Draft version August 17, 2023

Typeset using LATEX twocolumn style in AASTeX631

Stable Coronal X-Ray Emission Over Twenty Years of XZ Tau

Steven M. Silverberg,1 Hans Moritz Günther,2 Pragati Pradhan,3 David A. Principe,2 P. C. Schneider,4 and
Scott J. Wolk1
1 Smithsonian Astrophysical Observatory, MS 70, 60 Garden St., Cambridge, MA 02138
2 MIT Kavli Institute for Astrophysics and Space Research, 77 Massachusetts Avenue, Cambridge, MA 02139, USA
3 Embry-Riddle Aeronautical University, Prescott, AZ, USA
4 Hamburger Sternwarte, Gojenbergsweg 112, D-21029, Hamburg, Germany
arXiv:2308.07981v1 [astro-ph.SR] 15 Aug 2023

ABSTRACT
XZ Tau AB is a frequently observed binary YSO in the Taurus Molecular Cloud; XZ Tau B has
been classified as an EXOr object. We present new Chandra/HETG-ACIS-S observations of XZ
Tau AB, complemented with variability monitoring of the system with XMM-Newton, to constrain
the variability of this system and identify high-resolution line diagnostics to better understand the
underlying mechanisms that produce the X-rays. We observe two flares with XMM-Newton, but find
that outside of these flares the coronal X-ray spectrum of XZ Tau AB is consistent over twenty years
of observations. We compare the ensemble of XZ Tau X-ray observations over time with the scatter
across stars observed in point-in-time observations of the Orion Nebula Cluster and find that both
overlap in terms of plasma properties, i.e., some of the scatter observed in the X-ray properties of
stellar ensembles stems from intrinsic source variability.

Keywords: Young star, X-ray, flare star Facilities: Chandra, XMM-Newton

1. INTRODUCTION or even the “consumption” of tidally disrupted proto-

Stars and their circumstellar disks form as a result of planets.
gravitational contraction of molecular clouds. The early X-ray emission is ubiquitous among low-mass pre-MS
stages of low-mass (< 2M⊙ ) stellar evolution are charac- stars due to their convective zones and fast rotational
terized by violent accretion events, large molecular out- periods which generate strong magnetic dynamos. In
flows and jets as star-disk systems clear their surround- particular, several distinct physical mechanisms are ca-
ing environment. A subset of pre-MS stars undergoing pable of producing X-rays in pre-MS stars: magneti-
this violent accretion are part of an observationally rare cally heated coronae with characteristic temperatures of
but important class called FUor and EXor (after their ∼ 0.1 − 10.0 keV that result in continuum and line emis-
namesakes FU Ori and EX Lup). These objects undergo sion dependent on coronal abundances (e.g. Preibisch
extreme mass accretion events (∆Ṁ ∼ 102−4 ) where et al. 2005; Scelsi et al. 2007), shock-heated plasma (at
their optical brightness can increase by ∆V ∼ 3 − 5 characteristic temperatures of ∼ 0.3 keV) as a result of
(Audard et al. 2014). FUor and EXor objects are dis- mass funneled from the circumstellar disk along mag-
tinguished by their outburst intensity and frequency: netic field lines and onto the star (see review by Schnei-
FUors exhibit brighter ∆V ∼ 3 − 5 mag events which der et al. 2022), and star-disk magnetic reconnection
last years-centuries, while EXors undergo less intense events which can magnetically heat material to tem-
but more frequent outbursts on timescales of months- peratures in excess of ∼ 8.6 keV (Favata et al. 2005).
years. While it is clear FUor/EXors drastically increase A multiwavelength campaign to study the outburst of
in magnitude as a result of sudden accretion, it is de- FUor/EXor type object V1647 Ori detected a ∼ 2 order
bated as to why these objects undergo these intense of magnitude increase in X-ray flux associated with star-
mass accretion events. Several theories have been pro- disk magnetospheric interactions (Kastner et al. 2004).
posed including: disk fragmentation (Vorobyov 2013) A handful of cases have also been identified where an-
and/or perturbations from a binary component or mas- gular momentum is lost as jets launched from the sys-
sive planet causing disk instabilities (Clarke et al. 1990) tem during the Class II stage, which are also seen in
X-rays (Pravdo et al. 2001; Favata et al. 2002; Schnei-
2 Silverberg et al.

der et al. 2011), mostly from plasma with kT ≈ 0.3 keV the XZ Tau A system, typically a signature of youth
(e.g. Grosso et al. 2006; Bonito et al. 2007). This X-ray (Zapata et al. 2015; Arce & Sargent 2006). The 1.3mm
emission likely comes from shock-heated material trav- continuum detection of XZ Tau B indicates an unusu-
eling away from the source with velocities in excess of ally small circumstellar disk with a radius of only 3.6 AU
300 km s−1 (see, e.g., DG Tau, Güdel et al. 2008; Schnei- and an inner cavity of 1.3 AU that the authors attribute
der & Schmitt 2008). X-ray observations of FUor/EXor to ongoing planet formation (Osorio et al. 2016). Such
type objects have revealed indications of X-ray emitting a small disk may be unable to sufficiently shield itself
jets (Stelzer et al. 2009, Z Cma), X-ray bright accre- from intense X-ray irradiation and thus the impact on
tion hot spots (Hamaguchi et al. 2012), and magnetic planet formation should be investigated. The classifica-
reconnection events (Kastner et al. 2004). tion of XZ Tau B as an EXor object (Coffey et al. 2004;
Eventually, pre-MS stars clear out their surrounding Lorenzetti et al. 2009) in combination with its unusually
molecular envelopes and disks and thus lose many of the high accretion rate and small disk suggests this pre-MS
above physical mechanisms capable of producing X-rays star may be going through rapid stellar evolution and
even while they remain coronally active. Therefore, it its disk may not survive for much longer.
is imperative to investigate features during this young, XZ Tau AB has been observed with multiple X-ray
embedded stage of stellar evolution to understand how instruments between 1990 and the present and displays
stars evolve and, in particular, investigate what impact both long and short term variability. None of these ob-
these sources of X-rays have on circumstellar disks and servations are able to resolve the ∼ 0.3′′ binary; Chan-
the eventual formation of planets (Owen et al. 2011; dra’s PSF comes close but does not achieve it, while
Skinner & Güdel 2013; Cleeves et al. 2013). While nu- XMM-Newton is incapable of it, as seen in Figure 1.
merous studies have been published revealing a wealth Favata et al. (2003) reported variability during their
of data regarding star formation (Getman et al. 2005; ∼ 50 ks observation of XZ Tau AB. While the count
Güdel et al. 2007), the vast majority of the observa- rate increased linearly during the observation, NH de-
tional work has been limited to single snapshots in time creased from 1.06 × 1022 cm−2 to 0.26 × 1022 cm−2 while
for a variety of objects, rather than tracking particular the plasma temperature increased by a similar factor.
sources over longer time periods. One potential source An emission measure ratio of ∼ 1300 between the soft
to analyze over this longer time frame is the YSO binary (0.14 keV) and hard (2.29 keV) temperature compo-
system XZ Tau. nent at the onset of the count rate increase strongly
suggest the presence of either accretion or jet emission.
1.1. The XZ Tau AB system In 2006, a five day monitoring campaign of XZ Tau AB
XZ Tau AB1 is a close separation binary with a ∼ 0.3′′ displayed only low levels of NH ∼ 0.1 × 1022 cm−2 with
or 42 au (Haas et al. 1990) composed of an M1 and an kTX1 , kTX2 = 0.84 keV and 4.6 keV respectively, al-
M2 pre-MS star. Each member hosts its own circumstel- though spectral analysis was complicated by extensive
lar disk (Zapata et al. 2015), and their small separation background emission during the observation. Giardino
likely induces disk disruptions leading to mass accretion et al. (2006) presented a re-analysis of the Favata et al.
events. Optical spectroscopy resolving the binary has (2003) data, showing that the variable X-ray spectrum
demonstrated that each member is a highly accreting could be fit with either a low or high NH depending on
source (White & Ghez 2001). Moreover, the spectrum the coronal abundances assumed in the model. However,
of XZ Tau B is heavily veiled and shows spectral fea- none of these observations incorporate high-resolution
tures similar to that of DG Tau, a star-disk system with gratings spectra, potentially capable of breaking degen-
intense accretion and an observed jet resolved at both eracies between X-ray emission mechanisms by enabling
optical and X-ray wavelengths (Güdel et al. 2008). Both resolution of temperature- and density-sensitive line ra-
XZ Tau A and XZ Tau B exhibit jets at optical wave- tios (e.g. Mg XI, Ne IX, OVII; Huenemoerder et al.
lengths (Krist et al. 2008) while a complex bubble-like 2007; Kastner et al. 2002; Brickhouse et al. 2010).
system encompasses both stars. In this paper, we present new observations of XZ Tau
ALMA 1.3mm continuum observations have detected AB with Chandra/HETG, complemented with observa-
dust emission associated with both circumstellar disks. tions with XMM-Newton collected as part of a larger
12CO emission traces collimated outflows surrounding X-ray monitoring campaign on the Taurus star-forming
region to constrain variability observed in the Chandra
observations. Using these data, we analyze the present-
1 For clarity, we refer to the combined binary system as XZ Tau day state of XZ Tau AB, in comparison to previous ob-
AB, while each individual component is referred to as XZ Tau A
or XZ Tau B, respectively. servations. We also present contemporaneous ground-
 A Stable Coronal Spectrum on XZ Tau 3

