Page 1 of 2 12 LastLast
Results 1 to 30 of 33

Thread: Amateurs measuring parallax

  1. #1
    Join Date
    May 2002
    Location
    My own private Nogero
    Posts
    183

    Amateurs measuring parallax

    How easy would it be for amateur astronomers to measure parallax using off-the-shelf technology? What kind of precision could they get?

  2. #2
    Join Date
    Jul 2005
    Location
    Massachusetts, USA
    Posts
    21,808
    If it were just based on diffraction limited viewing, and you wanted to do some finding-the-centroid work, you could probably do it with a very good 3-inch telescope. A 10-inch telescope could (in theory) get you 0.5 second resolution. You can buy 20+ inch telescopes on a largish amateur budget, which would get you (in theory) 0.25 second or better. Issues with your mount, clock-drive, weather, optics, etc will degrade that performance. There are several stars in the sky that 0.25 second resolution would be enough to show parallax.
    Forming opinions as we speak

  3. #3
    Join Date
    Sep 2006
    Posts
    1,796
    Recall that the first measurements of stellar parallax were made with small telescopes and the naked eye. Bessel's telescope had a diameter of 6.2 inches, though its special design helped him greatly to measure the parallax of 61 Cygni. Struve used a nine-inch refractor to measure the parallax of Vega. Henderson used a 4-inch mural circle to measure the parallax of alpha Centauri.

    The issue isn't so much telescope size as method, patience, and skill.

  4. #4
    Join Date
    Dec 2004
    Posts
    14,765
    The question of this thread fits right in with something I've
    supposed for decades without ever trying to confirm before:

    Until the advent of CCDs and images from HST and other
    high-resolution telescopes became available online, I'd guess
    that far less than a thousand people -- maybe even less than
    a hundred -- had ever made an actual measurement of stellar
    parallax. Does that seem right to you?

    If my guess was right, I think it means that using the parsec
    as the standard unit of astronomical distances was an especially
    bad choice. It is a unit that almost nobody has ever calculated
    from measurements they made, and nobody has ever made a
    direct measurement of distance in parsecs (it isn't possible).
    The measurement which can be used to calculate a distance in
    parsecs only works for observers on one particular planet, when
    observing the stars closest to that planet.

    Now that tools and images are available to make measurements
    easier, I suppose there might be more people measuring stellar
    distances, so the parsec might no longer be quite as inappropriate
    as it was a few years ago.

    -- Jeff, in Minneapolis
    http://www.FreeMars.org/jeff/

    "I find astronomy very interesting, but I wouldn't if I thought we
    were just going to sit here and look." -- "Van Rijn"

    "The other planets? Well, they just happen to be there, but the
    point of rockets is to explore them!" -- Kai Yeves

  5. #5
    Join Date
    Mar 2004
    Posts
    3,127
    Dennis Di Cicco did a nice series of CCD astrometric measurements for Barnard's Star using 11-16-inch telescopes, and got a very solid measurement of its parallax. One place showing his plot is here.

  6. #6
    Join Date
    Sep 2006
    Posts
    1,796
    Quote Originally Posted by Jeff Root View Post
    Until the advent of CCDs and images from HST and other
    high-resolution telescopes became available online, I'd guess
    that far less than a thousand people -- maybe even less than
    a hundred -- had ever made an actual measurement of stellar
    parallax. Does that seem right to you?
    Actually, I'd wager that the opposite is true: yes, before CCDs
    were invented, probably fewer than 1,000 people had ever
    measured a parallax themselves. Since the advent of CCDs,
    but, more importantly, computers and measuring engines,
    I'd guess that even _fewer_ people have measured a stellar
    parallax themselves. These days, thousands of parallax
    measurements are made by pipelines of software, fed images
    which have been taken by machines and reduced automatically.

    If my guess was right, I think it means that using the parsec
    as the standard unit of astronomical distances was an especially
    bad choice.
    Hmmm. I don't see how any unit of distance of stellar scale
    is better or worse than any other, really.
    But different squids for different kids.

  7. #7
    Join Date
    Dec 2004
    Posts
    14,765
    Setting up your machine to measure something counts for me
    as measuring it yourself, in this case.

    A parsec is a unit that is completely disconnected from anything
    that is actually observed. It can only be calculated from what
    can be observed. If virtually no-one makes those observations,
    it is doubly disconnected.

    While the light-year and related units are also based on Earth-
    centric base units (the year and the second), they are intuitively
    obvious in application: light travels a certain distance in a given
    time. And it applies to any distance in any location, not just to
    the stars nearest Earth.

