Thursday, December 3, 2009

Technological progress

A couple of videos that caught my eyes. First is one on an upside of technological progress -- cars today are enormously safer in a crash than cars 50 years ago. This video shows the collision, and the driver crash test dummies, between a 1959 and 2009 car. The 2009 car undoubtedly weighed far less than the 1959. Superior engineering is the key -- a point that Consumer Reports routinely winds up making in their vehicle reviews.
Crash test video.

Digressing a second: It occurs to me that Consumer Reports is probably the popularly available magazine that does the most consistent job of displaying a scientific approach. The typical review article shows what they were testing, how they tested it, adds information about how significant the test differences are like, and so on.

The second video is one on a topic that weather and climate folks are probably more than a little tired of. Namely, the accusation that we didn't realize that there's such a thing as an urban heat island effect. I've never taken up the search seriously, but a few years ago, an urban heat island reference was in the Bulletin of the American Meteorological Society's '50 years ago' column. So, well-known (the referenced article was clearly not the discovery of the effect, just another illustration) by the early 1950s.

The video is from Peter Sinclair's Climate Crock of the week. In it, he carries out a good practice for science -- suppose an argument is correct, then look for observations that will confirm or reject that argument. The argument is that the urban heat island is producing the observed warming trend. Ok, says Peter, if that's the case, we should see that the trend is the strongest (most positive) in urban areas. Now, it isn't hard to figure out where the urban areas are. Nor is it hard to map out what the trends are for different areas of the globe. Compare the two.

In truth, as he illustrates, the warmings are highest in areas that have very few people -- Siberia, the Arctic, and Hudson Bay being leading zones. His figure is for the 2008 anomalies -- after the 'decade of cooling' (what cooling?), rather than the 30 year trends ending that date. If anything, the trend map is worse for the urban heat island fans, as it shows large trends across northern Canada as well.

Nothing obvious connecting the two videos. But the thing is, I have a lot of confidence in engineers to solve engineering problems. One such problem is car safety. Others would be things like more efficient cars, new and better ways of producing energy, and so on. In the 1950s and 60s, it was an article of faith in the car industry that customers did not care about safety. And that if they were forced to engineer safety, they'd go out of business (it would be too expensive). Instead, we have vastly safer cars today, and tens of thousands of people are still alive because of it. The engineers were more than up to the challenge. On the other hand, if the engineers aren't allowed to work on a solution, they won't find it.

I'm not taking up geoengineering in this; that's a topic for a lengthier post of its own. I'm just minded that there are quite a few climate-related technology issues that exist regarding efficiency of old technologies, or new technologies to develop, that we're being told would drive companies out of business, cost jobs, and other alarmist statements -- as there were in the 50s and 60s regarding automobile safety. Yet the engineers found ways of improving safety even as we drove more, drove lighter cars, and so on. And the companies didn't go out of business, indeed make quite a lot of money.

9 comments:

John Mashey said...

Thanks for mentioning the big boost in auto safety! You might find it interesting to see how that happened (or part of it):

In the early, 1990s, auto companies used a few Crays to run "Crash-codes", but they were expensive. Then Silicon Graphics started building scalable 64-bit microprocessor systems (first Power Challenges around 1994, then Origin 2000s in 1996) that were well-suited to this, and much cheaper, which led to most pre-production crashes being inside computers.

In addition, since we did high-end visualization, we helped the car companies move more design off clay models and onto computer, and then made it easier to visualize the crashes.

For example, virtual crash test dummies would be highlighted with intense colors to show stress. An especially striking example was one the Ford folks showed us about a new Taurus design. Within a narrow range of crash angles (not quite headon, at one side), something bad happened:

a) Crash dummy pitched forward. (OK)
b) Runs into airbag (OK).
c) Bounces back (OK)
d) And right then, the left-hand roof strut buckles inward and smashes its skull (bright yellow flash, instantly visible, surprise. Not OK.)

The example was interesting because:

a) Nobody could afford to do the variety of tests in the real world, so people often had to over-engineer otherwise, and the bad behavior was actually pretty rare.

b) It turned out they fixed it, if I recall right, by *weakening* the strut. Quite often, with good design, one could reduce weight and build better crumple zones.... but they needed to do a *lot* of simulations to be safe.

At one point, I think all but one of the world's ~30ish biggest auto companies used these systems or our followons. In any case, the 1990s saw a huge change in the ways auto companies designed cars.

Google: silicon graphics origin crash code
still gets a fair number of hits.

Those were fun days, especially when talking to the Volvo engineers.

Of course, unlike climate simulations, much cheaper computers are now often "good enough" for crash codes, i.e., cars only have so many parts, and at some point the grid elements are good enough that one need not make them smaller.

But still, it was very satisfying to have helped design these systems, work with the third-party software vendors, and then talk to enthusiastic auto engineers about ways to help them build better cars.

Anonymous said...

It's not true - the old 1959 did not weigh much more than the 2009. The difference is only 179 pounds. Not enough to make a difference. They basically crashed 2 cars more or less the same weight. Today's cars are just as heavy as the old stuff, some SUVs and trucks considerably heavier.

Alastair said...

In a paper I found this sentence

"This phenomenon, known as an urban heat island (UHI), can raise temperatures in a city from 2 to 8F (Bornstein, 1987; Chandler, 1965; Landsberg, 1981; Oke, 1987)."

So knowledge of the UHI preceded the growth of scepticism in the early 1990s.

Also, in Rudolf Geiger, 1941 (translated in 1950) a chapter headed "The City Climate" refers to A. Kratzer, 1937 and also states:

"Fig. 172 shows the temperature distribution of a July evening in Kalsruhe. In the centre of the city it is as much as 7 degrees warmer than in open country, ..."

So you are right about the UHI, but I feel you are missing an important point regarding the safety of cars. If one manufacturer had introduced improvements it would have priced him out the market. It was only when regulations were introduced and all manufacturers had an even playing field that the improvements could be made.

Unless the US government, including the Senate, agrees to global regulation of the use of oil and other fossil fuels then there is no more prospect of avoiding the Climate Crunch than there was of the inevitable Credit Crunch, also caused by lack of regulation.

Gareth Rees said...

enormously safer in a crash

... for the occupants, yes. But no progress has been made in protecting people outside the car.

Hank Roberts said...

Gareth, not much, but some:

http://www.google.com/search?q=pedestrian+car+collision+hood+impact

Oh, and for our libertarian friends -- this is a case of people gathering together and organizing for their mutual benefit, also known as "government."

"... some automakers are now also putting the focus on people outside the vehicle, in part because regulations overseas, particularly in Europe, require such designs."
http://www.edmunds.com/ownership/safety/articles/122731/article.html

RBH said...

Having run on a volunteer emergency squad and heavy rescue team for 35 years, I can attest to the fact that practically no one eats the windshield or steering wheel any more. The mechanisms of injury in auto accidents has changed a whole lot over the years.

Anna Haynes said...

> In the 1950s and 60s, it was an article of faith in the car industry...that if they were forced to engineer safety, they'd go out of business (it would be too expensive).

Do we have quotes & references for this?
(and for other such episodes, e.g. ozone, acid rain, ...)

Penguindreams said...

For the car industry, Lee Iacoca, including in his 1980s autobiography, was one of the people most outspoken about the 'public's lack of interest in safety' and, I believe, the damaging costs of providing seat belts.

On acid rain, I think the major speakers of such things were the coal mining companies (including Illinois', as Illinois coal is high sulfur) and electric companies. Not a full source, but a lead.

Cahya said...

Nice blog. Nice to known u.