STAR FORMATION IN ORION'S L1630 CLOUD: AN INFRARED AND MULTI-EPOCH X-RAY STUDY

The Spitzer InfraRed Array Camera (IRAC) archival data for L1630 analyzed here comprises a subset of observations carried out in 2004 October (PI: G. Fazio; see Fang et al. 2009). The observations were obtained using IRAC's High Dynamic Range mode, yielding images with 10.4 s and 0.4 s exposure times. The MOsaicker and Point source EXtractor (MOPEX) and the Astronomical Point source EXtractor (APEX)⁶ were used, respectively, to create mosaics of L1630 using archival Spitzer IRAC observations and for performing aperture photometry on the resulting mosaics. Two mosaics were made, one for the long (10.4 s) exposures (Figure 1, left) and one for the short (0.4 s) exposures, so as to detect faint sources as well as bright sources that might have been saturated in the long exposure. Before mosaicking the images, overlap correction was performed via the MOPEX overlap correction module. The MOPEX sliding box routine was employed for background subtraction. Aperture photometry was performed using an aperture radius of 10 pixels.

Zoom In Zoom Out Reset image size

A [3.6–4.5] versus [5.8–8.0] color–color diagram (Figure 2) was created to classify YSOs as Class 0/I, Class II, or "transition disks."  The last category represents objects whose SEDs indicate that they are in transition between Class II (disk-bearing) and Class III (apparently diskless) YSOs (Fang et al. 2009; Allen et al. 2004). In general, this classification scheme is useful for determining the evolutionary state of a YSO because disk dissipation leads to a sharp decrease in ∼1–10 μm emission. The Class 0/I, Class II, and transition disk YSOs were classified according to the Spitzer-based color–color criteria originally used in Fang et al. (2009). These classification criteria are as follows: (a) 0.4 ⩽ [5.8]–[8.0] ⩽ 1.4 and 0.8 ⩽ [3.6]–[4.5] ⩽ 2.0 (Class 0/I); (b) 0.4 ⩽ [5.8]–[8.0] ⩽ 1.1 and 0.1 ⩽ [3.6]–[4.5] ⩽ 0.8 (Class II); (c) 0.2 ⩽ [5.8]–[8.0] ⩽ 1.0 and −0.1 ⩽ [3.6]–[4.5] ⩽ 0.2 (transition disk). Class III sources (defined as having little or no detectable IR excess above photospheric emission) were classified using X-ray observations and a Two Micron All Sky Survey (2MASS) color–color diagram (see below). Only sources with a signal-to-noise ratio >3 were considered.

Zoom In Zoom Out Reset image size

Some sources display excess 8 μm emission in Figure 2, placing them outside the standard YSO classification regions in color–color space. After individual inspection, four of these infrared sources were found to be positionally correlated with diffuse, bow-shock-like infrared emission (Figure 3). All four of these sources are also X-ray emitters. Since the bow-shock-like extended infrared emission in the Spitzer observations contaminates the point-source flux at 8 μm, they have been tentatively classified assuming a smaller amount of intrinsic 8 μm emission (i.e., by reducing their values of [5.8–8.0] in Figure 2).

Zoom In Zoom Out Reset image size

Chandra X-ray sources (Section 2.2) and Spitzer infrared Chandra sources were positionally matched with the Tool for Operations on Catalogues and Tables, version 3.9. X-ray and infrared sources were considered coincident if separated by <3'' (to account for the point-spread function (PSF) for off-axis detections and the pointing accuracy of Spitzer and Chandra) and if they met the aforementioned requirements for both infrared and X-ray detection. The number of infrared counterparts to X-ray sources is listed in Table 1 for each Chandra observation, and the identifications and infrared colors of these counterparts are listed in Table 2. The infrared luminosity L_IR was determined by integrating the 2MASS and Spitzer flux from 1.23 μm to 8.0 μm. Infrared SEDs created using 2MASS, Spitzer, and WISE photometry for all X-ray sources with infrared counterparts are presented in the Appendix. Some SEDs in the Appendix show WISE excesses >8 μm that could potentially be signatures of a cool dust component not accounted for in our (2MASS/Spitzer-based) classifications. However, many of these sources have been flagged by WISE pipeline reduction procedures and are likely either spurious detections, contaminated by a diffraction spike or scattered light halo from a nearby bright source, or are highly infrared-variable.

