Aligning HST images to an Absolute Reference Catalog

This notebook requires creating and activating a virtual environment using the requirements file in this notebook's repository. Please also review the README file before using the notebook.

Table of Contents

Introduction
Import Packages

1. Download the Observations with astroquery
    1.1 Check image header data
    1.2 Inspect the alignment

2. Query the Reference Catalogs
    2.1 Identify Coordinates
    2.2 SDSS Query
    2.3 Gaia Query

3. Align images with TweakReg
    3.1 SDSS Alignment
    3.2 Inspect the shift file and fit residuals
    3.3 Gaia Alignment
    3.4 Inspect the shift file and fit residuals
    3.5 Overplot matched sources on the image
    3.6 Check the significance of proper motions
    3.7 Update header with optimal parameters

4. Combine the images using AstroDrizzle

Conclusions
About this Notebook

Introduction

The alignment of HST exposures is a critical step in image stacking or combination performed with software such as AstroDrizzle. Generally, a relative alignment is performed to align one or multiple images to another image that is designated as the reference image. The reference image is generally the deepest exposure and/or that covering the largest area of all the exposures. This process aligns the images to each other, but the pointing error of the observatory can still cause the images to have incorrect absolute astrometry. When absolute astrometry is desired, the images can be aligned to an external catalog with an absolute world coordinate system (WCS). In this example, we will provide a workflow to query catalogs such as SDSS and Gaia using the astroquery package, and then align the images to that catalog using TweakReg.

The workflow in this notebook for aligning images to Gaia is based on WFC3 ISR 2017-19: Aligning HST Images to Gaia: a Faster Mosaicking Workflow and contains a subset of the information and code found in the gaia_alignment repository. For more information, see the notebook in that repository titled Gaia_alignment.ipynb.

For more information about TweakReg, see the other notebooks in this repository or the TweakReg Documentation.

For more information on Astroquery, see the other notebooks in this repository or the Astroquery Documentation.

Import Packages

Table of Contents

---

The following Python packages are required to run the Jupyter Notebook:

• os - change and make directories

• glob - gather lists of filenames

• shutil - remove directories and files

• numpy - math and array functions

• matplotlib - make figures and graphics

• astropy - file handling, tables, units, WCS, statistics

• astroquery - download data and query databases

• drizzlepac - align and combine HST images

import os
import glob
import shutil
import numpy as np
import matplotlib.pyplot as plt
from collections import defaultdict
from IPython.display import Image, clear_output
import astropy.units as u
from astropy.io import ascii, fits
from astropy.table import Table
from astropy.units import Quantity
from astropy.time import Time
from astropy.coordinates import SkyCoord
from astropy.visualization import ZScaleInterval
from astroquery.mast import Observations
from astroquery.sdss import SDSS
from astroquery.gaia import Gaia
from drizzlepac import tweakreg, astrodrizzle
from drizzlepac.processInput import getMdriztabPars

%config InlineBackend.figure_format = 'retina' # Improves the resolution of figures rendered in notebooks.
Gaia.MAIN_GAIA_TABLE = 'gaiadr3.gaia_source' # Change this to the desired Gaia data release.
Gaia.ROW_LIMIT = 100000 # Show all of the matched sources in the printed tables.

# Set the locations of reference files and retrieve the MDRIZTAB recommended drizzle parameters.
os.environ['CRDS_SERVER_URL'] = 'https://hst-crds.stsci.edu'
os.environ['CRDS_SERVER'] = 'https://hst-crds.stsci.edu'
os.environ['CRDS_PATH'] = './crds_cache'
os.environ['iref'] = './crds_cache/references/hst/wfc3/'

1. Download the Observations with astroquery

Table of Contents

For this notebook, we will download WFC3/UVIS images of NGC 6791 from program 12692 in the F606W filter acquired in Visit 01. The three FLC images have been processed by the HST WFC3 pipeline (calwf3), which includes bias subtraction, dark current and CTE correction, cosmic-ray rejection, and flat-fielding:

        ibwb01xqq_flc.fits
        ibwb01xrq_flc.fits
        ibwb01xxq_flc.fits

---

MAST queries may be done using query_criteria, where we specify:

    \(\rightarrow\) obs_id, proposal_id, and filters

MAST data products may be downloaded by using download_products, where we specify:

    \(\rightarrow\) products = calibrated (FLT, FLC) or drizzled (DRZ, DRC) files

    \(\rightarrow\) type = standard products (CALxxx) or advanced products (HAP-SVM)

Depending on your connection speed this cell may take a few minutes to execute.

obs_ids = ['ibwb01*']
props = ['12692']
filts = ['F606W']

obsTable = Observations.query_criteria(obs_id=obs_ids, proposal_id=props, filters=filts)
products = Observations.get_product_list(obsTable)

data_prod = ['FLC']    # ['FLC','FLT','DRC','DRZ']                  
data_type = ['CALWF3'] # ['CALACS','CALWF3','CALWP2','HAP-SVM']    

Observations.download_products(products, productSubGroupDescription=data_prod, project=data_type, cache=True)

Downloading URL https://mast.stsci.edu/api/v0.1/Download/file?uri=mast:HST/product/ibwb01xqq_flc.fits to ./mastDownload/HST/ibwb01xqq/ibwb01xqq_flc.fits ...

 [Done]

Downloading URL https://mast.stsci.edu/api/v0.1/Download/file?uri=mast:HST/product/ibwb01xrq_flc.fits to ./mastDownload/HST/ibwb01xrq/ibwb01xrq_flc.fits ...

 [Done]

Downloading URL https://mast.stsci.edu/api/v0.1/Download/file?uri=mast:HST/product/ibwb01xxq_flc.fits to ./mastDownload/HST/ibwb01xxq/ibwb01xxq_flc.fits ...

 [Done]

Table length=3

Local Path	Status	Message	URL
str47	str8	object	object
./mastDownload/HST/ibwb01xqq/ibwb01xqq_flc.fits	COMPLETE	None	None
./mastDownload/HST/ibwb01xrq/ibwb01xrq_flc.fits	COMPLETE	None	None
./mastDownload/HST/ibwb01xxq/ibwb01xxq_flc.fits	COMPLETE	None	None

Move to the files to the local working directory:

for flc in glob.glob('./mastDownload/HST/*/*flc.fits'):
    flc_name = os.path.basename(flc)
    os.rename(flc, flc_name)
    
if os.path.exists('mastDownload'):
    shutil.rmtree('mastDownload')

1.1 Check image header data

The cell below retrieves values for specific keywords from the first and second extensions of the image header. We see that the 1st exposure is 30 seconds and the 2nd and 3rd exposures are 360 seconds in duration. The 3rd exposure is dithered by approximately 82 arcseconds in the y-direction which is approximately the width of one WFC3 UVIS detector chip. A full list of header keywords is available in Section 2.4 of the WFC3 Data Handbook.

paths = sorted(glob.glob('*flc.fits'))
data = []
keywords_ext0 = ["ROOTNAME", "ASN_ID", "TARGNAME", "DETECTOR", "FILTER", "exptime", 
                 "ra_targ", "dec_targ", "postarg1", "postarg2", "DATE-OBS"]
keywords_ext1 = ["orientat"]

for path in paths:
    path_data = []
    for keyword in keywords_ext0:
        path_data.append(fits.getval(path, keyword, ext=0))
    for keyword in keywords_ext1:
        path_data.append(fits.getval(path, keyword, ext=1))
    data.append(path_data)
    
keywords = keywords_ext0 + keywords_ext1
table = Table(np.array(data), names=keywords, dtype=['str', 'str', 'str', 'str', 'str', 'f8', 'f8', 'f8', 'f8', 'f8', 'str', 'f8'])
table['exptime'].format = '7.1f'
table['ra_targ'].format = table['dec_targ'].format = '7.4f'
table['postarg1'].format = table['postarg2'].format = '7.3f'
table['orientat'].format = '7.2f'
table

Table length=3

ROOTNAME	ASN_ID	TARGNAME	DETECTOR	FILTER	exptime	ra_targ	dec_targ	postarg1	postarg2	DATE-OBS	orientat
str32	str32	str32	str32	str32	float64	float64	float64	float64	float64	str32	float64
ibwb01xqq	NONE	NGC-6791	UVIS	F606W	30.0	290.2248	37.8027	0.000	0.000	2011-10-20	-56.27
ibwb01xrq	NONE	NGC-6791	UVIS	F606W	360.0	290.2248	37.8027	0.000	0.000	2011-10-20	-56.27
ibwb01xxq	NONE	NGC-6791	UVIS	F606W	360.0	290.2248	37.8027	0.000	81.600	2011-10-20	-56.25

1.2 Inspect the Alignment

Check the active WCS solution in the image header. If the image is aligned to a catalog, then list the number of matches and the fit RMS in milli-arcseconds. Next, convert the fit RMS values to pixels for comparison with the alignment results performed later in this notebook using Tweakreg.

ext_0_keywords = ['DETECTOR', 'EXPTIME'] # extension 0 keywords.
ext_1_keywords = ['WCSNAME', 'NMATCHES', 'RMS_RA', 'RMS_DEC'] # extension 1 keywords.

