Hubble SWEEPS field (APOD 13 Oct 2006)

lex · Post by **lex** » Fri Oct 13, 2006 8:28 am

Okay, so extra-solar planets can be detected by this method when they pass in front of their parent stars, this only works though when we are side-on to the orbits of the planet though (ie when they pass between their stars and ourselves as we observe) how many degrees does that allow the orbital plane to be in for this method to work? between 1 and 5 maybe?

What about all those systems where we are not side-on to the orbital plane of the planets though? Do we have to detect planets around those stars by how much the stars appear to wobble?

Just wondering hypothetically how much higher the number of stars that have planets could be going by this data.

harry · Post by **harry** » Fri Oct 13, 2006 8:35 am

Hello Lex

If the origin of stars is smilar to our sun.

We maybe expect a high proportion of stars having planets and in some cases other stars. Thats a real solar system.

astro_uk · Post by **astro_uk** » Fri Oct 13, 2006 8:58 am

Hi Lex

There are several methods that can be used, the first is the transit method you describe, the viewing angle that allows you to see a transit depends on the inclination of the system, the distance to the system and the radius of the orbit of the planet,.

Another is the doppler method, which allows you to detect the doppler shift caused by the planet pulling back on the star. But this method also only works if some component if the stars motion due to the planets gravity is in the line of sight, so it wont work if you are looking down on the system from "above".

The other is simple in premise but vey difficult in practice, you block the light of the star and try to find the light from the planet. This will work for all configurations of observer and system. In practice the star will be outshining the planet by a factor of a million or more, so it is incredibly difficult.

andyrint · Post by **andyrint** » Fri Oct 13, 2006 11:38 am

Hmm, I'm curious, how do they know it's a gas giant?

I would have thought that a gas giant orbiting that quickly and close to a star would get evaporated (or something similar) over time.

ckam · Post by **ckam** » Fri Oct 13, 2006 2:00 pm

how can gas evaporate?
then, theres so many gas that it was already enough to form a planet in that location. what could possibly happen to it afterwards?

Bob_Scopatz · Post by **Bob_Scopatz** » Fri Oct 13, 2006 3:06 pm

Could the transit method work to detect smaller planets in more distant orbits?

I had two thoughts on this:
1) If the detectors are sensitive enough, a small change in detected light from the star might indicate the presence of a small planet -- just like it does for close-orbit large planets.

2) if, from our vantage point, the small planet in turn "eclipses" the large planet that is eclipsing the sun. We should see changes in the change in light due to the large planet's transit across the sun. At some point in its path, the smaller planet would block light as it's "disk" lines up with the edge of the larger planet's "disk" and suddenly, we'd see a "increase" in the shadow" that would disappear as the large planet moves along...then reappear at the other side... then maybe become indetectably small again...

So...we'd need to catch two planets whose orbits just happenened to line up perfectly with Earth, and which happen to line up as we were viewing that day...

Small chances of seeing it, but...would this work?

Again...what's the smallest object detectible using this method at the distances involved?

iamlucky13 · Post by **iamlucky13** » Sat Oct 14, 2006 12:26 am

Bob, for your second hypothesis, unless I misunderstand it somehow, the intensity would actually increase slightly as the smaller planet lined up with the big planet and the star. The total shadowed area would be the area of just the big planet, as opposed to the area of the big planet plus the little planet as the alignment breaks a little bit.

For your first hypothesis, you are correct. In fact, the transit method probably has the best chance for any method of detecting small planets because they don't have enough mass to induce a noticeable wobble. The Kepler space observatory will actually be a dedicated mission lasting several years using the same method as this 7 day Hubble observation, and has the potential to find thousands of planets, some as small as the earth.

For those interested, I believe 6 of these planets have been confirmed using the wobble method.

Andyrint, I believe a planet's magnetic field has a significant effect on how fast gas "evaporates" (actually is blown away by the solar wind) from a gas giant. A planet with a strong magnetic field should be able to survive for a long time.

Bob_Scopatz · Post by **Bob_Scopatz** » Sat Oct 14, 2006 2:15 am

This is amazing stuff.

Thanks for the answer!

Makes sense.

Nereid · Post by **Nereid** » Tue Oct 31, 2006 1:34 pm

For those interested, this Hubblesite "Fast Facts" page has a scale bar of the SWEEPS field featured in the APOD, and a pair of orientation arrows.

Keldor314 · Post by **Keldor314** » Sun Nov 05, 2006 12:17 pm

There would be a limit to how small a planet you could pick out by the occulusion method, simply because every star has an active changing surface (some more so that others).

Imagine an earth sized planet orbiting a sun sized star. Now, if we could actually resolve the star's disc, we would see the planet as a tiny speck on it. Now, imagine the star having its own solar flares, prominaces, and so forth. These don't really change the brightness all that much, but against a pinprick sized planet, the planet will tend to get lost in the noise.

iamlucky13 · Post by **iamlucky13** » Mon Nov 06, 2006 5:24 pm

That's a good point, although it may still be possible to pick out relatively small signals because planetary occlusions are very periodic. An alien civillization watching the sun from hundreds of light years away would see it dim by a tiny amount every ~365.25 days. The sun spots and solar flares may have a bigger effect (I'm not sure), but are more random.

But eventually even that signal becomes insignificant.