In order to ensure classification of very red (potential Class 0 or Class I) sources, we required detection at 5.8 μm for all YSOs classified in this study other than Class III. This requirement led to the inclusion of some sources that were not included in the photometric study of Fang et al. (2009), who required detection in both 3.6 μm and 4.5 μm IRAC bands. This condition also potentially leads to the exclusion of many Class III YSOs from Figure 2. To include these X-ray luminous objects, we compiled a list of 2MASS counterparts to X-ray sources and extracted their photometry in the J, H and K bands using the NASA/IPAC Infrared Science Archive (Table 2). We then constructed a near-infrared color–color diagram (Figure 4) to identify Class III YSOs. The MS line and reddening vectors are taken from Bessell & Brett (1988) and Cohen et al. (1981) respectively. The classical T Tauri star (cTTS) locus is the area in the plot where cTTSs suffering little or no reddening are located (Meyer et al. 1997). Given that pre-MS stars are typically ∼10³ brighter in X-rays than MS stars (Feigelson & Montmerle 1999), we conclude that all X-ray sources in L1630 that appear as reddened MS stars are likely Class III pre-MS stars.

Zoom In Zoom Out Reset image size

The analysis is focused on 11 Chandra X-Ray Observatory observations obtained with the Advanced CCD Imaging Spectrometer array (ACIS-I), spanning four years, of the star forming region L1630. Individual exposure times were 18–30 ks. V1647 Ori, a well-studied highly variable YSO, was the target in 9 of these 11 observations. Details concerning results for V1647 Ori itself can be found in Teets et al. (2011), Hamaguchi et al. (2012), and Kastner et al. (2006). The date and exposure time of each observation included in the present study are listed in Table 1. X-ray data reduction was performed using Chandra's Interactive Analysis and Observation (CIAO)⁷ version 4.3. The images were cropped to contain only data from the ACIS-I imaging array. All 11 observations were merged with CIAO's dmmerge and reproject-events to generate the master merged event file and merged exposure map used for source detection. The merged image is shown in Figure 1 (right) with the respective fields of view of the 11 component exposures represented by colored boxes.

All Chandra X-ray observations, including the ∼240 ks merged observation, were filtered into soft (0.5–2.0 keV), hard (2.0–8.0 keV), and broad (0.5–8.0 keV) energy bands. We removed photons with energies <0.5 keV and >8.0 keV so as to avoid spurious and background events. The CIAO process fluximage was used to create exposure maps for each observation. All exposure maps were created with an effective energy of 1.6 keV. Tests of exposure maps constructed with effective energies larger than 1.6 keV do not significantly affect the results. The effective exposure time is calculated for each source using a merged exposure map that accounts for instrumental effects such as effective area, quantum efficiency, and telescope pointing motion. To determine count rates and other source properties, CIAO's wavdetect source detection algorithm was employed in each band in all observations, including the final merged observation, with the detection probability threshold set to P = 10⁻⁶. The wavdetect scales (i.e., wavelet radii in pixels) used were 1, 2, 4, 8, and 16 in order to optimize detection of point sources well off-axis. The wavdetect source extraction region is based on the Chandra PSF and chip position and is calculated considering a PSF energy of 1.4967 keV and an encircled energy fraction of 40%. The faintest source detected by wavdetect in any of our observations had four events; therefore, at least four events are needed to be considered a reliable on-axis detection in this study.