based optical observations to assess whether XZ Tau AB and HL Tau systems, we requested observations of
AB was in outburst during the most recent Chandra XZ Tau AB and HL Tau from the Association of Am-
and XMM-Newton observations. ateur Variable Star Observers (AAVSO) over the time
We discuss the observations and data reduction meth- periods of observation by Chandra and XMM. These
ods in Sect. 2. In Section 3 we summarize the charac- observations, distributed across multiple observers, pro-
teristics of our observed X-ray data. In Section 4 we vide low-cadence optical monitoring over the course of
outline the implications of these observations, and com- the observations. While the AAVSO data do not resolve
pare our assessments to past work. We summarize our the separate components of XZ Tau A and B, Krist et al.
work in Section 5. (2008) resolves the two components with HST and find
that XZ Tau A is relatively stable (R magnitude changes
2. OBSERVATIONS AND DATA REDUCTION ∼ 0.6 mag between 1995 and 2004), while XZ Tau B can
We summarize our new observations in Table 1. Be- exhibit wide variations (R magnitude changes ∼ 3.23
low, we briefly describe the observations and data re- mag between 2001 and 2004), which suggests that the
duction. bulk of the intrinsic variability is due to XZ Tau B.

3. ANALYSIS
3.1. Identifying Components
2.1. Chandra
We attempted to determine if the two components of
XZ Tau AB was observed by Chandra eleven times XZ Tau AB were resolvable in the Chandra data, using
over the span of three weeks, from 2018 Octo- the positions for the stars recorded by ALMA (Ichikawa
ber 24 through 2018 November 12, with the High- et al. 2021). While XZ Tau A and B are separated by
Energy Transmission Grating Spectrograph (HETGS) ∼ 0.3′′ (Ichikawa et al. 2021), the Chandra and XMM
(Canizares et al. 2005). The aimpoint was centered be- point spread functions (PSFs) are too wide to precisely
tween XZ Tau AB and HL Tau, with the goal of observ- pinpoint whether the X-rays are coming from one star,
ing both sources in parallel. Data were reduced with the or both–the center of the X-ray source in the Chandra
Chandra Interactive Analysis software (CIAO; v. 4.14). data is offset from both of the two ALMA sources (0′′ .35
The observations were energy filtered (0.5-8.0 keV) and north of XZ Tau A, 0′′ .26 east of XZ Tau B). We thus
time-filtered on good time intervals to reduce flaring treat all of our observations as the combined spectrum
particle background. Zeroth-order and gratings spectra of both components.
were extracted with standard procedures in CIAO.
3.2. Was XZ Tau AB in Outburst During the New
2.2. XMM-Newton X-Ray Observations?
XZ Tau AB was observed by the X-ray Multimirror We considered the full light curve for XZ Tau pro-
Mission (XMM-Newton) observatory six times over the vided by the AAVSO, and used it to generate color light
course of 33 days from 2020 August 18 through 2020 curves for each observation. Analyzing color as a func-
September 19, as part of a larger campaign to monitor tion of time as well as the observed brightness should
variability of young stellar objects in Taurus (PI Schnei- mitigate uncertainty in recorded brightness due to dif-
der). The observations used the medium thickness opti- ferences in observing setup for each observer. It also
cal blocking filter. XZ Tau and HL Tau were extracted allows for evaluation of the extinction of the star as a
using standard procedures in SAS version 19.1.0. Be- function of time, which yields insight into the behavior
cause of the close proximity of XZ Tau AB and HL Tau, of the circumstellar dust. To supplement the AAVSO
we defined custom extraction regions to ensure minimal light curves and serve as a “ground truth” measure of
contamination of each source by the other source. The XZ Tau AB’s brightness, we also considered the V-band
observations were energy filtered (0.3-8.0 keV) and time- light curve of XZ Tau from the ASAS-SN catalog of vari-
filtered on good time intervals to reduce flaring particle able stars (Shappee et al. 2014; Jayasinghe et al. 2019).
background. Based on Hubble Space Telescope images of the re-
solved components of the XZ Tau system, Giardino et al.
2.3. AAVSO (2006) found that while XZ Tau B was potentially in
The two components of XZ Tau AB are known to be outburst when the XZ Tau system was first observed
variable in the optical over the course of years, indi- with XMM-Newton in 2000, it had clearly subsided by
cating potential outbursts expected of an ExOr object the time of the 2004 XMM-Newton observations. Krist
(Krist et al. 2008). To track the state of the XZ Tau et al. (2008) found similar results.
4 Silverberg et al.

Figure 1. XZ Tau and its environs; all three panels are on the same spatial scale. Left: 2004 HST/ACS/F625W image of XZ
Tau AB. Middle: 2018 Chandra/ACIS observations (obsid 21948 shown here) do not resolve XZ Tau AB, but have a tight PSF
with low background. Right: 2020 XMM-Newton observations (obsid 0865040201 shown here) have more counts, but a much
wider PSF.

Table 1. Summary of New Observations of XZ Tau AB

obsid Telescope/Instrument Start (MJD) Start Date (UTC) Duration (ks)

21946 Chandra/ACIS/HETG 58415.68640 2018-10-24 12.0
20160 Chandra/ACIS/HETG 58417.26343 2018-10-26 41.5
21947 Chandra/ACIS/HETG 58418.28612 2018-10-27 12.0
21948 Chandra/ACIS/HETG 58418.62041 2018-10-27 56.0
20161 Chandra/ACIS/HETG 58419.90608 2018-10-28 48.5
21950 Chandra/ACIS/HETG 58421.71834 2018-10-30 14.0
21951 Chandra/ACIS/HETG 58422.54435 2018-10-31 36.5
21952 Chandra/ACIS/HETG 58424.40864 2018-11-02 12.5
21953 Chandra/ACIS/HETG 58425.19608 2018-11-03 36.5
21954 Chandra/ACIS/HETG 58433.72045 2018-11-11 25.5
21965 Chandra/ACIS/HETG 58434.31770 2018-11-12 25.5
0865040201 XMM-Newton/EPIC 59079.67552 2020-08-18 36.8
0865040301 XMM-Newton/EPIC 59083.65674 2020-08-22 40.0
0865040401 XMM-Newton/EPIC 59089.81463 2020-08-28 33.0
0865040601 XMM-Newton/EPIC 59095.94792 2020-09-03 33.0
0865040701 XMM-Newton/EPIC 59104.26799 2020-09-12 33.0
0865040501 XMM-Newton/EPIC 59110.54296 2020-09-18 47.9

The full AAVSO+ASAS-SN V -magnitude light curve burst based on HST imaging that resolved both compo-
for XZ Tau AB is presented in Figure 2, with the time- nents of the system.
frames of various X-ray observations (including the 2018 Over the twenty-year timeframe we consider, we see
Chandra and 2020 XMM-Newton observations) high- periods of quiescence of duration ∼ one year, in 2022.
lighted. We estimate the baseline “quiescent” level of We also observe clear brightening events of durations
the total V -band light from XZ Tau AB outside of out- that would suggest multiple EXOr-like outbursts, in-
burst from the AAVSO data contemporaneous with the cluding ongoing outbursts during the 2018 Chandra and
XMM-Newton observations presented in Giardino et al. 2020 XMM-Newton observations. The 2017 Chandra
(2006), as these observations were taken outside of out- observations (Skinner & Güdel 2020) appear to take
place during a local minimum of optical brightness, but
 A Stable Coronal Spectrum on XZ Tau 5

2005-01-01 2009-01-01 2013-01-01 2017-01-01 2021-01-01

12.0
X-ray Obs.
12.5 AAVSO V
13.0 ASAS-SN V
Magnitude

13.5
14.0
14.5
15.0
53000 54000 55000 56000 57000 58000 59000 60000
MJD
Figure 2. V -band light curve for XZ Tau AB from the AAVSO and ASAS-SN. The black dashed line represents the estimated
baseline level for when XZ Tau B is not in an elevated state. Gray windows highlight times of X-ray observations in 2004, 2017,
2018, and 2020. “Zoomed-in” light curves for individual X-ray observations are shown in Figures 3 and 4.
Date We find that two of the eleven observations have signif-
2018-10-26 2018-11-02 2018-11-09
10 icantly elevated count rates compared to the remaining
B
12 V nine, as depicted in Figure 3. For initial consideration,
Magnitude

R we treated these two bright observations separately, and

14 I
combined the data from the remaining nine observations
16 for improved statistics.
We present light curves from XMM-Newton with a
0.006 Chandra
Net counts/s