# Define the detector plate scales in arcsec per pixel.
DETECTOR_SCALES = {
  'IR': 0.1283, 
  'UVIS': 0.0396, 
  'WFC': 0.05
}

formatted_data = {}
column_data = defaultdict(list)

for fits_file in sorted(glob.glob('*flc.fits')):
    column_data['filename'].append(fits_file)
    header0 = fits.getheader(fits_file, 0)
    header1 = fits.getheader(fits_file, 1)
    
    for keyword in ext_0_keywords:
        column_data[keyword].append(header0[keyword])
    for keyword in ext_1_keywords:
        if 'RMS' in keyword:
            value = np.around(header1[keyword], decimals=1)
        else:
            value = header1[keyword]
        column_data[keyword].append(value)
            
    for keyword in ['RMS_RA', 'RMS_DEC']:
        rms_value = header1[keyword] / 1000 / DETECTOR_SCALES[header0['DETECTOR']]
        column_data[f'{keyword}_pix'].append(np.round(rms_value, decimals=2))

wcstable = Table(column_data)
wcstable

Table length=3

filename	DETECTOR	EXPTIME	WCSNAME	NMATCHES	RMS_RA	RMS_DEC	RMS_RA_pix	RMS_DEC_pix
str18	str4	float64	str30	int64	float64	float64	float64	float64
ibwb01xqq_flc.fits	UVIS	30.0	IDC_2731450pi-FIT_IMG_GAIAeDR3	468	1.4	1.5	0.04	0.04
ibwb01xrq_flc.fits	UVIS	360.0	IDC_2731450pi-FIT_IMG_GAIAeDR3	475	3.4	3.0	0.09	0.08
ibwb01xxq_flc.fits	UVIS	360.0	IDC_2731450pi-FIT_IMG_GAIAeDR3	426	3.4	3.5	0.09	0.09

2. Query the Reference Catalogs

Table of Contents

Now that we have the FITS images, we will download the reference catalogs from both SDSS and Gaia using astroquery.

2.1 Identify Coordinates

We will first create a SkyCoord Object to point astroquery to where we are looking on the sky. Since our example uses data from NGC 6791, we will use the ra_targ and dec_targ keywords from the first image to get the coordinates of the object. We then use these RA and Dec values to an construct an astropy SkyCoord object, which we will later pass to the SDSS query. Additionally, we use an astropy Quantity object to create a radius for the SDSS query in physical units. For reference, the UVIS detector field of view is approximately 2.7’\(\times\)2.7’ and a y-dither of 82 arcseconds covers a total area on the sky of approximately 2.7’\(\times\)4.1’. For SDSS the maximum search radius is 3 arcminutes, which covers the majority of our field of view in this example. In the Gaia query, we set the radius to 5 arcminutes to fully cover the area of our observations.

RA = table['ra_targ'][0]
Dec = table['dec_targ'][0]

coord = SkyCoord(ra=RA, dec=Dec, unit=(u.deg, u.deg))

2.2 SDSS Query

Next, we perform the query via the SDSS.query_region method of astroquery.sdss. The spectro=False keyword argument means we want to exclude spectroscopic objects, as we are looking for objects to match with an image. In the fields parameter, we specify a list of fields we want returned by the query. In this case we only need the position, as well as a g-band magnitude if we want to remove very dim and/or bright objects from the catalog, as those are likely measured poorly. Details on selecting objects by magnitude may be found in the Gaia_alignment.ipynb. Many other fields are available in the SDSS.query_region function, which are documented on this webpage.

SDSS.clear_cache()
radius = Quantity(3., u.arcmin)

# If the query fails due to connection issues, proceed with the catalog included in the repository.
try:
    # raise ValueError('Using offline mode.') # Allows users to work offline.
    sdss_query = SDSS.query_region(coord, radius=radius, spectro=False, fields=['ra', 'dec', 'g'])
    print('The query passed. Proceeding with the SDSS query catalog.')
except Exception:
    print('The query failed. Proceeding with the repository catalog.')
    sdss_query = Table.read('sdss.ecsv', format='ascii.ecsv')

# Handle the case where 'objID' is included as the first column by some versions of astroquery.
try:
    sdss_query.remove_column('objID')
except KeyError:
    print('The table does not contain an objID column... proceeding.')

# sdss_query.write('sdss.ecsv', format='ascii.ecsv', overwrite=False) # Used to create a static catalog.
    
sdss_query

The query passed. Proceeding with the SDSS query catalog.

Table length=2148

ra	dec	g
float64	float64	float64
290.203062469049	37.8284558064513	18.85147
290.202835080385	37.827934351748	-9999.0
290.202824930811	37.8279347406866	18.84776
290.217936958777	37.8312822178273	23.9199
290.210249524598	37.832152102827	17.15361
290.209206019376	37.835720854937	-9999.0
290.209215663888	37.8357188564505	-9999.0
290.209232479979	37.8357102695015	18.01884
290.210223106997	37.8343053865869	17.76188
...	...	...
290.219187514438	37.7561787854031	18.63054
290.224721509389	37.7546515877439	18.48179
290.225512938821	37.754014018543	18.63468
290.225471553079	37.7540325645363	18.3785
290.221998493145	37.7551307981833	16.79066
290.224692002974	37.7546710154955	18.49708
290.220817707434	37.7532424875039	18.05285
290.2215207565	37.7538111894451	17.45435
290.221586956686	37.7537172867002	18.62096
290.222789411519	37.7534535032111	18.85183

We then need to save the catalog to use with TweakReg, which by default expects that the first two columns are RA and Dec. The returned value of the query is an astropy.Table, so saving the file is straightforward:

sdss_query.write('sdss.cat', format='ascii.commented_header', overwrite=True)

2.3 Gaia Query

Similar to SDSS, we can query Gaia catalogs for our target via astroquery.gaia. This may be preferable in many cases because the spaced-based astrometry from Gaia is generally very accurate and precise. We can use the same coord search parameters from our earlier SDSS query, along with a slightly larger radius that encompasses the full field of view of these observations.

radius = Quantity(5., u.arcmin)

# If the query fails due to connection issues, proceed with the catalog included in the repository.
try:
    # raise ValueError('Using offline mode.') # Allows users to work offline.
    gaia_query = Gaia.query_object_async(coordinate=coord, radius=radius)
    print('The query passed. Proceeding with the Gaia query catalog.')
except Exception:
    print('The query failed. Proceeding with the repository catalog.')
    gaia_query = Table.read('gaia.ecsv', format='ascii.ecsv')

# gaia_query.write('gaia.ecsv', format='ascii.ecsv', overwrite=False) # Used to create a static catalog.
    
gaia_query

INFO: Query finished. [astroquery.utils.tap.core]
The query passed. Proceeding with the Gaia query catalog.

