Finding X-ray observations for a sample of SNe 30 days either side of discovery

This case study searches for X-ray observations (from XMM, Chandra, and Swift; but it is applicable to all DAXA missions) of Type 1a Supernovae locations, within a time window around each SNe’s discovery date. This process is applicable to any search for an observation of a specific place at a specific time (or a sample of places at a set of different times).

Import Statements

[1]:

import numpy as np
import pandas as pd
from astropy.coordinates import SkyCoord
from datetime import datetime, timedelta

from daxa.mission import XMMPointed, Chandra, Swift
from daxa.archive import Archive

Sample

We read in a demonstrative sample, which consists of 1000 randomly selected Type 1a Supernovae from the transient name server (TNS); there are many columns, but the only ones we require are position (Right-Ascension and Declination) and the discovery date (which we will search around):

[2]:

samp = pd.read_csv("samp_files/sn1a_samp.csv")
samp.head(5)

[2]:

	objid	name_prefix	name	ra	declination	redshift	typeid	type	reporting_groupid	reporting_group	...	source_group	discoverydate	discoverymag	discmagfilter	filter	reporters	time_received	internal_names	creationdate	lastmodified
0	133295	SN	2023ock	232.474130	66.075011	0.039	3.0	SN Ia	48.0	ZTF	...	ZTF	2023-07-26 08:16:47.000	19.2400	110.0	g	C. Fremling (Caltech) on behalf of the Zwicky ...	2023-07-28 07:24:04	ZTF23aauqmys, ATLAS23pvm	2023-07-28 07:24:05	2023-08-06 05:49:07
1	74545	SN	2021dj	208.777227	54.304902	0.070	3.0	SN Ia	48.0	ZTF	...	ZTF	2021-01-01 11:24:00.000	19.2500	111.0	r	C. Fremling (Caltech) on behalf of the Zwicky ...	2021-01-03 22:01:11	ZTF21aaabucr, ATLAS21aza, PS21xg	2021-01-03 22:01:13	2021-02-07 11:04:35
2	56011	SN	2020enj	155.237795	0.573764	0.104	3.0	SN Ia	18.0	ATLAS	...	ATLAS	2020-03-16 10:59:31.200	19.2350	72.0	orange	J. Tonry, L. Denneau, A. Heinze, H. Weiland, H...	2020-03-16 18:48:30	ATLAS20hza, PS20agg, ZTF20aaubotx	2020-03-16 18:48:32	2020-03-16 18:48:32
3	74416	SN	2021E	62.978428	16.803008	0.050	3.0	SN Ia	74.0	ALeRCE	...	ZTF	2021-01-01 03:34:56.997	19.9476	110.0	g	F. Forster, F.E. Bauer, A. Munoz-Arancibia, G....	2021-01-01 16:43:48	ZTF21aaaalaf, ATLAS21ash	2021-01-01 16:43:53	2021-01-13 00:38:47
4	49744	SN	2019xck	97.070105	23.608591	0.035	3.0	SN Ia	74.0	ALeRCE	...	ZTF	2019-12-19 09:51:42.000	19.3648	110.0	g	F. Forster, F.E. Bauer, G. Pignata, J. Arredon...	2019-12-19 16:00:58	ZTF19aczeomw, ATLAS19bdqm	2019-12-19 16:01:09	2019-12-19 16:01:09

5 rows × 21 columns

We read out the coordinates into an astropy coordinate object, and set up the time windows we will be searching for each SNe - for this demonstration we will search for X-ray observations 30 days either side of the disovery date:

[3]:

pos = SkyCoord(samp['ra'].values, samp['declination'].values, unit='deg')
start_times = np.array([datetime.strptime(dd, "%Y-%m-%d %H:%M:%S.%f") - timedelta(days=30)
                        for dd in samp['discoverydate'].values])
end_times = np.array([datetime.strptime(dd, "%Y-%m-%d %H:%M:%S.%f") + timedelta(days=30)
                      for dd in samp['discoverydate'].values])

Defining missions

We create instances of the XMM, Chandra, and Swift missions in order to search their archives - Swift is the most likely to have many matching observations, as it acts as a transient follow-up telescope, but XMM and Chandra are workhorses and may have some observations that we might want to explore:

[4]:

xm = XMMPointed()
ch = Chandra()
sw = Swift()

/Users/dt237/code/DAXA/daxa/mission/xmm.py:83: UserWarning: 140 of the 17697 observations located for this mission have been removed due to NaN RA or Dec values
  self._fetch_obs_info()
/Users/dt237/code/DAXA/daxa/mission/swift.py:101: UserWarning: 598 of the 353616 observations located for Swift have been removed due to all instrument exposures being zero.
  self._fetch_obs_info()
/Users/dt237/code/DAXA/daxa/mission/swift.py:101: UserWarning: 17 of the 353616 observations located for Swift have been removed due to all chosen instrument (XRT, BAT) exposures being zero.
  self._fetch_obs_info()

Searching for observations

We will make use of the DAXA filtering method that allows us to search for observations of a particular coordinate, within a particular time frame, for a whole sample. It is called the same way for the three missions we are using, we pass the positions, start times, and end times, and the keyword arguments have the following meanings:

return_obs_info - If True, a dataframe is returned from the method to allow the user to link specific ObsIDs to particular entries in our original sample table. The dataframe contains a ‘pos_ind’ column, which contains indexes corresponding to the input positions (i.e. the 4th entry of pos would have index 3), it also contains ObsIDs matched to that coordinate and time window.
over_run - If True, observations that start or end outside of the specified time window are accepted. If False, they are not.