1000-s cadence in Figure 4. The XMM-Newton data

0.004 have more counts than the Chandra observations, en-
abling assessment of source variability on timescales
0.002
shorter than the observation itself. Four of the XMM-
58415 58420 58425 58430 58435 Newton observations are fairly stable in both low- and
MJD
high-energy bands (defined as below and above 1 keV
Figure 3. bottom: Zeroth-order observation-averaged net respectively). However, two observations are much
count rates for each Chandra observation of XZ Tau AB in brighter. The light curves of the six XMM-Newton ob-
Table 1. Error bars in the time dimension indicate the dura- servations clearly indicate that the observations on 2020
tion of each observation. Two observations (obsids 20161 August 28-29 and 2020 September 3-4 are flaring, with
and 21951) show enhanced count rates compared to the peak brightnesses ∼ 12× the median for all observations
other nine observations. top: Contemporaneous AAVSO
in the hard band, and ∼ 5.5× the median for all obser-
data shows some rolling variability with no obvious corre-
lation to the Chandra variability. vations in the soft band, even after correcting for the
high background at points during these observations.
at an elevated optical brightness compared to the base- 3.4. Modeling spectra
line set in 2004.
We assume that the X-ray data are adequately de-
scribed by two optically thin, collisionally excited emis-
3.3. X-ray Time Series sion components (APEC models Foster et al. 2012) in
The zeroth-order (non-dispersed) count rate for XZ collisional equilibrium, assumed to be “cool” and “hot,”
Tau AB in the individual Chandra observations is too multiplied by a photoelectric absorption phabs compo-
low to produce a light curve with detailed time resolu- nent to model interstellar and circumstellar absorption.
tion. We instead look at the count rates for each obser- We also set neon and iron abundances as free parame-
vation as a whole. ters, and pin the abundances of nickel, silicon, calcium,
6 Silverberg et al.

1.2
1.0
0.8
Cts/s

0.6
0.4
0.2
0.0
79.8 80.0 83.8 84.0 89.9 90.1 96.0 96.2 104.3 104.6 110.7 111.0
10
11
12
Magnitude

13
14
15 B R
V I
16
75 80 85 90 95 100 105 110 115 120
MJD-59000
Figure 4. Broad-band light curves of XZ Tau AB from XMM-Newton (top) and AAVSO (bottom). The upper plots depict the
individual XMM observations in chronological order, presenting the hard- (> 1 keV, black) and soft-energy (< 1 keV, red) count
rates from the EPIC-PN. The bottom light curve depicts observations of XZ Tau AB from AAVSO, with the XMM-Newton
observing windows highlighted in gray. Error bars are smaller than plot symbols. There is no apparent correlation in the X-ray
and optical variability.

and magnesium (which have similar first ionization po- number of counts. We therefore fit the combined zeroth-
tentials to iron) to the iron abundance. As a baseline, we order spectrum from all eleven observations. In addition
fit the four “quiescent” observations from XMM-Newton to fitting all spectra with free Ne and Fe abundances, we
jointly, and use the temperatures, normalizations, ab- fixed Ne and Fe abundances at the best-fit values from
sorption column density, and abundances from that fit the joint fit to the four quiescent XMM-Newton obser-
as the initial guess for fitting each individual observa- vations and then fit all observations again; we list the
tion. We list the best-fit parameters for these models best-fit parameters for these models with fixed abun-
with free abundances in Table 5. dances in Table 2.
Because of the diminished effective area of ACIS at We plot the PN spectra for one typical quiescent ob-
low energies caused by contamination, we fixed NH for servation (obsid 0865040501) and the two flaring obser-
the Chandra observations at the best-fit NH for the qui- vations in Figure 5, with best fit models. As expected,
escent XMM-Newton observations to mitigate degener- the two flare spectra are brighter than the quiescent
acy between soft plasma and hydrogen absorption. We spectrum. Notably, the high-energy slopes of the flar-
fit the zeroth-order spectrum for each “bright” Chan- ing spectra are different from the quiescent spectrum,
dra observation, but did not attempt to find individual indicating a different high-energy state for those spec-
fits for each of the faint Chandra observations. Rather, tra. We explore time-resolved spectroscopy of the flares
we fit the combined zeroth-order spectrum for the other in Section 3.5. This recalls a similar difference in spec-
nine observations. We find that the model fits to the tral shape from 2000 to 2004 identified in Giardino et al.
bright observations show no significant difference from (2006)—a hard plasma component readily apparent in
the fit to the combined faint observations due to the the XMM-Newton spectrum from 2000 was no longer
substantial uncertainties that result from the limited seen in 2004. Also notable in the flare observations is
 A Stable Coronal Spectrum on XZ Tau 7

the presence of 6.7 keV emission from the Fe emission due to the likely blend of data in the RGS from XZ Tau
line. A closer look at the 6-7 keV range of the data AB and the nearby (separation ∼ 12′′ ) HL Tau. Follow-
(Figure 5, right) shows clear emission from the 6.7 keV ing the method presented in Pradhan et al. (2021), we
line, but no significant evidence of emission from the flu- fit the Chandra grating spectra with a two-temperature
orescent line at 6.4 keV, limited by the count rate and model as obtained from XMM-Newton fits, while allow-
thus the bin widths adopted. It is notable that the two- ing the spectral parameters to vary over the range al-
temperature, fixed-abundance model here underpredicts lowed by XMM-Newton and fixing the abundances to
the Fe emission at 6.7 keV, albeit with marginal statisti- the XMM-Newton values. We then fit the individual re-
cal significance. This can be explained by a change in Fe gions line-by-line, grouping the Si XIII, Mg XII, Mg XI,
abundance from the quiescent to flaring observations, or Ne X, Ne IX by a factor of 2. We list our line fluxes for
by a lack of adequate high-energy emission in the simple the gratings data in Table 3. Due to the low signal in the
2-T model we adopt here. Mg XI and Ne IX triplet features, lines typically used
to differentiate low-density coronal plasma from high-
density plasma associated with accretion shocks, we are
unable to constrain the plasma density.
3.5. Time-resolved Spectra of X-ray Flares
Franciosini et al. (2007) present a time-resolved spec-
troscopic analysis of an observation of XZ Tau AB orig- 4. DISCUSSION
inally presented in Favata et al. (2003) and reanalyzed 4.1. X-ray Evolution Over Time, Or Lack Thereof
in Giardino et al. (2006). In this observation, XZ Tau We present the best-fit temperatures, emission mea-
AB monotonically rises over the course of the ∼ 55 ks sures, and abundances for each observation in Table
observation. By contrast, in obsid 0865040401 (here- 5. We note that within uncertainties the two com-
after referred to as 401), XZ Tau AB monotonically de- ponents exhibit remarkable consistency in temperature
cays, and in observation 0865040601 (hereafter referred over time. The cool component in particular shows lit-
to as 601), it exhibits a sharp rise and then decay. The tle variation in temperature. The hot component ex-
short timescales involved indicate that these are typi- hibits significantly more uncertainty in each observation,
cal flares, rather than long time-scale increases in X-ray such that while the best-fit temperatures can differ by
brightness associated with magnetic reconnection flares ∼ 1 keV, the temperatures are generally consistent with
in FUor/EXor outbursts (Kastner et al. 2004). each other within the uncertainties. The significant vari-
Following the technique of Franciosini et al. (2007), ation in brightness during flares appears to come solely
we divide observations 401 and 601 each into blocks of from the emission measure itself, which exhibits mod-
time using the bayesian blocks method as provided erate increase in the cool component during a flare and
in AstroPy, and fit models to the spectra from each of an increase on order ∼ 30 in the hot component. This
these blocks of time to better understand the flare evolu- is generally consistent with a model of occasional flares
tion. We assume in these fits that (a) the Ne and Fe-like superposed on top of a stable underlying stellar corona.
abundances match those jointly fit to the quiescent ob- To consider the hypothesis of a stable underlying stel-
servations, and that (b) the hydrogen column density lar corona, we compare our models to those of previ-
does not change over the course of one observation (i.e. ous work analyzing X-ray spectra from XZ Tau AB. We
it remains fixed at the best-fit value from fitting the present our results and previous work as a function of
spectrum of the full-duration observation). We present time in Figure 7, and briefly discuss the relevant param-
the best-fit parameters in Table 4, and plot the evolu- eters below. The parameters from previous work are
tion in parameters as a function of time along with the summarized in the Appendix.
observation light curves in Figures 8 and 9. We discuss Skinner & Güdel (2020) jointly fit their two observa-
the evolution of the flares observed in these observations tions of XZ Tau AB in 2017 December and 2018 January
in more detail in Section 4.2. with a two-component model with a fixed hydrogen col-
umn density of NH = 0.2cm2 . They find best-fit model
3.6. Gratings Data temperatures that are quite cool (0.23 keV and 1.28
The combined first (positive and negative) order grat- keV, respectively) compared to those we find. Skinner &
ings data from both the High-Energy and Medium- Güdel (2020) also present a norm-weighted average tem-
Energy Gratings (HEG and MEG, respectively) from perature of these two components of 0.78 keV, consistent
the eleven Chandra observations are presented in Figure with our cool temperature, and the normalizations of
6. We do not consider gratings data from XMM-Newton, those two components are similarly consistent. This in-
8 Silverberg et al.

