Slashdot Mirror
Origin of Cosmic Rays Confirmed
cats-paw writes in with news of research that seems to confirm and support current theories of how cosmic rays are created. The prevailing thinking has been that cosmic rays are generated in the regions where supernovas' shock waves interact with the interstellar medium. The new research used the variability in X-ray emissions from a supernova remnant to estimate the strength of the magnetic fields present in that environment. The results lend support to the possibility of protons and nucleii being accelerated in supernova remnants to energies of 1 PeV (10^15 eV) and beyond. Here is the abstract from Nature.
Multiply that with Avogadro's number, and you get the energy of a regular bullet with 'cosmic ray' speed:
6x10^20 J. That, amazingly, equals the total enery production on earth in one year. http://en.wikipedia.org/wiki/World_energy_resources_and_consumption
don't cut it off www.mgmbill.org
I believe it's an upper limit to how fast this particular mechanism can accelerate a proton, not how fast a proton can be accelerated under any circumstances.
ResidntGeek
Additionally, the abstract says their research "provide[s] a strong argument" for a theory. I suppose these statements are way too hard-line for Real Science. Sheesh. These are people who know very well they're doing inference rather than deduction - including the submitter! - and you take them to task for jumping to conclusions.
You say:
The hypothesize/predict/experiment cycle isn't nearly as boolean as you make it out, even though we teach it that way in school.
If a result doesn't disprove a theory, it actually increases its probability among other possibilities. Bayesian statistics models this quite well, and scientists think about it that way but without such a rigorous foundation. For example, in all forces, we've measured the differential relationships among position, velocity and acceleration to ridiculous precision. Doesn't this increase the probability that we've got it right? In this area, if there's a conflict between predicted and expected outcomes, we regard the explanation that the theory is wrong as the less probable one - much less probable.
Part of the problem is classical statistics. Null hypotheses and tests against them are kludgy nonsense, everyone knows it, and everyone has their own way of doing it "properly". (Think about it this way: Pr(null hypothesis), where the null hypothesis has a continuous component - and this is done all the time - is ZERO.) Doing inference without priors is a misguided attempt at objectivity. These mindsets are well-preserved in scientific philosophy, and they've got to go. Nobody actually thinks about real inference the classical way. It'd be ridiculous to try it on any hypothesis of moderate complexity.
I got my Linux laptop at System76.
First of all, the summary (but also the article) refer to "cosmic rays", as if they are all the same. Most, actually, come from the sun. The nature abstract talks about "galactic cosmic rays", which better, but there are thought to be many flavors of these as well, as there are many ways to accelerate charged partcles.
The poster child of uber-freaked out cosmic rays is a crazy bugger detected in 1991 that had an energy of 3.2 x 10^20 eV. One scientist compared it to dropping a brick on your toe! Cosmic rays with this much energy are too enegetic to fit the supernova shock wave model nicely. They might come from gamma ray bursts or black holes on a feeding frenzy.
Those people who think they know everything are a great annoyance to those of us who do. (Isaac Asimov)
Can you propose a mechanism for star formation that would give the sun a sufficiently large positive charge to accelerate ions at large distances? That voltage difference had to come from somewhere, and large charge imbalances counteract the gravitational attraction needed for nuclear fusion.
Maybe I'm wasting my time. It seems clear that "alternative cosmologies" means the electric universe theory, which doesn't make any useful, testable predictions.
"Your notation sucks!" -- Serge Lang (1927-2005)