Hydrodynamics of the Ever Given's blockage of the Suez Canal

Readers of this blog ought to be following the news of the Taiwanese-operated cargo vessel Ever Given's blockage of the Suez Canal.  Earlier this week, the Financial Times' blog featured an article by Brendan Greeley, speculating on the hydrodynamics of how this event occurred.  Though his is not the final word on this episode, there is one passage in his piece worth memorializing here:

Sailors talk about hydrodynamics the way CEOs talk about macroeconomics: they either treat it with mystical reverence, or they claim to understand it and are wrong. Unlike with macroeconomics, though, if you know what you’re doing you can test the propositions of hydrodynamics on actual, physical models in a lab. As in: you build little boats and then you drag them through the water, in a towing tank. Hydrodynamics is what a five-year old would do, if a five-year old had a PhD. 

Bravo, Mr. Greeley!

 

Attention graduate students: try an internship while in school.

Last week, Nature published an excellent article by three Norwegian graduate students on their experiences doing summer internships in industry, while studying in the life sciences at Norwegian graduate schools. Graduate students in any STEM discipline should heed their advice.  I myself did two summer internships in industry while in graduate school.  They helped me try out different kinds of work and understand my strengths, weaknesses, and interests better.  They also allowed me to try out living in different parts of the country.

A few points made by each of the students resonated with me.  The second student, Kathleen, noted her surprise when the marketing team asked her how a new medical device would sell, and who the target was.  "I became aware that, for a product to hit the market properly, the product maker must understand the end user and their needs. A solution on paper might not always be a marketable one."  This is something I actually learned vividly while engaged in a lengthy conversation with a French post-doc I met in San Diego at a fluid dynamics conference.  He was working with a start-up company developing a new immunoassay.  Its engineering was innovative, but the product's value proposition for potential customers was unclear.  What do I gain by adopting the new technology over what I am doing now?  The existing technique was cheap, relatively fast, easy, and familiar to technicians.  The new product would improve the quality of measurement, but this improvement might be too marginal given the cost to adopt the new method and retrain the workforce.  

The third student, Nancy, described one of her lessons as "Quality Management is king" and I fully endorse this perspective as well.  She emphasizes the commercial and regulatory benefits, but I would argue further that there are scientific ones as well.  Quality Management is an attitude that more academic labs should adopt, because it contributes very directly to doing (and documenting) reproducible research.

The first student, Erik, listed one lesson as "Done is the new Perfect."  It is a lesson well earned and worthy.  Ironically, during my first internship, I learned an opposite lesson that I will have to tell on another occasion.  Suffice it to say that sometimes, a product must be absolutely bulletproof when it is ready to ship, especially when it has life-or-death consequences.

I learned other lessons as well.  One was, there is no such thing as a Physics problem, Math problem, or Computer Science problem.  The problems arrive unclassified, and you need to figure out, learn, and deploy whatever relevant knowledge is needed to solve it.

The two companies I worked for were a contrast.  The first one was a small company, and while they owned licenses to software such as Matlab and others, they were not always installed on all the computers in the office, in order to save memory.  Small companies often exist at the edge of survival; there is no largesse.  The second company I worked for was a Fortune 500 company, that provided subsidized housing and a shuttle van (occasionally a stretch limo when the van wasn't available) to take us interns to and from work.  On one evening they took us on a dinner cruise.  The intern manager had $4 million to spend on us interns (most were undergrads, but I was part of the graduate student cohort)  including pay...a situation which did not last in later years.

On the other hand, at a small company, all I had to do to get a decision made was to walk into the boss' office (when he was in town).  At the large company, often I had to fill out forms and collect multiple signatures.

I did end up working for the second company full-time, for about 7 years after I graduated.  But both internships gave me something concrete to talk about at job interviews when I was leaving school....not just academic research that might be of limited interest to employers.  I considered both my internships to be formative experiences that influenced the kind of scientist I am today.

