Accessing Roman Catalogs: Leveraging Ancillary Tables and Visualization to Apply Data Quality Cuts

Accessing Roman Catalogs: Leveraging Ancillary Tables and Visualization to Apply Data Quality Cuts#

Learning Goals#

By the end of this tutorial, you will:

Understand how to launch and programatically control MAST Aladin to view targets on the sky from within a notebook, including overlaying a set of targets and observation footprints on the sky.
Understand how to use the MAST catalog schema browser to discover ancillary tables.
Understand how to “decode” catalog bitflags, aided by a bitflag “helper” class and the bitflag definition documentation (or database flag definition tables).

Table of Contents#

Introduction#

This notebook demonstrates how to leverage catalog queries together with MAST Aladin (to view objects on the sky) and the MAST Catalog Schema Browser to examine a selected sample, investigate data quality issues, and identify quality flags that can be used to refine the sample for analysis.

MAST will provide access to a range of Roman catalog products. In addition to analysis-ready measurements of observed and derived physical quantities, Roman catalogs will include ancillary information about, e.g., data quality, processing flags, and data provenance (such as the primary/overlap designation bit flag in the flagged_spatial_id identifiers in WFI multiband photometric source catalogs). This notebook demonstrates how to query candidate stars in the Galactic Bulge using multiband photometry, view the targets in MAST Aladin, and then examine data quality flag information from ancillary tables/columns to determine how to exclude data artifacts and spurious sources from the analysis.

Imports#

This tutorial makes use of the following packages:

astroquery.mast Catalogs, Observations for querying the MAST catalog database and observation holdings
astropy.coordinates SkyCoord for defining sky coordinates
astropy.units to specify angular units
numpy for numerical calculations
matplotlib.pyplot for plotting catalog data
mast_aladin to view targets on the sky
sidecar Sidecar to display MAST-Aladin alongside the notebook
time to enable time delays when flipping through objects with MAST-Aladin
astropy.table Table to construct tables from flag values
A simple “flag helper class” PanSTARRSFlagHelper to retrieve definitions (from a table via a MAST TAP service), represent, and help parse data quality flags encoded as bitflags for easier readibility.

from astroquery.mast import Catalogs, Observations
from astropy.coordinates import SkyCoord
from astropy.table import Table
import astropy.units as u
import numpy as np

from mast_aladin import MastAladin
from sidecar import Sidecar
import matplotlib.pyplot as plt
from flag_helper import PanSTARRSFlagHelper
import time

Identification and quality checks on candidate stars in the Galactic Bulge#

In this notebook, we will construct a color-magnitude diagram (\(r-i\) vs \(z\)) of stars in the Milky Way bulge near the Galactic Center, from which specific categories of stars can be selected for further analysis.

For demonstration purposes (as Roman catalogs are not yet available), PanSTARRS will be used as the example dataset. Similar to Roman (particularly the WFI multiband photometric source catalogs), PanSTARRS features a rich set of ancillary tables containing data quality and processing flags — providing a reasonable analog to demonstrate MAST data access and visualization tools that will be available for use with Roman.

(In the interest of simplicity, this example will omit crossmatching to Gaia to leverage Gaia-derived distances to exclude fore/background objects.)

Identifying candidate stars in the Galactic Bulge#

In this example, we aim to identify stars in the Galactic Bulge by constructing a \(r-i\) vs \(z\) color magnitude diagram. We begin with the following selection criteria:

Located within 0.1 degree of the Galactic center
Are point sources (excluding extended objects, such as galaxies).
Have measured \(r\), \(i\), \(z\) magnitudes

Querying the PanSTARRS catalog#

First, we consult the PanSTARRS DR2 catalogs documentation and the MAST Catalog Schema Browser to investigate the available tables and columns.

We determine that we can obtain all necessary information from the PanSTARRS DR2 “mean” table, which we can access using astroquery.mast.

To construct our query, we begin by obtaining the sky coordinates for the Galactic center using astropy.

# Use the Astropy SkyCoord name resolver to get 
# the position of the galactic center in the ICRS frame
gc = SkyCoord.from_name("Galactic center")
gc

<SkyCoord (ICRS): (ra, dec) in deg
    (266.41500889, -29.00611111)>

Based on PanSTARRS documentation (from astroquery.mast and the PanSTARRS DR2 documentation), we see the MeanObject catalog provides the relevant photometric information (in the columns rMeanPSFMag, iMeanPSFMag, zMeanPSFMag), as well as position information (RA/Dec).

Using the Catalogs module of astroquery.mast, we query the PanSTARRS mean_object table to perform a cone search of 0.1 degree around the Galactic Center, restricting the results to objcts with \(r\), \(i\), \(z\) magnitudes > -999 (the placeholder used for missing values in the database).

# Query the PanSTARRS DR2 Mean catalog aroung the 
# Galactic center, with available z,r,i magnitudes
sample = Catalogs.query_criteria(
    collection="ps1_dr2",
    catalog="mean_object",
    coordinates=gc,
    radius=0.1*u.deg,
    rMeanPSFMag=">-999",   # missing magnitudes are -999
    iMeanPSFMag=">-999",   # missing magnitudes are -999
    zMeanPSFMag=">-999",   # missing magnitudes are -999
)

(Note: the returned table column names are all lower case, which we use below.)

Next, to distinguish stars (point sources) from extended objects, we will make a cut using the PSF and Kron magnitudes in the i band, as point sources typically have iMeanPSFMag-iMeanKronMag <= 0.05 (as recommended in the PanSTARRS documentation).

sample = sample[
    sample['imeanpsfmag']-sample['imeankronmag'] <= 0.05
]

Finally, we will sort our sample by descending “objID” for consistent ordering.

sample.sort("objid", reverse=True)

Our resulting sample has 2906 candidate stars:

