KepCurve: A Kepler Light Curve Tool

Installation

Download Setup.exe

PLEASE UNINSTALL ANY PREVIOUS VERSION. I've changed installers again and now use InstallShield. This has allowed me to correctly bundle the Limb Darkening coefficients and to fix the file associations. However, it does mean you will have manually download updates.

Download and run Setup.exe. The usual Windows instructions apply.

The first time it is run, it will prompt you for the FTP details. Normally the only one you need to change will be the FTP password. If you are using anonymous access, supply your email address. If you have a guest observer username and password, you will need to enter both. Note: if you need to change these in the future, use the Tools >> Options menu item.

Starting a New Project

Initiate your download request at MAST
When you get successful confirmation, click the "New" button or select "New" in the file menu
Enter a project name (I usually use "KID" + the number)
Check "Download files" (if you want to start the project again from scratch, unchecking this will use whatever FITS files are in the project directory)
Enter the last part of the directory name (e.g. "anonymous07324") into the Sub-directory box
Click through the rest of the wizard.

When new data is released, an existing project can be updated using the "Update Data" option in the "File" menu. This option will download any new curves from the specified directory, will delete any existing merged curves, but will retain all data about transits, etc.

Overall Flow of a Project

Generally, a project characterising planets proceeds roughly as:

Create the project and download the files
Create a merged curve containing all quarters
Mark the first and last transit
From the start time of these and the number of transits, compute the period (Note: see new controls on "Planets" tab)
Create a "planet", add missing transits and customise marker appearance
On the "Time Plots" tab, zoom to first transits and adjust start and end times, zoom to next and repeat until all transits adjusted
On the "Planets", click "Recalculate planets".
Detrend the curves
Fine tune the period using the "nudge" buttons (use "Epoch++/--" to centre and "Period++/--" to adjust) on the "XY Plots" tab
Look for contamination - rain plots, phased centroids, etc
If there is contamination, look at the vector to dientify rough direction of contaminating star
Look for Transit Timing Variations (select the "Epoch Error" plot on the "Time Plots" tab)

Merging Curves

Select the curves you want in the list on the "System" tab
Click on the "Merge" button

Detrending Curves

Select the curves you want from the list on the "System" tab, or select a curve from the drop-down box in the toolbar.
Click on the "Detrend" button
Enter the required period (it defaults to 0.2 days or the transit width, whichever is greater)
Click through the wizard (this can be very slow for merged curves)

Note: plots that require detrended data will be blank if you haven't detrended a curve.

Marking a Transit

Make sure both rulers are visible
Position vertical rulers at start and end of transit
Position horizontal rulers at top and bottom of transit
Click "Mark Transit" button and it should appear in the list

Creating a Planet

Select a merged curve containing all quarters
Go to "Planets" tab
Select first and last marked transit
Set # transits to the the total number of transits included (and including) between first and last
Click "Calculate Approx Period"
Set an appropriate "Identifier" - convention is to start with "b"
Set a marker style - i.e. colour, shape and size
Click "Add" - this will create a "planet"
If you haven't marked all transits, click "Add Missing"

Reviewing a Planet

Once a planet has been created and all transits have been created, often you'll want to adjust the individual transits
Go to "Time plots" tab
Zoom to the first transit by clicking next
Adjust the X rulers to mark the start and end of the transit
Uncheck the "Complete" check box if there are no samples or an incomplete set of samples for the transit
Click "Update" and then "OK" to update the start and end times of the transit
Click "Next >>" to zoom to the next transit; repeat until all transits adjusted
If you have any excess / duplicate transits (that may result from significant TTVs etc), select these and click "Delete"
Go to the "Planets" tab and click "Recalculate Planets". This will work out the average period for each planet and calculate epoch errors from this period, assuming the first transit is correct.

Reviewing the Period of a Planet

TBD

Checking for Contamination

TBD

Checking for Transit Timing Variations

TBD

Checking Individual Target Pixels

Select a single quarter light curve
Click the "Pixels" button on the "Star" tab. A new window should appear showing the aperture on the left.
Click on an aperture pixel.