Physicists and astronomers honored on US postage stamps

Today I received an order of US postage stamps, including a sheet of the new Forever stamp honoring Chinese-American nuclear physicist Chien-Shiung Wu.  Adrian Cho wrote about this stamp last month in Science.  The stamp was issued on the International Day of Women and Girls in Science.  As far as I know, Dr. Wu is the third woman physicist honored on a US postage stamp, and the first Chinese-American one.  The end of Cho's article talks about other physicists honored on US postage stamps -- there are many fewer of these than American Nobel Physics Laureates!  Cho mentions several, including some aerospace scientists, though he says the US Postal Service doesn't actually track how many there are.

This got me thinking about that very question.  How many physicists and astronomers are honored on US postage stamps?  There are just over a dozen, by my count. The overlap with mathematicians is considerable (since there are far fewer of those), so I'll include them too in this post.

The most honored American physicist on our nation's postage stamps is, of course, Benjamin Franklin.  It may be no coincidence that Franklin was also the nation's first Postmaster General.  I haven't even tried to count how many US stamps have honored Franklin.

At a distant second place is Albert Einstein, honored on two stamps:  an 8 cent stamp in the Prominent Americans series in 1966, and a 15 cent stamp in 1979 issued on the centenary of his birth. All other physicists on US stamps have a single stamp in their honor.  These include Robert Millikan, on a 37 cent stamp issued in 1982 as part of the Great Americans series, and Enrico Fermi on a 34 cent stamp issued in 2001.

The American Scientists series, which began in 2005, was a boon for physicists.  Josiah Willard Gibbs (a polymath claimed equally by mathematicians and chemists) and Richard P. Feynman were included in the first round (37 cents), while John Bardeen and astronomer Edwin Hubble appear in the 2008 second round (41 cents).  Maria Goeppert Mayer was honored in the third round (2011) with a Forever stamp.  Unfortunately, the series appears to have been discontinued after the third set.  (Did they just not sell well?)  The first round in 2005 also included mathematician John von Neumann, who made colossal contributions to physics, computer science, and meteorology, among other fields.  

The new stamp honoring C.-S. Wu is the second to honor a physicist in the decade since the last set of American Scientists was issued.  It follows the 2018 issue of a Forever stamp honoring Sally K. Ride, astronaut, physicist, and stamp collector.  Of the honored physicists mentioned, Einstein, Millikan, Feynman, and Mayer were Nobel laureates, with Bardeen being a double Nobel laureate in physics.  (Cho's article disucsses that many feel that Wu should have received a Nobel as well.)

Also worthy of note is Benjamin Bannker, honored in 1980 with a 15 cent stamp in the Black Heritage series.  Banneker was a surveyor, mathematician, and astronomer, who like Franklin, published his own almanacs.  A math teacher, Jaime Escalante, was honored with a Forever stamp in 2016.  Along with Wu, Banneker and Escalante are the only non-whites honored by stamps among those discussed in this post.

Since Cho's article mentions aerospace scientists and inventors, let's look at them too.  The Wright Brothers and their achievements were honored on the 25th, 46th, 75th, and 100th anniversaries of their first powered flight.  The first stamp (2 cents) was issued as one of a pair honoring of the 1928 International Civil Aeronautics Conference, in that year.  It featured the Wright Flyer.  Three airmail stamps honored both the Wrights and their aircraft, in 1949 (6 cents) and two in 1978 (31 cents).  Finally on the centenary of their first flight, another stamp featuring the Wright Flyer was issued in 2003 (37 cents).  Other aerospace pioneers include Robert Goddard (1964, 8 cents), Igor Sikorsky (1988, 36 cents airmail), and Theodore von Karman (1992, 29 cents).  As far as I know, von Karman is the only fluid dynamicist honored on a US postage stamp! 

Unfortunately, I cannot think of another American fluid dynamicist who should next be honored with a US postage stamp, for reasons similar to those I wrote about in an earlier post. But what about other American physicists, astronomers, and mathematicians?  Who would you nominate next?