sample

Table length=2906

objname	objaltname1	objaltname2	objaltname3	objid	uniquepspsobid	ippobjid	surveyid	htmid	zoneid	tessid	projectionid	skycellid	randomid	batchid	dvoregionid	processingversion	objinfoflag	qualityflag	rastack	decstack	rastackerr	decstackerr	rameanerr	decmeanerr	epochmean	posmeanchisq	cx	cy	cz	lambda	beta	l	b	nstackobjectrows	nstackdetections	ndetections	ng	nr	ni	nz	ny	gqfperfect	gmeanpsfmag	gmeanpsfmagerr	gmeanpsfmagstd	gmeanpsfmagnpt	gmeanpsfmagmin	gmeanpsfmagmax	gmeankronmag	gmeankronmagerr	gmeankronmagstd	gmeankronmagnpt	gmeanapmag	gmeanapmagerr	gmeanapmagstd	gmeanapmagnpt	gflags	rqfperfect	rmeanpsfmag	rmeanpsfmagerr	rmeanpsfmagstd	rmeanpsfmagnpt	rmeanpsfmagmin	rmeanpsfmagmax	rmeankronmag	rmeankronmagerr	rmeankronmagstd	rmeankronmagnpt	rmeanapmag	rmeanapmagerr	rmeanapmagstd	rmeanapmagnpt	rflags	iqfperfect	imeanpsfmag	imeanpsfmagerr	imeanpsfmagstd	imeanpsfmagnpt	imeanpsfmagmin	imeanpsfmagmax	imeankronmag	imeankronmagerr	imeankronmagstd	imeankronmagnpt	imeanapmag	imeanapmagerr	imeanapmagstd	imeanapmagnpt	iflags	zqfperfect	zmeanpsfmag	zmeanpsfmagerr	zmeanpsfmagstd	zmeanpsfmagnpt	zmeanpsfmagmin	zmeanpsfmagmax	zmeankronmag	zmeankronmagerr	zmeankronmagstd	zmeankronmagnpt	zmeanapmag	zmeanapmagerr	zmeanapmagstd	zmeanapmagnpt	zflags	yqfperfect	ymeanpsfmag	ymeanpsfmagerr	ymeanpsfmagstd	ymeanpsfmagnpt	ymeanpsfmagmin	ymeanpsfmagmax	ymeankronmag	ymeankronmagerr	ymeankronmagstd	ymeankronmagnpt	ymeanapmag	ymeanapmagerr	ymeanapmagstd	ymeanapmagnpt	yflags	astrometrycorrectionflag	pmra	pmdec	pmraerr	pmdecerr	decmean	ramean
																			deg	deg	mas	mas	mas	mas	d					deg	deg	deg	deg										mag	mag	mag		mag	mag	mag	mag	mag		mag	mag	mag				mag	mag	mag		mag	mag	mag	mag	mag		mag	mag	mag				mag	mag	mag		mag	mag	mag	mag	mag		mag	mag	mag				mag	mag	mag		mag	mag	mag	mag	mag		mag	mag	mag				mag	mag	mag		mag	mag	mag	mag	mag		mag	mag	mag				mas	mas	mas	mas	deg	deg
object	object	object	object	int64	int64	int64	uint8	int64	int32	uint8	int16	uint8	float64	int64	int32	uint8	int32	uint8	float64	float64	float32	float32	float32	float32	float64	float32	float64	float64	float64	float64	float64	float64	float64	int16	int16	int16	int16	int16	int16	int16	int16	float32	float32	float32	float32	int16	float32	float32	float32	float32	float32	int16	float32	float32	float32	int16	int32	float32	float32	float32	float32	int16	float32	float32	float32	float32	float32	int16	float32	float32	float32	int16	int32	float32	float32	float32	float32	int16	float32	float32	float32	float32	float32	int16	float32	float32	float32	int16	int32	float32	float32	float32	float32	int16	float32	float32	float32	float32	float32	int16	float32	float32	float32	int16	int32	float32	float32	float32	float32	int16	float32	float32	float32	float32	float32	int16	float32	float32	float32	int16	int32	uint8	float32	float32	float32	float32	float64	float64
PSO J266.4148-28.9061	-999	-999	-999	73312664147912990	3526718000013439	511109698172032	0	11710823714680	7331	1	693	70	0.764368522692815	3526718	119002	3	436527104	52	266.41478676	-28.90610598	0.001	0.001	0.0070586307	0.0070586307	55927.67218421359	2.977626	-0.05474229746111264	-0.8736995987831238	-0.48337593232895626	266.847371042176	-5.506067272511191	0.03010305808887179	0.00834130681494409	-999	5	43	0	8	10	13	12	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	0	114720	0.998891	21.3498	0.061316	0.219186	5	21.2942	21.7458	21.3105	0.105665	0.609897	4	21.2082	0.080092	0.766996	5	115000	0.999104	19.8697	0.072825	0.111847	4	19.8166	19.9951	19.965	0.095545	0.159543	4	19.8827	0.103618	0.178087	4	115000	0.998817	19.1733	0.022056	0.194126	7	19.1342	19.6215	19.3604	0.065533	0.351007	6	19.227	0.037813	0.12218	6	114992	0.998956	18.9296	0.039735	0.120504	7	18.7681	19.0617	19.7013	0.31707	0.589191	6	19.7993	0.01751	0.336853	7	115000	7	-5.892208	-21.073942	5.3919663	5.3919663	-28.90612260697708	266.4147770200182
PSO J266.4113-28.9084	-999	-999	-999	73312664113250267	3526718000012839	511109698171432	0	11710823704335	7331	1	693	70	0.489421554365588	3526718	119002	3	503635968	60	266.41132436	-28.90838389	0.001	0.001	0.0002902839	0.0002907789	56064.83036509671	61.33877	-0.05479397074219825	-0.8736771513453905	-0.4834106494351431	266.8443757444756	-5.508410676770176	0.02658374167397246	0.009753797166211999	-999	5	63	17	10	12	12	12	0.999828	16.7477	0.006301	0.014903	10	16.7284	16.7706	16.7989	0.003505	0.017162	11	16.7298	0.005469	0.01737	11	115000	0.998943	15.7687	0.006629	0.011305	6	15.7531	15.7818	15.8248	0.007763	0.014294	7	15.7577	0.010988	0.02169	7	115000	0.999453	15.2556	0.000177	0.00558	5	15.2452	15.259	15.314	0.002996	0.010585	8	15.2517	0.005702	0.010646	8	115000	0.99985	14.9433	0.003342	0.009762	7	14.9253	14.9518	14.986	0.005434	0.030208	9	14.927	0.008166	0.02463	9	114992	0.999285	14.7855	0.005587	0.04294	10	14.7197	14.8379	14.8201	0.012729	0.010257	11	14.7877	0.009123	0.056917	11	115000	7	0.37823653	-11.727919	0.21489641	0.21511844	-28.908394866718794	266.4113096737104
PSO J266.4065-28.9078	-999	-999	-999	73312664065420964	3526718000011980	511109698170573	0	11710823889157	7331	1	693	70	0.519699596352595	3526718	119002	3	503635968	60	266.40653886	-28.90780678	0.001	0.001	0.00045743337	0.00045955813	55968.43169296153	28.340628	-0.05486724274815632	-0.8736774656835025	-0.4834017704043577	266.8401521219629	-5.50792945698319	0.02490079151480977	0.013636306219625104	-999	5	62	17	8	12	13	12	0.999702	17.8105	0.008651	0.018045	11	17.7764	17.8315	17.8824	0.009793	0.027138	11	17.7978	0.010912	0.026471	11	115000	0.999453	16.6078	0.003683	0.007635	5	16.6037	16.6179	16.6577	0.003096	0.01514	6	16.5928	0.008547	0.013387	6	115000	0.998781	15.9779	0.006093	0.011615	9	15.964	15.9932	16.0281	0.003923	0.006577	10	15.9685	0.003957	0.00653	10	115000	0.999352	15.5657	0.005643	0.015029	5	15.5429	15.5823	15.6124	0.010004	0.028709	6	15.5631	0.00974	0.031454	6	114992	0.999014	15.3297	0.009369	0.024988	7	15.2954	15.3601	15.4045	0.012496	0.033631	7	15.316	0.00941	0.028294	7	115000	7	-2.850673	0.9082483	0.32712114	0.32724363	-28.907813722987918	266.40652463391956
PSO J266.3928-28.9082	-999	-999	-999	73312663928400552	3526718000009518	511109698168111	0	11710823843175	7331	1	693	70	0.173465513785803	3526718	119002	3	436527104	52	266.39283765	-28.90815399	0.001	0.001	0.0027350765	0.0027070804	55928.41677632242	4.3475966	-0.05507598317521397	-0.8736613841739089	-0.48340709746611416	266.8281194934244	-5.508582914594351	0.01835232180924498	0.023685916930894287	-999	5	59	14	10	11	12	12	0.999531	21.3144	0.045576	0.148403	9	21.0593	21.4456	21.4877	0.092855	0.202394	7	21.4779	0.208816	0.521667	9	115000	0.999025	19.5711	0.005976	0.045692	6	19.5254	19.6567	19.6594	0.046031	0.155451	6	19.5611	0.033074	0.065438	6	115000	0.999369	18.6801	0.015781	0.018193	9	18.6627	18.7101	18.7583	0.021113	0.063553	9	18.6798	0.022948	0.043303	9	115000	0.998996	18.1486	0.010729	0.033271	7	18.0777	18.1695	18.2848	0.021113	0.131731	7	18.1951	0.01431	0.032569	7	114992	0.999842	17.9907	0.032032	0.127972	11	17.764	18.1452	18.221	0.010875	0.089552	9	18.123	0.016246	0.307358	11	115000	7	6.268403	2.1076665	1.980687	1.9685138	-28.908162385488396	266.39282319368255