Table length=4865

solution_id	designation	source_id	random_index	ref_epoch	ra	ra_error	dec	dec_error	parallax	parallax_error	parallax_over_error	pm	pmra	pmra_error	pmdec	pmdec_error	ra_dec_corr	ra_parallax_corr	ra_pmra_corr	ra_pmdec_corr	dec_parallax_corr	dec_pmra_corr	dec_pmdec_corr	parallax_pmra_corr	parallax_pmdec_corr	pmra_pmdec_corr	astrometric_n_obs_al	astrometric_n_obs_ac	astrometric_n_good_obs_al	astrometric_n_bad_obs_al	astrometric_gof_al	astrometric_chi2_al	astrometric_excess_noise	astrometric_excess_noise_sig	astrometric_params_solved	astrometric_primary_flag	nu_eff_used_in_astrometry	pseudocolour	pseudocolour_error	ra_pseudocolour_corr	dec_pseudocolour_corr	parallax_pseudocolour_corr	pmra_pseudocolour_corr	pmdec_pseudocolour_corr	astrometric_matched_transits	visibility_periods_used	astrometric_sigma5d_max	matched_transits	new_matched_transits	matched_transits_removed	ipd_gof_harmonic_amplitude	ipd_gof_harmonic_phase	ipd_frac_multi_peak	ipd_frac_odd_win	ruwe	scan_direction_strength_k1	scan_direction_strength_k2	scan_direction_strength_k3	scan_direction_strength_k4	scan_direction_mean_k1	scan_direction_mean_k2	scan_direction_mean_k3	scan_direction_mean_k4	duplicated_source	phot_g_n_obs	phot_g_mean_flux	phot_g_mean_flux_error	phot_g_mean_flux_over_error	phot_g_mean_mag	phot_bp_n_obs	phot_bp_mean_flux	phot_bp_mean_flux_error	phot_bp_mean_flux_over_error	phot_bp_mean_mag	phot_rp_n_obs	phot_rp_mean_flux	phot_rp_mean_flux_error	phot_rp_mean_flux_over_error	phot_rp_mean_mag	phot_bp_rp_excess_factor	phot_bp_n_contaminated_transits	phot_bp_n_blended_transits	phot_rp_n_contaminated_transits	phot_rp_n_blended_transits	phot_proc_mode	bp_rp	bp_g	g_rp	radial_velocity	radial_velocity_error	rv_method_used	rv_nb_transits	rv_nb_deblended_transits	rv_visibility_periods_used	rv_expected_sig_to_noise	rv_renormalised_gof	rv_chisq_pvalue	rv_time_duration	rv_amplitude_robust	rv_template_teff	rv_template_logg	rv_template_fe_h	rv_atm_param_origin	vbroad	vbroad_error	vbroad_nb_transits	grvs_mag	grvs_mag_error	grvs_mag_nb_transits	rvs_spec_sig_to_noise	phot_variable_flag	l	b	ecl_lon	ecl_lat	in_qso_candidates	in_galaxy_candidates	non_single_star	has_xp_continuous	has_xp_sampled	has_rvs	has_epoch_photometry	has_epoch_rv	has_mcmc_gspphot	has_mcmc_msc	in_andromeda_survey	classprob_dsc_combmod_quasar	classprob_dsc_combmod_galaxy	classprob_dsc_combmod_star	teff_gspphot	teff_gspphot_lower	teff_gspphot_upper	logg_gspphot	logg_gspphot_lower	logg_gspphot_upper	mh_gspphot	mh_gspphot_lower	mh_gspphot_upper	distance_gspphot	distance_gspphot_lower	distance_gspphot_upper	azero_gspphot	azero_gspphot_lower	azero_gspphot_upper	ag_gspphot	ag_gspphot_lower	ag_gspphot_upper	ebpminrp_gspphot	ebpminrp_gspphot_lower	ebpminrp_gspphot_upper	libname_gspphot	dist
				yr	deg	mas	deg	mas	mas	mas		mas / yr	mas / yr	mas / yr	mas / yr	mas / yr																	mas				1 / um	1 / um	1 / um								mas					deg								deg	deg	deg	deg			electron / s	electron / s		mag		electron / s	electron / s		mag		electron / s	electron / s		mag							mag	mag	mag	km / s	km / s								d	km / s	K	log(cm.s**-2)	dex		km / s	km / s		mag	mag				deg	deg	deg	deg															K	K	K	log(cm.s**-2)	log(cm.s**-2)	log(cm.s**-2)	dex	dex	dex	pc	pc	pc	mag	mag	mag	mag	mag	mag	mag	mag	mag
int64	object	int64	int64	float64	float64	float32	float64	float32	float64	float32	float32	float32	float64	float32	float64	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	int16	int16	int16	int16	float32	float32	float32	float32	int16	bool	float32	float32	float32	float32	float32	float32	float32	float32	int16	int16	float32	int16	int16	int16	float32	float32	int16	int16	float32	float32	float32	float32	float32	float32	float32	float32	float32	bool	int16	float64	float32	float32	float32	int16	float64	float32	float32	float32	int16	float64	float32	float32	float32	float32	int16	int16	int16	int16	int16	float32	float32	float32	float32	float32	int16	int16	int16	int16	float32	float32	float32	float32	float32	float32	float32	float32	int16	float32	float32	int16	float32	float32	int16	float32	object	float64	float64	float64	float64	bool	bool	int16	bool	bool	bool	bool	bool	bool	bool	bool	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	float32	object	float64
1636148068921376768	Gaia DR3 2051293319932402944	2051293319932402944	1710414812	2016.0	290.2248611356283	0.12343551	37.80192950361359	0.15534069	1.2409154633064086	0.15976872	7.7669487	6.0644784	-0.6226543407618474	0.15474372	6.0324290792879545	0.18924683	-0.033001468	0.16417459	-0.027278472	-0.020170886	0.07942047	-0.046242356	-0.19703287	-0.07685185	0.0051713893	0.050814763	394	0	389	5	0.407836	413.48303	0.0	2.0511602e-15	31	False	1.3016587	--	--	--	--	--	--	--	45	26	0.26493552	49	22	0	0.01895016	14.319122	0	0	1.0138804	0.17569481	0.20240329	0.10288032	0.089386284	-95.13182	77.09379	2.6807811	-30.180996	False	385	492.16788454743386	0.9903239	496.97668	18.957083	39	106.26453488321661	5.3855076	19.731573	20.272572	39	549.1994295106254	7.0852404	77.51317	17.898571	1.3317894	0	2	0	2	0	2.3740005	1.3154888	1.0585117	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.98761891051129	10.914199739506339	302.04196574777654	59.01190915812673	False	False	0	False	False	False	False	False	True	False	False	1.4003378e-13	5.356044e-13	0.99996775	3653.592	3591.5752	3744.3992	4.8552	4.4992	4.9965	-0.4967	-0.7199	-0.1315	653.0443	491.9481	1283.5112	0.7812	0.6499	0.9202	0.5305	0.44	0.6266	0.3102	0.2581	0.3649	MARCS	0.0007449136479797354
1636148068921376768	Gaia DR3 2051293324218263808	2051293324218263808	214974978	2016.0	290.22327972754084	1.0491472	37.80283944795903	1.4068799	-0.6069077460064267	1.4172788	-0.42822045	3.8347042	-0.49694994391852737	1.433404	-3.802367186693702	2.183717	-0.32559526	0.44083008	0.040013984	-0.36232764	-0.18673699	-0.29855755	0.5145219	0.16053891	-0.3253903	-0.32209665	133	0	132	1	2.3624072	187.53317	3.4946198	1.7356544	95	False	--	1.2583821	0.3532982	-0.10796609	-0.06725511	-0.030691369	-0.22952616	-0.12757729	16	14	3.2688706	18	6	1	0.061227437	43.226418	5	17	1.1496774	0.23548113	0.3243092	0.35748923	0.4221966	-95.12672	74.99657	-8.526287	-16.614477	False	132	88.7375766294378	1.1992922	73.99162	20.817099	0	--	--	--	--	0	--	--	--	--	--	--	--	--	--	0	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.9878943903652	10.915721621139165	302.0400990046131	59.01311977354264	False	False	0	False	False	False	False	False	False	False	False	7.759076e-08	3.7895185e-09	0.9999943	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.0012108975043825237
1636148068921376768	Gaia DR3 2051293324222972288	2051293324222972288	164456380	2016.0	290.22261994880597	0.032841645	37.80270752474709	0.039459	0.3053094896365258	0.041127697	7.423452	2.5961652	-0.7012623275324616	0.041221105	-2.4996610433857405	0.047550496	-0.047555298	0.16546503	0.021742897	-0.077867456	0.066915035	-0.08765522	-0.26260835	-0.059347805	-0.087691456	-0.0013310307	383	0	381	2	-2.1703348	333.38327	0.0	0.0	31	False	1.4413656	--	--	--	--	--	--	--	44	25	0.06704042	48	21	0	0.015037518	12.04822	0	0	0.9210467	0.13890518	0.18901813	0.10900513	0.0641752	-88.36165	72.302956	5.923747	-33.787945	False	381	4179.254659077206	2.2517574	1855.9968	16.63462	43	1717.7830387470858	6.2028065	276.93643	17.251122	39	3494.1204079650242	10.008291	349.1226	15.889551	1.2470893	0	25	0	23	0	1.3615704	0.61650085	0.7450695	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.98754100993882	10.916132432650413	302.0390547900942	59.01313187059839	False	False	0	True	False	False	False	False	True	True	False	1.0228448e-13	9.3310026e-11	0.9999741	4770.3447	4749.6045	4790.818	4.1776	4.1596	4.1902	0.3546	0.3268	0.3805	2185.9316	2147.3625	2239.1926	0.3511	0.3259	0.3735	0.2701	0.2505	0.2875	0.1406	0.1303	0.1496	PHOENIX	0.001721026645501702
1636148068921376768	Gaia DR3 2051293324222972416	2051293324222972416	1599905588	2016.0	290.2248309701555	0.052941296	37.80058611405895	0.064896055	0.21837043824644525	0.067057766	3.2564526	2.3448853	-0.5827200539782851	0.066188715	-2.2713266001611636	0.0787358	-0.076830104	0.14136231	-0.025055349	-0.07297541	0.056257915	-0.07904705	-0.24559797	-0.07446872	-0.087429665	-0.03352803	395	0	392	3	-0.037620842	401.51434	0.0	0.0	31	False	1.4891585	--	--	--	--	--	--	--	45	25	0.110738866	49	22	0	0.0153436065	15.8542	0	0	0.99778724	0.15293275	0.21477312	0.12928022	0.08397196	-94.649994	70.08022	6.0968003	-29.722376	False	389	1751.0366076397277	1.4950225	1171.2443	17.579128	46	817.0065234341104	6.4622293	126.42797	18.057978	44	1347.155230450357	8.58573	156.90633	16.92435	1.2359318	0	32	0	29	0	1.1336269	0.4788494	0.6547775	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.98637845348246	10.913632562752337	302.0412242635996	59.01062093567249	False	False	0	True	False	False	False	False	True	True	False	6.110852e-11	5.145215e-13	0.9999132	4831.901	4812.959	4859.14	4.661	4.6412	4.6768	-0.4447	-0.5218	-0.3786	1595.5184	1545.6223	1641.4886	0.0087	0.0021	0.021	0.0069	0.0017	0.0166	0.0036	0.0009	0.0088	PHOENIX	0.0020867793124382814
1636148068921376768	Gaia DR3 2051293319934791936	2051293319934791936	127740795	2016.0	290.2225602638809	0.098567784	37.80397847004457	0.12238157	0.18608923760796978	0.12619987	1.4745597	2.3556678	-0.20591766034379622	0.12384657	-2.346650638048337	0.14836667	-0.0034372064	0.17569284	-0.028029243	-0.023688797	0.0831415	-0.04713238	-0.21456409	-0.08000817	-0.019006517	0.07798844	386	0	383	3	0.34466058	403.29874	0.0	0.0	31	False	1.4745288	--	--	--	--	--	--	--	44	26	0.20897423	48	21	0	0.02525984	16.266506	0	0	1.011676	0.16451268	0.180577	0.079406776	0.08114059	-89.370926	80.65579	0.7836158	-30.156015	False	377	665.1667185914472	1.124127	591.71844	18.630041	40	296.01604570644196	5.8455243	50.63978	19.160254	41	517.3148686523093	6.1276155	84.42352	17.963509	1.2227474	0	0	0	0	0	1.1967449	0.5302124	0.6665325	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.98868349525281	10.916731652994313	302.03962583571473	59.014369244472704	False	False	0	False	False	False	False	False	True	False	False	1.0210282e-13	5.259736e-13	0.99999475	4590.2783	4575.766	4603.0244	4.8255	4.7981	4.858	-1.2317	-1.4162	-1.1009	1664.2876	1569.1196	1730.915	0.0102	0.0026	0.0257	0.008	0.002	0.02	0.0043	0.0011	0.0107	PHOENIX	0.0021977590777958536
1636148068921376768	Gaia DR3 2051293324216543232	2051293324216543232	496974628	2016.0	290.22681959687196	0.5800434	37.80112772770517	0.8518114	0.9866119688573637	0.8487393	1.162444	4.5651155	-0.44243766262405015	1.042491	-4.543625170019005	1.2715639	-0.03034182	-0.13035429	0.09905615	0.024711896	-0.035745695	-0.13054979	0.34666777	-0.069062054	-0.28526154	-0.1078611	189	0	189	0	-0.84486425	177.7449	0.0	1.4856265e-15	95	False	--	1.5521348	0.21582161	-0.0062133768	0.2804249	0.035700142	-0.42101106	0.17818592	22	15	1.8150634	25	8	0	0.023252329	30.515776	0	0	0.954371	0.20877054	0.28841907	0.1927084	0.43569437	-76.9166	84.362335	-3.7503886	-19.752798	False	189	100.17345952653844	0.81711596	122.59393	20.685486	16	34.777743067191494	8.345614	4.1671877	21.485289	19	89.51229139692614	5.5230875	16.20693	19.868189	1.2407482	0	0	0	0	0	1.6170998	0.7998028	0.817297	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.98757543345182	10.912457430119012	302.0444461542871	59.01072318960171	False	False	0	False	False	False	False	False	False	False	False	1.5372552e-09	1.1275478e-09	0.99999505	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.002222460820739603
1636148068921376768	Gaia DR3 2051293324222974464	2051293324222974464	1798436105	2016.0	290.2275249308142	0.24079925	37.803472775139035	0.3074705	0.44506652441996963	0.3134235	1.4200164	1.903844	-0.7741499247735619	0.3071843	-1.7393429825276292	0.37454808	0.020487515	0.17073557	-0.036337543	0.041650914	0.12431544	-0.0061035734	-0.18864702	-0.06549308	0.060529068	0.123350315	382	0	380	2	0.5945536	412.03198	0.9015461	0.9578774	95	False	--	1.3147619	0.081987195	-0.016766502	0.118190974	0.037860535	-0.0898746	-0.10861043	44	25	0.52942884	48	21	0	0.027762208	20.245789	0	4	1.02092	0.18400808	0.1789637	0.073698275	0.10250232	-90.673706	85.22672	-9.854181	-30.444962	False	380	207.20873617672416	0.7419642	279.2705	19.896347	9	91.39950015947561	7.5014834	12.18419	20.436182	9	224.20190021064204	11.064253	20.263628	18.871298	1.5231086	0	0	0	0	0	1.5648842	0.539835	1.0250492	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.98997089407423	10.912983986333725	302.04670590522085	59.01283416072846	False	False	0	False	False	False	False	False	False	False	False	1.8206678e-09	2.9034798e-08	0.99998796	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.002298594469020544
