6 ms·

If I follow your argument about 16 bit cameras, why are hundreds of earth sized planets detected with the transit method? (< 2R)

7373737373 3y ago
I have no idea how, and would like to know. It may be that these are stars with smaller diameters, so the smaller planets occlude a larger fraction compared to sun-like stars. Otherwise I suspect the required hardware is likely out of reach for amateurs with current technology. Edit: Found these: https://en.wikipedia.org/wiki/MEarth_Project https://en.wikipedia.org/wiki/MEarth_Project > The MEarth Project is a United States NSF-funded, robotic observatory that is part of Fred Lawrence Whipple Observatory on Mt. Hopkins, Arizona, US. The project monitors the brightness of thousands of red dwarf stars with the goal of finding transiting planets. As red dwarf stars are small, any transiting planet blocks a larger proportion of starlight than transits around a Sun-like star would. This allows smaller planets to be detected through ground-based observations. https://www.eso.org/public/teles-instr/paranal-observatory/speculoos/ https://www.eso.org/public/teles-instr/paranal-observatory/s... > The mission for SPECULOOS is to detect terrestrial planets as they transit across small, cool stars in the solar neighbourhood It seems the terrestrial planets https://en.wikipedia.org/wiki/TRAPPIST https://en.wikipedia.org/wiki/TRAPPIST detected are also around dwarf stars. So I guess the summary is: finding and detecting earth-like planets around dwarf stars is possible, however around sun-like stars not quite yet using these two methods
- rich_sasha 3y ago
  I suppose it depends how regular the signal is otherwise. If a dip happens super-regularly (which I guess it would with single star planet orbits), it might be small relative to pointwise noise of the detector, but large enough to be detected over time (if you have the observations). Your point on period length stands. And anyway I'm just waxing lyrical, I'm a statistician and amateur stargazer not a pro :)
- pfdietz 3y ago
  How often is the bucket filled during the transit? Noise in each emptying will average out, so a dip could be detected even if the dip in a single reading of the bucket could not be.
  - 7373737373 3y ago
    Typical transits last a few hours, and exposures times are typically a few minutes. So you have up to hundreds of exposures per transit. I guess how "statistically recoverable" this data is depends beside the resolution also on the noise level. The atmosphere introduces a lot of variability in ground-based observations, and a star's brightness is not constant either.
    - 7373737373 3y ago
      https://astronomy.stackexchange.com/questions/51679/feasibility-of-detecting-earth-like-exoplanets-with-amateur-telescopes https://astronomy.stackexchange.com/questions/51679/feasibil...
- inoop 3y ago
  Damn those are some good names
- hotpotamus 3y ago
  Yeah, I was going to say it's probably to do with "sun-like" and angular size ratios. (I'm not religious, but the sun and the moon being the same angular size has always felt like an easter egg that a creator might leave as a bit of fun). Sun-like is important in the search for life because despite putting out lots more energy, the suspicion is that yellow stars tend to be less temperamental in their flare behavior (so they don't scour the surface of any potential planets with large amounts of radiation every once in awhile). And there is evidence that our sun is even a particularly calm yellow star. The habitable zone is also in a region that doesn't tidally lock the planet; a day-night cycle seems important though not neccesarily a requirement for habitability. And yellow stars' lifecycle seems to be long enough to allow for life to form and evolve. Essentially they seem most likely to produce habitable planets - it's probably not mere happenstance that we live around one.
contravariant 3y ago
Take measurements across multiple transits and average them. Periodic signals are quite easy to identify even amongst noise.
- 7373737373 3y ago
  In the case of an earth-like planet (considering not just composition and size, but also orbit + host star similarity) that's difficult to do when the transit only occurs every 365 days Even finding just one such planet will require a lot of observation: The "geometric probability" of the orbital plane being aligned for an earth-like orbit seems to be around 0.5% [0]. Assuming every star has an earthlike planet [1], only 1 in 200 observed stars will be correctly aligned. So observing 365*200=73,000 stars over an entire day (also not possible for ground-based telescopes due to the day/night cycle) should statistically result in one transit observation. The transit duration is also a problem, for an earth-like exoplanet it is 13 hours, so if the observation time is too short, one would only catch the beginning or end of the transit [0] https://www.researchgate.net/figure/Transit-Properties-for-Solar-System-Objects_tbl1_226590554 https://www.researchgate.net/figure/Transit-Properties-for-S... [1] The current hypothesis seems to be that every star has at least one planet https://en.wikipedia.org/wiki/Planet-hosting_star https://en.wikipedia.org/wiki/Planet-hosting_star), but it's all a bit up in the air as not many exoplanets (and especially earth-like ones) have been observed yet. The sample size is too low.
  - zuminator 3y ago
    > The current hypothesis seems to be that every star has at least one planet I think you mean, that stars have at least one planet on average. E.g. in the unlikely event that only one third of stars had planets, but each one of those stars had on average 3 stars, then it would fit the hypothesis of averaging at least one planet per star. It's not a stretch at all since our rather boring star has at least 8 planets, and of the 4,051 stars we can currently detect with exoplanets, they have a total of 5,438 planets, meaning over 1.34 planets per star not including the unknown but unquestionably vast number of planets we currently don't have the capability of detecting. Still, just as we suspect large numbers of rogue planets that have lost their stars, there must be at least some stars out there that have lost their planets, whether to passing neutron stars or perhaps indigestion. [0] https://en.wikipedia.org/wiki/Lists_of_exoplanets#:~:text=These%20are%20,systems https://en.wikipedia.org/wiki/Lists_of_exoplanets#:~:text=Th...

