1. Introduction
T Tauri stars (TTS) are young, low-mass (
$M_\star \leq 2\ M_\odot$
) pre-main-sequence stars (PMS), characterised by variability, strong chromospheric activity, and emission lines such as H
$\alpha$
. They can be classified into Classical T Tauri stars (CTTS) and Weak-line T Tauri stars (WTTS) based on the presence of the circumstellar disc. WTTS lack prominent circumstellar discs (Ghez, Neugebauer, & Matthews Reference Ghez, Neugebauer and Matthews1993), making them crucial for studying early stellar evolution without the complications of disc accretion. Unlike CTTS, WTTS do not exhibit obvious signs of accretion or possess optically thick discs, suggesting they represent a later evolutionary stage where the circumstellar disc has either become optically thin or dispersed. Despite this, some WTTS are located in the same region of the Hertzsprung-Russell diagram as CTTS, which indicates that there is no specific preferred disc lifetime (Walter et al. Reference Walter, Brown, Mathieu, Myers and Vrba1988; Alcala et al. Reference Alcala, Krautter, Covino, Neuhaeuser, Schmitt and Wichmann1997; Petrov Reference Petrov2003, Reference Petrov2021). WTTS emit X-rays through magnetic reconnection events, which heat the stellar plasma to extremely high temperatures (Fernández et al. Reference Fernández2004; Feeney-Johansson et al. Reference Feeney-Johansson2021). X-ray flares in WTTS are characterised by rapid decay and lower detection frequency (e.g. Stelzer, Neuhäuser, & Hambaryan Reference Stelzer, Neuhäuser and Hambaryan2000). Thus, detecting them via targeted observations is very challenging. Despite this, investigating WTTS flares remains crucial for understanding the nature of magnetic activity in the later phases of star formation.
Details of X-ray and optical observations. The last column shows the net exposure time for each observation in the corresponding energy bands used for the spectral analysis.
X-ray surveys from various missions provide ample opportunities to study X-ray emission from diverse stellar samples. The extended ROentgen Survey with an Imaging Telescope Array (eROSITA; Predehl et al. Reference Predehl2021), the primary instrument onboard the Spektrum-Roentgen-Gamma (SRG; Sunyaev et al. Reference Sunyaev2021) mission, is critical in this context as it is designed to conduct a highly sensitive all-sky survey in the soft to medium X-ray bands (0.2–10 keV). The wide field of view, repeated sky scans, and spectral capabilities make eROSITA well-suited to discover and study X-ray flares from the WTTS population. In a quest to study the X-ray sources in the Orion Complex, we identified the WTTS KM Ori in a very high X-ray flux state while cross-matching the APOGEE-2 survey (Kounkel et al. Reference Kounkel2018) catalogue with the SRG/eROSITA all-sky survey (eRASS1) Data Release 1 (DR1; Merloni et al. Reference Merloni2024) catalogue.
KM Ori is a young, K5-type WTTS within the Orion star-forming complex (Cruzalèbes et al. Reference Cruzalèbes2019). Its optical spectrum shows weak H
$\alpha$
emission, a characteristic of WTTS (Herbst et al. Reference Herbst, Herbst, Grossman and Weinstein1994; Szegedi-Elek et al. Reference Szegedi-Elek, Kun, Reipurth, Pál, Balázs and Willman2013). Spectroscopic studies (e.g. Lavail et al. Reference Lavail, Kochukhov, Hussain, Alecian, Herczeg and Johns-Krull2017; Sokal et al. Reference Sokal, Johns-Krull, Mace, Nofi, Prato, Lee and Jaffe2020) reveal a moderate surface magnetic field with a strength of approximately 1.9 kG. KM Ori exhibits significant X-ray emission (Getman et al. Reference Getman2005; Prisinzano et al. Reference Prisinzano2008), characterised by flaring events (Getman et al. Reference Getman, Feigelson, Broos, Micela and Garmire2008). Observations, including those from the Chandra Orion Ultradeep Project (COUP), have been pivotal in characterising X-ray variability and luminosity in KM Ori (e.g. Getman et al. Reference Getman2005; Bustamante et al. Reference Bustamante, Mern, Bouy, Manara, Ribas and Riviere-Marichalar2016). Additionally, the rotational modulation of X-ray emission, as seen in the COUP data (Flaccomio et al. Reference Flaccomio, Micela, Sciortino, Feigelson, Herbst, Favata and Harnden2005), suggests a spatially inhomogeneous distribution of magnetically active regions on its surface.