1636148068921376768	Gaia DR3 2051293324216539648	2051293324216539648	1649457609	2016.0	290.2242320409821	0.5271274	37.80039510739463	0.76655334	1.3632756584241434	0.6728546	2.0261073	2.2253373	-0.9708645882906146	0.6578662	-2.0023856302519407	0.98862535	-0.33465353	0.08556003	-0.2735162	0.2602703	0.161614	0.21899539	-0.26542807	0.033500012	0.1458301	-0.23663591	243	0	243	0	1.2024069	273.66382	0.0	0.0	95	False	--	1.6818463	0.19114311	-0.26075235	0.11348202	-0.08333635	-0.06321393	-0.15212661	28	18	1.4304484	32	14	1	0.0707902	34.375954	0	15	1.0543019	0.24560685	0.4438982	0.22968337	0.165783	-110.25776	67.43046	5.994361	-23.29821	False	244	109.8552156027335	0.8030061	136.80496	20.585316	0	--	--	--	--	0	--	--	--	--	--	--	--	--	--	2	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.98599242480627	10.91397427439119	302.0402394379523	59.01056325889003	False	False	0	False	False	False	False	False	False	False	False	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.0023210548670645637
1636148068921376768	Gaia DR3 2051293324225336064	2051293324225336064	398090960	2016.0	290.22302939666463	0.47650018	37.8003404115494	0.57163584	1.0465539617532613	0.60573983	1.7277284	0.3426481	0.016525679687419553	0.6219764	-0.34224935447443744	0.7205138	-0.11358616	0.1529165	0.07183364	-0.104466826	0.054135017	-0.07860708	-0.1758327	0.10776901	-0.16631071	-0.16472225	373	0	365	8	86.80703	12487.478	6.598426	424.3264	95	False	--	1.2976544	0.14454299	-0.057665557	0.052674405	0.033202734	0.010857275	-0.07922697	43	25	1.0361372	50	50	0	0.19814193	21.407808	63	0	5.6149945	0.1693135	0.21580335	0.10183408	0.09376588	-92.04377	70.39044	0.82470447	-23.783632	False	298	836.206629484886	5.043918	165.78514	18.381582	43	492.85805036589596	5.0838623	96.945595	18.606737	41	916.8429900655037	6.7629867	135.56776	17.342157	1.6858286	0	2	0	2	0	1.2645798	0.22515488	1.0394249	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.98551833187275	10.914804515213548	302.0384324817961	59.01076433315906	False	False	0	False	False	False	False	False	False	False	False	9.514203e-11	1.056036e-05	0.99998164	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.0027187891427339723
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1636148068921376768	Gaia DR3 2051294870411535616	2051294870411535616	430955701	2016.0	290.1710542799103	0.089351624	37.874210138824154	0.10865699	0.3694586266163912	0.113834515	3.2455766	9.114665	-6.31871143886455	0.11210035	-6.568942750936787	0.13436429	-0.007841255	0.11561305	-0.01169015	-0.03635325	0.11398082	-0.05428327	-0.21885028	-0.10007106	-0.08220788	0.061571144	399	0	392	7	0.13260055	406.64862	0.22923899	0.41862753	31	False	1.4996165	--	--	--	--	--	--	--	46	27	0.18900031	49	22	0	0.011167029	1.7622564	0	0	1.0039074	0.26786533	0.10760525	0.10404573	0.08249484	-87.45818	75.753624	-15.217908	-38.780235	False	384	779.8054003243237	1.0609461	735.00946	18.457401	46	362.8243922628937	6.342421	57.20598	18.9393	45	563.964440036763	5.256008	107.299	17.869766	1.1884873	0	0	0	0	0	1.0695343	0.48189926	0.58763504	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	70.03485933670954	10.984050062538527	301.9998719659017	59.09291303817852	False	False	0	False	False	False	False	False	True	False	False	3.5958955e-11	5.3623333e-13	0.9999949	4889.5947	4878.3716	4901.4575	4.8392	4.8272	4.8515	-1.6225	-1.7671	-1.5237	1828.6827	1778.3691	1860.6678	0.0045	0.0011	0.0108	0.0036	0.0009	0.0086	0.002	0.0005	0.0047	MARCS	0.0831808340919906
1636148068921376768	Gaia DR3 2051288234680628736	2051288234680628736	275359989	2016.0	290.1228125225418	0.12832773	37.7819163631625	0.16770653	0.45711970927190115	0.16442272	2.7801495	15.87947	-6.491827713462847	0.16428103	-14.491851015938131	0.21764661	0.002764468	0.12523003	-0.10028199	0.050657623	0.06666436	0.005579139	-0.37272593	-0.118829906	0.0050995317	0.1139823	409	0	409	0	-1.3663982	383.07782	0.0	1.9997444e-15	31	False	1.3034092	--	--	--	--	--	--	--	47	27	0.3113583	48	23	0	0.010293381	2.4593005	0	0	0.95150584	0.17092009	0.1387889	0.05828934	0.13473459	-116.178055	83.69419	2.6145937	-36.532135	False	407	450.44352750937213	0.8725948	516.2116	19.053267	45	103.68771152397954	5.136524	20.186357	20.299225	45	502.45009105705196	3.6524675	137.56456	17.995163	1.3456466	0	1	0	0	0	2.304062	1.2459583	1.0581036	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.9333371194973	10.977945740748442	301.88063476010194	59.014107067896354	False	False	0	False	False	False	False	False	False	False	False	3.8000676e-13	7.05338e-13	0.9998841	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.08322243598195858
1636148068921376768	Gaia DR3 2051294870411542272	2051294870411542272	1651247147	2016.0	290.1750101665401	0.06694008	37.87602842402639	0.092485614	0.20603759179502212	0.08579865	2.4014082	2.2515461	-0.3932337230718288	0.08297268	-2.216941094221096	0.118552975	-0.06297449	0.14318395	-0.033632368	0.03577277	-0.034115147	0.00034698079	-0.4515389	-0.10937831	0.07231181	0.009728736	377	0	374	3	0.06470177	384.27502	0.0	0.0	95	False	--	1.4371054	0.023171378	-0.16051838	0.025070293	-0.010794556	-0.07607864	0.028111283	43	26	0.17073497	48	22	1	0.01102003	161.05574	0	15	1.0014795	0.2962618	0.18218416	0.14427209	0.11568721	-84.172356	75.77043	-18.850288	42.884228	False	374	1241.344401787872	1.2767242	972.2886	17.952637	0	--	--	--	--	0	--	--	--	--	--	--	--	--	--	2	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	70.0379165005182	10.982037682530848	302.00667588191897	59.093831343557795	False	False	0	False	False	False	False	False	False	False	False	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.08322879189943044
1636148068921376768	Gaia DR3 2051295381516623744	2051295381516623744	1667872848	2016.0	290.2882629014902	0.14793906	37.86913015536792	0.21439508	0.6608720254317446	0.1987435	3.3252509	8.816373	-2.6884990926127106	0.19940175	-8.39645178415166	0.26135257	-0.12174883	0.10869827	0.06210663	0.0064050634	0.07006799	-0.034789	-0.36298344	0.027279017	-0.00647773	-0.083680145	339	0	338	1	0.5362549	369.1253	0.41999847	0.5260669	95	False	--	1.3389693	0.05077461	-0.087569006	0.18131016	-0.04555531	-0.13339908	-0.11010846	39	23	0.3739153	42	20	0	0.029168483	178.31316	0	0	1.019872	0.2148467	0.20623815	0.12674522	0.11732818	-75.89927	79.33952	-25.199	-39.338097	False	346	420.7538148013615	1.0356286	406.27866	19.127296	39	140.64232775208788	7.4752154	18.814486	19.968252	39	478.4602734701604	9.109383	52.523895	18.04828	1.4714129	0	22	0	22	0	1.9199715	0.84095573	1.0790157	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	70.0714830724915	10.89864899377486	302.1707078793191	59.063253212461184	False	False	0	False	False	False	False	False	False	False	False	8.597156e-11	5.9048273e-09	0.9999822	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.08324003047068324
1636148068921376768	Gaia DR3 2051288406479389696	2051288406479389696	856711248	2016.0	290.1194354085638	0.1685125	37.80348115009846	0.21949896	0.24010742276483804	0.20961256	1.1454821	6.186268	-2.2508290547655534	0.21421075	-5.762263219114228	0.27950856	-0.04189369	0.055101015	-0.16017503	0.114143774	0.15277705	0.08570732	-0.41045648	-0.013016244	-0.025973871	0.046091143	369	0	369	0	-0.34031537	371.16574	0.0	4.2135146e-15	31	False	1.4944085	--	--	--	--	--	--	--	42	25	0.40078756	43	21	0	0.009520529	19.020708	0	0	0.9865019	0.20752354	0.112608805	0.099987626	0.1787911	-113.13249	83.67482	12.821432	-33.98683	False	367	332.1875262158305	0.86817735	382.62634	19.38391	38	136.82832781506718	7.0689607	19.356216	19.998102	40	250.45730102656773	5.909497	42.38217	18.75106	1.1658645	0	0	0	0	0	1.2470417	0.61419296	0.63284874	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.95190401428546	10.989775662227835	301.88677645351646	59.035605221497256	False	False	0	False	False	False	False	False	False	False	False	5.805757e-10	2.1388297e-10	0.9999928	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.0832529718271768
1636148068921376768	Gaia DR3 2051107128798566528	2051107128798566528	53785428	2016.0	290.33007288300433	0.5063185	37.799005196546766	0.60861003	2.59582303334833	0.72787225	3.5663168	15.07725	-9.90581987549699	0.8522273	-11.36653796175356	0.78330284	-0.037682395	-0.031865645	-0.0061560404	-0.0959621	0.24388646	-0.024079807	-0.085402675	0.35197842	-0.08016355	-0.08673279	257	0	257	0	2.9779897	345.6605	2.6002293	2.6564052	95	False	--	1.3103058	0.1573422	-0.10415734	0.033458136	-0.09948709	-0.2720693	0.04197427	31	22	1.2133433	33	13	1	0.020845002	138.17416	0	0	1.1344311	0.18316215	0.09937623	0.23592351	0.23159805	-121.595795	82.81833	-15.507539	-25.80699	False	253	128.70082874164865	0.86452293	148.86919	20.413414	6	39.717968158819254	8.353689	4.7545424	21.341074	10	145.18349404606658	15.55031	9.336373	19.343102	1.4366765	0	0	0	0	0	1.9979725	0.92766	1.0703125	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	70.02207192846973	10.83820692421412	302.19596551414276	58.98684034920937	False	False	0	False	False	False	False	False	False	False	False	2.8843332e-12	1.4122469e-09	0.99999326	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.08326359931304438
1636148068921376768	Gaia DR3 2051287964103728384	2051287964103728384	1638894158	2016.0	290.14278652187744	0.03494116	37.75037525259858	0.04241879	0.18366243526469542	0.043460846	4.2259283	2.248427	-0.4281612111184363	0.04497176	-2.2072837830215937	0.050941683	0.0351151	0.055712797	-0.14853671	-0.051824972	0.13105968	-0.06969542	-0.3188419	-0.1026289	-0.11638979	0.09372311	428	0	426	2	-0.6057206	420.23416	0.0	0.0	31	False	1.5003259	--	--	--	--	--	--	--	49	28	0.07342281	51	24	0	0.03158952	8.174233	0	0	0.97841215	0.12762302	0.13048583	0.06985256	0.09125381	-109.3513	84.420715	-1.0416744	-31.609655	False	426	3374.6423730218935	1.91779	1759.6516	16.866798	48	1618.891181717306	6.9249954	233.77505	17.315498	49	2463.8001524600472	6.911771	356.4644	16.268883	1.2098145	0	2	0	1	0	1.0466156	0.44869995	0.59791565	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.9114824714278	10.949949126990399	301.8939100069244	58.97950356750648	False	False	0	True	False	False	False	False	True	True	False	1.02206966e-13	5.105279e-13	0.99998176	5064.985	5040.8657	5095.124	4.4424	4.4217	4.4671	-0.2906	-0.3312	-0.2518	1867.0823	1788.1481	1945.3528	0.018	0.0052	0.037	0.0144	0.0042	0.0297	0.0076	0.0022	0.0157	PHOENIX	0.08328823877757062
1636148068921376768	Gaia DR3 2051286933311580032	2051286933311580032	135441617	2016.0	290.17127999981165	0.04414948	37.73091082720331	0.05403776	0.2357502713121104	0.056459904	4.1755347	2.2787905	-0.29109566200012765	0.055603746	-2.260121567186946	0.06337706	-0.017252762	-0.013258554	-0.13981561	-0.051438995	0.18330099	-0.08995912	-0.30687386	-0.12603593	-0.08434859	0.07638833	414	0	412	2	0.1777575	428.2643	0.0	0.0	31	False	1.5031	--	--	--	--	--	--	--	47	26	0.09099819	50	24	0	0.032246273	6.7542543	1	0	1.0054163	0.08898983	0.14224733	0.105735905	0.10415334	-109.958275	77.31491	6.73328	-30.988867	False	407	2336.0785755551096	1.5201434	1536.7488	17.266148	46	1127.3346931358217	8.7009325	129.56482	17.70841	47	1705.0323755571399	7.5450583	225.98001	16.668564	1.2124451	0	2	0	0	0	1.0398464	0.44226265	0.5975838	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.9037037562353	10.921169975727066	301.9260051099228	58.95474875157075	False	False	0	True	False	False	False	False	True	True	False	1.0224553e-13	5.108376e-13	0.99997795	5002.6284	4981.4526	5036.2627	4.602	4.5511	4.6331	-0.8208	-0.8784	-0.7534	1583.614	1512.6202	1733.1886	0.0243	0.0074	0.052	0.0195	0.0059	0.0419	0.0104	0.0032	0.0225	PHOENIX	0.08330419708303734
1636148068921376768	Gaia DR3 2051286967664775424	2051286967664775424	954310899	2016.0	290.1971807329532	0.29303434	37.72225240386049	0.3611389	0.5080668310750222	0.37153193	1.3674917	2.9852633	-0.4817031435724317	0.36552507	-2.9461430747317072	0.44257566	0.14276683	0.072574675	-0.1556142	-0.05411414	0.09976298	-0.10295689	-0.3932338	-0.21745701	-0.03672264	0.23811138	345	0	343	2	0.5666112	368.38873	0.0	0.0	95	False	--	1.4355559	0.09823673	-0.033583496	0.060646504	0.054102458	-0.074938364	-0.10565697	40	23	0.66058314	42	19	0	0.017682025	156.029	0	0	1.0209081	0.26176822	0.10280499	0.15333837	0.21728502	-115.630585	-75.681816	-4.198842	-29.578873	False	341	175.07776030520455	0.8039993	217.7586	20.079288	18	81.09285212765688	9.768502	8.301462	20.566086	21	184.31333266013976	7.4463835	24.75206	19.084003	1.5159332	0	0	0	0	0	1.4820824	0.48679733	0.99528503	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	NOT_AVAILABLE	69.9049148304103	10.898961260981507	301.95982243002953	58.94095155723347	False	False	0	False	False	False	False	False	False	False	False	4.541147e-10	2.3747997e-08	0.9999982	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--		0.08333124886811095

