5 ms·

> The hypothesis have been extremely successful and predictive. Epicycles were successful and predictive. But they were later found to be the wrong way to unde

> The hypothesis have been extremely successful and predictive. Epicycles were successful and predictive. But they were later found to be the wrong way to understand the celestial motions. The problem is these calculations with dark matter are just fitting observations with zillion parameters. It is busywork, yes sometimes it is useful, but it is not discovering new laws, just chugging along without having to challenge our concept of how star/galaxy motions really work.

sega_sai 5y ago
Epicycles were a good theory that approximated a ellipses by combinations of circles (essentially through a Fourier transform). When a simpler theory came that was easier to compute, was more predictive and came from a first principles calculations it was adopted instead. That's not the case with the modified gravity (at least right now). And please don't BS about zillion of parameters. That's not the case. The dark matter only simulations don't have that many parameters (other than the basic cosmological ones) . When you start putting baryons in, yes that's where the parameters start appearing (probably of the order of a 100), but it's hardly surprising give the need to describe the complexity of star formation, gas dynamics, black hole accretion etc. The modified gravity theories don't need as many parameters because nobody has ever run any modified gravity simulation that is even close in complexity to what Cold Dark Matter simulations do.
- posix_me_less 5y ago
  I agree MOND results aren't that great now. But Kopernik's results weren't that great either, compared to standard theory. They gave less precise results initially. It took time for the astronomers to accept them. What kind of initial condition about dark matter distribution are you assumming in your simulations?
  - sega_sai 5y ago
    Copernicus model was not that much of a change from a model complexity point of view, as it just removed some common epicycles by putting Sun in the center. The key changes were Kepler's results and Newton's, where one had provided drastically simpler model, and another had a theoretical/(first principles) derivation. Regarding initial conditions, I'll point you towards wikipedia. But basically the assumptions is that it's collisionless fluid with very small Gaussian density perturbations and scale invariant power spectrum. In total the specification of that model requires ~ 10 parameters.
    - posix_me_less 5y ago
      So I infer you are talking about large scale simulations of many galaxies/a region much greater than single galaxy, right? If so, do these kinds of simulations with 10 parameter model predict galactical rotation curves in agreement with observation? Or do we need more detailed model of dark matter distribution for the rotation curves?
      - sega_sai 5y ago
        There are different simulations focusing on various aspects of the universe. All the CDM simulations give (predict) a dark matter density profile close to the so called Navarro-Frenk-White (NFW) profile (i.e. here https://ned.ipac.caltech.edu/level5/March01/Battaner/node27.html https://ned.ipac.caltech.edu/level5/March01/Battaner/node27.... ) . This profile has essentially two parameters, mass and concentration. These are the two free parameters you work with for individual galaxies if you try to match the data.
        posix_me_less 5y ago
        Okay I can appreciate the fact that there are DM models that work with small number of parameters. But these are not working great for some galaxies, which pushes in direction of more parameters, or something different than DM. Quote from from the page you've linked [1]: > However, either the observations do not constitute a proof of the CDM models, or dynamic ingredients other than halo and disk density profiles are necessary to study the rotation of spirals. [1] https://ned.ipac.caltech.edu/level5/March01/Battaner/node27.html https://ned.ipac.caltech.edu/level5/March01/Battaner/node27....
        raattgift 5y ago
        Sure, dynamical ingredients like the jets carrying huge amount of luminous matter out of active galactic nuclei or the outflow from star forming activities and supernovae, for instance, have been raised as possible explanations for the apparently shallow increase in density of dark matter in galaxy cores (the "cuspiness" problem). Roughly, dynamical process in cores can -- simply by gravitational interaction -- throw dark matter back out towards the halo along with the luminous matter. One can also step up a scale level and consider that the distribution of peculiar velocities in galaxy clusters is consistent with plenty of dark matter inside galaxy clusters but outside galaxy halos. This DM is bound to be stirred up by the moving galaxies, even if only gravitationally, and such dynamical processes will also affect the density of gas and dust in the cores of galaxy clusters (the cores of which need not be occupied by galaxies) too. We need to explain peculiar motion of galaxy-cluster-members, and member-galaxy-vs-whole-cluster lensing behaviour too, and astronomical surveys of galaxies and clusters continue to provide plenty of data to test various theoretical approaches. Those are what propose observable consequences of self-interaction (is there dark chemistry? do dark matter particles collide and annihilate?) or is dark matter warmer than cold dark matter (if it's a naturally warm gas, then it would have a hard time gathering in the cores of galaxies and clusters, much like a warm gas has trouble gathering on the floor of a laboratory compared to a very cold gas). Or does it couple non-gravitationally but very weakly to baryons or electrons, such that it undergoes a phase change as one goes from the very dense cores of galaxies and gas-rich clusters towards the sparser edges? If one proposes atmosphere-like dynamical processes at these scales, what are the observables one expects from astronomic spectroscopy? Spirals are certainly interesting, but there are also elliptical galaxies where the outermost hydrogen gas is in arbitrary orbit in clouds of various sizes; unlike edge-on spirals, they don't show a advancing-side/receding-side red/blueshift dipole, and where the rotating elements are large enough to measure significant dipoles, they do not strongly correlate with oblateness. More intriguingly, the measurable red/blue shift