PSO J266.4596-28.9166	-999	-999	-999	73302664596120485	3526718000021334	511109698179927	0	11710820559351	7330	1	693	70	0.260685745869471	3526718	119002	3	503635968	60	266.45961358	-28.91654044	0.001	0.001	0.00029077148	0.0002833123	55911.3653421161	112.62389	-0.05405319632704648	-0.8736543842366525	-0.4835351785246916	266.8870419988373	-5.515510611639258	0.041643385717449015	-0.030585140109685893	-999	5	76	25	8	14	14	15	0.99986	16.7213	0.003431	0.010922	22	16.6994	16.7375	16.7896	0.003058	0.013453	25	16.7142	0.003194	0.013862	25	115000	0.998375	15.9174	0.003167	0.006655	4	15.9086	15.9236	15.9624	0.002209	0.006378	6	15.9078	0.005181	0.008166	6	115000	0.998823	15.5521	0.002937	0.007409	7	15.5422	15.5607	15.5947	0.003004	0.007486	9	15.5361	0.003611	0.010244	9	115000	0.99947	15.3304	0.00306	0.007746	6	15.3226	15.345	15.3816	0.007319	0.017477	9	15.3085	0.004628	0.011875	9	114992	0.999835	15.2338	0.015708	0.039021	12	15.164	15.2347	15.308	0.014448	0.036045	12	15.2448	0.019015	0.047709	12	115000	7	-0.16482891	-8.938035	0.19212946	0.18769103	-28.91654579525221	266.4596098804736
PSO J266.4557-28.9145	-999	-999	-999	73302664557452925	3526718000020667	511109698179260	0	11710820571234	7330	1	693	70	0.881059124023709	3526718	119002	3	436527104	52	266.45574003	-28.91451411	0.001	0.001	0.0011210607	0.0010858217	55833.99970269575	5.112461	-0.054113318695001314	-0.8736678421942318	-0.4835041367511638	266.8835872042989	-5.513564406856752	0.04161213428694835	-0.02663392285882768	-999	5	78	26	10	14	14	14	0.999836	19.6969	0.011879	0.06174	26	19.6308	19.8811	19.8548	0.022247	0.130861	26	19.8507	0.032595	0.238899	26	115000	0.998936	18.4268	0.00703	0.021514	8	18.388	18.441	18.5449	0.028993	0.060575	8	18.4552	0.019929	0.055427	8	115000	0.999606	17.7905	0.001956	0.032377	9	17.7363	17.8205	17.8402	0.01708	0.022406	9	17.7678	0.004221	0.032074	9	115000	0.999379	17.3464	0.007168	0.038244	10	17.3235	17.413	17.4524	0.015814	0.063112	10	17.3674	0.012681	0.055227	10	114992	0.999119	17.1154	0.007005	0.018891	8	17.0933	17.1425	17.2308	0.019522	0.081095	8	17.1196	0.019376	0.062722	8	114992	7	-0.81038105	-5.046962	0.78591174	0.7709051	-28.91451392670525	266.4557364604022
PSO J266.4551-28.9129	-999	-999	-999	73302664550964871	3526718000020552	511109698179145	0	11710820554420	7330	1	693	70	0.665235193134505	3526718	119002	3	436527104	52	266.45509699	-28.91288967	0.001	0.001	0.0010219155	0.0010122373	55914.46624291296	3.5007098	-0.05412396863838971	-0.8736809003583604	-0.48347934843986373	266.88297906670664	-5.511955811544087	0.0427034555252411	-0.02530613701194428	-999	5	81	26	10	14	16	15	0.999565	19.6135	0.010699	0.038455	24	19.5204	19.688	19.7671	0.012574	0.079666	26	19.6898	0.021447	0.093802	26	115000	0.998738	18.3532	0.009945	0.026508	9	18.3044	18.3881	18.4768	0.021128	0.044339	9	18.39	0.018448	0.032467	9	115000	0.999439	17.6833	0.003458	0.017937	8	17.6543	17.6944	17.774	0.00861	0.020461	8	17.6885	0.00536	0.014195	8	115000	0.999269	17.2662	0.006371	0.030518	8	17.2063	17.314	17.3748	0.015131	0.05203	9	17.2957	0.005803	0.019022	9	114992	0.999628	17.0213	0.043709	0.039336	11	16.9654	17.0923	17.1849	0.01342	0.097186	10	17.076	0.005821	0.047435	12	115000	7	-2.6508658	-9.321629	0.67698455	0.6729033	-28.912891413004118	266.4550935473552
PSO J266.4544-28.9167	-999	-999	-999	73302664543610230	3526718000020428	511109698179021	0	11710820580774	7330	1	693	70	0.491885742630703	3526718	119002	3	436527104	52	266.45436476	-28.91674831	0.001	0.001	0.008073305	0.008073305	55908.53265045844	2.7811184	-0.05413325808800559	-0.873647766004334	-0.48353817980012054	266.88242547582115	-5.515820998846939	0.0390796070127061	-0.026759075737074783	-999	3	39	0	0	10	12	17	-999.0	27.062	-999.0	-999.0	0	-999.0	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	0	16420	-999.0	23.207	-999.0	-999.0	0	-999.0	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	0	16420	0.999152	21.2671	0.046245	0.129863	8	21.093	21.4584	21.3633	0.171428	0.200328	5	21.3277	0.166256	0.520896	8	115000	0.999331	19.5779	0.021768	0.077072	7	19.5217	19.7204	19.7992	0.073163	0.203064	5	19.7308	0.085828	0.209303	7	114992	0.999847	18.38	0.011155	0.163366	15	18.2675	18.6709	18.4734	0.043479	0.320322	7	18.3411	0.105891	0.423534	11	115000	7	8.720977	4.3033338	5.703452	5.703452	-28.916742248674723	266.4543525512401
PSO J266.4507-28.9145	-999	-999	-999	73302664506962987	3526718000019818	511109698178411	0	11710820570706	7330	1	693	70	0.0436405875195196	3526718	119002	3	436527104	52	266.45069911	-28.91446245	0.001	0.001	0.0013823385	0.0013434916	55846.41452771297	7.5354714	-0.054190256561050476	-0.8736634905309293	-0.4835033830359002	266.8791548937404	-5.513613621764997	0.03936616894062861	-0.02283628161515323	-999	5	79	26	9	12	17	15	0.999391	19.9817	0.017403	0.080103	23	19.8589	20.1313	20.1519	0.025571	0.110551	23	20.1126	0.030303	0.166824	23	115000	0.999708	18.7017	0.002362	0.044273	6	18.6338	18.7091	18.8178	0.024338	0.064314	6	18.7326	0.024172	0.028244	6	115000	0.998669	18.0361	0.024	0.026817	7	17.9867	18.0538	18.1383	0.006168	0.035655	8	18.0599	0.012947	0.03296	8	115000	0.999615	17.6134	0.004802	0.020473	9	17.5725	17.6384	17.6752	0.008035	0.032264	10	17.5664	0.00839	0.03023	10	114992	0.99949	17.3175	0.012027	0.040276	12	17.2206	17.34	17.3019	0.022991	0.069831	12	17.3198	0.025668	0.111352	12	114992	7	4.4773464	-0.63228834	0.9526122	0.9410091	-28.91446459205714	266.45069249505644
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
PSO J266.4163-29.1059	-999	-999	-999	73072664163533243	3526962000014375	511092518303786	0	11710814721085	7307	1	693	70	0.551312843851105	3526962	118998	3	436527104	52	266.41634343	-29.10591786	0.001	0.001	0.0041510947	0.0041510947	56040.48076540702	6.2278366	-0.05461292574532506	-0.8720134735557573	-0.486425667783638	266.85375482834365	-5.705762461567769	359.8602729082378	-0.09690259025787783	-999	5	34	2	6	10	9	7	0.998088	21.9383	0.064084	0.091509	2	21.8832	22.0114	-999.0	-999.0	-999.0	2	21.5335	0.045851	0.06546	2	114992	0.998229	20.3468	0.10348	0.206601	2	20.3464	20.5534	20.6564	0.101168	-999.0	1	20.6943	2.8e-05	-999.0	2	115000	0.999706	19.3241	0.019993	0.068775	7	19.3023	19.4687	19.4758	0.055606	0.082484	7	19.3858	0.054322	0.082105	7	115000	0.999672	18.7931	0.016066	0.045283	8	18.7227	18.8524	18.9781	0.03174	0.168035	8	18.9241	0.030154	0.17095	8	114992	0.998822	18.407	0.051268	0.195363	5	18.1872	18.726	18.6028	0.011777	0.05122	3	18.5244	0.063972	0.168108	5	115000	7	0.13514704	-9.42143	3.194401	3.194401	-29.105920574402727	266.41632986593146
PSO J266.4148-29.1040	-999	-999	-999	73072664148225512	3526962000014026	511092518303437	0	11710814739237	7307	1	693	70	0.374451386159706	3526962	118998	3	503635968	60	266.41481665	-29.10403352	0.001	0.001	0.0003020419	0.00028558227	56225.0459685395	170.21924	-0.05463701025152244	-0.872027973396164	-0.48639696825263434	266.85236925496497	-5.703915992241392	359.8611839745609	-0.09478592941077839	-999	5	63	25	8	10	10	10	0.999766	16.1048	0.002397	0.008047	16	16.0907	16.1183	16.163	0.002045	0.008641	17	16.0969	0.001671	0.005784	17	115000	0.998803	15.3339	0.004701	0.00907	4	15.3238	15.3422	15.367	0.003965	0.006855	5	15.3288	0.009557	0.02528	5	115000	0.998551	14.9267	0.002652	0.014285	5	14.9046	14.9399	14.9864	0.002246	0.00524	5	14.9211	0.001909	0.00392	5	115000	0.999602	14.6859	0.001803	0.004734	4	14.6809	14.692	14.7884	0.003364	0.009755	7	14.7052	0.014339	0.034194	7	114992	0.998807	14.5414	0.00374	0.030501	5	14.5045	14.5825	14.7396	0.074895	0.148003	6	14.7238	0.10254	0.20984	6	115000	7	5.8455477	-8.572473	0.221526	0.21033978	-29.104038571850154	266.41481303420676
