said his beliefs had changed, but all we get instead is crapwaffles.

I often read Razib Khan, an old-school Burkean conservative who also writes a lot about science. Several years back the NY Times hired him and then quickly dismissed him - he had unwisely associated with some simply vile racists, and guilt by association was enough to deem him unacceptable. I disagree, but to think Stephens, whose range is from wrong to crapwaffle, is better just tells you something about the Times. I recently subscribed to the Washington Post instead.

So skip Stephens, and read Razib to see what a thoughtful conservative would say.

P.S. And fellow bloggers, a reminder to add "no follow" whenever you link to Stephens. I'm pretty good about that when linking to denialists


Wednesday, May 03, 2017

Renewable Energy and Creative Construction


One of the weirdest flips in an exceedingly weird year has been the usual suspects going into complete meltdown about there now being extended periods where there is so much renewable energy from wind and solar and hydro that they are not just giving it away, they are paying you to take some of it.

Electricity has become like zucchini at the end of the summer, when gardeners leave a few hundred pounds on your doorstep, ring the bell and run.  Remember when the fusion and fission folk were talking about too cheap to meter, it's now a "problem" for renewables.

In any case, when there is money blowing in the wind somebunny will make money while the sun shines, and indeed this is a classic capitalist system opportunity, that somehow all the Randians and Trumplets let alone your average garden variety Bret Stephens don't appear happy with.  If there is a price differential arbitraging the electricity price is a great way to get rich and the technologies already exist.

There has always been a price differential between wee hours of the morning and the peak demand daylight hours, a differential that many industries have taken advantage of.  The guys with the green plastic eyeshades are no bunny's fools
Kentucky Electric Steel spends a lot of time and money trying to control our electric bill, over $2 million spread over the past eight years. This has reduced energy intensity from 743 kWh per billet ton in 2002 to 480 kWh per billet ton today. That represents an annual savings of over $600k with just our night-time operations; the savings would be even more if we ran during on peak hours, except that the higher power cost would eat them up! 
Aluminum smelters in Germany are already lapping up some of the freibier by using the molten metal as an energy storage medium from whose cooling they can draw power
By varying the rate at which the metal is produced, the plant will be able to adjust the power consumption of the 290-megawatt smelter up and down by about 25 percent. Trimet can soak power from the grid when energy is cheap. It can then resell the power when demand is at its peak. The company can temporarily reduce its power consumption by slowing the electrolysis, cutting the energy drain.
Using stored thermal energy is really old technology.  Ice houses that lasted through the desert summer have existed like forever in Iran and storage of heat from the summer to use in the winter is also a Canadian reality (tip o the ears to Andy Skuce )
The first of its kind in North America, DLSC is heated by a district system designed to store abundant solar energy underground during the summer months and distribute the energy to each home for space heating needs during winter months.
For decades large building have built tons of ice at night when electricity is inexpensive and used the ice to cool the building during the day.  Going by the name of ice storage air conditioning, the technique is now moving into residential units.  Eli first became aware of it in the context of labs using large ice systems for to supply coolants for lasers.  Storage heaters are also coming back driven by the low cost of renewable thermal.

So the next time your electric provider tries to leave some zucchini on your doorstep, smile and use it to charge your batteries, heat or cool your house or some other creative construction.


Friday, April 28, 2017

The Data Lies. The Crisis in Observational Science and the Virtue of Strong Theory


The problem with data fetishists is their choking down a daily flagon of numerical drivel without analyzing the brew.  One of the things that a good scientist knows is how to interrogate the numbers, not waterboard them.  Truth is that useful models improve flaky data and the statistical treatment thereof.

An introduction  for Eli was a talk that Drew Shindell gave, twenty maybe more years ago with a title that ran, "Which should your trust, the data or the models?" about global temperature data in the late 19th century.  The useful conclusion was trust neither, but use them together to produce understanding and improve both.  Yes theory can improve measurements and data.