In this paper, we report the eROSITA detection of the WTTS KM Ori in its highest X-ray flaring state ever recorded. We characterise the properties of this observation and compare them with previous observations during flaring and quiescent states, conducted with Chandra and XMM-Newton, which we discuss in the subsequent sections. The multi-wavelength/mission data used for this work is explained in Section 2. The X-ray, optical, and SED analyses of the source are detailed in Section 3. A brief discussion, along with major conclusions, is included in the last section (Section 4).
2. Observations and data reduction
2.1. X-ray
This study utilises X-ray data from observations conducted with SRG/eROSITA, Chandra X-ray Observatory (CXO), and XMM-Newton. The details of the observations and the data reduction methods are presented in the following subsections. In order to identify the presence of any previous flare in the source, we obtained the Chandra observations of the source observed from 1999 to 2020. However, we observed a strong flare only in one of the observations during the COUP survey. We discuss this in detail in Section 3.1.1. Since our aim is to characterise the X-ray flare in the eROSITA epoch (X1), we compare the relevant Chandra observation conducted on 10 January 2003 (X2) and the lowest flux state observation with Chandra conducted on 27 November 2019 (X3) for further analysis. Also, we utilise the longest available XMM-Newton observation, conducted on 2 March 2006 (X4) to obtain high-quality spectrum. This long observation did not show the presence of any flaring events in the source. Hence, these three data (X2, X3 & X4) represent previous observations at flaring and quiescent epochs. Table 1 collates the details of these observations, including mission information, date of observation and exposure time.
2.1.1. SRG/eROSITA
For this study, we used the eROSITA data of KM Ori from the DR1 release observed during the eRASS1 survey. The data release encompasses observations of the Western Galactic hemisphere within the longitude range of
$359.94 \gt l \gt 179.94$
degrees. The data products of the observation of KM Ori (DR1 main catalogue; Merloni et al. Reference Merloni2024) were retrieved from the eROSITA-DE DR1 archive (eRODatFootnote b). The data products comprise the source and background spectra, light curves, and response files for the seven individual Telescope Modules (TM) and the merged TM combination. We used the science data products obtained for the combined telescope modules for the subsequent analyses. The spectrum was grouped to ensure a minimum of 20 counts per bin. The details of spectral and light curve analyses are given in Section 3.1.
2.1.2. Chandra/ACIS
The Orion Nebula Cluster (ONC) region has been extensively investigated (Flaccomio et al. Reference Flaccomio, Damiani, Micela, Sciortino and Harnden2003; Getman et al. Reference Getman2005) using the Advanced Camera for Imaging Spectroscopy (ACIS) onboard the Chandra. We retrieved data for all available observations of the field obtained with Chandra to see if KM Ori was flaring in any of these epochs. There are 87 observations spanning from 1999 to 2020, and KM Ori was detected in 60 observations. We obtained the level-2 data of these observations from the public Chandra Data Archive and reprocessed using the software package CIAO (Fruscione et al. Reference Fruscione, Silva and Doxsey2006) version 4.16, along with the calibration database (CALDB) 4.11.0. To the reprocessed level-2 data (event file), an energy filter of 0.3–8 keV is applied. These event files were then checked for flaring background using deflare task. We found background flaring in two observations, which were cleaned to contain only good time events. Certain observations were affected by readout streaks, which hindered the effectiveness of the background flare cleaning process. To resolve this, we utilised the acis_streak_map tool to create a detailed map of the readout streaks associated with bright sources. This map enabled us to identify and eliminate the streaks, thereby enhancing the reliability of our background analysis. We used the source detection algorithm wavdetect (Freeman et al. Reference Freeman, Kashyap, Rosner and Lamb2002) to detect the sources within the CCD. Wavelet scales ranging between 1 and 32 in steps of 2
$^{n}$
(where n = 0,1,2,3,4,5) were used. To ensure reliable detections and minimize false positives, a stringent significance threshold of 10
$^{-6}$
was set. A circular region with a radius of 6 arcsec was defined for the source to encompass 90% of the PSF at 1.5 keV, as determined by the wavdetect algorithm. A circular region of radius
$\sim$
30 arcsec devoid of sources on the same CCD chip was selected for background measurements.