    Parsecs are perfect for recording distances of the nearest stars
    meaured using parallax from Earth. For any other purpose, they
    have no advantage over other units which are far more readily
    understood.

    How easy is it to explain what a parsec is? How easy is it to
    explain what a light-second is?

    -- Jeff, in Minneapolis

    .
    Last edited by Jeff Root; 2014-Feb-21 at 03:11 PM.
    http://www.FreeMars.org/jeff/

    "I find astronomy very interesting, but I wouldn't if I thought we
    were just going to sit here and look." -- "Van Rijn"

    "The other planets? Well, they just happen to be there, but the
    point of rockets is to explore them!" -- Kai Yeves

  8. #8
    Join Date
    Mar 2004
    Posts
    3,127
    I take both points. Any unit is arbitrary if tied to the Earth. Parsecs, light-years, SI lengths. Parsecs seem to have won out in the research literature, although light-years may have appeared first and do have the advantage that we intuitively grasp the scale difference between a light-second and a light-year. (This is why I favor light-years in my nonmathematical intro classes).

    OTOH, a light-year is certainly no more directly connected to what we actually observe than is a parsec, once we get beyond the limits of ranging within the Solar System.

  9. #9
    Join Date
    Dec 2004
    Posts
    14,765
    Quote Originally Posted by ngc3314 View Post
    OTOH, a light-year is certainly no more directly connected
    to what we actually observe than is a parsec, once we get
    beyond the limits of ranging within the Solar System.
    It is more directly connected: We observe an object
    which is x light-time units away as it was x time units ago.
    There is no comparable relationship for parsecs except for
    the amount of annual wiggling of nearby stars.

    -- Jeff, in Minneapolis
    http://www.FreeMars.org/jeff/

    "I find astronomy very interesting, but I wouldn't if I thought we
    were just going to sit here and look." -- "Van Rijn"

    "The other planets? Well, they just happen to be there, but the
    point of rockets is to explore them!" -- Kai Yeves

  10. #10
    Join Date
    Mar 2004
    Posts
    3,127
    Quote Originally Posted by Jeff Root View Post
    A parsec is a unit that is completely disconnected from anything that is actually observed.
    More directly connected to something interesting, but beyond the limits of radar ranging and things like the time delay of orbital phenomena of planetary satellites, we don't directly observe the delay of seeing distant objects. (I don't really disagree with you, but your post brought up direct observation, which I see as a bit of a red herring compared to intuitive grasp of orders of magnitude).

  11. #11
    Join Date
    Sep 2006
    Posts
    1,796
    Quote Originally Posted by Jeff Root View Post
    A parsec is a unit that is completely disconnected from anything
    that is actually observed.
    So, the astronomers who measure the positions of stars and other objects
    over the course of several years, and who see some nearby objects moving
    relative to more distant objects, like this
    (taken from the VERA collaboration -- see http://lanl.arxiv.org/abs/0709.0820)

    vera_panel.gif

    or this (taken from the Hipparcos mission)

    vega_motion.gif

    aren't observing anything connected to parallax? Really?

    Gosh, what are they observing?

  12. #12
    Quote Originally Posted by StupendousMan View Post
    Recall that the first measurements of stellar parallax were made with small telescopes and the naked eye. Bessel's telescope had a diameter of 6.2 inches, though its special design helped him greatly to measure the parallax of 61 Cygni. Struve used a nine-inch refractor to measure the parallax of Vega. Henderson used a 4-inch mural circle to measure the parallax of alpha Centauri. The issue isn't so much telescope size as method, patience, and skill.
    According to the Wiki "Galaxy" article, already by around 500 A.D. we knew the Milky Way had no parallax. Is that true, and if so, how could we have known that without telescopes?

  13. #13
    Join Date
    Jan 2008
    Posts
    662
    Quote Originally Posted by Wikipedia article
    The Neoplatonist philosopher Olympiodorus the Younger (c. 495–570 AD) was scientifically critical of this view, arguing that if the Milky Way were sublunary (situated between the Earth and the Moon) it should appear different at different times and places on the Earth, and that it should have parallax, which it does not.
    This amount of parallax does not need a telescope to measure.

  14. #14
    Join Date
    Dec 2004
    Posts
    14,765
    Quote Originally Posted by Jeff Root View Post
    A parsec is a unit that is completely disconnected from anything
    that is actually observed.
    Put that sentence in the context of the rest of the paragraph it
    was in.