Stellar Parameters

KepCurve attempts to estimate an approximate stellar class from the photometry data. First it computes an effective temperature, using polynomials from the following references:

The infrared flux method and color calibration, Casagrande et al (2008). [PDF from IOP Science].
M dwarfs: effective temperatures, radii and metallicities, Casagrande et al (2008). [PDF from arxiv.org]. Not used yet.
An absolutely calibrated Teff scale from the infrared flux method, Casagrande et al (2010). [PDF from arxiv.org]. Used for estimation from 2MASS J - K magnitudes where (0.07 ≤ J - K ≤ 0.80)
A Revised Effective Temperature Scale for the Kepler Input Catalogue, Pinsonneault et al (2012). [PDF from arxiv.org]. Used for estimation from SDSS griz band magnitudes.

It then computes the class, mass, radius and log(g) from the data here. If the FITS files from MAST do not include all the mass, radius, log(g) and metallicity data, the values compute above are used (a metallicity of -0.2 is assumed)

Planned Changes

Below is a list of things I'd like to do. I make no promises about when or which order they will be implemented!

Measurement tools. In particular, a special ruler with a label indicating the difference (in time) between the two vertical rulers
Contamination vector plot centred on nominal RA / Dec with skyview background
FFI image support
Current CCD visualisation (with vectors for x, y, RA and Dec)
SAP_QUALITY visualisation (shade the background where pointing is dodgy, etc)

Known Issues

There are a few known issues. Some are harder to fix than others . . .

Ruler positions - issues when change visibility and / or zoom level
Ruler performance is awful, particularly on merged curves

Recent Changes

17 Apr 2013: Changed to use InstallShield.; Various bug fixes.
15 Dec 2012: Added "Stellar Parameters" tab. This shows the available photometry for the star and computes an approximate class for the star
28 Nov 2012: UI tweaks to add info about selected planet to "Star" tab and to display a/R* rather than R*/a; UI tweaks to only show mid-point of transits.; Bug fixes in file load to prevent overwriting of Rp/R* and R*/a values.; Bug fixes in proposal modelling, analyse transits, "add missing" and "re-calculate".; Bug fixes in transits at the start of a light curve.; Bug fixes in normalisation and detrending of SAP_FLUX; Bug fix in calculation of transit parameters when eccentricity > 0
26 Nov 2012: Added a "Model..." button to the "Time Plots" tab. This allows the planet proposal paramters to be manually tuned.; Upgraded the image capture for transit analysis and viewing individual pixels; Various bug fixes in calculation of eccentric anomaly and true anomaly
24 Nov 2012: The application will re-read the FITS files to extract the additional star data, so loading an existing project will be slower the first time.; Added Target Pixel File viewer.; Added additional fields from the FITS file to the star.; Various bug fixes in SkyGroup; Various bug fixes in adding planets and adding missing transits.; Code for transit fitting and MCMC analysis now complete but buggy ("Analyse..." button on the Phase tab).; Updated the XML file format to add planet proposal parameters for transits and planets; Updated the XML file format to fix "Temperatue" typo
10 Nov 2012: Added Sky Group visualisation.; Added Row and Column direction visualisation.; Added "Zoom" button on "XY Plots" tab.; Added "Modified" check on exit.
4 Nov 2012: Updated plots to differentiate between marked, orphaned and predicted transits.; Fixed bug in calculating number of transits in selection on Phase tab; Fixed bug where Y axis not updated correctly on zoom; Started preparation for adding Mandel and Algol curve fitting and wavelet noise analysis
1 Nov 2012: Updated plots to calculate whether "in transit" from transit start and end times rather than from period of planet. This matters when there are significant TTVs.; Added code for calculating planet radius and Teq
28 Oct 2012: Implemented Kwee vanWoerden method for finding transit minima; still a little flaky; Added ability to show cubic spline fit to transits; still a little flaky; Various minor UI tweaks
28 Oct 2012: Added "Update Data" option to file menu
23 Oct 2012: Improved "zoom to transit" functionality; Added TTV curves to Time plots (Epoch Error and Duration Error); Fixed bug in default colour and size for markers on new planets; Fixed bugs and improved pan left / right