Table 2. Summary of parameters for best fit models for X-ray spectra of XZ Tau AB with fixed Ne and Fe

NH kT (keV) EM (1052 cm−3 ) Flux (10−13 erg cm−2 s−1 ) Log(LX )

Source rstata (1022 cm−2 ) Cool Hot Cool Hot Absorbed Unabsorbed (erg s−1 )
Faint Chandra b 1.025 0.113 0.48+0.07
−0.08 2.0+0.3
−0.3 3.5+0.6
−0.7 3.0+0.6
−0.5 2.14 2.93 29.8
201610 0.3565 0.113 0.7+0.2
−0.5 > 2.2 9+2
−4 2.4+0.6
−0.9 4.12 5.44 30.1
219510 0.1261 0.113 0.7+0.4
−0.4 3+65
−3 13+4
−6 5+7
−4 6.06 8.12 30.3
All Chandra 0.7606 0.113 0.49+0.10
−0.05 2.3+0.4
−0.3 4.6+0.7
−0.6 3.6+0.5
−0.7 2.84 3.86 30.0
Faint XMM-Newton c 0.9135 0.113+0.007
−0.007 0.75+0.02
−0.02 3.2+0.5
−0.3 4.3+0.2
−0.2 1.7+0.2
−0.2 2.08 2.74 29.8
0865040201 0.6056 0.09+0.01
−0.01 0.72+0.04
−0.04 4.6+1.6
−0.9 4.3+0.3
−0.3 2.0+0.3
−0.3 2.48 3.06 29.9
0865040301 0.5751 0.077+0.009
−0.009 0.84+0.03
−0.03 5.4+14.4
−2.3 5.4+0.3
−0.4 0.9+0.3
−0.2 2.33 2.83 29.8
0865040401 0.7459 0.127+0.005
−0.005 0.87+0.02
−0.02 2.8+0.2
−0.1 21+1
−1 23+1
−1 16.3 20.8 30.7
0865040601 0.7041 0.105+0.005
−0.005 0.85+0.03
−0.03 3.8+0.2
−0.2 13.4+0.8
−0.8 31+1
−1 19.8 23.3 30.7
0865040701 0.6863 0.10+0.01
−0.01 0.77+0.04
−0.04 2.9+0.8
−0.5 4.0+0.4
−0.4 2.0+0.3
−0.4 2.15 2.75 29.8
0865040501 0.6152 0.15+0.02
−0.01 0.71+0.04
−0.04 2.9+0.7
−0.5 3.9+0.3
−0.3 1.6+0.3
−0.3 1.72 2.50 29.8
a Reduced χ2 using Gehrels weighting.
b Joint fit of obsids 20160, 21946, 21947, 21948, 21950, 21952, 21953, 21954, and 21965.
c Joint fit of obsids 0865040201, 0865040301, 0865040501, and 0865040701.

100
0.012
10 1
0.010
Counts/sec/keV

Counts/sec/keV

0.008
10 2

0.006
10 3
2020-09-12 0.004
2020-09-03/04 0.002
10 4
2020-08-28/29
100 101 0.000
6.0 6.2 6.4 6.6 6.8 7.0
Energy (keV) Energy (keV)
Figure 5. Spectrum of one quiescent observation with XMM-Newton (blue), in comparison to the two observations with flares
(orange and green). Left: The two flare spectra exhibit enhancement in overall brightness, and show the Fe 6.7 keV emission
line. Right: A detailed look at the 6-7 keV range shows that the Fe 6.7 keV line exhibited by the flare spectra is underfit by
models, and that flare spectra show no significant evidence for emission from the fluorescent Fe line at 6.4 keV.

dicates to us that the two-component model they adopt temperature plasma model similar to the parameteriza-
is essentially reproducing our cool component. Simi- tion we adopt. They present a fit to the “characteristic”
larly, with carefully chosen initial parameters for fitting non-flare interval in the data (also analyzed in Favata
our data, we can produce a fit to our faint XMM-Newton et al. 2003; Giardino et al. 2006), with best-fit temper-
data with similar (though somewhat worse) goodness-of- atures of kT1 = 0.75 keV and kT2 = 2.28 keV, respec-
fit with temperatures similar to the temperatures found tively, with emission measures EM1 = 4.00 × 1052 cm−3
in this work. We thus conclude that the two tempera- and EM2 = 3.76 × 1052 cm−3 , and a hydrogen column
tures found by Skinner & Güdel (2020) reproduce our density of NH = 0.24 ± 0.03 × 1022 cm2 . Franciosini
cool component. et al. (2007) split the observation used for XZ Tau AB in
Güdel et al. (2007) present multiple fit options to Güdel et al. (2007) into five time-resolved spectra using
all sources in the XMM-Newton Extended Survey of the Bayesian blocks method and fit a two-temperature
the Taurus Molecular Cloud (XEST), including a two- plasma to each of these. In this paradigm, the bright-
 A Stable Coronal Spectrum on XZ Tau 9

Si XIV
Si XIII

Mg XI

Ne X
20

15
Counts / bin

10

0
2 4 6 8 10 12 14
Wavelength (Å)
Figure 6. Chandra/HETG data for XZ Tau AB, combining both the MEG and HEG. Prominent lines in the spectrum are
labeled. We exclude the spectrum at wavelengths longer than 15 Åas there is minimal emission at these wavelengths due to the
contamination of the ACIS-S chip.
Table 3. Line Fluxes from HETGS Spectrum of XZ ening of the light curve corresponded to an increase in
Tau the emission measure of the hot plasma, while the cool
component held at constant temperature and (until the
Wavelength Flux last spectrum) emission measure. These data directly
Line (Å) (10−7 photons cm−2 s−1 ) correspond to our fits despite differences in the treat-
Si XIII (r) 6.648 3+3
ment of abundance. While Favata et al. (2003) identify
−2
a significant decrease in emission measure of the cool
Si XIII (i) 6.687 < 3.36
component for this observation, Giardino et al. (2006)
Si XIII (f) 6.740 3+2
−2
note that this is due to the spurious elevated NH level in
Mg XII 8.422 < 3.63
the first time bin of the time-resolved analysis. Table 4
Mg XI (r) 9.169 9+5
−4
in Giardino et al. (2006) (which supersedes the Table in
Mg XI (i) 9.230 < 2.13
Favata et al. 2003) shows that while the emission mea-
Mg XI (f) 9.314 3+4
−3 sure does increase, this increase is not statistically sig-
Ne X 12.135 50+21
−17 nificant. We find that our variability is consistent with
Ne IX (r) 13.447 < 29.7 the variability summary in Favata et al. (2003) as well;
Fe XIX 13.462 28+33
−22 our observed fluxes fall neatly in the range of observed
Fe XIX 13.518 14+24
−7 fluxes in this older data.
Ne IX (i) 13.552 < 18.1 Giardino et al. (2006) present five days of monitoring
Fe XIX 13.645 < 27.7 of a field that includes XZ Tau AB with XMM-Newton,
Ne IX (f) 13.699 17+24
−13 during a period of high background for the telescope.
Fe XX 13.767 < 1.32 They present single-temperature fits to these data, find-
Fe XVII 13.825 < 2.16 ing a stable plasma (albeit with changing hydrogen col-
umn density) at a temperature between 0.63 keV and
0.8 keV. Fits to two of their observations indicate a hot-
ter temperature; however, these also show an increase
in the observed flux, suggesting a flare. Indeed, the au-
10 Silverberg et al.

2000 2004 2017/12-2018/01 2018 2020

12

13
V mag.

14

15
1e 12
Flux (ergs 1 cm 2)

2 Absorbed
Unabsorbed
1

6 0.76 keV
Cool
kT (keV)

4 Hot
2
0
40
NH (1022cm2) EM (1052cm 3)

Cool
Hot
20

0
0.3
0.2
0.1
51796 53070 58120 58420 58435 59085 59100
MJD MJD MJD MJD MJD

Figure 7. Characteristics of best-fit models to various observations of XZ Tau AB over time, from published work and analysis
of unpublished publicly-available data. Rows display (top to bottom) optical brightness as represented by V magnitude, X-ray
flux (absorbed and unabsorbed), plasma temperatures, plasma emission measures, and NH . Each column spans one set of
observations, from 2000 (Güdel et al. 2007), 2004 (Giardino et al. 2006), 2017-2018 (Skinner & Güdel 2020), 2018, and 2020.
Dotted lines connect points representing obsids that were jointly fit.

thors note that some of these observations are better recent Chandra observations with limited sensitivity to
fit by a two-component model. We argue that this too soft X-rays) do not have much leverage to constrain
demonstrates the persistence of a cool component be- NH . The best-fit models consistently exhibit a cool tem-
tween 0.7 keV and 0.8 keV over time. We also note that perature between 0.7-0.8 keV, indicating a stable tem-
while Giardino et al. (2006) hypothesize that the change perature feature we interpret as the stellar corona(e).
in the XZ Tau AB X-ray spectrum is connected to the Due to different models assumptions (e.g. abundances),
end of XZ Tau B’s outburst, we find that the spectrum the emission measures and NH vary widely for different
remains similar to the 2004 result despite the apparent datasets and are not directly comparable. The ratio of
outburst of XZ Tau B during our observations. temperature components and variability within each set
From these data, we see that both in the raw observ- are robust, as can be seen in the flares in 2020. The con-
ables and in the adopted modeling paradigm of stellar X- sistency of the cool component over an interval of twenty
ray spectra as stacked single-temperature plasmas, XZ years bolsters our hypothesis of an underlying stable
Tau AB remains fairly stable over observations span- corona, with occasional flares superposed on top of this,
ning the 2000-2020 timeframe, with the few exceptions and recommends against the idea that flares completely
of likely flares, as shown in Figure 7. Observed fluxes re- reorganize the stellar coronal behavior.
main fairly stable over time apart from the observations
we identify as flares. NH varies around a consistently
low level, indicating that the observations (particularly
 A Stable Coronal Spectrum on XZ Tau 11