2.1.3. XMM-Newton/EPIC
XMM-Newton observed the region containing the source for a duration of about 94 ks on 2 March 2006. We retrieved the X-ray data from the European Photon Imaging Camera (EPIC; Strüder et al. Reference Strüder2001) PN and MOS detectors, which were observed in full window mode with a medium filter. We performed the data reduction using SAS version 21.0, employing the epproc and emproc tasks and utilising updated calibration files as of March 2024. We identified and eliminated the intervals of flaring particle background from the event lists based on threshold rates (0.5 counts s
$^{-1}$
for PN, 0.2 counts s
$^{-1}$
for MOS1, and 0.23 counts s
$^{-1}$
for MOS2) derived from high-energy light curves. Event lists were filtered to retain the highest quality events (FLAG=0) with PATTERN
$ \lt =$
4 (single and double events) for PN and PATTERN
$ \lt =$
12 (up to quadruples) for MOS. Spectra were extracted from circular regions with a radius of 24 arcsec centred on the source, while background spectra were extracted from multiple circular regions (28 arcsec radii) on the same CCD chip. We generated the spectra and response files (ARF and RMF) for the cleaned event lists using the xmmselect task. We further rebinned the spectra and linked them to the response and background files using the specgroup tool. This process ensured a minimum of 25 counts per bin, with an oversampling factor of 5 applied. The background corrected light curves were generated using the epiclccorr task in SAS.
2.2. Optical
We conducted low-resolution optical spectral observations of KM Ori using the HFOSC instrument on the 2.01 m Himalayan Chandra Telescope (HCT) at two epochs on 27 February 2024, and 25 March 2024. The observations were carried out with grism 7 (Gr7) covering the wavelength range of 3800–6840 Å, employing the 167l slit. Standard IRAF (Tody Reference Tody and Crawford1986) routines were used for spectral analysis. Spectral data obtained from both epochs exhibited a resolution of
$R\sim 1\,000$
.
3. Analysis and results
Building upon the APOGEE-2 survey of the Orion Complex (Kunkel et al. 2018), we have constructed a catalogue of 1030 sources with both eRASS1-detected X-ray counterparts and APOGEE-derived stellar parameters. The eROSITA flux in the 0.2–2.3 keV band (Fig. 1) reveals KM Ori as the most luminous X-ray source in our sample,
$\sim$
30 times the median flux (0.09710−12 erg cm−2 s−1) for the complex. The presence of a hard X-ray (2.3–5 keV) component in the eRASS1 catalogue (UID_Hard = 208309600003) identifies KM Ori as a particularly compelling target, motivating us to conduct an in-depth X-ray and optical investigation, which is presented below.
The figure shows the eRASS1 flux (0.2–2.3 keV) versus the stellar effective temperature of APOGEE-2 sources in the Orion Complex.
3.1. X-ray analysis
3.1.1. Light curves
The eRASS1 light curve, included in the standard data products, encompasses time series data across three distinct energy bands (0.2–0.6 keV, 0.6–2.3 keV, and 2.3–5 keV), binned at 10 s intervals. We utilised the python code eRebin provided by Arcodia et al. (Reference Arcodia2021, Reference Arcodia2024) to create rebinned source-only (background subtracted) light curves for different energy bands provided in the standard light curve. The code rebins the light curves for 40 s, which is the on-target exposure of an eROday (4 h). We added these light curves using lcmath to generate a total light curve in the 0.2–5 keV band. The gaps of
$\sim$
4 h seen in the eROSITA light curve correspond to one revolution of the spacecraft in all-sky survey mode. The light curve showed moderate variability with fractional rms variation
$F_{\mathrm{rms}}=0.20\pm0.07$
across the
$\sim$
86 ks observation.
We extracted light curves for all Chandra observations of KM Ori in the 0.2–5 keV band. The light curve spanning from 1999 to 2020 is shown in Fig. 2. The highest flux state within this long-term dataset was observed during the COUP observation with Obs ID 3744, which is highlighted in the inset of the figure. To compare the eROSITA flare observation with previous flare data from the source, we focused our analysis on this specific Chandra observation (X2). Additionally, we analysed the Chandra observation with the lowest count rate (X3) and the longest available XMM-Newton observation (X4). The Chandra light curve of observation X2 (
$F_{\mathrm{rms}}=0.32\pm0.02$
) clearly displayed flaring activity, consistent with previous reports by Getman et al. (Reference Getman2005, Reference Getman, Feigelson, Broos, Micela and Garmire2008). The flaring was observed within the initial
$\sim$
85 ks of the
$\sim$
165 ks long Chandra observation. On the other hand, the light curves for X3 (duration
$\sim$
48 ks) and X4 (duration
$\sim$
94 ks) did not exhibit significant variability (
$F_{\mathrm{rms}} \lt 0.1$
and 0.13
$\pm$
0.02, respectively).