PSO J266.4135-29.1027	-999	-999	-999	73072664135277120	3526962000013781	511092518303192	0	11710814740308	7307	1	693	70	0.165712189083651	3526962	118998	3	503635968	60	266.41351423	-29.10268024	0.001	0.001	0.0023281826	0.0022338706	56022.42966362217	4.0016465	-0.05465754563729599	-0.8720381583189122	-0.48637640068229737	266.85119457651757	-5.702594073247435	359.8617454071643	-0.09311305569772489	-999	4	34	0	6	9	8	11	-999.0	23.381	-999.0	-999.0	0	-999.0	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	0	16420	0.998223	20.454	0.069296	0.112992	6	20.314	20.6019	20.5586	0.083303	0.219656	6	20.4768	0.137471	0.295199	6	115000	0.999344	18.6592	0.031698	0.0264	6	18.6168	18.6832	18.8382	0.029083	0.044432	5	18.7126	0.026404	0.031134	5	115000	0.999638	17.4691	0.01401	0.062837	7	17.4529	17.5799	17.5687	0.009326	0.067788	5	17.65	0.078213	0.338006	7	114992	0.999519	16.657	0.006541	0.033099	8	16.649	16.7127	16.6753	0.029932	0.157221	8	16.637	0.012694	0.123787	8	115000	7	9.535638	1.5697489	2.062752	1.9609464	-29.102689852515642	266.41351083331534
PSO J266.4121-29.1036	-999	-999	-999	73072664120656079	3526962000013507	511092518302918	0	11710814745138	7307	1	693	70	0.686256891755626	3526962	118998	3	436527104	52	266.41204779	-29.10355202	0.001	0.001	0.0015238022	0.0014990311	56187.041710321566	17.621878	-0.05467946390609488	-0.8720294428831148	-0.48638956297592023	266.84993492214244	-5.7034954556026305	359.8603384722594	-0.09247737508697682	-999	5	68	23	10	10	12	13	0.999481	20.0238	0.019978	0.076223	18	19.9017	20.2065	20.1333	0.030687	0.1012	18	20.1173	0.038211	0.116103	18	115000	0.999324	18.6359	0.012922	0.028001	5	18.6254	18.671	18.7032	0.002522	0.029865	5	18.6022	0.009065	0.036291	5	115000	0.998768	17.8899	0.013051	0.022656	6	17.8621	17.9227	17.9911	0.003932	0.008708	5	17.8961	0.006496	0.008436	5	115000	0.999703	17.4223	0.012549	0.037071	7	17.3558	17.47	17.4908	0.007259	0.052571	7	17.3957	0.028963	0.083506	7	114992	0.999644	17.1387	0.029222	0.049898	9	17.0931	17.2227	17.265	0.008056	0.083565	9	17.1588	0.018722	0.096538	9	115000	7	7.074168	1.2938998	1.0947742	1.0652758	-29.103552968466015	266.4120406028719
PSO J266.4112-29.1036	-999	-999	-999	73072664112026015	3526962000013330	511092518302741	0	11710814745260	7307	1	693	70	0.919058672077077	3526962	118998	3	436527104	52	266.4111924	-29.10359046	0.001	0.001	0.0066784113	0.0066784113	56043.207410043695	6.0418587	-0.054692732637680624	-0.8720282240867459	-0.4863902562682965	266.8491581415507	-5.703549951542324	359.85990520923116	-0.0918349716895232	-999	4	20	0	4	6	6	4	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	0	16416	0.998448	20.7066	0.002969	0.072583	3	20.604	20.7099	20.8331	0.032097	0.103045	3	20.7162	0.008653	0.012254	3	115000	0.997551	19.6071	0.00098	0.03285	4	19.5902	19.6614	19.6907	0.017844	0.038877	4	19.5951	0.005067	0.013147	4	115000	0.99808	18.9271	0.011867	0.016789	2	18.9149	18.9386	18.8585	0.012745	0.018051	2	18.7269	0.035953	0.050861	2	16892208	0.998105	18.4949	0.074261	0.19951	3	18.4193	18.7563	18.1122	0.170505	1.90978	3	18.1511	0.077252	0.109252	3	16892216	7	4.630359	-5.8978257	4.81822	4.81822	-29.10359843123767	266.4111672070801
PSO J266.4008-29.1022	-999	-999	-999	73072664007687767	3526962000011458	511092518300869	0	11710814667589	7307	1	693	70	0.104828091435906	3526962	118998	3	436527104	52	266.40075211	-29.1021498	0.001	0.001	0.002369141	0.0022852665	55890.451726417145	6.8113775	-0.05485204825189232	-0.8720304474493089	-0.48636832907163075	266.8399783046498	-5.702343775201733	359.8563878132956	-0.08331909847796636	-999	5	68	18	10	12	14	14	0.999104	20.1944	0.046698	0.101868	8	20.0649	20.3609	20.112	0.073873	0.168596	8	19.9457	0.053272	0.146476	8	16892216	0.998755	18.9216	0.004803	0.011998	4	18.9083	18.9374	18.9755	0.031379	0.076098	4	18.8662	0.019075	0.062877	4	115000	0.997378	18.2275	0.000293	0.018876	3	18.2274	18.2542	18.3241	0.020769	0.0214	3	18.2375	0.009063	0.009773	3	115000	0.999631	17.8003	0.014252	0.086934	9	17.6716	17.9492	17.8871	0.021883	0.13519	8	17.7696	0.036284	0.11639	9	114992	0.999105	17.5306	0.017603	0.084385	10	17.3687	17.6212	17.3844	0.06027	0.22854	11	17.3833	0.055306	0.099612	11	115000	7	12.017826	-22.82726	1.5707254	1.5209632	-29.10216056111791	266.4007498140779
PSO J266.3870-29.1028	-999	-999	-999	73072663870577034	3526962000009027	511092518298438	0	11710812726148	7307	1	693	70	0.559619276374224	3526962	118998	3	436527104	52	266.3870477	-29.10276781	0.001	0.001	0.007317069	0.007317069	55965.94734317772	3.2577538	-0.055060418512013626	-0.8720120272660552	-0.486377810571809	266.8279563677007	-5.703269624074106	359.8496106784922	-0.07341820858572076	-999	4	35	0	2	13	12	8	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	-999.0	-999.0	0	-999.0	-999.0	-999.0	0	16416	0.998782	21.7853	0.079685	0.15807	2	21.6272	21.7866	21.6801	0.254074	-999.0	1	23.6265	0.000238	-999.0	2	115000	0.999735	20.456	0.026772	0.106254	9	20.2922	20.5343	20.6733	0.016541	0.196798	8	20.6198	0.072136	0.228521	9	115000	0.999795	19.3537	0.022218	0.170094	8	19.0808	19.6298	19.4186	0.04649	0.301055	7	19.2531	0.041038	0.326238	8	114992	0.998818	18.6821	0.010796	0.096289	8	18.5491	18.8102	18.6666	0.09549	0.304906	8	18.551	0.02856	0.178773	8	115000	7	6.258978	18.795458	5.204437	5.204437	-29.102782305518506	266.3870371246228
PSO J266.3863-29.1018	-999	-999	-999	73072663863428135	3526962000008899	511092518298310	0	11710814622234	7307	1	693	70	0.515454332824288	3526962	118998	3	436527104	52	266.38634245	-29.10182818	0.001	0.001	0.002171067	0.0021673369	56017.4555105817	15.218551	-0.05507162975089041	-0.8720192471729693	-0.48636359666042916	266.8273181783258	-5.702343108681077	359.8500961419852	-0.07240528190906456	-999	5	70	19	11	14	13	13	0.999826	20.8986	0.058931	0.146439	17	20.7088	21.1577	20.9007	0.093004	0.332157	16	20.8936	0.121467	0.31166	17	115000	0.999024	19.6292	0.019294	0.067861	9	19.5958	19.7479	19.7768	0.033642	0.167174	9	19.7042	0.024352	0.144579	9	115000	0.998899	18.9284	0.020745	0.033608	13	18.8624	18.9683	19.0069	0.015895	0.06393	13	18.9428	0.01902	0.072479	13	115000	0.999333	18.4165	0.012023	0.040184	9	18.3499	18.4589	18.4225	0.020362	0.07992	9	18.3285	0.018166	0.068451	9	114992	0.99965	18.0105	0.015318	0.153201	10	17.735	18.034	17.7682	0.040909	0.243463	10	17.6778	0.040975	0.185175	10	16892216	7	7.3734326	9.147327	1.5440685	1.5455455	-29.101850237140827	266.38633325346336
PSO J266.3820-29.1007	-999	-999	-999	73072663820319580	3526962000008172	511092518297582	0	11710812730948	7307	1	693	70	0.921869803147269	3526962	118998	3	436527104	52	266.38201553	-29.10063913	0.001	0.001	0.0015101681	0.0015128183	55948.55030485668	45.243683	-0.055138045861233984	-0.8720252345643736	-0.48634533634193877	266.8234856459903	-5.7012480446995	359.84913994666965	-0.06855454449437284	-999	5	50	5	9	14	13	9	0.998036	20.5914	0.001136	0.114785	3	20.5906	20.7537	20.5934	0.011114	0.184051	3	20.43	0.065058	0.127246	3	16892216	0.99972	19.1933	0.010836	0.020971	5	19.166	19.2049	19.3539	0.010426	0.041265	5	19.0154	0.038944	0.066019	5	115000	0.999161	18.3815	0.011011	0.025712	11	18.3345	18.4086	18.4751	0.015316	0.038207	11	18.3918	0.014222	0.034072	11	115000	0.998943	17.8421	0.009471	0.017642	10	17.8215	17.8739	17.9243	0.012117	0.06584	10	17.8152	0.017479	0.07322	10	16892208	0.999008	17.5881	0.011336	0.122282	8	17.4029	17.6061	17.768	0.021255	0.286346	8	17.6495	0.053613	0.478912	8	16892216	7	-11.991852	5.7948008	1.0459574	1.0523597	-29.100652840525328	266.3820115338496

