Quickstart

To begin our data reduction, we need to load in several FITS files into Python-ready data structures. Using specsuite, you can quickly load your calibration and science exposures into Numpy arrays! Let's do this for some sample data located on the GitHub repository...

In [1]:

import specsuite as ss

# Defines where to look for data
DATA_REGION = (700, 800)
CAL_PATH = "../data/KOSMOS/calibrations"
DATA_PATH = "../data/KOSMOS/target"
INSTRUMENT = "kosmos"

# Loads average of all 'bias' exposures
bias = ss.average_matching_files(
    path = CAL_PATH,
    tag = "bias",
    crop_bds = DATA_REGION,
    instrument = INSTRUMENT,
)

# Loads average of all 'flat' exposures
flat = ss.average_matching_files(
    path = CAL_PATH,
    tag = "flat",
    crop_bds = DATA_REGION,
    instrument = INSTRUMENT,
) - bias

# Loads average of all 'neon' exposures
arclamp = ss.average_matching_files(
    path = CAL_PATH,
    tag = "neon",
    crop_bds = DATA_REGION,
    instrument = INSTRUMENT,
) - bias

# Loads all 'toi3884' exposures as separate arrays
science = ss.collect_images_array(
    DATA_PATH,
    "toi3884",
    crop_bds=DATA_REGION,
    instrument = INSTRUMENT,
) - bias

import specsuite as ss # Defines where to look for data DATA_REGION = (700, 800) CAL_PATH = "../data/KOSMOS/calibrations" DATA_PATH = "../data/KOSMOS/target" INSTRUMENT = "kosmos" # Loads average of all 'bias' exposures bias = ss.average_matching_files( path = CAL_PATH, tag = "bias", crop_bds = DATA_REGION, instrument = INSTRUMENT, ) # Loads average of all 'flat' exposures flat = ss.average_matching_files( path = CAL_PATH, tag = "flat", crop_bds = DATA_REGION, instrument = INSTRUMENT, ) - bias # Loads average of all 'neon' exposures arclamp = ss.average_matching_files( path = CAL_PATH, tag = "neon", crop_bds = DATA_REGION, instrument = INSTRUMENT, ) - bias # Loads all 'toi3884' exposures as separate arrays science = ss.collect_images_array( DATA_PATH, "toi3884", crop_bds=DATA_REGION, instrument = INSTRUMENT, ) - bias

We can use our flatfield exposure to perform a standard calibration on our data!

In [2]:

science = ss.flatfield_correction(
    image = science,
    flat = flat,
    debug = True,
)

science = ss.flatfield_correction( image = science, flat = flat, debug = True, )

Using specsuite's "warp models," we can come up with an approximate model for how light is warped on the detector.

In [3]:

# Identifies the pixel positions of line emissions
locs, _ = ss.find_cal_lines(
    image = arclamp,
    std_variation = 200,
    debug = True,
)

# Models how line emissions are 'bent'
warp_model = ss.generate_warp_model(
    image = arclamp,
    guess = locs,
    debug = True,
)

# Identifies the pixel positions of line emissions locs, _ = ss.find_cal_lines( image = arclamp, std_variation = 200, debug = True, ) # Models how line emissions are 'bent' warp_model = ss.generate_warp_model( image = arclamp, guess = locs, debug = True, )

UserWarning: The figure layout has changed to tight

UserWarning: The figure layout has changed to tight

This "warp model" can then be used to perform a high-quality background extraction!

In [4]:

backgrounds = ss.extract_background(
    images = science,
    warp_model = warp_model,
    mask_region = (40, 80),
    debug = True,
)
calibrated_science = science - backgrounds

backgrounds = ss.extract_background( images = science, warp_model = warp_model, mask_region = (40, 80), debug = True, ) calibrated_science = science - backgrounds

While there are several functions available for performing a flux extraction, we can use a simple boxcar extraction to get a quick look at our data.

In [5]:

flux, error = ss.boxcar_extraction(
    images = calibrated_science,
    backgrounds = backgrounds,
    RN = 6.0,
    debug = True,
)

flux, error = ss.boxcar_extraction( images = calibrated_science, backgrounds = backgrounds, RN = 6.0, debug = True, )

There is more specsuite has to offer, please take a look through the documentation to get an in-depth explanation of each of these steps (+ several additional tools)!