This query returns a large number of columns, and TweakReg expects that the first two are RA and Dec, so we write these to a file.

gaia_query['ra', 'dec'].write('gaia.cat', format='ascii.commented_header', overwrite=True)

3. Align images with TweakReg

Table of Contents

With the catalogs downloaded and the headers populated, we now need to run TweakReg with each catalog passed into the refcat parameter. The steps below compare the astrometric residuals obtained from aligning to each refcat. In each case, we set the updatehdr parameter to False to avoid altering the WCS of the images until we are satisfied with the alignment by inspecting both the TweakReg shift file and the astrometric residual plots.

3.1 SDSS Alignment

refcat = 'sdss.cat'
wcsname = 'SDSS' # Specify the WCS name for this alignment.
cw = 3.5         # 2x the FWHM of the PSF = 3.5 pixels for ACS/WFC, WFC3/UVIS or 2.5 for WFC3/IR.

tweakreg.TweakReg('*flc.fits', # Pass the input images to tweakreg.
                  imagefindcfg={'threshold': 500., 'conv_width': cw},  # Detection parameters that vary for different datasets.
                  refcat=refcat,                # Use user supplied catalog (Gaia).
                  interactive=False,            # Don't show the interactive interface.
                  see2dplot=False,              # Suppress additional figures in this test.
                  shiftfile=True,               # Save output shift file to examine shifts later.
                  outshifts='SDSS_shifts.txt',  # Name of the shift file that will be saved.
                  wcsname=wcsname,              # Give our WCS a new name defined above.
                  reusename=True,               # Overwrite the WCS if it exists.
                  ylimit=1.5,
                  fitgeometry='rscale',         # Allow for translation, rotation, and plate scaling.
                  updatehdr=False)              # Don't update the header with new WCS until later.
clear_output()

# If the alignment is unsuccessful then stop the notebook.

with open('SDSS_shifts.txt', 'r') as shift:
    for line_number, line in enumerate(shift, start=1):
        if "nan" in line:
            raise ValueError('nan found in line {} in shift file'.format(line_number))
        else:
            continue

3.2 Inspect the shift file and fit residuals

We can look at the shift file to see how well the fit did (or we could open the output PNG images for more information). The columns are:

• Filename, X Shift [pixels], Y Shift [pixels], Rotation [degrees], Scale, X RMS [pixels], Y RMS [pixels]

shift_table = Table.read('SDSS_shifts.txt',
                         format='ascii.no_header', 
                         names=['file', 'dx', 'dy', 'rot', 'scale', 'xrms', 'yrms'])

formats = ['.2f', '.2f', '.3f', '.5f', '.2f', '.2f']
for i, col in enumerate(shift_table.colnames[1:]):
    shift_table[col].format = formats[i]
shift_table

Table length=3

file	dx	dy	rot	scale	xrms	yrms
str18	float64	float64	float64	float64	float64	float64
ibwb01xrq_flc.fits	1.47	0.58	0.002	0.99995	0.38	0.36
ibwb01xxq_flc.fits	1.45	-0.84	359.999	0.99999	0.35	0.40
ibwb01xqq_flc.fits	1.48	0.68	0.003	0.99993	0.37	0.34

Show the astrometric residual plots.