A nice example is how NIST's acoustic thermometer can be used to define the thermodynamic temperature scale.  Starting with the theoretical result for the speed of sound in an ideal gas as a function of temperature (theory), a carefully built device to measure the same can be used to build a model of the response of platinum resistance thermometers as a function of temperature and then by applying the model PRTs can be used to more accurately calibrate other thermometers.

How about statistics, well most of what passes for statistical analysis these day is unconstrained, so it can wander off into never never land where never is stuff like thermodynamics and conservation laws.  Bart had a nice example of this when discussing the usual nonsense about how observed temperature anomaly data could be explained as a random walk

As you can see, the theory is valid: My weight has indeed remained between the blue lines. And for the next few years, my weight will be between 55 and 105 kg, irrespective of what I eat and how much I sport! After all, that would be deterministic, wouldn’t it? (i.e. my eating and other habits determining my weight)

Wow, if that’s the case, then I’ll stop my carrot juice diet right now and run to the corner store for a box of mars bars!! And I’ll cancel further consultations with my dietician. Energy balance… such nonsense. Never thought I’d be so happy with a root!
The other side of this is the replication crisis hitting the social sciences, most prominently psychology, well, also other stuff.  To disagree with the first link, unlike physical sciences psychology has no well established theoretical consensus against which nutso outcomes can be evaluated. Science is about coherence (a no on that as Alice’s Queen would say) consilience (baskets full of papers having nothing to do with each other but taken together mutually supporting) and consensus (everybunny with a clue agrees on climate change or at least 97%).

So the question really is what should a lagomorphs's prior be for statistical validity.  Clearly, if all you have is the data, the standard of proof for any assertion about the data has to be very high.  Wrong answers at low levels of proof are a reason that out on the edge physicists demand 5 sigma data before accepting that a new particle has been found, that's saying that there is 1 chance in 3.5 million that the discovery was in error if that standard is met. 

On the other hand, in the well established interior of a field, where there is a lot of supporting, consilient work, a whole bunch of basic theory and multiple data sets, 5 chances in 100 can do the job or even 10 in a hundred.  Of course 30 in 100 is pushing it.

Andrew Gelman has a useful set of criteria for priors (same holds for frequentist approaches).  Among his recommendations are for weakly informative priors that
should contain enough information to regularize: the idea is that the prior rules out unreasonable parameter values but is not so strong as to rule out values that might make sense
and those priors should be
Weakly informative rather than fully informative: the idea is that the loss in precision by making the prior a bit too weak (compared to the true population distribution of parameters or the current expert state of knowledge) is less serious than the gain in robustness by including parts of parameter space that might be relevant. It's been hard for us to formalize this idea.

Sunday, April 23, 2017

March for Science - San Francisco

Together with in-laws and friends, including (unlike me) an actual scientist, we went on the march in San Francisco yesterday.






The most hand-written/individualized signs I've seen on any march. Quite a young crowd too, and I'd guess two-thirds female. We'll see if there's a long-term effect - I'm sure it depends on what people continue to do after the march.

UPDATE:  here's something to do about it - get organized, and train to run for office:
Overview: Join us for a day of building political power for the climate movement by training bold, progressive climate activists to elected office at all levels. Potential candidates need to be identified, recruited, trained, and supported in order to achieve elected office—and once there, held accountable by the climate movement.

This training is for you if:
  • You consider running for office yourself in the next 1-3 years
  • You want to help a friend run for office
  • You want to learn how a local electoral strategy could help your campaign

Friday, April 21, 2017

The Squeegee Kid Returns or Steve Koonin on Team B


Eli had a rockem sockem post cued up on the replication crisis (nononono, Eli and Ms. Rabett are much too old for that stuff) but the Squeegee Kid, Steve Koonin reappeared in the Wall Street Journal editorial swamp and duty calls.  When last see, Koonin was huffing off in anger because APS (American Physical Society) leadership had frustrated his designs on their public statement on Climate Change.