I surmise that those known among the general public would have the best chance of being honored on a stamp, compared to those best known just within the physics community.  So, Nikola Tesla seems an obvious choice.  You can guarantee that such a stamp would sell well.  Perhaps W. Edwards Deming, though he is best known for his work in statistics and management consulting, or J. Robert Oppenheimer.  However, I am delighted that less publicly celebrated figures like Gibbs, Millikan, Fermi, Mayer, Bardeen, and Wu have been honored.  There are thus many worthy choices for the next postage stamp.  If only the American Scientists series could be revived!

Among astronomers, I'm surprised that Carl Sagan hasn't already been so honored.  Among mathematicians, perhaps Benoit Mandelbrot (the maestro of fractals), John Nash (Nobel in economics, subject of A Beautiful Mind), and Katherine Johnson (of Hidden Figures fame) might be considered.

I am grateful to the website of the Mystic Stamp Company, and Wikipedia, sources of the information I have provided above.

Excitement, self-deception, and retraction

Yesterday, Davide Castelvecchi reported in Nature on the retraction of a 2018 paper claiming detection of a signature suggesting the existence of a Majorana fermion state.  It's a good article, and includes a link to the report of an independent investigation which concluded that, while no fraud had occurred, the authors of the paper essentially fooled themselves, implying that they fooled everyone else too, by publishing.  There are a couple excerpts from the report (by "experts" P. Brouwer, K. Ensslin, D. Goldhaber-Gordon, and P. Lee) that are worthy of reading here.

First, from Section 3:

And finally, from the Conclusion:

It should be noted that the expert investigators were provided with unpublished data, available to the authors before they published, that would have cast doubt on the original authors' conclusions, and that even the published data had features that a sharp reader could have used to cast such doubt.

The Mars Rover's driver

This weekend's Wall Street Journal features an interview with Vandi Verma, a naturalized US citizen from Indian, whom the author Tunku Varadarajan describes as "arguably the world's most experienced Martian robot operator".  Verma worked with Curiosity previously, and serves as chief engineer of robotic operations for Perseverance and Ingenuity.  DTLR commends the piece to all its readers.

More on machine learning in weather and climate modeling

 My last post commented on the use of machine learning in Earth science modeling.  The Philosophical Transactions of the Royal Society, Series A, has just published a special issue on this topic, based on a workshop held at Oxford University in September 2019.  Here is a link to the introductory paper, and a link to one of the contributions, that deals with physically-aware machine learning models.  Ten case studies are discussed in the latter.

Physically aware machine learning models

My last post resurrected some Eos articles from a few years ago, but today I'd like to discuss a piece in this month's issue.  Maskey et al. write about "A Data Systems Perspective on Advancing AI", reporting on a NASA-sponsored workshop held in January.  They describe "traditional" Earth science modeling as "top down", starting with first principles (laws of physics), while the machine learning approach is "bottom up", having algorithms that learn relationships empirically from historical data.  An inherent limitation of empirical modeling, they recognize correctly, is the inability for a model trained on historical data to extrapolate into regimes never before seen in the training data.  Yet this is precisely what Earth science is called to do, when dealing with extreme weather events or climate change, for example.  The writers propose that "physically aware machine learning models" could overcome this limitation, suggesting a melding of the "top down" and "bottom up" approaches.  The authors mainly write about using physics to constrain the machine learning models or their cost functions during training, claiming promising results already.  It is less clear to me that placing constraints on an empirical model would allow it to creditably extrapolate, only that such constraints should improve interpolation capability.

On this blog, it was noted previously that there have been demonstrated cases of deep neural networks actually being capable of generalizing beyond the training data, though such cases are not well understood, and are not convincing unless validated in independent data.  From the context, it did not seem like these deep learning models were of the "physically aware" variety that Maskey et al. describe.

These are early days in the efforts to apply machine learning to physical problems.  We still have much to learn about what is possible, and what remains limited, with such efforts.