Visualizing the candidate stars#

Next, we plot a preliminary \(r-i\) vs \(z\) color-magnitude diagram of our candidate stars in the Galactic Bulge.

f, ax = plt.subplots()
f.set_size_inches((4, 4))
ax.scatter(
    sample['rmeanpsfmag']-sample['imeanpsfmag'],
    sample['zmeanpsfmag'], 
    s=5, lw=0, 
    alpha=0.5
)

ax.set_xlabel("r-i, PSF Mag")
ax.set_ylabel("z, PSF Mag")

# Invert y axis limits to plot fainter -> brighter
ax.set_ylim(ax.get_ylim()[::-1])

(22.22845063209534, 7.958129072189331)

../../../_images/d8b71945a1926fb21d88295a1ca158b58ed2f1a3b9f701219714d440be071ff7.png

We see a lot of objects in our preliminary sample that are surprisingly red or blue in \(r-i\). This suggests there may be data quality cuts that need to be made for our final sample.

To investigate, we first define a color cut to roughly select these very blue or red candidates, where objects with \(r-i > 3\) or \(<0\) are flagged as potentially suspect (overplotted in red below).

f, ax = plt.subplots()
f.set_size_inches((4, 4))

# Plot full preliminary sample
ax.scatter(
    sample['rmeanpsfmag']-sample['imeanpsfmag'],
    sample['zmeanpsfmag'], 
    s=5, lw=0, 
    alpha=0.2
)

# Rough suspect target color cut
whsuspect = (
    (sample['rmeanpsfmag']-sample['imeanpsfmag'] > 3)
) | (
    (sample['rmeanpsfmag']-sample['imeanpsfmag'] < 0)
) 

# Overplot potentially suspect targets
ax.scatter(
    (sample['rmeanpsfmag']-sample['imeanpsfmag'])[whsuspect],
    sample['zmeanpsfmag'][whsuspect], 
    s=5, lw=0, 
    alpha=0.5,
    color='red'
)


ax.set_xlabel("r-i, PSF Mag")
ax.set_ylabel("z, PSF Mag")

ax.set_ylim(ax.get_ylim()[::-1])

(22.22845063209534, 7.958129072189331)

../../../_images/5c4b2c6c1e317b8fde26b88f6006c45f6ff484a45ee4f7d227f2f4fd022c7db4.png

Overall, this selection includes 65 (potentially) suspect objects.

len(sample[whsuspect])

Launching MAST-Aladin to inspect targets & outliers#

We’ll now launch MAST-Aladin to the side of our notebook, to visually examine these suspect targets over a sky map of PanSTARRS imaging to hunt for clues regarding what type of quality issues might be at play with these very red/blue objects.

# Launch MAST-Aladin
with Sidecar(title="Aladin Viewer", anchor="split-right"):
    mast_aladin = MastAladin()
    display(mast_aladin)

As we’re working with PanSTARRS data in the galactic bulge, we will also change the base survey imaging to the PanSTARRS1 imaging, and center the viewport at the center of the GBTDS imaging (with a 5 degree field of view).

(See the documentation for ipyaladin, on which MAST-Aladin is built, for examples of how to set the sky survey displayed and changing the center and FOV).

# Change base imaging survey
mast_aladin.survey = "P/PanSTARRS/DR1/color-i-r-g"

# Change viewport location and FOV (in degrees)
mast_aladin.target = gc
mast_aladin.fov = 0.1   # FOV in degrees

Adding targets to MAST-Aladin#

We then plot the full set of objects as an overlay to MAST-Aladin, for reference.

Note that we must specify the columns of our sample table the contain the RA/Dec information (raMean and decMean).

mast_aladin.add_table(
    sample,
    ra_field="ramean",
    dec_field="decmean",
    name="Sample",
    shape="circle",
    color="yellow",
)

{'type': 'table',
 'table': <Table length=2906>
        objname        objaltname1 ...       decmean             ramean      
                                   ...         deg                deg        
         object           object   ...       float64            float64      
 --------------------- ----------- ... ------------------- ------------------
 PSO J266.4148-28.9061        -999 ...  -28.90612260697708  266.4147770200182
 PSO J266.4113-28.9084        -999 ... -28.908394866718794  266.4113096737104
 PSO J266.4065-28.9078        -999 ... -28.907813722987918 266.40652463391956
 PSO J266.3928-28.9082        -999 ... -28.908162385488396 266.39282319368255
 PSO J266.4596-28.9166        -999 ...  -28.91654579525221  266.4596098804736
 PSO J266.4557-28.9145        -999 ...  -28.91451392670525  266.4557364604022
 PSO J266.4551-28.9129        -999 ... -28.912891413004118  266.4550935473552
 PSO J266.4544-28.9167        -999 ... -28.916742248674723  266.4543525512401
 PSO J266.4507-28.9145        -999 ...  -28.91446459205714 266.45069249505644
                   ...         ... ...                 ...                ...
 PSO J266.4163-29.1059        -999 ... -29.105920574402727 266.41632986593146
 PSO J266.4148-29.1040        -999 ... -29.104038571850154 266.41481303420676
 PSO J266.4135-29.1027        -999 ... -29.102689852515642 266.41351083331534
 PSO J266.4121-29.1036        -999 ... -29.103552968466015  266.4120406028719
 PSO J266.4112-29.1036        -999 ...  -29.10359843123767  266.4111672070801
 PSO J266.4008-29.1022        -999 ...  -29.10216056111791  266.4007498140779
 PSO J266.3870-29.1028        -999 ... -29.102782305518506  266.3870371246228
 PSO J266.3863-29.1018        -999 ... -29.101850237140827 266.38633325346336
 PSO J266.3820-29.1007        -999 ... -29.100652840525328  266.3820115338496,
 'options': {'ra_field': 'ramean',
  'dec_field': 'decmean',
  'name': 'Sample',
  'color': 'yellow',
  'shape': 'circle'}}

By going into the “Overlays” menu (upper left, 3 “layers” stacked symbol), this table overlay can be turned off and back on — which can be helpful when investigating the suspect targets.