    -- Jeff, in Minneapolis
    http://www.FreeMars.org/jeff/

    "I find astronomy very interesting, but I wouldn't if I thought we
    were just going to sit here and look." -- "Van Rijn"

    "The other planets? Well, they just happen to be there, but the
    point of rockets is to explore them!" -- Kai Yeves

  15. #15
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,359
    Quote Originally Posted by Jeff Root View Post
    It is more directly connected: We observe an object
    which is x light-time units away as it was x time units ago.
    There is no comparable relationship for parsecs except for
    the amount of annual wiggling of nearby stars.

    -- Jeff, in Minneapolis
    The light-year is a function of the speed of light and an arbitrarily chosen unit of time. The parsec is a function of the longest usable baseline and an arbitrarily chosen unit of angular measure. I would say it is a tossup in arbitrariness.

    The parsec is directly related to what we actually observe, which is the wiggle of the star as we go around our orbit. We do not measure the transit time of the light from the star. Thus I would say that the parsec is completely connected with what actually is observed.

  16. #16
    Join Date
    Dec 2004
    Posts
    14,765
    Nobody ever measures the wiggle of a star. What is measured
    is the offset of the position of a star in two different images.
    Then a set of calculations are applied which factor in various
    angles and distances and correctons for all sorts of motions and
    variables in the viewing environment. That is repeated several
    times, and the average value of the result is the distance in
    parsecs.

    The arbitrarily chosen unit of time is used for all kinds of things
    by everyone on Earth, constantly, so it, at least, is very familiar.
    The part that is not so familiar is the speed of light. But if you
    have made an intercontinental phone call that goes through a
    communications satellite in geosynchronous orbit, you'll have
    *some* familiarity with it.

    -- Jeff, in Minneapolis
    http://www.FreeMars.org/jeff/

    "I find astronomy very interesting, but I wouldn't if I thought we
    were just going to sit here and look." -- "Van Rijn"

    "The other planets? Well, they just happen to be there, but the
    point of rockets is to explore them!" -- Kai Yeves

  17. #17
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,359
    Quote Originally Posted by Jeff Root View Post
    Nobody ever measures the wiggle of a star. What is measured
    is the offset of the position of a star in two different images.
    Then a set of calculations are applied which factor in various
    angles and distances and correctons for all sorts of motions and
    variables in the viewing environment. That is repeated several
    times, and the average value of the result is the distance in
    parsecs.

    The arbitrarily chosen unit of time is used for all kinds of things
    by everyone on Earth, constantly, so it, at least, is very familiar.
    The part that is not so familiar is the speed of light. But if you
    have made an intercontinental phone call that goes through a
    communications satellite in geosynchronous orbit, you'll have
    *some* familiarity with it.

    -- Jeff, in Minneapolis
    When I said "wiggle", I was referring to the annual sinusoidal parallax component of the motions to which you correctly referred. Once again, we are actually observing variations in the angular positions of the stars, and the parsec is defined in terms of this variation along with the length of our base line. I stand by my previous statement.

  18. #18
    Join Date
    Mar 2009
    Posts
    1,676
    Are amateurs capable of detecting the negative parallaxes which make up 25% of the Tycho Main Catalogue?

  19. #19
    Join Date
    Sep 2006
    Posts
    1,796
    Anyone -- anyone --- who tries to measure the parallaxes of very distant stars will end up with lots and lots of negative values. Noisy measurements of a quantity close to zero, plus a calculation involving reciprocals = (false) negative values.

  20. #20
    Join Date
    Mar 2009
    Posts
    1,676
    Quote Originally Posted by StupendousMan View Post
    = (false) negative values.
    If a negative parallax value turned out to be real, what would that imply about the star?

  21. #21
    Join Date
    Jan 2008
    Posts
    662
    Quote Originally Posted by wd40 View Post
    If a negative parallax value turned out to be real, what would that imply about the star?
    It can't be negative by definition.

  22. #22
    Join Date
    Mar 2009
    Posts
    1,676
    Could it be said in theory that negative parallax stars are further away than the majority of stars that show no parallax?

  23. #23
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,359
    Quote Originally Posted by wd40 View Post
    Could it be said in theory that negative parallax stars are further away than the majority of stars that show no parallax?
    I would say that they are farther away than reference stars that have what should be observable parallax that somehow was not recognized. I think it would require a statistical fluke in the distribution of the stars in a particular field of view.

