Working with Observations

Examples and more docs to be added.

Required File Formats and Organization

These are by design precisely compatible with the organization scheme adopted by MCRE, the IDL Monte Carlo Results Explorer package by Perrin et al.

The observations summary index

All observations should be stored in one subdirectory. That directory must contain a file named summary.txt providing an index of the available files. This might look like:

hktau_sed.txt           SED
example1.fits           image   1.1
example2.fits           image   10
psf_image1.fits         psf     10
tinytim_psf_F110W.fits  psf     1.1

In other words, it lists file names, what those files contain (image/psf/SED/visibility) and (for all types other than SED and SPECTRUM) the wavelength. There can be images and PSFs at multiple wavelengths, but can be only one SED file. This file should be created by hand when you assemble the SED and images.

Allowed input file types are as follows:

• SED: SED as a text file (see below)
• SPECTRUM: Spectrum as text file. Same format as SED
• IMAGE: Image as a 2D FITS file.
• IMAGE_UNCERT: Uncertainty image, 2D FITS file, must match image in size
• MASK: Good/bad pixel mask; 2D FITS file, must match image in size

The following options from MCRE are not yet supported in the mcfost python package.

• POLIMAGE: Polarization percentage images, 2D FITS file, must match image in size
• PSF: PSF. Must match image in pixel scale but can be smaller in size
• V2DATA: OIR V^2 interferometry data
• MMUVDATA: Millimeter V^2 interferometry data

The SED file

The name of this file can be anything, as set in the summary.txt file. Fluxes and uncertainties must be given in Janskys. The file format should be as follows:

wavelength     flux in Jy     uncertainty in Jy     comment.

Lines starting with ‘#’ are interpreted as comments. This file is created manually when you assemble the SED from data or the literature. For example:

# WAVELEN FLUX   FLUXERR         COMMENTS/SOURCE
0.550000 0.0082 0.002           ACS_F555W
0.795000 0.0124 0.005           ACS_F814W
1.23500 0.019   0.005           2MASS_J
1.66200 0.029   0.005           2MASS_H
2.15900 0.050   0.015           2MASS_Ks
3.550   0.188   0.010           IRAC_1
4.439   0.182   0.010           IRAC_2
100     0.00    5.2             IRAS_100
<etc.>

To specify a point with an upper limit only, set the Flux column to zero, and the Uncertainy to the 1 sigma upper limit value, as in the last line of the above example.

Images

These should be FITS files, of the exact same file dimensions in X and Y as the MCFOST outputs. Image uncertainty files give the 1-sigma uncertainty for the corresponding image. Mask files are binary masks indicating which pixels should be included when calculating statistics.

Observation classes detailed descriptions

class mcfost.Observations(directory='.')

Class for observational data on a target

To the extent possible, the API is consistent with that of the ModelResults class. That is, the names and functionality of attributes and methods should be similar to the extent possible given the differences in the underlying data.

images = None

Container for model images.

image_wavelengths = []

List of wavelengths for which we have images

sed = None

Observed SED

describe()

Return a descriptive brief paragraph on what results are present

class mcfost.models.ObservedSED(filename=None, uncertainty=None, mask=None, format='no_header')

Observed SED class. Reads observations from disk; returns them as as astropy Units objects

flux

Flux in Janskys, as an astropy.Quantity

wavelength

Wavelength in microns, as an astropy.Quantity

frequency

Frequency in Hertz, as an astropy.Quantity

uncertainty

Uncertainty in flux in Janskys, as an astropy.Quantity

plot(title=None, overplot=False, alpha=0.75, marker='o', color='blue', linestyle='None', **kwargs)

Plot the observed SED.

Parameters :

overplot : bool

Should this overplot or make a new plot?

alpha : float

Transparency for plots

title : string

Title for plot.

Matplotlib keywords such as marker, linestyle, color, etc are also supported. :

class mcfost.models.ObservedImage(filename, uncertainty=None, mask=None)

Container class for an observed image, optionally along with associated uncertainty image and pixel mask.

plot(wavelength, overplot=False, cmap=None, ax=None, vmin=None, vmax=None, which='image')

Display one image

Parameters :

wavelength : float

desired wavelength to plot into

overplot : bool:q

cmap : matplotlib color map

desired comor map

ax : matplotlib axis

axis to plot into

vmin, vmax : floats

min and max for image display range.