# Choose one of the files to view.
Image(filename='residuals_ibwb01xqq_flc.png', width=500, height=300)
# Image(filename='residuals_ibwb01xrq_flc.png', width=500, height=300)
# Image(filename='residuals_ibwb01xxq_flc.png', width=500, height=300)

We can see in the plots above that the RMS is fairly large at about 0.5 pixels. This is generally considered a poor fit, with the desired RMS being < 0.1 pixels depending on the number of sources. This is likely because the SDSS astrometric precision is insufficient for a high-quality HST alignment. One approach would be to align the first image to SDSS and then align the remaining HST images to one another. This would improve both the absolute and relative alignment of the individual frames.

3.3 Gaia Alignment

refcat = 'gaia.cat'
wcsname = 'Gaia' # Specify the WCS name for this alignment.
cw = 3.5         # 2x the FWHM of the PSF = 3.5 pixels for ACS/WFC, WFC3/UVIS or 2.5 for WFC3/IR.

tweakreg.TweakReg('*flc.fits', # Pass the input images to tweakreg.
                  imagefindcfg={'threshold': 500., 'conv_width': cw},  # Detection parameters that vary for different datasets.
                  refcat=refcat,                # Use user supplied catalog (Gaia).
                  interactive=False,            # Don't show the interactive interface.
                  see2dplot=False,              # Suppress additional figures in this test.
                  shiftfile=True,               # Save output shift file to examine shifts later.
                  outshifts='Gaia_shifts.txt',  # Name of the shift file that will be saved.
                  wcsname=wcsname,              # Give our WCS a new name defined above.
                  reusename=True,               # Overwrite the WCS if it exists.
                  ylimit=0.2,                   # The ylimit for the residuals plot.
                  fitgeometry='rscale',         # Allow for translation, rotation, and plate scaling.
                  searchrad=0.1,                # With many sources use a smaller search radius for convergence.
                  updatehdr=False)              # Don't update the header with new WCS until later.
clear_output()

# If the alignment is unsuccessful then stop the notebook.

with open('Gaia_shifts.txt', 'r') as shift:
    for line_number, line in enumerate(shift, start=1):
        if "nan" in line:
            raise ValueError('nan found in line {} in shift file'.format(line_number))
        else:
            continue

3.4 Inspect the shift file and fit residuals

We can similarly look at the shift file from alignment to the Gaia catalog. As expected, the Gaia catalog does quite a bit better, with rms residuals ~0.05 pixels.

The columns are: Filename, X Shift [pixels], Y Shift [pixels], Rotation [degrees], Scale, X RMS [pixels], Y RMS [pixels]

shift_table = Table.read('Gaia_shifts.txt',
                         format='ascii.no_header', 
                         names=['file', 'dx', 'dy', 'rot', 'scale', 'xrms', 'yrms'])

formats = ['.2f', '.2f', '.3f', '.5f', '.2f', '.2f']
for i, col in enumerate(shift_table.colnames[1:]):
    shift_table[col].format = formats[i]
shift_table

Table length=3

file	dx	dy	rot	scale	xrms	yrms
str18	float64	float64	float64	float64	float64	float64
ibwb01xrq_flc.fits	-0.22	0.11	0.000	1.00000	0.03	0.05
ibwb01xxq_flc.fits	-0.22	0.10	360.000	1.00001	0.03	0.05
ibwb01xqq_flc.fits	-0.22	0.10	360.000	1.00000	0.03	0.03

Print the Number of Gaia Matches per image from the TweakReg output.

rootnames = ['ibwb01xqq', 'ibwb01xrq', 'ibwb01xxq']

for rootname in rootnames:
    match_tab = ascii.read(rootname+'_flc_catalog_fit.match') # load match file in astropy table
    print(rootname+': Number of Gaia Matches =', len(match_tab))

ibwb01xqq: Number of Gaia Matches = 565
ibwb01xrq: Number of Gaia Matches = 566
ibwb01xxq: Number of Gaia Matches = 545

Compare the RMS and number of matches from TweakReg with the MAST alignment results.

wcstable

Table length=3

filename	DETECTOR	EXPTIME	WCSNAME	NMATCHES	RMS_RA	RMS_DEC	RMS_RA_pix	RMS_DEC_pix
str18	str4	float64	str30	int64	float64	float64	float64	float64
ibwb01xqq_flc.fits	UVIS	30.0	IDC_2731450pi-FIT_IMG_GAIAeDR3	468	1.4	1.5	0.04	0.04
ibwb01xrq_flc.fits	UVIS	360.0	IDC_2731450pi-FIT_IMG_GAIAeDR3	475	3.4	3.0	0.09	0.08
ibwb01xxq_flc.fits	UVIS	360.0	IDC_2731450pi-FIT_IMG_GAIAeDR3	426	3.4	3.5	0.09	0.09

The new Gaia solution is better than the MAST solution because there are a larger number of matches sources with smaller RMS values in both RA and Dec. For more details on absolute astrometry in HST images, see Section 4.5 in the DrizzlePac Handbook. Finally, show one of the astrometric residual plots as an example that highlights the small scatter in the best fit shift for all sources.

# Choose one of the files to view.
Image(filename='residuals_ibwb01xqq_flc.png', width=500, height=300)
# Image(filename='residuals_ibwb01xrq_flc.png', width=500, height=300)
# Image(filename='residuals_ibwb01xxq_flc.png', width=500, height=300)

plt.close('all') # Close figures to reduce memory usage.

3.5 Overplot matched sources on the image

Table of Contents

Now, let’s plot the sources that were matched between all images on the “bottom” UVIS chip. This is referred to as chip 2 or SCI, 1 or extension 1. The cell below reads in the *_fit.match file as an astropy table. Unfortunately, it doesn’t automatically name columns so we look at the header to grab the columns.

To verify that TweakReg obtained an acceptable fit between matched source catalogs, it is useful to inspect the results before updating the image header WCS. In the figure below, sources that are matched with Gaia are overplotted on one of the input FLC frames with the two chips merged together.

It can be useful to check that TweakReg locked onto stars and not hot pixels or other detector artifacts before proceeding to update the image header.

# Choose one of the files to view.
# rootname = 'ibwb01xqq'
rootname = 'ibwb01xrq'
# rootname = 'ibwb01xxq'

plt.figure(figsize=(7, 7), dpi=140)
chip1_data = fits.open(rootname+'_flc.fits')['SCI', 2].data
chip2_data = fits.open(rootname+'_flc.fits')['SCI', 1].data
fullsci = np.concatenate([chip2_data, chip1_data])
z1, z2 = ZScaleInterval().get_limits(fullsci)
plt.imshow(fullsci, cmap='Greys', origin='lower', vmin=z1, vmax=z2)

match_tab = ascii.read(rootname+'_flc_catalog_fit.match')  # Load the match file in astropy table.
match_tab_chip1 = match_tab[match_tab['col15'] == 2]  # Filter the table for sources on chip 1.
match_tab_chip2 = match_tab[match_tab['col15'] == 1]  # Filter the table for sources on chip 2.
x_cord1, y_cord1 = match_tab_chip1['col11'], match_tab_chip1['col12']
x_cord2, y_cord2 = match_tab_chip2['col11'], match_tab_chip2['col12']

# Correct for 0 vs. 1 indexing on the figure by subtracting 1.
plt.scatter(x_cord1-1.0, y_cord1+2051-1.0, s=50, edgecolor='r', facecolor='None', label='Matched Sources')
plt.scatter(x_cord2-1.0, y_cord2-1.0, s=50, edgecolor='r', facecolor='None')
plt.title(f'Matched sources UVIS to Gaia: N = {len(match_tab)}', fontsize=14)
plt.show()

3.6 Check the significance of proper motions

Table of Contents

Prior to accepting the Gaia alignment solution derived above, there is one additional refinement users may wish to consider depending on their dataset and science goals. Specifically, stars move across the sky over time, with more nearby (and thus often brighter) stars moving more quickly. This movemement with respect to a fixed WCS is known as proper motion.

By default, Gaia eDR3 coordinates are provided for an epoch of 2016. This means that the stars in images taken long before or after 2016, or those in high proper motion fields, may be displaced from their catalog coordinates. In general, the positions provided by Gaia are exceptionally accurate, and the accuracy of the proper motions vary depending on the brightness of the star, and the number of observations taken to measure the proper motion.