velocities of (groups of) stars within such galaxies are mostly radial, which is very different from what's seen in spirals, where rotation dominates. That radial speed is different from what would be seen if the luminous elements (including that that is luminous in infrared, radio, and so on, or material that produces absorption lines rather than emission ones) were all that determined these orbits. But a cold dark matter halo explains -- or at least is in concord with -- those orbital velocity anomalies. And as galaxy surveys continue, neat things like giant low-surface-brightness galaxies (and their substructures) will be found more often. Their properties are good stress-tests of dark matter theories and alternatives, in that they should generate predictions for observations which will arrive within the next few years. They are also probes of largest-scale structure formation, as so far they are not found in or near dense galaxy clusters. If we start discovering GLSBs in dense regions, that would be a challenge to largest-scale distribution of dark matter as the driver of the cosmic web, which is the best current explanation of the distribution of galaxy clusters that are not gravitationally-bound with one another.
- dnautics 5y ago
  The simulations themselves don't have zillions of parameters, but the field itself does. When you say something like "the bullet cluster blah blah" you are talking about a custom parameterization of the lambda-cdm field. You're not talking about simulation, because simulations predict that bullet cluster speed collisions should not happen in our universe's lifetime with the number of galaxies observed.
  - sega_sai 5y ago
    I'll point you here on the topic of the cluster speed https://ui.adsabs.harvard.edu/abs/2007MNRAS.380..911S/abstract https://ui.adsabs.harvard.edu/abs/2007MNRAS.380..911S/abstra... I.e. it's not a problem.
    - dnautics 5y ago
      ok that's fine but it still doesn't change the fact that there is a distinction between parameterizing field vs parameterizing the simulation. You shouldn't have conflated the two in your defense of LCDM.
      - raattgift 5y ago
        What exactly are you talking about with "parameterizing field" and a couple comments up and "the lambda-cdm field" (emphasis yours) ? Are you referring to the Friedmann equations? The equation of state? The half-dozen free parameters of the model itself? What's your exact objection here? Because there isn't anything that I would quickly call a "lambda-cdm field" in the standard model of cosmology any more than there is a single "standard model field" in the standard model of particle physics. "The field itself does" -- again, what exactly do you mean here? Are you even talking about a mathematical or even physical object?
        dnautics 5y ago
        s/field/particle density map/g
dTal 5y ago
Epicycles weren't really predictive, in the sense that every time measurements were refined, the new data didn't fit and more epicycles had to be added - and this was only possible because combinations of epicycles can represent anything at all. So by analogy, you're saying that modern dark matter theories are so general that they a) could account for any conceivable observation, and b) fail to predict novel phenomena? I don't have the physics knowledge to evaluate that assertion, but it doesn't seem to be shared by the broader physics community.
- dnautics 5y ago
  Yes, and yes. Not "any conceivable" but "any conceivable gravitational observation". The trouble with dark matter is that it's unimodal: It exists to fix a problem where our theory of gravity disagrees with observation, but at the same time the only way to measure dark matter is by observing gravitational effects. The only true predictions made have been "if dark matter is actually x particle then ..." Or (weaker) "if dark matter has non-gravitational property X, then..." And thus far all prediction of those sort have been refuted.
- Keysh 5y ago
  That's not how epicycles were used. Ptolemy set up the standard geocentric model with epicycles (and deferents and equants), and these were used more or less unchanged for over a thousand years. When, in the late medieval period, some Arab astronomers began tinkering with modifications, these were motivated by a desire for better agreement with the Aristotelian ideal of uniform circular motion (i.e., to somehow replace Ptolemy's non-uniform equant device with some combination of uniform circular motion), not any desire to match the measurements better.
  - selimthegrim 5y ago
    Tusi wasn't Arab, unless you mean Ibn Shatir.
- sega_sai 5y ago
  Epicycles are predictive, because they just represent a Fourier approximation to an ellipse. They predict the future, just with finite precision, so to increase precision you need to add more parameters. Regarding dark matter, I'm not comparing it to epicycles. The cold dark matter theory is very predictive and it has made many predictions that were confirmed. But in the same time, I am perfectly willing to accept other theory which will explain all current observations without dark matter. Just so far nobody came up with one.
edflsafoiewq 5y ago
Feel free to propose a Keplerian model. Until then people will continue on with their epicycles.
dnautics 5y ago
In retrospect, calling epicycles "successful and predictive" is a category error. You could use epicycles to make orbits shaped like a drawn outline of Einsteins face, because epicycles are equivalent to Fourier series and can can be used to represent anything. So if you can say epicycles are successful and predictive, you could say "enumerating coordinates is successful and predictive". To be fair, Fourier theory was not known at the time.
- posix_me_less 5y ago
  I don't see the point. Fourier methods, whether their modern form or their ancient form in the Ptolemy's model are useful, predictive and successful. They were used to predict planet positions, and (for example) sea levels near shores. However Copernicus and others realized that there is a better point of view and more deeper understanding of what is going on. My point is, dark matter hypothesis is similar situation, we can bend it to explain very much, but there is no great revolutionary insight coming from it, we are adding tweaks, adding more independent quantities, like we did with epicycles. Perhaps there will be new Copernicus who will show a better point of view, a deeper understanding of the observations that lead people to introduce DM.