Long-term Chandra light curve (0.2–5 keV) of KM Ori, spanning from 12 October 1999 to 25 December 2020. The inset shows the light curve extracted for the highest X-ray flux state, which is observed in the COUP survey (ObsID: 3744). The red dashed vertical line represents the epoch of eROSITA observation.
3.1.2. Spectra
We analysed the X-ray spectra from observations X1, X2, X3, and X4 using XSPEC version 12.13.0c (Arnaud Reference Arnaud, Jacoby and Barnes1996) (HEASOFT version 6.31.1). We used the Astrophysical Plasma Emission Code (APEC) for optically thin thermal plasma models (Smith et al. Reference Smith, Brickhouse, Liedahl and Raymond2001) to fit the data. The abundances relative to solar values were set to values from Grevesse & Sauval (Reference Grevesse and Sauval1998). The correction for the Galactic absorption along the line-of-sight was accounted for using the Tuebingen-Boulder ISM absorption model (TBabs). We also checked other abundance libraries and phabs model to account for the Galactic absorption. We found that the
$N_H$
value, along with other parameters and flux, are within error bars. Since there is no evidence of circumstellar material in the source, we did not apply any model for intrinsic absorption.
We used the 0.2–5 keV band for eROSITA, 0.3–8 keV for Chandra and 0.3–6 keV for XMM-Newton, considering the instrument calibration and background dominance. The X-ray spectra for observations X2 and X3 were extracted using the CIAO script specextract and grouped to a minimum of 25 and 15 counts/bin, respectively.
$\chi^2$
statistic was applied as the fit statistic. The uncertainties on the model parameters are at a 90% confidence level.
We applied time-resolved spectroscopy to the Chandra X2 data and extracted the spectra for the flaring state depicted in Fig. 3. We required 3 APEC components (3-T model) to fit the XMM-Newton (X4) and Chandra (X2) spectra, whereas the lower quality spectra of eROSITA (X1) and Chandra (X3) were fitted well with a 2-T model. The absorbing column density (
$N_\mathrm{H}$
) was set as a free parameter for all the observations. The best-fit values of N
$_H$
fall within the range derived using the extinction law of Ryter (Reference Ryter1996), that is,
$N_\mathrm{H}\,[\mathrm{cm}^{-2}$
] =
$A_v$
[mag]
$\times$
2.0
$\times$
10
$^{21}$
, and were consistent among the observations. The cross-calibration differences between EPIC PN, MOS1 and MOS2 spectra were accounted for using CONSTANT model in xspec.
X-ray (0.2–5 keV) light curves (background-subtracted) and spectra for the eROSITA (X1), Chandra (X2 & X3), and XMM-Newton (X4) epochs of observations. The shaded region in the X2 light curve indicates the flaring interval observed in the Chandra observation for which we extracted the spectrum. For the X4 epoch, the EPIC/PN light curve is shown. The spectra were fitted with multi-temperature plasma models. In the upper panels of the spectral fitting plots, the data and best-fit models, including individual model components, are displayed. The lower panels show the residuals of the spectral fit. In the XMM-Newton spectra, the black, red, and green points represent the PN, MOS1, and MOS2 data, respectively.
We calculated the unabsorbed flux for the 0.2–5 keV band using the convolution model cflux in XSPEC. The corresponding flux and the distance (
$d = 394.2$
pc) were then used to determine the X-ray luminosity (L0.2-5 keV). We also calculated the emission measure (EM) of each plasma component from the normalisation (N) of the APEC model (
$EM = 10^{14} \times 4 \pi D^2 N$
). The details of the best-fit temperatures and the derived parameters are given in Table 2. We found that the luminosities and EMs are higher in the eROSITA observation than in other datasets. This indicates that the source was in a flaring state at the eROSITA epoch. The model components indicate the presence of thermal plasma with temperatures around 0.3 keV and 3 keV in the eROSITA spectrum. An additional plasma component with kT around 1 keV was observed in Chandra and XMM-Newton spectra.
Best-fit X-ray spectral parameters for the absorbed thermal plasma model and the luminosity in the 0.2–5 keV band.
Fig. 3 shows the light curves and spectral fitting plots for the X-ray observations with eROSITA, Chandra and XMM-Newton described above. Fig. 4 displays the X-ray (0.2–5 keV) luminosities for these observations.