Table 4. Summary of parameters for best fit models for time-resolved spectra of XZ Tau AB Flares

NH Cool Component Hot Component Flux (10−13 erg s−1 cm−2 ) Log(Lum.)

obsid rstat (1022 cm−2 ) kT (keV) EM (1052 cm−3 ) kT (keV) EM (1052 cm−3 ) Absorbed Unabsorbed log (erg s−1 )
FaintXMM 0.9135 0.113+0.007
−0.007 0.75+0.02
−0.02 4.3+0.2
−0.2 3.2+0.5
−0.3 1.7+0.2
−0.2 2.078 2.739 29.8
0865040401A 0.5724 0.127 0.95+0.06
−0.05 21+3
−3 3.0+0.3
−0.2 39+3
−3 23.98 28.71 30.8
0865040401B 0.7251 0.127 0.89+0.04
−0.04 25+2
−2 2.9+0.4
−0.3 23+2
−2 17.92 22.08 30.7
0865040401C 0.5207 0.127 0.87+0.04
−0.04 25+2
−3 3.1+1.0
−0.6 13+3
−3 13.92 17.49 30.6
0865040401D 0.6073 0.127 0.81+0.04
−0.03 17+2
−1 2.6+0.5
−0.4 11+2
−2 9.909 12.54 30.5
0865040601A 0.5601 0.105 0.82+0.10
−0.12 7+1
−1 9+9
−3 11+1
−1 9.291 10.73 30.4
0865040601B 0.6247 0.105 0.99+0.14
−0.15 11+4
−3 7+6
−2 25+3
−3 18.65 21.23 30.7
0865040601C 0.6304 0.105 0.84+0.05
−0.05 12+1
−1 3.8+0.2
−0.2 42+1
−1 25.03 29.13 30.8
0865040601D 0.7071 0.105 0.89+0.05
−0.05 23+3
−3 3.0+0.6
−0.4 22+3
−3 17.32 21.27 30.7
0865040601E 0.591 0.105 0.74+0.06
−0.07 16+2
−2 2.4+0.5
−0.4 20+3
−3 13.07 16.38 30.6

4.2. Short-term Coronal Variability while the emission measure increases, suggesting a hot-
Observation 401 shows the clear signatures of the de- ter plasma in the rise than in the decay.
cay phase of a flare. The hardness of the spectrum The detection of the flare peak in observation 601 al-
evolves over time–the count rates clearly show a de- lows for a fuller characterization of the coronal loops
crease in hard flux while soft flux remains stable over the produced by the flare based on the flare decay. Follow-
first two blocks of time (before roughly MJD 59090.03), ing the application of Reale et al. (2004) by Giardino
before the soft flux also begins to decay over the lat- et al. (2006) and Franciosini et al. (2007), we estimate
ter half of the observation. The best-fit temperatures, the flare loop semi-length L from the e-folding timescale
abundances, and hydrogen column densities remain sta- of the decay and the maximum temperatures Tmax via
ble across the observation, though there is substantial the equation
uncertainty in the temperature of the hot component in √
the third, shortest, block of time. The clearest change τLC Tmax
L= , (1)
over time is in the emission measures of the “hot” com- αF (ζ)
ponent, which we will refer to in this section as the flare
where α = 3.7 × 10−4 cm−1 s−1 K1/2 , τLC is the e-
component. The flare component’s emission measure
folding timescale of the decay, and Tmax is the maxi-
decreases by > 60% between the first and fourth blocks,
mum plasma temperature of the flare, while accounting
while the emission measure of the cool component stays
for heating during the decay√via ζ, the slope of the flare
stable (within uncertainties) over that time frame. This
decay in the log(T ) vs. log( EM ) plane, shown in Fig-
indicates that for this flare, the decay is solely driven
ure 10.
by the emission measure decreasing. It is notable that
These characteristics (including the relationship be-
the emission measures invert in the third block of time,
tween the best-fit peak temperature Tobs and Tmax )
albeit with substantial uncertainties.
are calibrated for the PN detector; however, Franciosini
Observation 601, on the other hand, shows both the
et al. (2007) found that these formulae would give order-
rise and the start of the decay for the flare, and the char-
of-magnitude results for the MOS detectors as well,
acteristics are not nearly as clear-cut as in 401. Qual-
given the similarities of the instruments and the width
itatively, as the hard flux peaks the soft flux levels off,
of the adopted spectral band. We thus use these equa-
leading to a decay phase similar to that seen in 401. The
tions for our temperatures and emission measures de-
increase in hard flux is steeper than in the soft, as there
rived from joint fits to the three detectors.
is less hard flux initially and more hard flux at the peak.
For the flare in 601, we find ζ = 1.2 ± 0.4. We find an
While the cool component maintains its characteristic
e-folding timescale based on fitting a line to the natural
behavior (albeit with substantial uncertainties) through-
logarithm of the decay phase light curve of 50±4 ks. The
out the flare, the hot component is far more variable.
best-fit maximum temperature Tobs = 44+3 −2 MK, which
During the rise phase, the temperature is elevated com-
following Equation (3) from Giardino et al. (2006) cor-
pared to both 401 and the decay phase of this flare,
responds to a maximum temperature of Tmax = 96+4 −3
12 Silverberg et al.

0.8 PN 0.8 PN
Soft ct. rate (cts/s) Hard ct. rate (cts/s)

Soft ct. rate (cts/s) Hard ct. rate (cts/s)

0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
Cool Cool
7.5 Hot 10 Hot
kT (keV)

kT (keV)
5.0
5
2.5
0.0 0
NH (1022 cm 2) EM (1052 cm 3)

NH (1022 cm 2) EM (1052 cm 3)
40 40

20 20

0 0
0.2 NH NH
0.2
0.1 0.1

0.0 0.0
59089.9 59090.0 59090.1 59096.0 59096.1 59096.2 59096.3
MJD MJD

Figure 8. Evolution of XZ Tau AB over the course of a Figure 9. Evolution of XZ Tau AB over the course of a
flare decay in observation 0865040401. The decrease in the flare rise and decay in observation 0865040601. The change
emission measure of the “hot” component over time clearly in the emission measure of the “hot” component over time
tracks the decrease in count rate for photons at energies > 1 clearly tracks the rise in count rate for photons at energies
keV, while the temperature of the “hot” component remains > 1 keV. The “cool” component stays fairly stable over time,
stable. indicating that most of the change in brightness is due to the
hot component.
MK. Inputting this information into Equation 1 yields a
loop length of ∼ 10R⊙ , corresponding to ∼ 6R⋆ for XZ (e.g. Favata et al. 2005); however, there is no appar-
Tau A and ∼ 9R⋆ for XZ Tau B. These lengths are such ent difference in flare energy or flare peak energy be-
that the flare could reach into the disk, though we do tween flares on disk-hosting and diskless pre-main se-
not have direct measurement of the inner disk radius for quence stars (Getman & Feigelson 2021), and further
either star; ALMA data do not resolve the dust emis- work suggests that flares in disk-hosting pre-main se-
sion for either disk and thus leave the radius as ≲ 15 quence stars also exhibit loops with both footprints in
au (Ichikawa et al. 2021). Such extended flare sizes are the stellar surface, rather than a flare that extends from
not uncommon in pre-main sequence disk-hosting stars star to disk (Getman et al. 2021). The data in hand do
 A Stable Coronal Spectrum on XZ Tau 13