sega_sai 5y ago

Epicycles were a good theory that approximated a ellipses by combinations of circles (essentially through a Fourier transform). When a simpler theory came that was easier to compute, was more predictive and came from a first principles calculations it was adopted instead. That's not the case with the modified gravity (at least right now). And please don't BS about zillion of parameters. That's not the case. The dark matter only simulations don't have that many parameters (other than the basic cosmological ones) . When you start putting baryons in, yes that's where the parameters start appearing (probably of the order of a 100), but it's hardly surprising give the need to describe the complexity of star formation, gas dynamics, black hole accretion etc. The modified gravity theories don't need as many parameters because nobody has ever run any modified gravity simulation that is even close in complexity to what Cold Dark Matter simulations do.

posix_me_less 5y ago
I agree MOND results aren't that great now. But Kopernik's results weren't that great either, compared to standard theory. They gave less precise results initially. It took time for the astronomers to accept them. What kind of initial condition about dark matter distribution are you assumming in your simulations?
- sega_sai 5y ago
  Copernicus model was not that much of a change from a model complexity point of view, as it just removed some common epicycles by putting Sun in the center. The key changes were Kepler's results and Newton's, where one had provided drastically simpler model, and another had a theoretical/(first principles) derivation. Regarding initial conditions, I'll point you towards wikipedia. But basically the assumptions is that it's collisionless fluid with very small Gaussian density perturbations and scale invariant power spectrum. In total the specification of that model requires ~ 10 parameters.
  - posix_me_less 5y ago
    So I infer you are talking about large scale simulations of many galaxies/a region much greater than single galaxy, right? If so, do these kinds of simulations with 10 parameter model predict galactical rotation curves in agreement with observation? Or do we need more detailed model of dark matter distribution for the rotation curves?
    - sega_sai 5y ago
      There are different simulations focusing on various aspects of the universe. All the CDM simulations give (predict) a dark matter density profile close to the so called Navarro-Frenk-White (NFW) profile (i.e. here https://ned.ipac.caltech.edu/level5/March01/Battaner/node27.html https://ned.ipac.caltech.edu/level5/March01/Battaner/node27.... ) . This profile has essentially two parameters, mass and concentration. These are the two free parameters you work with for individual galaxies if you try to match the data.
      - posix_me_less 5y ago
        Okay I can appreciate the fact that there are DM models that work with small number of parameters. But these are not working great for some galaxies, which pushes in direction of more parameters, or something different than DM. Quote from from the page you've linked [1]: > However, either the observations do not constitute a proof of the CDM models, or dynamic ingredients other than halo and disk density profiles are necessary to study the rotation of spirals. [1] https://ned.ipac.caltech.edu/level5/March01/Battaner/node27.html https://ned.ipac.caltech.edu/level5/March01/Battaner/node27....
        raattgift 5y ago
        Sure, dynamical ingredients like the jets carrying huge amount of luminous matter out of active galactic nuclei or the outflow from star forming activities and supernovae, for instance, have been raised as possible explanations for the apparently shallow increase in density of dark matter in galaxy cores (the "cuspiness" problem). Roughly, dynamical process in cores can -- simply by gravitational interaction -- throw dark matter back out towards the halo along with the luminous matter. One can also step up a scale level and consider that the distribution of peculiar velocities in galaxy clusters is consistent with plenty of dark matter inside galaxy clusters but outside galaxy halos. This DM is bound to be stirred up by the moving galaxies, even if only gravitationally, and such dynamical processes will also affect the density of gas and