3.1.3. Contamination from nearby sources
The angular resolution of eROSITA is
$\sim$
16 arcsec, and that of EPIC cameras onboard XMM-Newton telescope is
$\sim$
6 arcsec. The ACIS camera onboard Chandra has an angular resolution of
$\sim$
0.5 arcsec. The distinct resolution of each mission is evident from Fig. 5. Due to the limited angular resolution of eROSITA and XMM-Newton, the FOV of KM Ori appears crowded. In the case of eROSITA, a spurious flag value of 1 (flag_sp_scl=1) is assigned to the KM Ori data, indicating a crowded region. This crowding raises the possibility of contamination in KM Ori from nearby sources. Interestingly, three distinct sources (labelled as S1, S2, and S3 in Fig. 5) are seen in the Chandra image. Because of the lower spatial resolution of eROSITA and XMM-Newton, it is challenging to separate nearby sources from KM Ori. Sources S1 and S2 are within the extraction region of the eROSITA data, while source S3 is outside this extraction region. Hence, we calculated the percentage of flux due to the sources S1 and S2 that could be contributing to the flux of KM Ori. Using the Chandra data, we first measured the flux of the sources KM Ori, S1 and S2 individually. The sources S1 and S2 are very faint (spectral counts < 40), so the flux cannot be obtained from model fitting. Therefore, we used PIMMS to calculate their flux from the count rate in the energy range of 0.5–8 keV. Next, we measured the total flux from the three objects KM Ori, S1 and S2 considering a circular extraction region of 30 arcsec (approximately the size of extraction regions considered for eROSITA and XMM-Newton). We find the percentage of flux (% F
$_{\mathrm{cont}}= \frac{F_{KM\ Ori}}{F_{S1+S2}}$
$\times$
100) contributed by sources S1 and S2 to be less than 2.5% for the Chandra data. The contribution of flux from the sources S1 and S2 is not significant and does not influence the XMM-Newton and eROSITA spectra of KM Ori.
3.2. Confirming the weak-line nature of KM Ori
3.2.1 Optical spectra
The primary aim of the optical spectral analysis was to evaluate the H
$\alpha$
emission strength of KM Ori. We observed no significant variation in the H
$\alpha$
equivalent width (EW) between the two epochs O1 (
$-5.1$
Å) and O2 (
$-4.9$
Å) (see Fig. 6 [Left]). This value falls within the expected range for WTTS, as mentioned by Gras-Velázquez & Ray (Reference Gras-Velázquez and Ray2005). Martin (1998) found a correlation between spectral type and H
$\alpha$
EW, concluding that K5 spectral type stars have H
$\alpha$
EW equal to or less than
$-5$
Å. Based on this analogy, KM Ori has a spectral type of K5, which supports the spectral type identified in the literature (Cruzalèbes et al. Reference Cruzalèbes2019).
Plot showing the changes in X-ray luminosities obtained with eROSITA (X1), Chandra (X2 & X3) and XMM-Newton (X4) epochs.
3.2.2 Spectral energy distribution
We constructed the spectral energy distribution (SED) of KM Ori from the available photometry in the Gaia G,
$G_{BP}$
and
$G_{RP}$
bands (Gaia Collaboration et al. 2023b), the 2MASS J, H and K
$_s$
bands (Skrutskie et al. Reference Skrutskie2006), the Spitzer IRAC channels Ch2, Ch3, Ch4 (Fazio et al. Reference Fazio2004) and Gaia synthetic photometry in the Johnson-Kron-Cousins BVRI, the SDSS griz, the PANSTARRS y, and the HST-ACS-WFC F606W and F814W bands (Gaia Collaboration et al. 2023a). SED templates corresponding to the queried photometric bands are derived from the BT-Settl model grid (Allard et al. Reference Allard, Guillot, Ludwig, Hauschildt, Schweitzer, Alexander, Ferguson and Martn2003; Allard, Homeier, & Freytag Reference Allard, Homeier, Freytag, Johns-Krull, Browning and West2011, Reference Allard, Homeier and Freytag2012), spanning from
$10^3$
to
$10^7$
Å, allowing
$T_{\mathrm{eff}}$
as a free parameter (2 700 to 70 000 K). The best fit was determined by minimising the
$\chi^2$
for each instance by using a Python routine (Arun et al. Reference Arun2021; Shridharan et al. Reference Shridharan2022; Bhattacharyya et al. Reference Bhattacharyya2022). We fixed the
$\log g$
parameter in the model grid to a value of 4, as derived from literature for KM Ori (Kounkel et al. Reference Kounkel2018). The observed photometry is corrected for extinction using the derived optical extinction (
$A_V$
= 1.05 mag) from Green’s map (Green 2019). The best-fit value of
$T_{\mathrm{eff}}$
is 4 900 K, which lies in the expected range for a K3-type star. The SED and best-fit theoretical model for KM Ori are displayed in Fig. 6 [Right]. We do not see any excess emission at IR wavelengths, which suggests the lack of dust in the circumstellar environment around the source.