As seen in Figure 11, the XZ Tau AB data traces

8.4 across the scatter in the COUP dataset. While the cool
temperature component does stay steady as a function
8.2 of unabsorbed X-ray luminosity, the hot component in-
8.0 creases in temperature as a function of luminosity for
log T

both the COUP dataset and XZ Tau AB, an expected

7.8 result of flare activity. XZ Tau AB over time similarly
follows the trend across the COUP snapshot of increas-
7.6 ing ratio of hot emission measure to cool as a function
of X-ray luminosity (albeit with a fair amount of scat-
7.4 ter). This suggests to us the possibility that much of the
26.6 26.7 26.8 observed variation in the X-ray characteristics of YSOs
log EM is due primarily to the chance timing of the observa-
tion, rather than to fundamental differences in the stars
themselves.
Figure 10. log(T ) of the five time-resolved
√ spectra of Ob-
servation 601, as a function of log( EM ). Arrows indicate 5. SUMMARY
the time order of observations. The orange line reflects the
In this paper, we present CCD-resolution and gratings
slope ζ of the flare decay in this space.
spectra of the pre-main sequence binary system XZ Tau
AB from both Chandra and XMM-Newton. We find the
not make it more or less likely that either star produced following.
the flare, in our opinion; while XZ Tau B is thought to
be an ExOR object, both XZ Tau A and B are young • The Chandra and XMM-Newton data are
stars with disks, making flare events likely from both. both consistent with a two-temperature plasma-
We also attempted using the available data from the emission model.
flare during observation 401 to provide a lower limit for • The cool component of the two-temperature model
the coronal loops in that flare as well. For that flare, remains stable across both the Chandra and
we find a lower limit decay e-folding time of 33 ± 1 ks XMM-Newton observations, and is consistent with
and a lower limit maximum temperature of 72+5 −4 MK. previous observations of the system with both ob-
However, for this flare ζ ∼ 0.1, well below the validity servatories spanning more than a decade. This
threshold for the relation of 0.35 < ζ ≤ 1.6. suggests a rotationally-induced magnetic field
4.3. A YSO Coronal Spectrum Over Time Looks like a corona that remains stable over this timeframe,
Snapshot of a Young Cluster with recurring aperiodic flare activity superposed
on top of this stable feature.
Many studies of young stars, such as the Chan-
dra Orion Ultradeep Project (COUP; Getman et al. • The scatter in model fits of XZ Tau AB over time is
2005), and the XMM-Newton Extended Survey of Tau- comparable to the scatter in two-component model
rus (XEST; Güdel et al. 2007), look at entire young clus- fits of the sources in COUP.
ters, leveraging the ability to study many YSOs at once
• Time-resolved spectra of two flares observed with
rather than focusing on a single source. Because of the
XMM-Newton show that the primary change dur-
multiple observations of XZ Tau AB over time discussed
ing the flare is an increase in the volume emission
in Section 4.1, we can compare the ensemble of states of
measure of the “hot” component. Detailed anal-
XZ Tau AB over time against other studies which have
ysis of the decay of one flare indicated a plasma
taken snapshots of an entire cluster at a given point in
loop length of ∼ 10R⊙ .
time.
To test this, we compared the temperatures and emis- • XZ Tau AB is likely undergoing a long-term out-
sion measures of our models of XZ Tau AB and those burst typical of EXor objects during our X-ray ob-
from previous work as a function of X-ray luminosity servations based on the optical light curves. How-
against the same characteristics for a subset of the data ever, the change in X-ray brightness we see dur-
from COUP with two-temperature models that produce ing several observations is more consistent with
good fits (i.e. not meeting the “marginal” or “poor” flags short-duration coronal flares than the long-term
in their paper), following the comparison of Preibisch X-ray brightness increase seen in other EXOr ob-
et al. (2005) with the initial COUP dataset. jects (e.g. V1647 Ori; Hamaguchi et al. 2012).
14 Silverberg et al.

1.5 COUP
1.0 XZ Tau
1.0
0.5 0.5
log(kT (keV))

0.0

EM2/EM1
0.0
0.5
0.5
1.0
COUP Cool
COUP Hot 1.5
1.0 XZTau Cool
XZTau Hot 2.0
28.0 28.5 29.0 29.5 30.0 30.5 31.0 31.5 32.0 28 29 30 31 32
log(LX) log(LX)

Figure 11. Comparison of models of various observations of XZ Tau AB over time to model fits to the Chandra Orion Ultradeep
Project (COUP). Left: As with COUP model fits, XZ Tau AB over time exhibits a steady cool component and variable hot
component that increase in brightness with increasing absorption-corrected X-ray luminosity. Right: As with COUP model fits,
the ratio of emission measures of warm and cool components increases (i.e. the warm component increases in emission) as a
function of absorption-corrected X-ray luminosity.

We thank the anonymous referee for insightful com-

ments that were invaluable to the science presented
in this work. S.M.S., S.J.W., and D.A.P. acknowl-
edge support from the National Aeronautics and Space
Administration through Chandra Award Number SAO
GO8-19013X issued by the Chandra X-ray Observa-
tory Center, which is operated by the Smithsonian As-
trophysical Observatory for and on behalf of the Na-
tional Aeronautics Space Administration under con-
tract NAS8-03060. HMG acknowledges support from
NASA grant 80NSSC21K0849. PCS acknowledges sup-
port from DLR 50OR2102.
This research has made use of data obtained from the
Chandra Data Archive and the Chandra Source Cata-
log, and software provided by the Chandra X-ray Center
(CXC) in the application packages CIAO and Sherpa.
Based on observations obtained with XMM-Newton,
an ESA science mission with instruments and contribu-
tions directly funded by ESA Member States and NASA.
We acknowledge with thanks the variable star obser-
vations from the AAVSO International Database con-
tributed by observers worldwide and used in this re-
search.

Facilities: CXO, XMM-Newton, AAVSO, ASAS-SN

Software: AstroPy (Astropy Collaboration et al.
2013, 2018), NumPy (Van Der Walt et al. 2011; Harris
et al. 2020), SciPy (Jones et al. 2001; Virtanen et al.
2020), Matplotlib (Hunter 2007), pandas (McKinney
2010; Reback et al. 2020), CIAO (Fruscione et al. 2006),
Sherpa (Freeman et al. 2001)
 A Stable Coronal Spectrum on XZ Tau 15

APPENDIX

A. FITTING NEW CHANDRA AND XMM-NEWTON DATA WITH VARIABLE ABUNDANCES

While we adopt the case that the Ne and Fe abundances are not significantly varying over the timeframe of our
observations, we present here our best-fit models for these data, with Ne and Fe allowed to vary, for comparison.
 16

Table 5. Summary of parameters for best fit models for X-ray spectra of XZ Tau AB with freely-varying Ne and Fe-like abundances

NH Cool Component Hot Component Flux (10−13 ergs−1 cm−2 ) Log(Lum.)

obsid rstata (1022 cm−2 ) kT (keV) EM 1052 cm−3 Ne (Ne⊙ ) Fe (Fe⊙ ) kT (keV) EM (1052 cm−3 ) Absorbed Unabsorbed log (erg s−1 )
All Chandra 0.568 0.11 0.83+0.1
−0.6 1.0+0.6
−0.4 7.6+4.8
−4.2 0.9+0.4
−0.3 2.2+0.2
−0.2 3.0+1.3
−0.4 2.5 3.1 29.9
201610 0.378 0.11 0.29+1
−0.3 15+649
−14 0.6+31.6
−0.6 0.9+7.8
−0.9 3.2+nan
−1.1 4+2
−3 7.4 11.6 30.5
219510 0.131 0.11 0.73+0.5
−0.7 8+22
−7 2.6+8.0
−2.6 0.2+0.9
−0.2 2.55+nan
nan 6+7
−5 5.7 7.5 30.3
Faint Chandra b 0.769 0.11 0.39+0.8
−0.2 0.5+0.6
−0.3 7.4+7.6
−3.8 1.0+0.7
−0.4 1.8+0.2
−0.2 2.6+0.6
−0.6 1.9 2.4 29.8
Faint XMM-Newton c 0.915 0.11+0.01
−0.01 0.75+0.02
−0.02 4.7+0.6
−0.6 1.4+0.2
−0.2 0.11+0.02
−0.01 3.2+0.7
−0.4 1.9+0.3
−0.3 2.1 2.7 29.9
0865040201 0.61 0.10+0.02
−0.02 0.74+0.04
−0.05 5.1+1.2
−1.0 1.3+0.3
−0.3 0.10+0.03
−0.02 5.0+3.8
−1.2 2.1+0.4
−0.5 2.5 3.1 29.9
0865040301 0.573 0.08+0.01
−0.01 0.80+0.05
−0.05 6.1+1.1
−1.2 1.6+0.4
−0.3 0.09+0.02
−0.02 5.5+nan
−2.6 1.0+0.6
−0.4 2.3 2.9 29.9
0865040401 0.716 0.10+0.01
−0.01 0.85+0.03
−0.02 10.3+1.6
−1.3 1.8+0.3
−0.3 0.26+0.04
−0.03 2.44+0.08
−0.09 30.4+0.9
−0.8 16.3 19.6 30.7
0865040601 0.684 0.08+0.01
−0.01 0.78+0.03
−0.04 6.0+1.2
−0.9 2.6+0.4
−0.4 0.26+0.04
−0.04 3.5+0.1
−0.1 35.8+0.8
−0.8 19.8 22.5 30.8
Silverberg et al.

0865040701 0.688 0.10+0.02

−0.02 0.79+0.05
−0.05 4.2+1.5
−1.3 1.1+0.5
−0.4 0.13+0.05
−0.03 2.7+1.1
−0.5 2.3+0.6
−0.7 2.1 2.7 29.8
0865040501 0.619 0.15+0.02
−0.02 0.71+0.04
−0.05 4.2+1.1
−1.1 1.3+0.4
−0.3 0.12+0.04
−0.02 2.9+0.9
−0.5 1.8+0.4
−0.4 1.7 2.5 29.8
a Reduced χ2 using Gehrels weighting.
b Joint fit of obsids 20160, 21946, 21947, 21948, 21950, 21952, 21953, 21954, and 21965.
c Joint fit of obsids 0865040201, 0865040301, 0865040501, and 0865040701.
 A Stable Coronal Spectrum on XZ Tau 17

B. SUMMARY OF PARAMETERS FROM in addition to our analysis here. While the methodolo-
PREVIOUS OBSERVATIONS gies between each paper are different enough to not be
We compiled fit parameters for previous observations directly comparable on a numerical basis (e.g. different
of XZ Tau from Güdel et al. (2007); Franciosini et al. treatments of metallicity), they provide the backbone
(2007); Giardino et al. (2006); Skinner & Güdel (2020), for the comparison presented in Figure 7.