dust in the cores of galaxy clusters (the cores of which need not be occupied by galaxies) too. We need to explain peculiar motion of galaxy-cluster-members, and member-galaxy-vs-whole-cluster lensing behaviour too, and astronomical surveys of galaxies and clusters continue to provide plenty of data to test various theoretical approaches. Those are what propose observable consequences of self-interaction (is there dark chemistry? do dark matter particles collide and annihilate?) or is dark matter warmer than cold dark matter (if it's a naturally warm gas, then it would have a hard time gathering in the cores of galaxies and clusters, much like a warm gas has trouble gathering on the floor of a laboratory compared to a very cold gas). Or does it couple non-gravitationally but very weakly to baryons or electrons, such that it undergoes a phase change as one goes from the very dense cores of galaxies and gas-rich clusters towards the sparser edges? If one proposes atmosphere-like dynamical processes at these scales, what are the observables one expects from astronomic spectroscopy? Spirals are certainly interesting, but there are also elliptical galaxies where the outermost hydrogen gas is in arbitrary orbit in clouds of various sizes; unlike edge-on spirals, they don't show a advancing-side/receding-side red/blueshift dipole, and where the rotating elements are large enough to measure significant dipoles, they do not strongly correlate with oblateness. More intriguingly, the measurable red/blue shift velocities of (groups of) stars within such galaxies are mostly radial, which is very different from what's seen in spirals, where rotation dominates. That radial speed is different from what would be seen if the luminous elements (including that that is luminous in infrared, radio, and so on, or material that produces absorption lines rather than emission ones) were all that determined these orbits. But a cold dark matter halo explains -- or at least is in concord with -- those orbital velocity anomalies. And as galaxy surveys continue, neat things like giant low-surface-brightness galaxies (and their substructures) will be found more often. Their properties are good stress-tests of dark matter theories and alternatives, in that they should generate predictions for observations which will arrive within the next few years. They are also probes of largest-scale structure formation, as so far they are not found in or near dense galaxy clusters. If we start discovering GLSBs in dense regions, that would be a challenge to largest-scale distribution of dark matter as the driver of the cosmic web, which is the best current explanation of the distribution of galaxy clusters that are not gravitationally-bound with one another.
dnautics 5y ago
The simulations themselves don't have zillions of parameters, but the field itself does. When you say something like "the bullet cluster blah blah" you are talking about a custom parameterization of the lambda-cdm field. You're not talking about simulation, because simulations predict that bullet cluster speed collisions should not happen in our universe's lifetime with the number of galaxies observed.
- sega_sai 5y ago
  I'll point you here on the topic of the cluster speed https://ui.adsabs.harvard.edu/abs/2007MNRAS.380..911S/abstract https://ui.adsabs.harvard.edu/abs/2007MNRAS.380..911S/abstra... I.e. it's not a problem.
  - dnautics 5y ago
    ok that's fine but it still doesn't change the fact that there is a distinction between parameterizing field vs parameterizing the simulation. You shouldn't have conflated the two in your defense of LCDM.
    - raattgift 5y ago
      What exactly are you talking about with "parameterizing field" and a couple comments up and "the lambda-cdm field" (emphasis yours) ? Are you referring to the Friedmann equations? The equation of state? The half-dozen free parameters of the model itself? What's your exact objection here? Because there isn't anything that I would quickly call a "lambda-cdm field" in the standard model of cosmology any more than there is a single "standard model field" in the standard model of particle physics. "The field itself does" -- again, what exactly do you mean here? Are you even talking about a mathematical or even physical object?
      - dnautics 5y ago
        s/field/particle density map/g