CTTS show variability in the optical band and IR excess in the SED due to the presence of dust in the disc. Our observations of KM Ori have consistently shown an average H
$\alpha$
strength of
$-5$
Å. This finding aligns with the typical H
$\alpha$
strength observed in WTTS, and no significant variability was detected between the observations conducted on 27 February and 25 March 2024. Furthermore, modelling the SED yields the best fit at
$T_{\mathrm{eff}} = 4\,900\ K$
and indicates negligible infrared excess up to 5
$\mu$
m. This absence of excess emission is consistent with a minimal circumstellar disc, further solidifying the classification of KM Ori as a WTTS.
4. Discussion and summary
During our analysis of the APOGEE-2 survey of the Orion Complex with the eROSITA DR1 dataset, we identified the K5-type WTTS KM Ori exhibiting an exceptionally elevated X-ray flux state compared to other WTTS objects. The X-ray luminosity observed during the eROSITA observation suggests the occurrence of a significant and unforeseen high X-ray flare state in KM Ori. We generated the long-term X-ray light curve using Chandra observations spanning from 1999 to 2020. Our analysis of the 20-year-long Chandra light curve revealed a high flaring state in KM Ori in one of the observations obtained during the COUP survey (Getman et al. Reference Getman2005). This X-ray flare lasted for
$\sim 85$
ks with a peak luminosity (
$L_\mathrm{pk}$
; 0.5–8 keV) of around
$10^{32} \mathrm{erg\ s}^{-1}$
during the 2003 COUP survey (Getman et al. Reference Getman, Feigelson, Broos, Micela and Garmire2008). The eROSITA observation reveals an X-ray luminosity similar to that of the 2003 COUP flaring observation, suggesting that eROSITA captured the source during a flaring episode. Further, we conducted a systematic X-ray analysis of the source utilising eROSITA data and three previous observations from Chandra and XMM-Newton, as outlined in Section 3.1.2. We used these three observations to represent the flaring (X2) and quiescent (X3, X4) states, allowing us to make a comparison with the eROSITA flaring observation of the source.
Our analysis revealed a higher X-ray luminosity in the eROSITA observation compared to other epochs. Remarkably, the X-ray (0.2–5 keV) luminosity exceeds
$3.9^{+8.2}_{-1.2}$
times more than that of the Chandra observation obtained for the time-averaged flaring interval. This clearly indicates that eROSITA caught the object in its most intense X-ray flaring state recorded to date. The ratio of X-ray to bolometric luminosity, log(
$L_X/L_{\mathrm{bol}}$
), an indicator of the coronal X-ray activity, comes around
$-2.4$
to
$-2.9$
for different energy bands. The range of values is similar to those reported during the flaring state of T Tauri stars (e.g. Imanishi, Koyama, & Tsuboi Reference Imanishi, Koyama and Tsuboi2001; Wolk et al. 2005). X-ray flaring events have also been reported in other WTTS sources (e.g. Tsuboi et al. Reference Tsuboi, Koyama, Murakami, Hayashi, Skinner and Ueno1998; Stelzer et al. Reference Stelzer2003; Fernández et al. Reference Fernández2004; Jensen, Cohen, & Gagné Reference Jensen, Cohen and Gagné2009; Baldovin-Saavedra et al. Reference Baldovin-Saavedra, Audard, Duchêne, Güdel, Skinner, Paerels, Ghez and McCabe2009; Uzawa et al. Reference Uzawa2011; Günther et al. Reference Günther, Wolter, Robrade and Wolk2013; Skinner & Güdel Reference Skinner and Güdel2021). However, coronal X-ray activity higher than that of KM Ori has been reported in only a few WTTS, such as TWA-7 (Uzawa et al. Reference Uzawa2011) and V773 Tau (Tsuboi et al. Reference Tsuboi, Koyama, Murakami, Hayashi, Skinner and Ueno1998).
Left – 120
$\times$
120 arcsec Chandra ACIS-I CCD image of KM Ori and nearby sources (S1, S2, and S3). The image is binned to a pixel size of 0.98 arcsec. Middle – 120
$\times$
120 arcsec XMM-Newton EPIC PN CCD image of KM Ori and nearby sources (S1, S2, and S3). The image is binned to a pixel size of 0.8 arcsec. Right – 120 arcsec x 120 arcsec eROSITA CCD image of KM Ori observed using all 7 telescope modules. The image is binned to a pixel size of 0.8 arcsec. The blue dashed line circle denotes the extraction region (
$\sim$
30 arcsec) of KM Ori for the eROSITA and XMM-Newton data. The blue solid line circle denotes the extraction region (
$\sim$
6 arcsec) of KM Ori for the Chandra data. The nearby sources S1, S2, and S3 are marked with yellow crosses. Their coordinates in XMM-Newton and eROSITA images are fixed at Chandra coordinates.