Next we add the potentially suspect candidates as another overlay table.

mast_aladin.add_table(
    sample[whsuspect],
    ra_field="ramean",
    dec_field="decmean",
    name="Outliers",
    shape="circle",
    color="cyan",
)

{'type': 'table',
 'table': <Table length=65>
        objname        objaltname1 ...       decmean             ramean      
                                   ...         deg                deg        
         object           object   ...       float64            float64      
 --------------------- ----------- ... ------------------- ------------------
 PSO J266.4656-28.9322        -999 ...        -28.93224898       266.46558802
 PSO J266.4586-28.9277        -999 ...  -28.92773489101772 266.45859674997916
 PSO J266.4568-28.9370        -999 ...  -28.93704063251521 266.45683174983714
 PSO J266.4984-28.9447        -999 ... -28.944662715413212  266.4984285545497
 PSO J266.4794-28.9423        -999 ...    -28.942277513343  266.4793867087948
 PSO J266.4765-28.9445        -999 ... -28.944538035176077  266.4765353570233
 PSO J266.4765-28.9446        -999 ... -28.944550047982744 266.47653568833647
 PSO J266.4801-28.9564        -999 ...        -28.95638239       266.48011513
 PSO J266.5115-28.9589        -999 ... -28.958921355911773 266.51148801741255
                   ...         ... ...                 ...                ...
 PSO J266.4038-29.0648        -999 ...  -29.06479528659513 266.40378791522704
 PSO J266.4812-29.0680        -999 ...  -29.06799782313279  266.4812231985844
 PSO J266.3901-29.0709        -999 ... -29.070926639866524  266.3901078931627
 PSO J266.4504-29.0787        -999 ... -29.078672733721874 266.45037142120907
 PSO J266.4043-29.0785        -999 ... -29.078561806869555  266.4042465928765
 PSO J266.3988-29.0876        -999 ...  -29.08761376271378 266.39881294292843
 PSO J266.3651-29.0918        -999 ...  -29.09180716355119  266.3651002579534
 PSO J266.3526-29.0886        -999 ... -29.088616273886714  266.3525890683952
 PSO J266.4243-29.0978        -999 ... -29.097772833999297  266.4243431315527,
 'options': {'ra_field': 'ramean',
  'dec_field': 'decmean',
  'name': 'Outliers',
  'color': 'cyan',
  'shape': 'circle'}}

Adding observation footprints to MAST-Aladin#

It’s also possible to add footprints corresponding to individual observations (or reduced mosaic tiles) to MAST-Aladin, using those observations’ associated s_region metadata.

In this case, all objects in our sample fall within a single PanSTARRS tile per band. The tile boundaries can be seen by zooming out.

(This visualization of where objects fall within images can be useful when investigating whether entire images or regions of images are driving data quality issues.)

First, we query for all PanSTARRS images overlapping our cone search region.

ps_obs = Observations.query_criteria(
    obs_collection="PS1",
    coordinates=gc,
    radius=0.1*u.deg,
)

We then add these image footprints to MAST-Aladin as overlays.

colors = ['green', 'blue', 'red', 'yellow', 'orange']
for i in range(len(ps_obs)):
    mast_aladin.add_graphic_overlay_from_stcs(
        ps_obs['s_region'][i], name=ps_obs['filters'][i], 
        color=colors[i]
    )

Setting the Viewport to center a target#

In addition to panning and zooming in MAST-Aladin, and then clicking on the region circle to obtain information on a specific object (the full table row), it is also possible to programatically change the viewport center and field of view (FOV) to go to a specific target.

For example, we can center and zoom the viewport to the first potentially suspect target using it’s RA/Dec from our sample table:

ind = 0
mast_aladin.target = SkyCoord(
    sample["ramean"][whsuspect][ind],
    sample["decmean"][whsuspect][ind],
    unit=u.deg,
)
mast_aladin.fov = 30*u.arcsec.to(u.deg)  # FOV in degrees

We can also iterate through a subset of these suspect objects using a for loop going to every object together with a short time delay, as follows:

# Change viewport location for set of suspect objects:
for ind in range(20):
    mast_aladin.target = SkyCoord(
        sample["ramean"][whsuspect][ind],
        sample["decmean"][whsuspect][ind],
        unit=u.deg
    )
    time.sleep(2)

(It is also possible to delay until a key input by using instead, e.g., input("Press any key to continue")).

Flipping through the suspect objects, we see that these very red or blue targets include objects that appear to be very faint and/or irregular, or have very extreme colors (which may indicate cosmic rays or hot pixels in one or more filters); that are in a saturated region (and thus the colors may be unreliable); and also some objects that could indeed be real.

Examining quality flag information#

To avoid throwing out real stars in these very red/blue regions, we’ll now examine the PanSTARRS catalog columns and flagging flag documentation to see if there are relevant quality cuts that can be made.

From the Mean object table columns (also as seen in the MAST catalog schema browser, searching for “flag”), we see there is a qualityFlag column which contains a bitflag. The description for the flags in each bit is described in the documentation for the PanSTARRS/PS1 object flags.

There are also flag columns for each of the PanSTARRS filters, including zFlags, rFlags, iFlags.

We construct a quick list of all MeanObject table columns containing the word “flag” (case-insensitive)

# Get a quick list of all mean table columns containing "flag" (case-insensitive)
flags = []
for key in sample.keys():
    if "flag" in key.lower():
        flags.append(key)
print(flags)

['objinfoflag', 'qualityflag', 'gflags', 'rflags', 'iflags', 'zflags', 'yflags', 'astrometrycorrectionflag']

Using this column name list, we inspect the values of these flags for our suspect objects, also displaying the “objName” column.

sample[whsuspect][
    "objname",
    *flags
]

Table length=65

objname	objinfoflag	qualityflag	gflags	rflags	iflags	zflags	yflags	astrometrycorrectionflag
object	int32	uint8	int32	int32	int32	int32	int32	uint8
PSO J266.4656-28.9322	335622144	40	4	4	16	4	4	0
PSO J266.4586-28.9277	503635968	60	16420	114724	115000	114992	114992	7
PSO J266.4568-28.9370	503635968	60	16420	115768	115000	114992	115248	7
PSO J266.4984-28.9447	436527104	52	16420	114724	115000	114992	115000	7
PSO J266.4794-28.9423	436527104	52	16420	114724	115000	114992	16892216	7
PSO J266.4765-28.9445	369418240	44	16777488	272	280	4	4	7
PSO J266.4765-28.9446	436527104	52	115000	115000	114992	114736	118832	7
PSO J266.4801-28.9564	1510268928	180	16416	16794672	16892216	16892208	16416	0
PSO J266.5115-28.9589	436527104	52	16420	114724	115000	114992	115000	7
...	...	...	...	...	...	...	...	...
PSO J266.4129-29.0609	503635968	60	16420	16420	115000	114992	115000	7
PSO J266.4038-29.0648	503635968	60	16420	16420	115000	114992	115000	7
PSO J266.4812-29.0680	436527104	52	16420	16420	115000	115248	114992	7
PSO J266.3901-29.0709	503635968	60	16420	16420	115000	16892208	16892216	7
PSO J266.4504-29.0787	436527104	52	16420	16420	115000	114992	114992	7
PSO J266.4043-29.0785	436527104	52	16416	16793648	115000	114992	114720	5
PSO J266.3988-29.0876	503635968	60	16420	16420	115000	114992	16892216	7
PSO J266.3651-29.0918	436527104	52	16420	114724	115000	114992	115000	7
PSO J266.3526-29.0886	503635968	60	16420	16892464	115000	114992	114992	7
PSO J266.4243-29.0978	503635968	60	16420	16420	16793904	16892208	115000	7

While the flag description documentation provides details on the flag values and descriptions, we will also create an instance of a bitflag “helper” class PanSTARRSFlagHelper — which will help us interpret the bitflag values of individual entries, and create masks to select objects with a specific flag (or sets of flags).

(This “helper” class is defined in the accompanying python file with this notebook.)

We first obtain the flag information for thequalityFlag column, which is available in the “ObjectQualityFlags” table via the MAST PS1 DR2 TAP service.

objqalflags = PanSTARRSFlagHelper(
    tap_url="https://mast.stsci.edu/vo-tap/api/v0.1/ps1dr2", 
    table="ObjectQualityFlags", 
)

/home/runner/micromamba/envs/ci-env/lib/python3.13/site-packages/pyvo/dal/query.py:403: DALOverflowWarning: Results truncated due to server limits. Consider setting a maxrec value.
  warn("Results truncated due to server limits. Consider "

Let’s examine the flag definition info table from this helper:

objqalflags.flag_info_table

Table length=9

description	hexadecimal	name	value
object	object	object	int32
Initial value; resets all bits.	0x00000000	DEFAULT	0
Extended in our data (eg; PS).	0x00000001	QF_OBJ_EXT	1
Extended in external data (eg; 2MASS).	0x00000002	QF_OBJ_EXT_ALT	2
Good-quality measurement in our data (eg; PS).	0x00000004	QF_OBJ_GOOD	4
Good-quality measurement in external data (eg; 2MASS).	0x00000008	QF_OBJ_GOOD_ALT	8
Good-quality object in the stack (> 1 good stack measurement).	0x00000010	QF_OBJ_GOOD_STACK	16
The primary stack measurements are the best measurements.	0x00000020	QF_OBJ_BEST_STACK	32
Suspect object in the stack (no more than 1 good measurement, 2 or more suspect or good stack measurement).	0x00000040	QF_OBJ_SUSPECT_STACK	64
Poor-quality stack object (no more than 1 good or suspect measurement).	0x00000080	QF_OBJ_BAD_STACK	128

These look quite promising as far as narrowing down our sample!