Those looking for a primer about Koonin's understanding of climate science could read the short version from Ben Santer who had the pleasure of dealing with him in the red team/blue team exercise that Koonin put together for the APS panel

Another source of real frustration is that Dr. Koonin had a real opportunity to listen. To consult experts in many different aspects of climate science. To do a deep dive into the science. To seek understanding of complex scientific issues. He did not make use of this opportunity. His op-Ed is not a deep dive - it is a superficial toe-dip into a shallow puddle, rehashing the same tired memes (the "warming hiatus" points toward fundamental model errors, climate scientists suppress uncertainties, there's a lack of transparency in the IPCC process, climate always varies naturally, etc.)
or somewhat less pithy though pointed ones from Andy Lacis and Ray Pierrehumbert.

Suffice it to say there is little new in Koonin's latest jeremiad which is merely a continuation of the House Science Committee farce w/o Mike Mann.  As the Weasel has pointed out we have had over 30 years of real red team evaluations of climate science
Well to start with it isn’t necessarily totally stupid, unless it is being run by a group of ideologues with a fore-ordained conclusion for which they’re desperately searching for evidence. How likely is that? Secondly, this is language from a different area (the military; business) being imported into science. If it was being done by the pols, you could simply put it down to ignorance. That it is being done by scientists in an effort to sell their ideas to pols I think you put down to something rather different. But the military and business are areas with rigid hierarchies and enforced obedience and suppression of dissent. C+C are trying to tell the pols that science is like that; and it isn’t. Science already provides all the internal red teams that it needs.

Could the idea actually be of any use? In the present context, I think that’s doubtful. Suppose they did it anyway, what happens? Probably, C+C and their ilk get thrown some taxpayers money to attack their should-be-colleagues, which would be galling but minor in the great scheme of things. They would fail to do anything of scientific use, and that failure might ultimately be revealing, and therefore good. But in the meantime they get a platform to spout nonsense. Ah well, these are difficult times, you cannot expect to choose amongst different good outcomes.
Among the many red team exercises in the US there have been multiple NAS reports on climate change and particular issues involved with climate change.  A major outcome of one was to put the wood to Spencer and Christie's UAH satellite record which was claiming global cooling because of errors.  Then, of course, the Jason (Koonin is one of them) model from the early 1980s as well.  IPCC reports are also massive red team exercises with open commenting.

But as the Weasel points out what the worthies want is not a red team exercise, but Team B.  Team B was a politically motivated operation run by Richard Pipes and populated by ideologues whose reason for existing was to exaggerate the threat from the Soviet Union.  There was a long campaign to impugn the CIA analysis, resulting in the formation of Team B under Pipes leadership.  Their report was a major impetus to the dangerous arms race of the 1980s including the fictional Star Wars programs pushed by the late, and not lamented George Marshall Institute.  As one critic of Team B, Anne Hessing Cahn, wrote of their report "I would say that all of it was fantasy. ... if you go through most of Team B's specific allegations about weapons systems, and you just examine them one by one, they were all wrong."

Joshua Rovner, in his book, Facixing the Facts, National Security and the Politics of Intelligence"points out where the Pipes Team B exercise went
The fundamental criticism of Team B was that the intelligence [read climate science-ER] community relied almost exclusively on "hard data" about capabilities. . . 
an eerie prequel to where Koonin, Curry and Christie want to go. Following Rovner, Eli can also tell you what Team B's report will be based on
Team B also defeated the purpose of the exercise by relying on open source publication rather than classified intelligence.  Although the panelists were cleared to evaluate the same data that went into the NIE [National Intelligence Assessment -ER] the Team B report contained very few references to intelligence.
Perhaps they will also cite Rabett Run, but more likely all the nonsense in WUWT and Curry's blog.  If anybunny wants to save money, of course, we also have any number of publications from the Heartland Institute that can be had for a penny or two. 

Okay, now that the bunnies have done their assigned reading Eli can flip the blog and let them loose in the comments