Users can asses the importance of stellar proper motions for their observations by taking the default Gaia catalog coordinates and proper motions, and calculating the epoch-corrected position of each star for the time of their observations. In the figure below, we show the proper motions of stars from the Gaia catalog created earlier, with motions in milliarcseconds per year (mas yr\(^{-1}\)).

fig, ax = plt.subplots(1, 3, figsize=(1.6*6.4, 4.8), layout='constrained')
ax[0].set_title('Gaia Proper Motions')
ax[1].set_title('Proper Motion (RA)')
ax[2].set_title('Proper Motion (Dec)')
ax[0].set_ylabel('Number of Stars')
ax[0].hist(gaia_query['pm'], bins=50)
ax[1].hist(gaia_query['pmra'], bins=50)
ax[2].hist(gaia_query['pmdec'], bins=50)
ax[0].axvline(np.ma.median(gaia_query['pm']), c='black', ls=':', label=''+str(round(np.ma.median(gaia_query['pm']), 2))+' mas/yr')
ax[1].axvline(np.ma.median(gaia_query['pmra']), c='black', ls=':', label=''+str(round(np.ma.median(gaia_query['pmra']), 2))+' mas/yr')
ax[2].axvline(np.ma.median(gaia_query['pmdec']), c='black', ls=':', label=''+str(round(np.ma.median(gaia_query['pmdec']), 2))+' mas/yr')
ax[0].set_xlim([0, 80])
ax[1].set_xlim([-75, 75])
ax[2].set_xlim([-75, 75])
for idx in [0, 1, 2]:
    ax[idx].set_yscale('log')
    ax[idx].set_xlabel('Proper Motion (mas yr$^{-1}$)')
    ax[idx].legend(handlelength=1, loc='upper right')
plt.show()

In this case, the median proper motion is \(\approx\) 2.42 mas yr\(^{-1}\). The date of observation for the example dataset of NGC 6791 is 2011.8, which is \(\approx\) 4.2 years before the Gaia epoch of 2016. This means the median stellar motion is \(\approx\) 10 mas, or approximately 0.25 UVIS pixels. Some stars will have moved by less than this amount, while others will have moved more based on their measured proper motions. Because the median motion of 0.25 pixels is larger than the earlier alignment solution RMS uncertainties of \(<\) 0.1 pixels, it may be worth considering the effects of proper motions for these observations.

First, we define the epoch of Gaia (gaia_epoch), the epoch of the observations (data_epoch), and we remove stars without accurate proper motion measurements. Numerous different quality cuts can be applied to the Gaia catalog using the large array of provided measurements, including the proper motion uncertainties themselves. For this example, we’ll apply a single cut on the re-normalised unit weight error, or RUWE. This overall goodness of fit parameter indicates how well each source was fit as a single star, with values of \(<\) 1.4 excluding most suspected binary star systems, and \(<\) 1.25 providing clean, well fit samples of stars. See the report GAIA-C3-TN-LU-LL-124-01 and Penoyre et al. (2022) for more information.

data_epoch = Time(table['DATE-OBS'][0]).decimalyear
gaia_epoch = gaia_query['ref_epoch'][0]
epoch_diff = round(np.abs(data_epoch - gaia_epoch), 2)
print(f'\nThe Gaia and Data epochs are {gaia_epoch} and {data_epoch}, respectively, resulting in a baseline of {epoch_diff} years.')
gaia_string = str(int(gaia_epoch))
data_string = str(int(data_epoch))

# Remove catalog entries with large fit uncertainties and empty data values.
gaia_query_masked = gaia_query[(gaia_query['ruwe'] < 1.25) & ~gaia_query['ruwe'].mask]

# Create a table with RA, Dec, and their proper motions.
gaia_table = Table([gaia_query_masked['ra'], gaia_query_masked['dec'], gaia_query_masked['pmra'], gaia_query_masked['pmdec']])

# Update the column names with their epoch.
gaia_table.rename_column('ra', 'ra_'+gaia_string)
gaia_table.rename_column('dec', 'dec_'+gaia_string)

print(f'\nThe quality mask removed {len(gaia_query)-len(gaia_query_masked)} sources, reducing the Gaia source catalog from {len(gaia_query)} to {len(gaia_query_masked)} objects.\n')

The Gaia and Data epochs are 2016.0 and 2011.8, respectively, resulting in a baseline of 4.2 years.

The quality mask removed 499 sources, reducing the Gaia source catalog from 4865 to 4366 objects.

Next, we create an Astropy SkyCoord object with the Gaia 2016 catalog coordinates and proper motions. We then use the apply_space_motion function to calculate the coordinates of each star at the epoch of observation, which is late 2011. See the Evolving Coordinate Positions Between Epochs Astropy tutorial and the Time from Epoch Formats documentation for additional information. As noted in those resources, the RA proper motions already contain the cosine of declination term, so these coordinates can be passed directly to SkyCoord.

# Create the SkyCoord object with RA, Dec, proper motions, and time of observation.
c = SkyCoord(ra=gaia_table['ra_'+gaia_string],
             dec=gaia_table['dec_'+gaia_string],
             pm_ra_cosdec=gaia_table['pmra'],
             pm_dec=gaia_table['pmdec'],
             obstime=Time(gaia_epoch, format='decimalyear'))

# Calculate coordinates for each epoch of observation.
gaia_coord = SkyCoord(ra=gaia_table['ra_'+gaia_string], dec=gaia_table['dec_'+gaia_string])
data_coord = c.apply_space_motion(Time(data_epoch, format='decimalyear'))

# Extract the separations from the resulting catalogs.
idx, sep_data, _ = data_coord.match_to_catalog_sky(gaia_coord)

# Create a list of the stellar separations between the two epochs in milliarcseconds.
separations_list = [round(((a).to(u.arcsec)*1000.0).value, 5) for a in sep_data]
print('The median proper motion is', round(np.nanmedian(np.nan_to_num(separations_list)), 1), 'milliarcseconds.')

The median proper motion is 10.1 milliarcseconds.

/home/runner/micromamba/envs/ci-env/lib/python3.12/site-packages/erfa/core.py:133: ErfaWarning: ERFA function "pmsafe" yielded 4365 of "distance overridden (Note 6)", 1 of "binary logical OR of the above warnings"
  warn(f'ERFA function "{func_name}" yielded {wmsg}', ErfaWarning)

# Convert the results to a table and write to a file.
table_multi_epoch = Table([gaia_table['ra_2016'],
                           gaia_table['dec_2016'],
                           data_coord.to_table()['ra'],
                           data_coord.to_table()['dec'],
                           separations_list],
                          names=('ra_'+gaia_string, 'dec_'+gaia_string, 'ra_'+data_string, 'dec_'+data_string, 'separation'))

# Save the table to a file with proper formatting.
table_multi_epoch.write('gaia_multi_epoch.cat', format='ascii.fast_commented_header', overwrite=True)

table_multi_epoch

Table length=4366

ra_2016	dec_2016	ra_2011	dec_2011	separation
deg	deg	deg	deg
float64	float64	float64	float64	float64
290.2248611356283	37.80192950361359	290.22486205497364	37.801922466008946	25.46998
290.22327972754084	37.80283944795903	290.22328046129354	37.80284388390957	16.10523
290.22261994880597	37.80270752474709	290.2226209842266	37.802710440923306	10.90354
290.2248309701555	37.80058611405895	290.22483183052253	37.80058876385366	9.8482
290.2225602638809	37.80397847004457	290.22256056792554	37.80398120771446	9.89348
290.22681959687196	37.80112772770517	290.2268202501215	37.80113302842852	19.17286
290.2275249308142	37.803472775139035	290.2275260738658	37.8034748043064	7.99588
290.2242320409821	37.80039510739463	290.2242334744281	37.80039744343509	9.34611
290.22212330415425	37.800777820408136	290.22212495883775	37.800781650999006	14.57126
...	...	...	...	...
290.1710542799103	37.874210138824154	290.17106361859726	37.87421780234066	38.28035
290.1228125225418	37.7819163631625	290.1228221050927	37.78193326977083	66.6916
290.1750101665401	37.87602842402639	290.17501074773094	37.8760310103734	9.45619
290.2882629014902	37.86913015536792	290.28826687466113	37.86913995090914	37.02756
290.1194354085638	37.80348115009846	290.11943873196947	37.80348787251983	25.98148
290.33007288300433	37.799005196546766	290.3300875083115	37.799018457074794	63.32238
290.14278652187744	37.75037525259858	290.1427871536142	37.750377827679095	9.44309
290.17127999981165	37.73091082720331	290.1712804292001	37.7309134639259	9.57061
290.1971807329532	37.72225240386049	290.1971814434193	37.72225584091542	12.5377

Finally, we can bring these results together and examine how well the UVIS images are fit to 1) the default 2016 Gaia catalog from earlier, 2) the 2016 Gaia catalog with poorly fit sources removed, and 3) the proper motion corrected 2011 Gaia catalog with poorly fit sources removed. In the cell below, we loop over these three options by iterating through the keyword arguments (kwargs) for TweakReg. The *.match files are over-written each time, so we append the results to a table and add comments for clarity.

kwargs = [{'refcat': 'gaia.cat', 'refxcol': 1, 'refycol': 2},
          {'refcat': 'gaia_multi_epoch.cat', 'refxcol': 1, 'refycol': 2},
          {'refcat': 'gaia_multi_epoch.cat', 'refxcol': 3, 'refycol': 4}]

for iteration, kwarg in enumerate(kwargs):
    
    tweakreg.TweakReg('ibwb01xrq_flc.fits',         # Pass the input images to tweakreg.
                      imagefindcfg={'threshold': 500., 'conv_width': 3.5},  # Detection parameters that vary for different datasets.
                      interactive=False,            # Don't show the interactive interface.
                      see2dplot=False,              # Suppress additional figures in this test.
                      shiftfile=True,               # Save output shift file to examine shifts later.
                      outshifts='Gaia_shifts.txt',  # Name of the shift file that will be saved.
                      wcsname='Gaia',               # Give our WCS a new name defined above.
                      reusename=True,               # Overwrite the WCS if it exists.
                      ylimit=0.2,                   # The ylimit for the residuals plot.
                      fitgeometry='rscale',         # Allow for translation, rotation, and plate scaling.
                      searchrad=0.3,                # With many sources use a smaller search radius for convergence.
                      updatehdr=False, **kwarg)     # Do not update the header, and pass in the keyword arguments.
    