Left: Representation of the H
$\alpha$
profiles observed on 27 February 2024 (O1) and 25 March 2024 (O2). Right: SED of KM Ori fitted with the theoretical BT-Settl model with
$A_V$
= 1.05, log(g) = 4, and
$T_\mathrm{eff}$
= 4 900 K. The SED shows that KM Ori does not have any IR excess.
Pre-main sequence stars have been reported to exhibit X-ray superflares/megaflares with peak luminosities in the range log(L
$_X$
; erg s
$^{-1}$
)
$\sim$
31–33 and total radiated energies log(E
$_X$
; erg)
$\sim$
34–38 (Getman, Feigelson, & Garmire Reference Getman, Feigelson and Garmire2021). Nevertheless, such flaring events are difficult to detect with normal pointing observations. During the flaring interval of Chandra observation of KM Ori, we observed a time-averaged X-ray (0.5–8 keV) luminosity of approximately 5 × 1031 erg s−1. The peak X-ray luminosity for the same observation is
$L_\mathrm{pk} \mathrm{(0.5-8\ keV)} \sim 1.5 \times {\mathrm{10^{32}\ erg\ s^{-1}}}$
(Getman et al. Reference Getman, Feigelson, Broos, Micela and Garmire2008). Interestingly, in the eROSITA observation, the source showed a time-averaged X-ray luminosity of
$1.6 \times {\mathrm{10^{32}\ erg\ s^{-1}}}$
in the same energy band, indicating a superflare/megaflare state. Additionally, Getman & Feigelson (Reference Getman and Feigelson2021) found that the duration of these flares can range from 3 ks to 154 ks. Though we observe marginal variability in the eROSITA light curve, it is likely that eROSITA has not captured the complete evolution of the flare in KM Ori, that is, quiescent, rise, and decay phases (see Fig. 3). The high X-ray luminosity (log(
$L_\mathrm{0.5-8\ keV}$
; erg s
$^{-1}$
)=32.32) and the time-averaged energy (log(
$E_\mathrm{0.5-8\ keV}$
; erg)=37.24) at the eROISTA epoch suggest the source to be in a megaflare state that persisted longer than the Chandra flaring episode. The eROSITA flare may be classified as a slow-rise top-flat (SRTF) flare (Getman et al. Reference Getman, Feigelson, Broos, Micela and Garmire2008), where variations appear to occur more slowly than in most flares. We cannot entirely rule out the occurrence of repeated short-term flares during the observation, although detecting them is challenging due to the low cadence of eROSITA. However, such intense, repeated flaring events were not observed in previous Chandradata, suggesting that KM Ori may have remained in an SRTF flare state throughout the eROSITA observation (
$\sim$
86 ks). These flares, characterised by slow rises, long-duration peaks, and/or very long decays, are similar to the slow-rise long-duration flares observed in sources like COUP 1268 and COUP 597, with durations of 100 ks and 150 ks, respectively (Getman et al. Reference Getman, Feigelson, Broos, Micela and Garmire2008; Favata et al. Reference Favata, Flaccomio, Reale, Micela, Sciortino, Shang, Stassun and Feigelson2005). This is consistent with our source KM Ori, though we do not have a defined flare shape (rise and decay) in the eROSITA light curve. Such long-duration flares are often attributed to X-ray emission from large, hot coronal loops, with sizes ranging from 1.5 R
$_{\star}$
to 5.5 R
$_{\star}$
(Favata et al. Reference Favata, Flaccomio, Reale, Micela, Sciortino, Shang, Stassun and Feigelson2005).
Our analysis of the X-ray spectra of KM Ori at various epochs reveals the presence of multiple thermal plasma components within its corona with temperatures ranging between
$\sim$
0.3 keV (
$\sim$
3 MK) and
$\sim$
5 keV (
$\sim$
60 MK). The emission measures for all thermal components are notably higher during the eROSITA epoch. According to the solar-flare model (Shibata & Yokoyama Reference Shibata and Yokoyama1999), the increase in the emission measure could be attributed to heightened evaporation of chromospheric plasma due to the heating of deposited particles (Waterfall et al. Reference Waterfall, Browning, Fuller and Gordovskyy2019). Hence, a higher emission measure indicates KM Ori to be a highly flaring/variable star.
