Weather on other worlds

It's not all climbing you know. The last 40 hours I've been travelling to Sydney to do some observing on an old favourite telescope of mine. Over the last few years, various astronomers across the globe have been working hard to directly measure the brightness of planets orbiting other stars - so called exoplanets. This is relatively 'easy' at infrared wavelengths but is incredibly difficult in optical light, where the planet is much, much, much fainter than the star it orbits. This matters because the best way of measuring the planet's brightness is to look for the minuscule dip in brightness of the combined light from the star and planet, when the planet disappears behind the star.

The dip in brightness is so small because the planet is so faint - the often used analogy is that it's like trying to see the change in overall brightness when a firefly disappears behind a searchlight. Nevertheless, over the last few years we have honed our techniques and we are starting to routinely measure the brightness of exoplanets in optical light - like the example below.

Light vs time for the exoplanet system WASP-12. The dip (a fraction of one percent) is caused by the planet disappearing behind the star.

The thing is, as we have more and more measurements like this, a puzzle is starting to emerge. It's becoming quite common for different groups to disagree on the brightnesses of exoplanets. An example is shown below - look at the furthest left datapoint. There's two measurements of brightnesses for the same planet by different groups, and they don't agree.

Brightness of planet as a function of wavelength. The furthest left points are two measurements of the optical brightness

There are only really two possibilities here. The first is that we are not as good at measuring the brightness of exoplanets as we think we are. The apparent disagreement in this case is caused by us overstating how confident we are in our measurements. The second possibility is that the planets really are changing in brightness over time - or, to put it another way - we could be seeing weather on the surface of planets around other stars. My money is on the first explanation - but there's a really obvious way to check. That's to measure the same event from many different telescopes. Since it's the same event all the different telescopes should give the same answer, and if they don't then our techniques are not as good as they thought they were.

That's what I should have been doing tomorrow night, but looking at the weather it looks like I might be watching movies on iTunes instead. Now, Sydney is a long way to come to work one night and find it is cloudy, but I'm not too depressed because I'm taking a two week climbing holiday afterwards, so if the science doesn't play ball I've got this to look forward to...