REFERENCES
Arce, H. G., & Sargent, A. I. 2006, ApJ, 646, 1070, Freeman, P., Doe, S., & Siemiginowska, A. 2001, in Society
doi: 10.1086/505104 of Photo-Optical Instrumentation Engineers (SPIE)
Argiroffi, C., Drake, J. J., Maggio, A., et al. 2004, ApJ, Conference Series, Vol. 4477, Astronomical Data
609, 925, doi: 10.1086/420692 Analysis, ed. J.-L. Starck & F. D. Murtagh, 76–87,
Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., doi: 10.1117/12.447161
et al. 2013, A&A, 558, A33, Fruscione, A., McDowell, J. C., Allen, G. E., et al. 2006, in
doi: 10.1051/0004-6361/201322068 Society of Photo-Optical Instrumentation Engineers
(SPIE) Conference Series, Vol. 6270, Society of
Astropy Collaboration, Price-Whelan, A. M., Sipőcz, B. M.,
Photo-Optical Instrumentation Engineers (SPIE)
et al. 2018, AJ, 156, 123, doi: 10.3847/1538-3881/aabc4f
Conference Series, ed. D. R. Silva & R. E. Doxsey,
Audard, M., Ábrahám, P., Dunham, M. M., et al. 2014, in
62701V, doi: 10.1117/12.671760
Protostars and Planets VI, ed. H. Beuther, R. S. Klessen,
Garcı́a-Alvarez, D., Drake, J. J., Lin, L., Kashyap, V. L., &
C. P. Dullemond, & T. Henning, 387,
Ball, B. 2005, ApJ, 621, 1009, doi: 10.1086/427721
doi: 10.2458/azu uapress 9780816531240-ch017
Getman, K. V., & Feigelson, E. D. 2021, ApJ, 916, 32,
Bonito, R., Orlando, S., Peres, G., Favata, F., & Rosner, R. doi: 10.3847/1538-4357/ac00be
2007, A&A, 462, 645, doi: 10.1051/0004-6361:20065236 Getman, K. V., Feigelson, E. D., & Garmire, G. P. 2021,
Brickhouse, N. S., Cranmer, S. R., Dupree, A. K., Luna, ApJ, 920, 154, doi: 10.3847/1538-4357/ac1746
G. J. M., & Wolk, S. 2010, ApJ, 710, 1835, Getman, K. V., Flaccomio, E., Broos, P. S., et al. 2005,
doi: 10.1088/0004-637X/710/2/1835 ApJS, 160, 319, doi: 10.1086/432092
Canizares, C. R., Davis, J. E., Dewey, D., et al. 2005, Giardino, G., Favata, F., Silva, B., et al. 2006, A&A, 453,
PASP, 117, 1144, doi: 10.1086/432898 241, doi: 10.1051/0004-6361:20053663
Clarke, C. J., Lin, D. N. C., & Pringle, J. E. 1990, Grosso, N., Feigelson, E. D., Getman, K. V., et al. 2006,
MNRAS, 242, 439, doi: 10.1093/mnras/242.3.439 A&A, 448, L29, doi: 10.1051/0004-6361:200600004
Cleeves, L. I., Adams, F. C., & Bergin, E. A. 2013, ApJ, Güdel, M., Skinner, S. L., Audard, M., Briggs, K. R., &
772, 5, doi: 10.1088/0004-637X/772/1/5 Cabrit, S. 2008, A&A, 478, 797,
Coffey, D., Downes, T. P., & Ray, T. P. 2004, A&A, 419, doi: 10.1051/0004-6361:20078141
593, doi: 10.1051/0004-6361:20034316 Güdel, M., Briggs, K. R., Arzner, K., et al. 2007, A&A,
468, 353, doi: 10.1051/0004-6361:20065724
Favata, F., Flaccomio, E., Reale, F., et al. 2005, ApJS, 160,
Haas, M., Leinert, C., & Zinnecker, H. 1990, A&A, 230, L1
469, doi: 10.1086/432542
Hamaguchi, K., Grosso, N., Kastner, J. H., et al. 2012,
Favata, F., Fridlund, C. V. M., Micela, G., Sciortino, S., &
ApJ, 754, 32, doi: 10.1088/0004-637X/754/1/32
Kaas, A. A. 2002, A&A, 386, 204,
Harris, C. R., Millman, K. J., van der Walt, S. J., et al.
doi: 10.1051/0004-6361:20011387
2020, Nature, 585, 357, doi: 10.1038/s41586-020-2649-2
Favata, F., Giardino, G., Micela, G., Sciortino, S., & Huenemoerder, D. P., Kastner, J. H., Testa, P., Schulz,
Damiani, F. 2003, A&A, 403, 187, N. S., & Weintraub, D. A. 2007, ApJ, 671, 592,
doi: 10.1051/0004-6361:20030305 doi: 10.1086/522921
Foster, A. R., Ji, L., Smith, R. K., & Brickhouse, N. S. Hunter, J. D. 2007, Computing In Science & Engineering,
2012, ApJ, 756, 128, doi: 10.1088/0004-637X/756/2/128 9, 90, doi: 10.1109/MCSE.2007.55
Franciosini, E., Pillitteri, I., Stelzer, B., et al. 2007, A&A, Ichikawa, T., Kido, M., Takaishi, D., et al. 2021, ApJ, 919,
468, 485, doi: 10.1051/0004-6361:20066536 55, doi: 10.3847/1538-4357/ac0dc3
 18 Silverberg et al.

Table 6. Comparison of Present Work to Previous Observations

Abundance Fit Boundsb NH c kT (keV) EM (1052 cm−3 ) Flux (10−13 erg cm−2 s−1 )