Given that KM Ori lacks an accreting disc (weak-line nature) and has no evidence of binarity (Herbst et al. Reference Herbst, Herbst, Grossman and Weinstein1994; Szegedi-Elek et al. Reference Szegedi-Elek, Kun, Reipurth, Pál, Balázs and Willman2013; Cruzalèbes et al. Reference Cruzalèbes2019), it suggests that neither of these factors is responsible for generating significant flares in the source. The low-mass PMS stars exhibit pronounced X-ray emission originating from magnetically confined plasma structures in the corona (Lang et al. Reference Lang, Jardine, Donati, Morin and Vidotto2012), driven by magnetic fields generated by internal dynamos. Evolutionary tracks for stars of this mass and age suggest KM Ori be fully convective (Lawson et al. Reference Lawson, Feigelson and Huenemoerder1996), indicating that a turbulent dynamo might be operating in this star. The higher occurrence of large flares in WTTS could be due to the more frequent explosive magnetic reconnection, which leads to plasma heating and higher temperatures (Waterfall et al. Reference Waterfall, Browning, Fuller and Gordovskyy2019). The elevated levels of magnetic activity in the source may result from the early dissipation of the interacting discs, leading to rapid rotation and strong magnetic dynamos. A similar mechanism has been suggested for another young TTS (age
$\sim$
1 Myr) V773 Tau (HD 283447), which exhibits strong X-ray emissions during quiescent states (1031 erg s−1) and frequent day-long flares peaking at around 1032 erg s−1 (Feigelson et al. Reference Feigelson, Welty, Imhoff, Hall, Etzel, Phillips and Lonsdale1994; Skinner et al. Reference Skinner, Guedel, Koyama and Yamauchi1997; Tsuboi et al. Reference Tsuboi, Koyama, Murakami, Hayashi, Skinner and Ueno1998).
The identification of the most intense X-ray flare in KM Ori underscores the capability of SRG/eROSITA to detect X-ray flaring occurrences within the WTTS population. Such flares are infrequent and challenging to detect through targeted observations. Investigating larger samples of WTTS from such surveys holds promise for elucidating the fundamental mechanisms governing magnetic activity and topology in young stellar systems. This work is likely the first study to analyse a long-term X-ray light curve of a WTTS. In future work, we explore the long-term X-ray variability of a sample of WTTS, including additional multi-wavelength observations (Anilkumar et al. in preparation). These observations will provide a better understanding of the variability and nature of their X-ray emissions, as well as the physical characteristics of the emitting regions.
Acknowledgement
We are grateful to the anonymous reviewer for their encouraging evaluation and recommendation for acceptance of this manuscript. This work is based on data from eROSITA, the soft X-ray instrument aboard SRG, a joint Russian-German science mission supported by the Russian Space Agency (Roskosmos), in the interests of the Russian Academy of Sciences represented by its Space Research Institute (IKI), and the Deutsches Zentrum für Luft- und Raumfahrt (DLR). The SRG spacecraft was built by Lavochkin Association (NPOL) and its subcontractors, and is operated by NPOL with support from the Max Planck Institute for Extraterrestrial Physics (MPE). The development and construction of the eROSITA X-ray instrument was led by MPE, with contributions from the Dr. Karl Remeis Observatory Bamberg & ECAP (FAU Erlangen-Nuernberg), the University of Hamburg Observatory, the Leibniz Institute for Astrophysics Potsdam (AIP), and the Institute for Astronomy and Astrophysics of the University of Tübingen, with the support of DLR and the Max Planck Society. The Argelander Institute for Astronomy of the University of Bonn and the Ludwig Maximilians Universität Munich also participated in the science preparation for eROSITA. This research has made use of data obtained from the Chandra Data Archive provided by the Chandra X-ray Center (CXC). BM, SB, and SSK acknowledge the financial support from CHRIST (Deemed to be University, Bangalore) through the SEED money projects (No: SMSS-2335, 11/2023 & SMSS-2220,12/2022). The authors acknowledge the support provided by the Department of Science and Technology (DST) under the ‘Fund for Improvement of S & T Infrastructure (FIST)’ programme (SR/FST/PS-I/2022/208). BM and SSK acknowledge the support provided by the SERB project (CRG/2023/005271). SE thanks Dr T. Saha for the discussions on the data. VJ and SSK thank the Inter-University Centre for Astronomy and Astrophysics (IUCAA), Pune, India, for the Visiting Associateship.
Data availability statement
The data underlying this article will be shared on reasonable request.
Open access