[5]:

xm_assoc = xm.filter_on_positions_at_time(pos, start_times, end_times, return_obs_info=True, over_run=True)
ch_assoc = ch.filter_on_positions_at_time(pos, start_times, end_times, return_obs_info=True, over_run=True)
sw_assoc = sw.filter_on_positions_at_time(pos, start_times, end_times, return_obs_info=True, over_run=True)

/Users/dt237/code/DAXA/daxa/mission/base.py:1389: UserWarning: Every value in the filter array is False, meaning that no observations remain.
  self.filter_array = np.full(self.filter_array.shape, False)
/Users/dt237/code/DAXA/daxa/mission/base.py:1075: UserWarning: Chandra FoV are difficult to define, as they can be strongly dependant on observation mode; as such take these as very approximate.
  fov = self.fov
/Users/dt237/code/DAXA/daxa/mission/base.py:1389: UserWarning: Every value in the filter array is False, meaning that no observations remain.
  self.filter_array = np.full(self.filter_array.shape, False)
/Users/dt237/code/DAXA/daxa/mission/base.py:97: UserWarning: There are multiple chosen instruments XRT, BAT for swift with different FoVs, but they observe simultaneously. As such the search distance has been set to the largest FoV of the chosen instruments.
  any_ret = change_func(*args, **kwargs)
/Users/dt237/code/DAXA/daxa/mission/base.py:1389: UserWarning: Every value in the filter array is False, meaning that no observations remain.
  self.filter_array = np.full(self.filter_array.shape, False)

Identified observations

We will now use the returns from the filtering methods to highlight the observations which have been identified as fulfilling our criteria:

[6]:

xm_assoc['sn_name'] = samp.loc[xm_assoc['pos_ind'].values.astype(int), 'name'].values
xm_assoc

[6]:

	pos_ind	pos_ra	pos_dec	ObsIDs	sn_name
30	414	158.42781249083353	39.49061071686476	0824030101	2018hus

[7]:

ch_assoc['sn_name'] = samp.loc[ch_assoc['pos_ind'].values.astype(int), 'name'].values
ch_assoc

[7]:

	pos_ind	pos_ra	pos_dec	ObsIDs	sn_name
10	45	219.741280784	51.0827058066	21697	2020hyi
41	142	234.553792	39.732811	22528	2019rmq
44	160	161.2694701	2.31901835	22494	2020kxf
114	429	186.495176661	7.23543298389	23771	2021ita
166	664	206.843520969	26.384687742	27023,27026,27021,27022,27024,27029,27020,27806	2023hrk
174	706	222.24025	18.326689	22659	2021min
196	783	193.82175	2.897311	23563	2021pkz
219	894	222.905086576	18.9261358766	26962	2023jgq

[8]:

sw_assoc['sn_name'] = samp.loc[sw_assoc['pos_ind'].values.astype(int), 'name'].values
sw_assoc

[8]:

	pos_ind	pos_ra	pos_dec	ObsIDs	sn_name
1	1	208.77722506346083	54.30490205010956	00013608061,00013608056,00013608059,0001360805...	2021dj
2	2	155.23780171544678	0.5737592936995243	00095611001,00075041008	2020enj
7	7	0.8142941780660411	16.14571038076381	00014427002,00014427004,00014427001,0001442700...	2021rhu
10	10	250.60306782108816	78.91498537452249	00095179001	2019kyz
17	17	219.49858839000424	9.388540565458925	00013556003,00013556005,00013556002,00013556004	2020lil
...	...	...	...	...	...
873	955	267.9480465961222	44.90420978372445	00074924063,00074924061,00075754012	2022ucs
879	961	295.4186780442415	-21.26268560353943	00016144004,00016144003,00016144001,00016144002	2023mvl
887	970	39.01071994163291	43.472059484607946	00010346005,00010346003,00010346008,0001034601...	2017hjy
905	990	347.3879247185516	15.65927267376877	00013713001,00013713003,00013713002,0001371300...	2020szr
909	995	355.94687610671747	51.241493712166886	00081310002,00035031184,00035031190,0003503118...	2020rsi

119 rows × 5 columns

Defining an archive

The filtered missions can then be used to define an archive containing the selected data:

[9]:

arch = Archive('sne_search', [xm, ch, sw])
arch.info()

Downloading XMM-Newton Pointed data: 100%|██████████████████████████████████████| 1/1 [00:22<00:00, 22.25s/it]
Downloading Chandra data: 100%|███████████████████████████████████████████████| 15/15 [00:38<00:00,  2.56s/it]
Downloading Swift data: 100%|███████████████████████████████████████████████| 454/454 [07:33<00:00,  1.00it/s]


-----------------------------------------------------
Number of missions - 3
Total number of observations - 470
Beginning of earliest observation - 2016-02-21 03:25:58
End of latest observation - 2024-01-05 07:08:51

-- XMM-Newton Pointed --
   Internal DAXA name - xmm_pointed
   Chosen instruments - M1, M2, PN
   Number of observations - 1
   Fully Processed - False

-- Chandra --
   Internal DAXA name - chandra
   Chosen instruments - ACIS-I, ACIS-S, HRC-I, HRC-S
   Number of observations - 15
   Fully Processed - False

-- Swift --
   Internal DAXA name - swift
   Chosen instruments - XRT, BAT
   Number of observations - 454
   Fully Processed - False
-----------------------------------------------------