Modelling Transits from Ground and Space Instruments

Don Pollacco, Frederic Pont, Suzanne Aigrain

Project Description:

In this project you will be given transit light curves to analyse, and a python program which generates model transit light curves. You will be tasked with measuring the planet's radius from the light curves. You will need to write your own cost function, and use a built-in least-squares optimization routine (part of the numpy package) to perform the fit. You will be encouraged to pay particular attention to correlated noise, star spots occulted by the planet, and (if time permits) unocculted star spots.

Recommended Reading:

• Transits and occultations by Josh Winn (chapter of the graduate-level textbook, EXOPLANETS, ed. S. Seager, University of Arizona Press)

Software:

• You will need to have a working installation of python on your laptop to carry out this project. Please visit the Python setup page and follow the instructions there before the start of the School.
• In the attachments below you will find 3 files containing python source code. The file ma02.py contains routines to compute the flux occulted by a planet according to the formalism of Mandel & Agol (2002), and was kindly provided by Ian Crossfield (UCLA). The file orbit.py contains routines to compute the relative positions of the star and planet, and lightcurve.py contains a routine that uses both to compute a transit light curve, and an example of how to simulate a fake transit dataset and fit it using scipy's built in downhill simplex optimizer.

Data:

The data you will use are also in the attachments below:

• XO2lightcurvedata.txt: A light curve of XO-2b obtained with the Philip Wetton Telescope on the roof of the Oxford Physics Department (courtesy of E. Bacchus & F. Clarke). The appropriate quadratic limbd-darkening coefficients to use are c1 = 0.5484, c2 = 0.1742
• ACSdata.phot: Hubble Space Telescope / ACS transit light curves of HD189733b with two wavelength channels ("red" and "blue"), from Pont et al. (2008). The non-linear limb-darkening coefficients that should be used to fit this data are included at the top of the file.