dTal 5y ago

Epicycles weren't really predictive, in the sense that every time measurements were refined, the new data didn't fit and more epicycles had to be added - and this was only possible because combinations of epicycles can represent anything at all. So by analogy, you're saying that modern dark matter theories are so general that they a) could account for any conceivable observation, and b) fail to predict novel phenomena? I don't have the physics knowledge to evaluate that assertion, but it doesn't seem to be shared by the broader physics community.

dnautics 5y ago
Yes, and yes. Not "any conceivable" but "any conceivable gravitational observation". The trouble with dark matter is that it's unimodal: It exists to fix a problem where our theory of gravity disagrees with observation, but at the same time the only way to measure dark matter is by observing gravitational effects. The only true predictions made have been "if dark matter is actually x particle then ..." Or (weaker) "if dark matter has non-gravitational property X, then..." And thus far all prediction of those sort have been refuted.
Keysh 5y ago
That's not how epicycles were used. Ptolemy set up the standard geocentric model with epicycles (and deferents and equants), and these were used more or less unchanged for over a thousand years. When, in the late medieval period, some Arab astronomers began tinkering with modifications, these were motivated by a desire for better agreement with the Aristotelian ideal of uniform circular motion (i.e., to somehow replace Ptolemy's non-uniform equant device with some combination of uniform circular motion), not any desire to match the measurements better.
- selimthegrim 5y ago
  Tusi wasn't Arab, unless you mean Ibn Shatir.
sega_sai 5y ago
Epicycles are predictive, because they just represent a Fourier approximation to an ellipse. They predict the future, just with finite precision, so to increase precision you need to add more parameters. Regarding dark matter, I'm not comparing it to epicycles. The cold dark matter theory is very predictive and it has made many predictions that were confirmed. But in the same time, I am perfectly willing to accept other theory which will explain all current observations without dark matter. Just so far nobody came up with one.

edflsafoiewq 5y ago

Feel free to propose a Keplerian model. Until then people will continue on with their epicycles.

dnautics 5y ago

In retrospect, calling epicycles "successful and predictive" is a category error. You could use epicycles to make orbits shaped like a drawn outline of Einsteins face, because epicycles are equivalent to Fourier series and can can be used to represent anything. So if you can say epicycles are successful and predictive, you could say "enumerating coordinates is successful and predictive". To be fair, Fourier theory was not known at the time.

posix_me_less 5y ago
I don't see the point. Fourier methods, whether their modern form or their ancient form in the Ptolemy's model are useful, predictive and successful. They were used to predict planet positions, and (for example) sea levels near shores. However Copernicus and others realized that there is a better point of view and more deeper understanding of what is going on. My point is, dark matter hypothesis is similar situation, we can bend it to explain very much, but there is no great revolutionary insight coming from it, we are adding tweaks, adding more independent quantities, like we did with epicycles. Perhaps there will be new Copernicus who will show a better point of view, a deeper understanding of the observations that lead people to introduce DM.