7373737373 3y ago

I have no idea how, and would like to know. It may be that these are stars with smaller diameters, so the smaller planets occlude a larger fraction compared to sun-like stars. Otherwise I suspect the required hardware is likely out of reach for amateurs with current technology. Edit: Found these: https://en.wikipedia.org/wiki/MEarth_Project https://en.wikipedia.org/wiki/MEarth_Project > The MEarth Project is a United States NSF-funded, robotic observatory that is part of Fred Lawrence Whipple Observatory on Mt. Hopkins, Arizona, US. The project monitors the brightness of thousands of red dwarf stars with the goal of finding transiting planets. As red dwarf stars are small, any transiting planet blocks a larger proportion of starlight than transits around a Sun-like star would. This allows smaller planets to be detected through ground-based observations. https://www.eso.org/public/teles-instr/paranal-observatory/speculoos/ https://www.eso.org/public/teles-instr/paranal-observatory/s... > The mission for SPECULOOS is to detect terrestrial planets as they transit across small, cool stars in the solar neighbourhood It seems the terrestrial planets https://en.wikipedia.org/wiki/TRAPPIST https://en.wikipedia.org/wiki/TRAPPIST detected are also around dwarf stars. So I guess the summary is: finding and detecting earth-like planets around dwarf stars is possible, however around sun-like stars not quite yet using these two methods

rich_sasha 3y ago
I suppose it depends how regular the signal is otherwise. If a dip happens super-regularly (which I guess it would with single star planet orbits), it might be small relative to pointwise noise of the detector, but large enough to be detected over time (if you have the observations). Your point on period length stands. And anyway I'm just waxing lyrical, I'm a statistician and amateur stargazer not a pro :)
pfdietz 3y ago
How often is the bucket filled during the transit? Noise in each emptying will average out, so a dip could be detected even if the dip in a single reading of the bucket could not be.
- 7373737373 3y ago
  Typical transits last a few hours, and exposures times are typically a few minutes. So you have up to hundreds of exposures per transit. I guess how "statistically recoverable" this data is depends beside the resolution also on the noise level. The atmosphere introduces a lot of variability in ground-based observations, and a star's brightness is not constant either.
  - 7373737373 3y ago
    https://astronomy.stackexchange.com/questions/51679/feasibility-of-detecting-earth-like-exoplanets-with-amateur-telescopes https://astronomy.stackexchange.com/questions/51679/feasibil...
inoop 3y ago
Damn those are some good names
hotpotamus 3y ago
Yeah, I was going to say it's probably to do with "sun-like" and angular size ratios. (I'm not religious, but the sun and the moon being the same angular size has always felt like an easter egg that a creator might leave as a bit of fun). Sun-like is important in the search for life because despite putting out lots more energy, the suspicion is that yellow stars tend to be less temperamental in their flare behavior (so they don't scour the surface of any potential planets with large amounts of radiation every once in awhile). And there is evidence that our sun is even a particularly calm yellow star. The habitable zone is also in a region that doesn't tidally lock the planet; a day-night cycle seems important though not neccesarily a requirement for habitability. And yellow stars' lifecycle seems to be long enough to allow for life to form and evolve. Essentially they seem most likely to produce habitable planets - it's probably not mere happenstance that we live around one.

contravariant 3y ago

Take measurements across multiple transits and average them. Periodic signals are quite easy to identify even amongst noise.

7373737373 3y ago
In the case of an earth-like planet (considering not just composition and size, but also orbit + host star similarity) that's difficult to do when the transit only occurs every 365 days Even finding just one such planet will require a lot of observation: The "geometric probability" of the orbital plane being aligned for an earth-like orbit seems to be around 0.5% [0]. Assuming every star has an earthlike planet [1], only 1 in 200 observed stars will be correctly aligned. So observing 365*200=73,000 stars over an entire day (also not possible for ground-based telescopes due to the day/night cycle) should statistically result in one transit observation. The transit duration is also a problem, for an earth-like exoplanet it is 13 hours, so if the observation time is too short, one would only catch the beginning or end of the transit [0] https://www.researchgate.net/figure/Transit-Properties-for-Solar-System-Objects_tbl1_226590554 https://www.researchgate.net/figure/Transit-Properties-for-S... [1] The current hypothesis seems to be that every star has at least one planet https://en.wikipedia.org/wiki/Planet-hosting_star https://en.wikipedia.org/wiki/Planet-hosting_star), but it's all a bit up in the air as not many exoplanets (and especially earth-like ones) have been observed yet. The sample size is too low.
- zuminator 3y ago
  > The current hypothesis seems to be that every star has at least one planet I think you mean, that stars have at least one planet on average. E.g. in the unlikely event that only one third of stars had planets, but each one of those stars had on average 3 stars, then it would fit the hypothesis of averaging at least one planet per star. It's not a stretch at all since our rather boring star has at least 8 planets, and of the 4,051 stars we can currently detect with exoplanets, they have a total of 5,438 planets, meaning over 1.34 planets per star not including the unknown but unquestionably vast number of planets we currently don't have the capability of detecting. Still, just as we suspect large numbers of rogue planets that have lost their stars, there must be at least some stars out there that have lost their planets, whether to passing neutron stars or perhaps indigestion. [0] https://en.wikipedia.org/wiki/Lists_of_exoplanets#:~:text=These%20are%20,systems https://en.wikipedia.org/wiki/Lists_of_exoplanets#:~:text=Th...