Tuesday, September 01, 2009

See, then discover

Recently I read Hermann Bondi's "Relativity and Common Sense," an improbably slim Dover books with a strange sort of hourglass on its cover. I enjoyed Bondi's methodical explanation, aimed at the common man--why does it seem like people had more faith in that person back then?--especially with the extremely retro drawings of Adam, Bill, Charles, and occasionally David, flying in their respective UFOs or sitting at home in their giant-receiver-equipped houses, flashing signals at each other and measuring different times on their watches. Bondi points out that, in Einstein's day, the only way people had of thinking about the speed of light were extremely fast trains passing each other. What a luxury, he says, for those of us in the modern era (the book was written in the early 60s) to be able to think about jet planes and and even space craft.

The phrase "Ah, I see," encapsulates how important visualization is to the way we understand. But physics often deals in things that we will never "see"--even with microscopes, telescopes, or particle accelerators--and often can only even imagine by analogy(like the fifth dimension, for instance, never mind the other six that might be out there).

The strange hourglass on the cover of Bondi's book is what's called a light cone. Shaped by the path of light through space from a single point, it's the chunk of space-time (mapped from four dimensions onto a three-dimensional space, since that's all we can visualize in what's called a Minkowski diagram) representing the chunk of space-time that can ever reach a given point in it. Even as it illustrates the fundamental truth that nothing can travel faster than the speed of light—and thus anything outside this light cone can ever effect the point at its infinitely slender waist—it embodies the hobbles of our three-dimensional sight.

Living in a spatially three-dimensional world may limit what we can understand through seeing, but supercomputers, which can churn through voluminous datasets in at incredibly high speeds, have lately enhanced our imaginations, expanded what we can consider in a concrete form, putting even the most distant, miniscule, or simply mind-bending physical phenomena right before our eyes.

The WIRED Science blog grabbed the 10 best science visualization videos picked by DOE's SciDAC, which stands for Scientific Discovery through Advanced Computing. This one's a supernova explosion, rendered in beautiful, vivid technicolor (actually, the colors represent temperatures). WIRED writes:

Type Ia supernovae are thought to be white dwarf stars in binary systems that explode due to a thermonuclear runaway. This movie shows a simulation of Type Ia supernovae exploding from multiple ignition points. When the hot ash breaks through the surface of the star, it spreads rapidly across the stellar surface, converges at the opposite point and produces a jet-like flow that triggers a detonation. The simulation shows that multiple ignition points generate more nuclear burning and produce more expansion of the star than a single ignition point. As a result, less radioactive nickel is produced during the detonation phase, and the explosion is less luminous.

The work comes out of the University of Chicago's ASC/FLASH center, who first simulated a white dwarf explosion in 2007. The process in nature takes only about three seconds; "just one of the compiling jobs," states the U Chicago press release, "ran for 75 hours on 768 computer processors, for a total of 58,000 hours." Understanding how supernovas explode—and what could be a better understanding than visualization?—could help physicists puzzle out why the universe is expanding:
...the scientific payoff for logging these long, stressful hours is potentially huge. Astrophysicists value type Ia supernovas because they all seem to explode with approximately the same intensity. Calibrating these explosions according to their distance reveals how fast the universe has been expanding at various times during its long history.

Cosmic Variance's Mark Trodden writes:
The current understanding of this has been enough to discover the fact that the universe is accelerating, but our future plans are to exploit it further, to help provide insight into the origin of cosmic acceleration. A detailed understanding of how supernova explosions occur would be a valuable contribution to this quest.
Now, one of my colleague, who will remain anonymous to protect him from hate mail, claims that simulations done entirely on a computer "isn't real physics." (In fact, when I was writing about blast waves and brain injury a few days ago, doctors were very hesitant about saying too much about conclusions drawn from a simulation.) But I think the DOE's terms "Scientific Discovery through Advanced Computing" actually say a lot. If a computer can recreate how a process unfolds based on the laws of physics, it can also reveal features and details that we would never be able to intuit otherwise. At the risk of inciting full-on warfare, I'll ask you to weigh in. Should we depend as much on computation as we do? Are people, like my colleague, who demand either working entirely theoretically or entirely experimentally, merely unimaginative and stuck in the past?


  1. Scientific progress is based on creativity. A new theory is developed when a scientist goes beyond the current model and proposes new ideas that further explain the universe. Much like a song that you can't get out of your head, an animation of an electron orbiting around the nucleus, for instance, still pervades the imaginations of many students. A visualization destroys the imagination of students and stymies the creativity necessary for new ideas. Try watching the movie version of your favorite novel. You will never get back the characters you imagined while reading it.

  2. Great comment. When I was studying physics I would always look for pictures, ways of "seeing" the phenomenon I was trying to understand. But perhaps I should have accept that there are some things in nature that we just don't know how to visualize in a simple way - for instance, the planet-around-the-sun picture of the electron you mention doesn't tell us anything about how an electron really behaves. But it seems like making computational models based on current theory might help us at least see that theory's limitations.

  3. Relativity is necessary for analysis of nanoscale examples, and the discussion of visualizations leads to a good paradigm in the RQT (relative quantum topological) atomic video function. This method of graphic data point mapping for the force, energy, and electrons of an atom labeled psi (Z) achieves 3D picoyoctometric interactive video model imagery by constructing the function as a combination of the relativistic Lorenz-Einstein transform functions for time, mass, and energy with the workon quantized electromagnetic wave equations for frequency and wavelength.
    The nucleus is designed as a sphere radiating forcons with valid joule values by {e=m(c^2)} transformation of nucleoplastic surface mass. The equation is written as a series mass differential within spacetime limits of {gravity-time} to build the GT integral atomic topological function. Quantum symmetry numbers are assigned along the series of orders of mass transform to give 3D topology to the solutions. Psi pulsates by cycles of nuclear emission and absorption of force at the frequency {Nhu=e/h}.
    When the atom's internal momentum function is rearranged to the photon gain rule and integrated for GT boundaries a series of 26 topological wavefunctions is found. Each is the particle function of a type of energy intermedon
    of the 5/2 kT J internal heat capacity energy cloud, accounting for all of them.. Those 26 energy values intersect the sizes of the fundamental physical constants: h, h-bar, delta, nuclear magneton, beta magneton, k (series). Each is displayed as a picoyoctometric 3D image. The result is the exact 3D interactive video model image of the atom.
    Images of the h-bar magnetic energy waveparticle of ~175 picoyoctometers are available online at http://www.symmecon.com with the complete guide to RQT atomic imaging titled The Crystalon Door. TCD conforms to the unopposed motion of disclosure in U.S. District (NM) Court of 04/02/01 titled The Solution to the Equation of Schrodinger, U.S. copyright TXu1-266-788.