View Single Post
Old 18-04-2010, 07:06 AM
Registered User

binofied is offline
Join Date: May 2005
Location: Auckland New Zealand
Posts: 44
Thanks, your right about it being hard to get people interested. My hope is that the idea that any ameteur can actually be part of a real planet discovery will spark someone elses imagination. It is very cutting edge stuff and very few others have been there before you. The first planet discovered by Microlensing was only about five years ago!

I have also been doing a lot of astrophotography over the last few years and to be honest this stuff is very easy in comparison. All you need to do are five minute guided exposures. You don't need to do any autoguider dithering or multiple filters. The feilds are always busy being in the galactic bulge meaning you always have a bright guide star. You don't have to do anything with your images other than dark subtract and flat feilding. So no aligning, stacking let alone colour combining and balancing or stretching, masks etc.

The data extraction is done by data processing pipelines at Ohio State University.

A typical event starts with an E-Mail from Ohio State who are analysing all the possible candidates found by the big MOA and OGLE survey scopes. It says we thing MB077 will possibly go high mag in two days. They only have data from MOA of what the event has done up until that time and have no real idea of how it will finally unfold. As more data comes in they can get a bit more of the graph and see if it's going higher or lower than the prediction possibly signalling an anomaly. The flurry of E-Mails as the Pros discuss possible models of the event are spell binding as it unfolds. There are four possible outcomes. It is a normal star with no planet, it is a binary star, the data is so sparse or ambiguous they can not deconvolve the outcome (this is rare). It is a planet!! On several occasions the results show that it is actually two planets around a star!

It has taken me about two years to get to grips with the science behind, even in crude terms, how they deconvolve the light curve to figure out what combinations of source/ planet mass ratio rotational speed and often the planets actual mass. So if you like high end physics it doesn't get any better than this. This idea was dismissed my Einstien as being impossible to ever see happen, he refused for several year to even publish a paper on it as he knew how hard it would be to ever prove. If you had 1,000,000 observers all taking very accurate measurements of stars as dim as mag 22 doing 10 observations each night, ten of them may detect a microlensing event once a year. Every ten years one of those events would yeild a planet!

Attached is the data from a recent event that has been proven to be a binary.

Last edited by binofied; 18-04-2010 at 07:19 AM.
Reply With Quote