  24. #24
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,359
    Addendum: Let me elaborate a bit. On any astrometric photograph many of the background stars will have small amounts of parallax that is barely detectable with the measuring instruments, and failure to allow for it will cause the parallax of a nearby star to be somewhat underestimated. Most photos will have a typical average amount that has been estimated statistically by the observers. If a field has a sparse, unrecognized cluster at some intermediate distance, that cluster might bias the statistically assigned residual parallax. A truly distant supergiant seen through this anomalous field would then appear to have a negative parallax. If we could somehow get truly distant supergiants, or better yet quasars, in every sample the problem would go away.

    I am no expert, but I think I am on the right track here.

  25. #25
    Join Date
    Sep 2006
    Posts
    1,796
    It's actually quite simple: suppose that you measure the positions of 10 stars in a field, and each measurement has an uncertainty of +/- 0.1 arcseconds. Suppose that all 10 stars are so far away (say, 10000 pc) that their actual parallax (of order 0.0001 arcsec) is swamped by the uncertainty in the measurements.

    You use 9 of the stars to define a reference frame. Okay. Then, you measure the motion of the tenth star relative to that reference frame.
    All you are really doing is "measuring" the noise. Since the noise can make the star appear to move any random direction, there's a fifty-ish
    percent chance that the star will appear to move in the direction opposite to that which parallax would cause it move. Bing! An apparent
    negative parallax.

  26. #26
    Join Date
    Mar 2009
    Posts
    1,676
    Over 1 million objects are listed in the Tycho Main Catalogue, and they state: "The trigonometric parallax is expressed in units of milliarcsec. The estimated parallax is given for every star, even if it appears to be insignificant or negative (which may arise when the true parallax is smaller than its error)."

    25% have negative parallax, 29% positive parallax and 46% assumed zero parallax.

    Tests were done to see if the stars moving across the instrument slit were directionally different in the northern celestial hemisphere to what they were in the southern celestial hemisphere. Of the non-zero-parallax stars in the northern celestial hemisphere, 45% of them had a negative parallax, and in the southern celestial hemisphere, 46% of non-zero objects had a negative parallax.

    Could this very symmetrical distribution be a naturally occurring phenomenon?
    Last edited by wd40; 2014-Feb-24 at 01:59 AM.

  27. #27
    Join Date
    Jan 2005
    Location
    Anzakistan
    Posts
    10,866
    My underline:

    Quote Originally Posted by wd40 View Post
    Over 1 million objects are listed in the Tycho Main Catalogue, and they state: "The trigonometric parallax is expressed in units of milliarcsec. The estimated parallax is given for every star, even if it appears to be insignificant or negative (which may arise when the true parallax is smaller than its error)."

    25% have negative parallax, 29% positive parallax and 46% assumed zero parallax.

    Tests were done to see if the stars moving across the instrument slit were directionally different in the northern celestial hemisphere to what they were in the southern celestial hemisphere. Of the non-zero-parallax stars in the northern celestial hemisphere, 45% of them had a negative parallax, and in the southern celestial hemisphere, 46% of non-zero objects had a negative parallax.

    Could this very symmetrical distribution be a naturally occurring phenomenon?
    I don't understand why it wouldn't be "natural". Why would there be fewer/lower errors North than South or vice versa?

    What is it you think is going on?
    Measure once, cut twice. Practice makes perfect.
    Hey hey, night fights day.

  28. #28
    Join Date
    Mar 2009
    Posts
    1,676
    Negative parallax is generally viewed as being impossible or incompatible eg unlike the Tycho Catalogue, in the Hipparcos Catalogue it has been artificially weighted about the zero value by restricting the negative range, negative parallaxes being dismissed as statistical errors.

    63 Ophiuci:
    "Uncertain negative parallax measurements of –0.77 0.40 mas suggest that this extremely luminous star may be located about 4000 light-years away."

    What if its negative parallax was found to be certain rather than uncertain?
    Last edited by wd40; 2014-Feb-24 at 12:46 PM.

  29. #29
    Join Date
    Sep 2006
    Posts
    1,796
    Your question carries about as much meaning as someone asking, "2+2 = 4 most of the time, sure, but what would happen if 2 + 2 = -4?"

  30. #30
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,359
    Quote Originally Posted by wd40 View Post
    Negative parallax is generally viewed as being impossible or incompatible eg unlike the Tycho Catalogue, in the Hipparcos Catalogue it has been artificially weighted about the zero value by restricting the negative range, negative parallaxes being dismissed as statistical errors.

    63 Ophiuci:
    "Uncertain negative parallax measurements of –0.77 0.40 mas suggest that this extremely luminous star may be located about 4000 light-years away."

    What if its negative parallax was found to be certain rather than uncertain?
    Please tell us, in appropriate geometric detail, what you mean by "certain negative parallax".

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •