Exoplanet periods

This afternoon I was researching the confirmed Exoplanets using the Open Exoplanet Catalogue, to get a feel for the length of planetary periods. I'm developing WEASA - WolfMagick Exoplanet Adaptive Search Algorithm. ...I'll talk about that technology later on the observatories thread.

Anyway...
The average period of an extra solar planet is on the order of 723 days, a bit less than two years

Over 730 days: 212 10.62%
365 – 730: 109 5.46%
270 – 364: 42 2.10%
180 – 270: 45 2.25%
90 – 180: 80 4.0%
1 – 90: 1331 66.68%
0 – 1: 13 0.65%

Its a bit interesting that most 66.68%, 2/3 of exoplanets are close in and fast moving with a period less than 90 days. While the average is still at 723 days, but those 212 slow planets can be rather slow...

I still have to do more analysis on the 1 - 90 day period planets...


Of the slow planets the average period is 5717.72 days. This average is inflated like this due to 2 planers; 1) HR8799b at 164250 days, and 2) Fomalhautb at 320000 days.

originally posted by: tanka418
This afternoon I was researching the confirmed Exoplanets, to get a feel for the length of planetary periods. I'm developing WEASA - WolfMagick Exoplanet Adaptive Search Algorithm. ...I'll talk about that technology later on the observatories thread.

Anyway...
The average period of an extra solar planet is on the order of 723 days, a bit less than two years

Over 730 days: 212 10.62%
365 – 730: 109 5.46%
270 – 364: 42 2.10%
180 – 270: 45 2.25%
90 – 180: 80 4.0%
1 – 90: 1331 66.68%
0 – 1: 13 0.65%

Its a bit interesting that most 66.68%, 2/3 of exoplanets are close in and fast moving with a period less than 90 days. While the average is still at 723 days, but those 212 slow planets can be rather slow...

I still have to do more analysis on the 1 - 90 day period planets...


Of the slow planets the average period is 5717.72 days. This average is inflated like this due to 2 planers; 1) HR8799b at 164250 days, and 2) Fomalhautb at 320000 days.

You also have to consider tanka, the selection bias of various exoplanet detection techniques.

For instance radial velocity and transit photometric searches both require the planet to make several orbits around it's star before it is confirmed as real.

So the bias here is towards planets which orbit close to their star as they make more frequent transits.

There's also the bias due to the amount of planets around red dwarfs, the most common type of star in the universe. Planets around these stars are interesting because they can have more frequent transits yet exist in the stars habitable zone. And on top of this, because the star they orbit is dimmer they are easier seen in transit searches and are more sought after as potential future targets for direct imaging space telescope missions.

So yeah, never forget selection bias.

originally posted by: JadeStar
You also have to consider tanka, the selection bias of various exoplanet detection techniques.

For instance radial velocity and transit photometric searches both require the planet to make several orbits around it's star before it is confirmed as real.

So the bias here is towards planets which orbit close to their star as they make more frequent transits.

There's also the bias due to the amount of planets around red dwarfs, the most common type of star in the universe. Planets around these stars are interesting because they can have more frequent transits yet exist in the stars habitable zone. And on top of this, because the star they orbit is dimmer they are easier seen in transit searches and are more sought after as potential future targets for direct imaging space telescope missions.

So yeah, never forget selection bias.

Oh yes...I know, that wee bit of analysis is from a sample of 1996 planets, a rather small sample...but, gotta start somewhere.

This analysis will help to build the methods, algorithms that I use to initialize a search. It also appears, that a thought I had of possibly identifying the stars in a given view and measuring (recording data) on a target star, even when the telescope is tasked with a different job. A method possibly to improve the sample rate, and regularity (duty cycle) for targets.

And, as I'm sure you know, I'll be using the transit method, So I'll be looking primarily for larger planets. But, it also gives me a small advantage in that once a target is identified in a view, I only need to capture magnitude data for processing, which is much simpler and does not require any tracking of the star...only enough time in "view" to measure magnitude.

As I've already stated, my observatory project is providing an incredible learning experience, and as such, I don't always get to analyzing and understand all things at once. With this Exoplanet thing I was originally thinking that it should be possible, but didn't know much about the actual process. When Jade told me this equipment should be able to "see" transits, I was pleasantly surprised...though not really surprised because I strongly suspected the equipment should have the requisite sensitivity and resolution.

As I am learning, I'm finding that this equipment, if handled properly, may be capable of some rather fantastic things...like the detection of exoplanets.

So, given the optical sensor (CCD) I'm planning to use, and after watching that really nice video, I'm wondering IF there is anything preventing me from using Radial Velocity, since, thins is typically detected by Doppler shift.

My system will be using an adapted version of OSCAAR, a differential photometry system. This is used to "stabilize" the magnitude of a star in a field of view, so that the magnitude of another may be accurately measured. If looking for "transits" a change in the target star's magnitude is used to produce a "light curve", and ultimately the "transit" may be detected and measured.

The same sort of differential photometry can also be used to stabilize color shifts cased by atmospheric distortions, not unlike the stabilization of magnitudes. Thus it should be possible to do a color analysis of a target star, and over time produce a sort of "light curve" for the color properties of the target (color curve), and via that method, discover planets using Radial Velocity.

Of course some of the questions I have would be; "how small of a color change would there be due to Doppler shift?" I think about the actual hardware that MUST be involved with any such search, and I'm just not sure that 24 bit color is enough. But, on the other hand, I've not heard of a camera / CCD that does better.

It is true that most discovered exoplanets are large and quite near their own suns, because the method used to detect them works best on this exact type of planet. Still, it is interesting to me how we have found lots of big planets, both earth-like and gas-like, very near their suns, when in our own solar system, we have four earth-like planets nearer the sun, and four gas-like planets farther out. I'm not sure what implications this has for things like how solar systems and planets are formed, but I do find it interesting that we have found solar systems that differ dramatically from our own in this manner. It's impossible to know what the most 'common' type of solar system is without more data because, as was said, the data is biased due to the method of detection, but I hope we'll learn someday!

originally posted by: DragonsDemesne
It is true that most discovered exoplanets are large and quite near their own suns, because the method used to detect them works best on this exact type of planet. Still, it is interesting to me how we have found lots of big planets, both earth-like and gas-like, very near their suns, when in our own solar system, we have four earth-like planets nearer the sun, and four gas-like planets farther out. I'm not sure what implications this has for things like how solar systems and planets are formed, but I do find it interesting that we have found solar systems that differ dramatically from our own in this manner. It's impossible to know what the most 'common' type of solar system is without more data because, as was said, the data is biased due to the method of detection, but I hope we'll learn someday!

Indeed! Of the 1996 planets in the open exoplanet catalog about 1200 were found using the "transit" method. This requires a substantial planet, on the order of 2% or so of its Star's size. And, of those 1200, 1139 have a period of less than 90 days...that's like 57% of discovered planets so far share a discovery method, and have a period less than 90 days.

Most of the rest were discovered by the Radial Velocity method...interestingly; 45 planets have been detected by direct imaging, though with them it sees the data is less complete.

You can browse the open exoplanet catalog at my observatory...under the database->open exoplanet catalog menu item. Sorry, I haven't finished the search capabilities yet.