The helper class has a method to expand a bitflag into a table, with separate columns for each flag. We’ll use this to expand the qualityFlag entries for our suspect objects to investigate whether some of these flags can be used to refine our sample.

objqalflags.expand_to_table(
    sample["qualityflag"][whsuspect], 
    ids=sample["objname"][whsuspect],
)

Table length=65

id	QF_OBJ_EXT	QF_OBJ_EXT_ALT	QF_OBJ_GOOD	QF_OBJ_GOOD_ALT	QF_OBJ_GOOD_STACK	QF_OBJ_BEST_STACK	QF_OBJ_SUSPECT_STACK	QF_OBJ_BAD_STACK
object	bool	bool	bool	bool	bool	bool	bool	bool
PSO J266.4656-28.9322	False	False	False	True	False	True	False	False
PSO J266.4586-28.9277	False	False	True	True	True	True	False	False
PSO J266.4568-28.9370	False	False	True	True	True	True	False	False
PSO J266.4984-28.9447	False	False	True	False	True	True	False	False
PSO J266.4794-28.9423	False	False	True	False	True	True	False	False
PSO J266.4765-28.9445	False	False	True	True	False	True	False	False
PSO J266.4765-28.9446	False	False	True	False	True	True	False	False
PSO J266.4801-28.9564	False	False	True	False	True	True	False	True
PSO J266.5115-28.9589	False	False	True	False	True	True	False	False
...	...	...	...	...	...	...	...	...
PSO J266.4129-29.0609	False	False	True	True	True	True	False	False
PSO J266.4038-29.0648	False	False	True	True	True	True	False	False
PSO J266.4812-29.0680	False	False	True	False	True	True	False	False
PSO J266.3901-29.0709	False	False	True	True	True	True	False	False
PSO J266.4504-29.0787	False	False	True	False	True	True	False	False
PSO J266.4043-29.0785	False	False	True	False	True	True	False	False
PSO J266.3988-29.0876	False	False	True	True	True	True	False	False
PSO J266.3651-29.0918	False	False	True	False	True	True	False	False
PSO J266.3526-29.0886	False	False	True	True	True	True	False	False
PSO J266.4243-29.0978	False	False	True	True	True	True	False	False

Using this flag-expanded table, we see that PanSTARRS quality flags that seem relevant, such as QF_OBJ_GOOD (False = bad) or QF_OBJ_SUSPECT_STACK or QF_OBJ_BAD_STACK (True = suspect/bad), appear to have “good” values for many of these potentially suspect objects, so these won’t be particularly useful in this case.

Instead, data quality issues for the individual bands may be driving these outliers. Thus, we’ll next examine the flags for the individual bands: zFlags, rFlags, iFlags

From the documentation for the PanSTARRS/PS1 object flags, we see this flag information is in the “ObjectFilterFlags” metadata table.

We then create a new helper class instance from this metadata, and inspect the list of flags and their descriptions:

filterflags = PanSTARRSFlagHelper(
    tap_url="https://mast.stsci.edu/vo-tap/api/v0.1/ps1dr2", 
    table="ObjectFilterFlags", 
)

We’ll display the flag descriptions in two chunks as there are 23 total flags.

# Split up as there are many flag entries:
filterflags.flag_info_table[0:10]

Table length=10

description	hexadecimal	name	value
object	object	object	int32
Initial value; resets all bits.	0x00000000	DEFAULT	0
Used within relphot: skip star.	0x00000001	SECF_STAR_FEW	1
Used within relphot: skip star.	0x00000002	SECF_STAR_POOR	2
Synthetic photometry used in average measurement.	0x00000004	SECF_USE_SYNTH	4
Ubercal photometry used in average measurement.	0x00000008	SECF_USE_UBERCAL	8
PS1 photometry used in average measurement.	0x00000010	SECF_HAS_PS1	16
PS1 stack photometry exists.	0x00000020	SECF_HAS_PS1_STACK	32
Tycho photometry used for synthetic magnitudes.	0x00000040	SECF_HAS_TYCHO	64
Synthetic magnitudes repaired with zeropoint map.	0x00000080	SECF_FIX_SYNTH	128
Average magnitude calculated in 0th pass.	0x00000100	SECF_RANK_0	256

filterflags.flag_info_table[10:23]

Table length=13

description	hexadecimal	name	value
object	object	object	int32
Average magnitude calculated in 1st pass.	0x00000200	SECF_RANK_1	512
Average magnitude calculated in 2nd pass.	0x00000400	SECF_RANK_2	1024
Average magnitude calculated in 3rd pass.	0x00000800	SECF_RANK_3	2048
Average magnitude calculated in 4th pass.	0x00001000	SECF_RANK_4	4096
PS1 stack photometry comes from primary skycell.	0x00004000	SECF_STACK_PRIMARY	16384
PS1 stack best measurement is a detection (not forced).	0x00008000	SECF_STACK_BESTDET	32768
PS1 stack primary measurement is a detection (not forced).	0x00010000	SECF_STACK_PRIMDET	65536
This photcode has SDSS photometry.	0x00100000	SECF_HAS_SDSS	1048576
This photcode has HSC photometry.	0x00200000	SECF_HAS_HSC	2097152
This photcode has CFH photometry (mostly Megacam).	0x00400000	SECF_HAS_CFH	4194304
This photcode has DES photometry.	0x00800000	SECF_HAS_DES	8388608
Extended in this band.	0x01000000	SECF_OBJ_EXT	16777216
bad stack object due to mis-aligned warp images in this skycell, all magnitudes set to -999, see stackObjectThinBadSkycells or meanObjectBadSkycells for original magnitudes, data from this skycell will be corrected in a future release	0x40000000	BAD_STACK	1073741824

We note that there are flags about whether or not synthetic photometry was used (SECF_USE_SYNTH) and whether PS1 stack photometry exists (SECF_HAS_PS1) or whether the stack best measurement is detected and not forced SECF_STACK_BESTDET. There is also a BAD_STACK flag.

As above, we’ll use this flag helper class to expand the flags for the z, r, i bands. This time we’ll expand only the flags we’ve noted above.

We’ll finish be aggregating the expanded columns into a single table for side-by-side inspection.

tfilts = None
filts = ['z', 'r', 'i']
for filt in filts:
    if tfilts is None:
        ids = sample["objname"][whsuspect]
    else:
        ids = None

    # Get expanded bitflag table for each band
    tabf = filterflags.expand_to_table(
        sample[f"{filt}flags"][whsuspect], 
        ids=ids,
        flags_to_include=["SECF_USE_SYNTH",
                          "SECF_HAS_PS1",
                          "SECF_STACK_BESTDET",
                          "BAD_STACK"],
    )

    # Create an aggregate table: list the "id":
    if tfilts is None:
        tfilts = Table({"id": tabf['id']})
        
    # Add the band/filter's flags to the aggregate table, 
    # with the filter name prepended:
    for key in tabf.keys():
        if key != "id":
            tfilts[f"{filt}_{key}"] = tabf[key]

# Reorder keys to list the same flag type together:
keys_out = ["id"]
for key in tabf.keys():
    for filt in filts:
        keys_out.append(f"{filt}_{key}")

tfilts = tfilts[keys_out]

tfilts

Table length=65

id	z_SECF_USE_SYNTH	r_SECF_USE_SYNTH	i_SECF_USE_SYNTH	z_SECF_HAS_PS1	r_SECF_HAS_PS1	i_SECF_HAS_PS1	z_SECF_STACK_BESTDET	r_SECF_STACK_BESTDET	i_SECF_STACK_BESTDET	z_BAD_STACK	r_BAD_STACK	i_BAD_STACK
object	bool	bool	bool	bool	bool	bool	bool	bool	bool	bool	bool	bool
PSO J266.4656-28.9322	True	True	False	False	False	True	False	False	False	False	False	False
PSO J266.4586-28.9277	False	True	False	True	False	True	True	True	True	False	False	False
PSO J266.4568-28.9370	False	False	False	True	True	True	True	True	True	False	False	False
PSO J266.4984-28.9447	False	True	False	True	False	True	True	True	True	False	False	False
PSO J266.4794-28.9423	False	True	False	True	False	True	True	True	True	False	False	False
PSO J266.4765-28.9445	True	False	False	False	True	True	False	False	False	False	False	False
PSO J266.4765-28.9446	False	False	False	True	True	True	True	True	True	False	False	False
PSO J266.4801-28.9564	False	False	False	True	True	True	True	False	True	False	False	False
PSO J266.5115-28.9589	False	True	False	True	False	True	True	True	True	False	False	False
...	...	...	...	...	...	...	...	...	...	...	...	...
PSO J266.4129-29.0609	False	True	False	True	False	True	True	False	True	False	False	False
PSO J266.4038-29.0648	False	True	False	True	False	True	True	False	True	False	False	False
PSO J266.4812-29.0680	False	True	False	True	False	True	True	False	True	False	False	False
PSO J266.3901-29.0709	False	True	False	True	False	True	True	False	True	False	False	False
PSO J266.4504-29.0787	False	True	False	True	False	True	True	False	True	False	False	False
PSO J266.4043-29.0785	False	False	False	True	True	True	True	False	True	False	False	False
PSO J266.3988-29.0876	False	True	False	True	False	True	True	False	True	False	False	False
PSO J266.3651-29.0918	False	True	False	True	False	True	True	True	True	False	False	False
PSO J266.3526-29.0886	False	False	False	True	True	True	True	True	True	False	False	False
PSO J266.4243-29.0978	False	True	False	True	False	True	True	False	False	False	False	False

Leveraging this breakdown, we now print aggregate counts for each flag with the number of objects with that flag set to True/False.

for key in tfilts.keys():
    if key != "id":
        strout = f"{key:>20}:   True: {np.sum(tfilts[key]):>3}"
        strout += f",  False: {np.sum(~tfilts[key]):>3}"
        print(strout)

    z_SECF_USE_SYNTH:   True:   9,  False:  56
    r_SECF_USE_SYNTH:   True:  46,  False:  19
    i_SECF_USE_SYNTH:   True:   0,  False:  65
      z_SECF_HAS_PS1:   True:  56,  False:   9
      r_SECF_HAS_PS1:   True:  19,  False:  46
      i_SECF_HAS_PS1:   True:  65,  False:   0
z_SECF_STACK_BESTDET:   True:  54,  False:  11
r_SECF_STACK_BESTDET:   True:  23,  False:  42
i_SECF_STACK_BESTDET:   True:  54,  False:  11
         z_BAD_STACK:   True:   0,  False:  65
         r_BAD_STACK:   True:   0,  False:  65
         i_BAD_STACK:   True:   0,  False:  65

We see that none of the suspect objects are marked as “bad” stacks, that some of the stack values are forced (and not detected, suggesting the objects are very faint in that filter), and also a decent number have no PS1 data & instead have synthetic data.

Revisiting the schema for the “mean” table, we note that there are columns nz, nr, ni, that contain the number of single-epoch detections in each filter.

sample[whsuspect]["objname", "nz", "nr", "ni"]

Table length=65

objname	nz	nr	ni
object	int16	int16	int16
PSO J266.4656-28.9322	0	0	1
PSO J266.4586-28.9277	8	0	9
PSO J266.4568-28.9370	6	3	10
PSO J266.4984-28.9447	11	0	11
PSO J266.4794-28.9423	13	0	14
PSO J266.4765-28.9445	13	3	8
PSO J266.4765-28.9446	1	8	4
PSO J266.4801-28.9564	3	1	6
PSO J266.5115-28.9589	11	0	13
...	...	...	...
PSO J266.4129-29.0609	12	0	10
PSO J266.4038-29.0648	14	0	10
PSO J266.4812-29.0680	4	0	3
PSO J266.3901-29.0709	12	0	4
PSO J266.4504-29.0787	7	0	6
PSO J266.4043-29.0785	10	1	7
PSO J266.3988-29.0876	12	0	9
PSO J266.3651-29.0918	13	0	13
PSO J266.3526-29.0886	11	1	11
PSO J266.4243-29.0978	9	0	2

By a quick aggregate count, we see that quite a few of these outliers (65 total) have either 0 or only 1 r detection, and some for z and i as well.

for filt in filts:
    for val in [0, 1]:
        strout = f"n{filt}: number = {val}: "
        strout += f"{len(np.where(sample[whsuspect][f'n{filt}'] == val)[0])}"
        print(strout)

nz: number = 0: 6
nz: number = 1: 6
nr: number = 0: 46
nr: number = 1: 9
ni: number = 0: 0
ni: number = 1: 7

With only 1 detection, the magnitudes in those bands could be driven by data/noise artifacts, while those with 0 detections could be impacted by issues in the synthetic magnitudes derived from other data.

Refining the sample with data quality cuts#

As astroquery.mast currently doesn’t support bitflag-based filtering on only a subset of bits, we’ll first examine whether restriting our sample to only objects with \(\geq 2\) detections in each band is sufficient to resolve these suspect outliers.

We make a mask to select only the subset of our sample with >= 2 detections in all 3 bands.

whmultdet = (
    (sample['nz'] >= 2)
    & (sample['nr'] >= 2)
    & (sample['ni'] >= 2)
)

We then replot our sample, marking the suspect outliers, in two ways:

Showing all objects in the preliminary sample, with objects without at least 2 detections in all three bands outlined in yellow (left panel).
Showing only objects with at least 2 detections in all 3 bands, with color-outlier objects marked in red as before (right panel).

# Rough selection of suspect targets
f, axes = plt.subplots(1, 2)
f.set_size_inches((8.8, 4))

ax = axes[0]
ax.set_title("Preliminary sample")
ax.scatter(
    (sample['rmeanpsfmag']-sample['imeanpsfmag']),
    sample['zmeanpsfmag'], 
    s=5, lw=0, 
    alpha=0.2
)

ax.scatter(
    (sample['rmeanpsfmag']-sample['imeanpsfmag'])[whsuspect],
    sample['zmeanpsfmag'][whsuspect], 
    s=5, lw=0, 
    alpha=0.5,
    color='red'
)

ax.scatter(
    (sample['rmeanpsfmag']-sample['imeanpsfmag'])[~whmultdet],
    sample['zmeanpsfmag'][~whmultdet], 
    s=7, lw=1, 
    alpha=0.5,
    edgecolor='yellow', 
    zorder=-10
)


ax.set_xlabel("r-i, PSF Mag")
ax.set_ylabel("z, PSF Mag")

ax.set_ylim(ax.get_ylim()[::-1])

xlim = ax.get_xlim()
ylim = ax.get_ylim()


ax = axes[1]
ax.set_title(r"Sample with $\geq2$ detections in all bands")
ax.scatter(
    (sample['rmeanpsfmag']-sample['imeanpsfmag'])[whmultdet],
    sample['zmeanpsfmag'][whmultdet], 
    s=5, lw=0, 
    alpha=0.2
)

ax.scatter(
    (sample['rmeanpsfmag']-sample['imeanpsfmag'])[whmultdet & whsuspect],
    sample['zmeanpsfmag'][whmultdet & whsuspect], 
    s=5, lw=0, 
    alpha=0.5,
    color='red'
)


ax.set_xlabel("r-i, PSF Mag")
ax.set_ylabel("z, PSF Mag")

ax.set_xlim(xlim)
ax.set_ylim(ylim)

(22.22845063209534, 7.958129072189331)

../../../_images/a42d5252a3008e06a6aec6bbd8c5d506fe77afe963ab59550f012f5a297f8114.png

Many of our “color-outlier” suspect targets — as well as some others we hadn’t flagged — are removed by requiring more than one detection in each relevant band! The resulting sample is much closer to a clean color magnitude diagram.

These constraints are something that could have be applied as part of the initial query.

Further, similar checks can reveal whether an even more refined sample could be obtained by filtering on the information from the flags explored here, or if additional flags from, e.g., the individual detection catalog. This is left as an exercise for interested readers!

Additional Resources#

Astroquery.mast#

Access to MAST holdings, including catalogs, through the Astropy-affiliated astroquery package.
A full list of available MAST Astroquery catalog services is available through the astroquery documentation.

PanSTARRS 1 DR 2#

See the PanSTARRS documentation for full details regarding PanSTARRS DR2 catalogs.

Citations#

If you use astropy for published research, please cite the authors. Follow these links for more information about citing astropy:

Citing astropy

If you use PanSTARRS data accessed through MAST for published research, please include the following acknowledgements, found at the following links:

About This Notebook#

The use of MAST-Aladin in this notebook draws from the demonstrations in “Intro to MastAladinLite” by Hatice Karatay.

Authors: Sedona Price
Keywords: Tutorial, astroquery, MAST-Aladin, bitflags
Last updated: August 2026

Top of Page