Source Type Uncertaintya Year (keV) (1022 cm−2 ) Cool Hot Cool Hot Absorbed Unabsorbed
XEST (1) 0.68 2000 0.5 - 7.3 0.24+0.03
−0.03 0.7497 2.278 4 3.76 2.078 3.38
FranciosiniA (1) 0.9 2000 0.3 - 7.3 0.245+0.061
−0.044 0.73+0.08
−0.11 3.44+2.31
−0.95 4.5+1.2
−0.9 3.1+0.9
−0.7 2.217 3.567
FranciosiniB (1) 0.9 2000 0.3 - 7.3 0.205+0.047
−0.039 0.82+0.15
−0.09 5.72+4.49
−1.71 4.9+1.6
−1.2 6.8+1.2
−0.9 4.735 6.324
FranciosiniC (1) 0.9 2000 0.3 - 7.3 0.191+0.039
−0.033 0.77+0.13
−0.12 6.03+2.06
−1.51 4.5+1.4
−1.4 12.5+1.6
−0.9 7.775 9.783
FranciosiniD (1) 0.9 2000 0.3 - 7.3 0.222+0.023
−0.021 0.78+0.13
−0.09 4.42+0.64
−0.45 4.9+1.2
−1.2 18.3+1.2
−1.2 9.662 12.63
FranciosiniE (1) 0.9 2000 0.3 - 7.3 0.256+0.029
−0.026 0.79+0.09
−0.09 3.5+0.52
−0.42 8.2+1.9
−1.9 22.3+2.1
−1.6 10.82 15.29
Giardino201 (2) 0.68 2004 0.3 - 7.5 0.29+0.06
−0.06 0.63+0.06
−0.06 ··· 30.2+13.1
−13.1 ··· 0.7543 6.388
Giardino301 (2) 0.68 2004 0.3 - 7.3 0.15+0.03
−0.03 0.78+0.04
−0.04 ··· 14.6+3.2
−3.2 ··· 2.215 3.465
Giardino401 (2) 0.68 2004 0.3 - 7.3 0.08+0.02
−0.02 0.77+0.04
−0.04 ··· 12.5+2.6
−2.6 ··· 2.294 2.952
Giardino501 (2) 0.68 2004 0.3 - 7.3 0.11+0.04
−0.04 0.72+0.05
−0.05 ··· 14.7+4.1
−4.1 ··· 2.362 3.364
Giardino601 (2) 0.68 2004 0.3 - 7.3 0.07+0.02
−0.02 0.78+0.04
−0.04 ··· 11.9+0.4
−5.2 ··· 2.265 2.824
Giardino901 (2) 0.68 2004 0.3 - 7.3 0.14+0.03
−0.03 0.76+0.05
−0.05 ··· 16.8+4.4
−4.4 ··· 2.572 3.944
Giardino1001 (2) 0.68 2004 0.3 - 7.3 0.09+0.02
−0.02 1.03+0.07
−0.07 ··· 14.5+2.6
−2.6 ··· 2.948 3.723
Giardino1101 (2) 0.68 2004 0.3 - 7.3 0.14+0.02
−0.02 1.00+0.04
−0.04 ··· 23.7+2.5
−2.5 ··· 4.240 6.046
Giardino1201 (2) 0.68 2004 0.3 - 7.3 0.13+0.02
−0.02 0.80+0.04
−0.04 ··· 18+3.3
−3.3 ··· 2.931 4.315
Skinner (3) 0.68 2017 0.3 - 8 0.2 0.26+0.04
−0.03 1.28+0.04
−0.04 13.37+3.752
−3.049 13.84+0.9381
−0.9381 3.446 5.895
ChandraFull (4) 0.68 2018 0.5 - 8 0.113 0.49+0.10
−0.05 2.3+0.4
−0.3 4.6+0.7
−0.6 3.6+0.5
−0.7 2.837 3.862
FaintXMM (5) 0.68 2020 0.3 - 9 0.113+0.007
−0.007 0.75+0.02
−0.02 3.2+0.5
−0.3 4.3+0.2
−0.2 1.7+0.2
−0.2 2.078 2.739
0865040201 (4) 0.68 2020 0.3 - 9 0.09+0.01
−0.01 0.72+0.04
−0.04 4.6+1.6
−0.9 4.3+0.3
−0.3 2.0+0.3
−0.3 2.477 3.06
0865040301 (4) 0.68 2020 0.3 - 9 0.077+0.009
−0.009 0.84+0.03
−0.03 5+14
−2 5.4+0.3
−0.4 0.9+0.3
−0.2 2.326 2.831
0865040401 (4) 0.68 2020 0.3 - 9 0.127+0.005
−0.005 0.87+0.02
−0.02 2.8+0.2
−0.1 21+1
−1 23+1
−1 16.31 20.84
0865040401A (4) 0.68 2020 0.3 - 9 0.127 0.95+0.06
−0.05 3.0+0.3
−0.2 21+3
−3 39+3
−3 23.98 28.71
0865040401B (4) 0.68 2020 0.3 - 9 0.127 0.89+0.04
−0.04 2.9+0.4
−0.3 25+2
−2 23+2
−2 17.92 22.08
0865040401C (4) 0.68 2020 0.3 - 9 0.127 0.87+0.04
−0.04 3.1+1
−0.6 25+2
−3 13+3
−3 13.92 17.49
0865040401D (4) 0.68 2020 0.3 - 9 0.127 0.81+0.04
−0.03 2.6+0.5
−0.4 17+2
−1 11+2
−2 9.909 12.54
0865040601 (4) 0.68 2020 0.3 - 9 0.105+0.005
−0.005 0.85+0.03
−0.03 3.8+0.2
−0.2 13.4+0.8
−0.8 31+1
−1 19.76 23.27
0865040601A (4) 0.68 2020 0.3 - 9 0.105 0.82+0.10
−0.12 9+9
−3 7+1
−1 11+1
−1 9.291 10.73
0865040601B (4) 0.68 2020 0.3 - 9 0.105 0.99+0.14
−0.15 7+6
−2 11+4
−3 25+3
−3 18.65 21.23
0865040601C (4) 0.68 2020 0.3 - 9 0.105 0.84+0.05
−0.05 3.8+0.2
−0.2 12+1
−1 42+1
−1 25.03 29.13
0865040601D (4) 0.68 2020 0.3 - 9 0.105 0.89+0.05
−0.05 3.0+0.6
−0.4 23+3
−3 22+3
−3 17.32 21.27
0865040601E (4) 0.68 2020 0.3 - 9 0.105 0.74+0.06
−0.07 2.4+0.5
−0.4 16+2
−2 20+3
−3 13.07 16.38
0865040701 (4) 0.68 2020 0.3 - 9 0.10+0.02
−0.01 0.77+0.04
−0.04 2.9+0.8
−0.5 4.0+0.4
−0.4 2.0+0.3
−0.4 2.146 2.752
0865040501 (4) 0.68 2020 0.3 - 9 0.15+0.02
−0.01 0.71+0.04
−0.04 2.9+0.7
−0.5 3.9+0.3
−0.3 1.6+0.3
−0.3 1.724 2.497

Note—(1) Fixed VAPEC model abundances described in Güdel et al. (2007), based on Telleschi et al. (2005), Argiroffi et al. (2004), Garcı́a-Alvarez et al. (2005),
and Scelsi et al. (2005). (2) Fixed metallicity of Z = 0.08Z⊙ based on average metallicity across fits to eleven individual observations. (3) Best fit metallicity of
Z = 0.14+0.03
−0.02 Z⊙ . (4) Fixed Ne and Fe values based on the best fit to a joint fit of the four XMM-Newton observations not during flares. (5) Best-fit Ne and Fe
values for this fit.
(a) Decimal uncertainties—i.e. 0.68 corresponds to 68% uncertainties in the following numbers.
(b) Energy range considered when fitting—i.e. data outside these bounds are not considered in the fit.
(c) Values listed without uncertainties are fixed at the quoted value in the fitting.
 A Stable Coronal Spectrum on XZ Tau 19

Jayasinghe, T., Stanek, K. Z., Kochanek, C. S., et al. 2019, Reback, J., McKinney, W., jbrockmendel, et al. 2020,
MNRAS, 486, 1907, doi: 10.1093/mnras/stz844 pandas-dev/pandas: Pandas 1.0.1, v1.0.1, Zenodo,
doi: 10.5281/zenodo.3644238
Jones, E., Oliphant, T., Peterson, P., & Others. 2001,
Scelsi, L., Maggio, A., Micela, G., Briggs, K., & Güdel, M.
SciPy: Open source scientific tools for Python.
2007, A&A, 473, 589, doi: 10.1051/0004-6361:20077792
http://www.scipy.org/ Scelsi, L., Maggio, A., Peres, G., & Pallavicini, R. 2005,
Kastner, J. H., Huenemoerder, D. P., Schulz, N. S., A&A, 432, 671, doi: 10.1051/0004-6361:20041739
Canizares, C. R., & Weintraub, D. A. 2002, The Schneider, P. C., Günther, H. M., & Schmitt, J. H. M. M.
Astrophysical Journal, 567, 434, doi: 10.1086/338419 2011, A&A, 530, A123,
doi: 10.1051/0004-6361/201016305
Kastner, J. H., Richmond, M., Grosso, N., et al. 2004,
Schneider, P. C., Günther, H. M., & Ustamujic, S. 2022,
Nature, 430, 429, doi: 10.1038/nature02747
arXiv e-prints, arXiv:2207.06886,
Krist, J. E., Stapelfeldt, K. R., Hester, J. J., et al. 2008, doi: 10.48550/arXiv.2207.06886
AJ, 136, 1980, doi: 10.1088/0004-6256/136/5/1980 Schneider, P. C., & Schmitt, J. H. M. M. 2008, A&A, 488,
Lorenzetti, D., Larionov, V. M., Giannini, T., et al. 2009, L13, doi: 10.1051/0004-6361:200810261
Shappee, B. J., Prieto, J. L., Grupe, D., et al. 2014, ApJ,
ApJ, 693, 1056, doi: 10.1088/0004-637X/693/2/1056
788, 48, doi: 10.1088/0004-637X/788/1/48
McKinney, W. 2010, in Proceedings of the 9th Python in Skinner, S. L., & Güdel, M. 2013, ApJ, 765, 3,
Science Conference, ed. S. van der Walt & J. Millman, 51 doi: 10.1088/0004-637X/765/1/3
– 56 —. 2020, ApJ, 888, 15, doi: 10.3847/1538-4357/ab585c
Osorio, M., Macı́as, E., Anglada, G., et al. 2016, ApJL, Stelzer, B., Hubrig, S., Orlando, S., et al. 2009, A&A, 499,
825, L10, doi: 10.3847/2041-8205/825/1/L10 529, doi: 10.1051/0004-6361/200911750
Telleschi, A., Güdel, M., Briggs, K., et al. 2005, ApJ, 622,
Owen, J. E., Ercolano, B., & Clarke, C. J. 2011, MNRAS,
653, doi: 10.1086/428109
412, 13, doi: 10.1111/j.1365-2966.2010.17818.x
Van Der Walt, S., Colbert, S. C., & Varoquaux, G. 2011,
Pradhan, P., Huenemoerder, D. P., Ignace, R., Pollock, Computing in Science & Engineering, 13, 22
A. M. T., & Nichols, J. S. 2021, ApJ, 915, 114, Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020,
doi: 10.3847/1538-4357/ac02c4 Nature Methods, 17, 261, doi: 10.1038/s41592-019-0686-2
Vorobyov, E. I. 2013, A&A, 552, A129,
Pravdo, S. H., Feigelson, E. D., Garmire, G., et al. 2001,
doi: 10.1051/0004-6361/201220601
Nature, 413, 708, doi: 10.1038/35099508
White, R. J., & Ghez, A. M. 2001, ApJ, 556, 265,
Preibisch, T., Kim, Y.-C., Favata, F., et al. 2005, ApJS, doi: 10.1086/321542
160, 401, doi: 10.1086/432891 Zapata, L. A., Galván-Madrid, R., Carrasco-González, C.,
Reale, F., Güdel, M., Peres, G., & Audard, M. 2004, A&A, et al. 2015, ApJL, 811, L4,
416, 733, doi: 10.1051/0004-6361:20034027 doi: 10.1088/2041-8205/811/1/L4