    # Read the shift file.    
    shift_table = Table.read('Gaia_shifts.txt',
                             format='ascii.no_header', 
                             names=['file', 'dx', 'dy', 'rot', 'scale', 'xrms', 'yrms'])
    
    match_table = ascii.read('ibwb01xrq_flc_catalog_fit.match')
    shift_table.add_column([len(match_table)], name='n_stars')
    
    formats = ['.2f', '.2f', '.3f', '.5f', '.2f', '.2f', '.0f']
    for i, col in enumerate(shift_table.colnames[1:]):
        shift_table[col].format = formats[i]

    # Copy shift_table on the first iteration and then append so it is not over-written.
    if iteration == 0:
        shift_table_final = Table(shift_table, copy=True)
    else:
        shift_table_final.add_row(shift_table[0])
    
    plt.close('all')
    clear_output()

shift_table_final.add_column(['Gaia 2016 default catalog', 'Gaia 2016 cleaned catalog', 'Gaia 2011 cleaned catalog'], name='comments')

shift_table_final

Table length=3

file	dx	dy	rot	scale	xrms	yrms	n_stars	comments
str18	float64	float64	float64	float64	float64	float64	int64	str25
ibwb01xrq_flc.fits	-0.21	0.11	0.000	1.00000	0.03	0.05	574	Gaia 2016 default catalog
ibwb01xrq_flc.fits	-0.21	0.12	0.000	1.00000	0.03	0.05	511	Gaia 2016 cleaned catalog
ibwb01xrq_flc.fits	0.02	0.02	0.000	1.00000	0.05	0.06	595	Gaia 2011 cleaned catalog

The table above shows the resulting shifts in X and Y, together with the rotation, scale, and RMS values we explored in earlier sections. In addition, the n_stars column lists the number of stars matched in the image from each repsective Gaia catalog. The first row duplicates our earlier TweakReg solution, with shifts in X and Y of -0.21 and 0.11 pixels, respectively. As shown in the second row, removing poorly fit sources from the Gaia catalog resulted in fewer matches, but the same overall shift solution and RMS values to within \(\sim\)0.01 pixels. In the third row, which uses the catalog accounting for proper motions, the X and Y shifts are now different by \(\approx\) 0.2 pixels in X and 0.1 pixels in Y. The RMS uncertainties have increased minutely, but are still a factor of 2-4 smaller than the shift values, with more sources matched, indicating the measured shift is physical.

Overall, the best-fit global WCS changed by \(\sim\)0.2 pixels after accounting for stellar proper motions. The effect will generally be smaller for observations closer to the Gaia reference epoch of 2016, and continually grow for observations earlier and later in time. Whether or not accounting for proper motions is necessary for a specific set of observations depends on the science goals of the analysis, and the date of observation.

As a general guide, differences in shifts that are larger than the RMS values of the fit usually indicate an improved WCS solution. In this example, we proceed with the epoch-corrected Gaia alignment solution.

3.7 Update header with optimal parameters

Table of Contents

Once you’ve decided on the optimal set of parameters for your alignment, it is safe to update the header of the input data, (assuming that the new solution is better than the MAST alignment). To apply these transformations to the image, we simply need to run TweakReg the same as before, but set the parameter updatehdr equal to True:

kwargs = {'refcat': 'gaia_multi_epoch.cat', 'refxcol': 3, 'refycol': 4}

tweakreg.TweakReg('*flc.fits', # Pass the input images to tweakreg.
                  imagefindcfg={'threshold': 500., 'conv_width': 3.5},  # Detection parameters that vary for different datasets.
                  interactive=False,            # Don't show the interactive interface.
                  see2dplot=False,              # Suppress additional figures in this test.
                  shiftfile=True,               # Save output shift file to examine shifts later.
                  outshifts='Gaia_shifts.txt',  # Name of the shift file that will be saved.
                  wcsname='Gaia',              # Give our WCS a new name defined above.
                  reusename=True,               # Overwrite the WCS if it exists.
                  ylimit=0.2,                   # The ylimit for the residuals plot.
                  fitgeometry='rscale',         # Allow for translation, rotation, and plate scaling.
                  searchrad=0.3,                # With many sources use a smaller search radius for convergence.
                  updatehdr=True, **kwargs)     # Update the header with new WCS solution, and pass in the keyword arguments.

plt.close('all') # Close figures to reduce memory usage.
clear_output()

4. Combine the images using AstroDrizzle

Table of Contents

While the three sets of FLC files are now aligned, in this example we drizzle together only the two long exposures. When exposures are very different lengths, drizzling them together doesn’t work well when ‘EXP’ weighting is used. For objects that saturate in the long exposures, the problem occurs at the boundary where the signal transitions from only being present in short exposures near the core to being present at larger radii in the longer exposures. The result is a discontinuity in the PSF radial profile and a resulting flux that is too low in some regions. For photometry of saturated objects, the short exposures should be drizzled separately from the long exposures.

Next, we combine the images through drizzling, and retrieve some recommended values for this process from the MDRIZTAB reference file. The parameters in this file are different for each detector and are based on the number of input frames. These are a good starting point for drizzling and may be adjusted based on your science goals.

input_images = sorted(glob.glob('ibwb01x[rx]q_flc.fits'))

mdz = fits.getval(input_images[0], 'MDRIZTAB', ext=0).split('$')[1]

print('Searching for the MDRIZTAB file:', mdz)

!crds sync --hst --files {mdz} --output-dir {os.environ['iref']}

Searching for the MDRIZTAB file: ubi1853qi_mdz.fits

CRDS - INFO -  Symbolic context 'hst-latest' resolves to 'hst_1349.pmap'
CRDS - INFO -  Reorganizing 0 references from 'instrument' to 'flat'
CRDS - INFO -  Reorganizing from 'instrument' to 'flat' cache,  removing instrument directories.

CRDS - INFO -  Syncing explicitly listed files.
CRDS - INFO -  0 errors
CRDS - INFO -  0 warnings
CRDS - INFO -  4 infos

def get_vals_from_mdriztab(input_images, 
                           kw_list=[
                               'driz_sep_bits', 
                               'combine_type', 
                               'driz_cr_snr', 
                               'driz_cr_scale', 
                               'final_bits']):
    
    'Get only selected parameters from the MDRIZTAB.'
    mdriz_dict = getMdriztabPars(input_images)
    
    requested_params = {}
    
    print('Outputting the following parameters:')
    for k in kw_list:
        requested_params[k] = mdriz_dict[k]
        print(k, mdriz_dict[k])
    
    return requested_params

selected_params = get_vals_from_mdriztab(input_images)

- MDRIZTAB: AstroDrizzle parameters read from row 2.
Outputting the following parameters:
driz_sep_bits 96
combine_type minmed
driz_cr_snr 3.5 3.0
driz_cr_scale 1.2 0.7
final_bits 96

Note that the parameter final_bits= ‘96’ is equivalent to final_bits= ‘64, 32’ because they are added in a bit-wise manner.

# To override any of the above values:
selected_params['driz_sep_bits'] = '256, 64, 32'
selected_params['final_bits'] = '256, 64, 32'
# selected_params['combine_type']  = 'median'
# selected_params['driz_cr_snr']   = '4.0 3.5'
# selected_params['driz_cr_scale'] = '1.5 1.2'

astrodrizzle.AstroDrizzle(input_images, 
                          output='f606w',
                          preserve=False,
                          clean=True, 
                          build=False,
                          context=False,
                          skymethod='match',
                          **selected_params)

clear_output()

Display the combined science and weight images. In this example there appear to be three detector chips due to the large dither that was used in the observations. We note that the exposures overlapped in the center region, as evidenced by the increased value of the weight map corresponding to a longer effective exposure time. In addition, the resulting mosaic is better cleaned of artifacts and cosmic rays in this region due to the multiple exposures, as seen by the uniform background in the science mosaic.

sci = fits.getdata('f606w_drc_sci.fits')
wht = fits.getdata('f606w_drc_wht.fits')

fig = plt.figure(figsize=(20, 20))
ax1 = fig.add_subplot(1, 2, 1)
ax2 = fig.add_subplot(1, 2, 2)

ax1.imshow(sci, vmin=-0.05, vmax=0.4, cmap='Greys_r', origin='lower')
ax2.imshow(wht, vmin=0.0, vmax=1000, cmap='Greys_r', origin='lower')

ax1.title.set_text('Science Image')
ax2.title.set_text('Weight Image')
ax1.set_xlabel('X (pixels)')
ax1.set_ylabel('Y (pixels)')
ax2.set_xlabel('X (pixels)')
ax2.set_ylabel('Y (pixels)')
plt.show()

For more details on inspecting the drizzled products after reprocessing, see Section 7.3 in the DrizzlePac Handbook.

Conclusions

Table of Contents

Many other services have interfaces for querying catalogs that can also be used to align HST images. In general, Gaia works very well for HST due to its high precision, but can have a low number of sources in some regions, especially at high galactic latitudes. Aligning images to an absolute frame provides a way to make data comparable across many epochs/detectors/observatories, and in many cases, makes the alignment easier to complete.

About this Notebook

Created: 14 Dec 2018;     V. Bajaj
Updated: 27 May 2025;     M. Revalski, V. Bajaj, & J. Mack
Updated: 27 Jul 2026;     M. Revalski

Source: GitHub spacetelescope/hst_notebooks

Additional Resources

Below are some additional resources that may be helpful. Please send any questions through the HST Help Desk, selecting the DrizzlePac category.

• WFC3 Website

• WFC3 Data Handbook

• WFC3 Instrument Handbook

Citations

When publishing results that use this notebook, please include a citation in your manuscript. A latex-ready bibliography entry can be copied from the HST Notebooks repository using the "Cite this repository" drop-down on GitHub or "Export > BibTex" on Zenodo.

---

If you use Python packages such as astropy, astroquery, drizzlepac, matplotlib, or numpy for published research, please cite the authors.

Follow these links for more information about citing various packages:

• Citing astropy

• Citing astroquery

• Citing drizzlepac

• Citing matplotlib

• Citing numpy

---

Top of Page