Source count rates were determined using exposure maps to account for the effective exposure time corresponding to the position of each source. Hardness ratios, defined as (H − S)/(H + S), were calculated using the hard (H) and soft (S) count rates determined for each source. Median X-ray energies (Table 2) were determined for the X-ray detected L1630 YSOs using the wavdetect position for each source and the CIAO tool dmstat. Errors on these median energies were calculated using the standard error on the mean as a proxy for standard error on the median. Values of N_H for all sources were estimated using the relationship between median energy and absorption in YSOs determined by Feigelson et al. (2005); for sources with more than 100 counts, N_H was calculated using the best-fit model flux. Assuming a distance to L1630 of 400 pc, intrinsic X-ray luminosity was calculated from the best-fit flux model. For sources with fewer than 100 counts, X-ray luminosity was not estimated.

The flux detection limit of the Chandra data set was determined using the Portable, Interactive Multi-Mission Simulator⁸ software developed by the NASA High Energy Astrophysics Science Archive Science Center. In a 234 ks observation, one X-ray photon corresponds to an energy flux of 2.4 × 10⁻¹⁷ erg s⁻¹ cm⁻², assuming an intrinsic source spectrum consisting of a 10 MK thermal plasma with a metal abundance of 0.4 times solar (Getman et al. 2005; Forbrich & Preibisch 2007), a representative intervening column density of 7.5 × 10²¹ cm⁻² (Table 2 and Section 3.1), and adopting an energy band of 0.5–8.0 keV. At our assumed distance of 400 pc to L1630, an on-axis four event source detection therefore corresponds to an intrinsic luminosity of 1.8 × 10²⁷ erg s⁻¹.

X-ray spectral analysis was performed on all sources with ≳100 counts in the merged observation. CIAO's specextract routine was run for each source to generate/extract spectra and associated spectral response files for each individual observation. If more than 10 events were detected from a source in any particular observation, a merged spectrum for that source was created using the specextract "combine" parameter. It is important to note that the resulting merged spectrum from multiple observations could be strongly influenced by source variability (see Section 4.2). Many of the sources did not have enough counts in an individual observation to constrain the spectral fitting; therefore, we only performed spectral analysis on the combined observations.

Models of absorbed one- or two-temperature thermal plasmas were fit to the merged spectra to estimate plasma temperature, intervening hydrogen absorbing column density, and flux. We performed our spectral analysis with HEASARC's X-ray Spectral Fitting Package using the wabs absorption model and the apec plasma model, assuming a uniform density plasma with a global pattern of abundances fixed to the value Z = 0.8 Z_☉. If the reduced χ² of a two-temperature fit was smaller than for a one-temperature fit, an F-test was performed to ensure the fit was improved statistically (i.e., that the F-test P < 0.05).

The X-ray spectral properties for each source are listed in Table 4, and mean values for each class are listed in Table 5. The count rate of each source listed in Table 4 was calculated from the merged observation and the exposure map corrected effective exposure time. The remaining parameters listed in Table 4 are the best-fit model parameters from spectral fitting. The listed values for F_X are observed (i.e., "absorbed") X-ray fluxes, while the values of L_X are intrinsic (i.e., "unabsorbed") X-ray luminosities.

To be considered for our variability analysis (Section 4.2), X-ray-emitting YSOs had to meet the following criteria: (1) be detected with Spitzer 5.8 μm emission or have little near-infrared excess (to preserve a Class III sample), (2) lie in the field of view (FOV) of at least 9 of the 11 Chandra observations, (3) be detected in at least two of the 11 Chandra observations, and (4) have a signal to noise ratio ⩾3 in both Chandra and Spitzer photometry. Each source's broad count rate was integrated over the entire exposure and accounts for effective exposure time. Upper limits for non-detected sources in individual exposures were calculated with the CIAO tool aprates. The location and shape of the source, as detected by wavdetect in the merged Chandra observation, was used to calculate upper limits on the count rate at the location of any source not detected in an individual observation.

The variance of each X-ray-emitting YSO in our variability analysis (s²) was calculated over the course of all 11 observations and normalized to its mean broad band count rate:

$$\begin{equation*} {{s^{2}}}=\frac{\sum (x-\bar{x})^{2}}{\bar{x}^{2}(N-1)}, \end{equation*}$$

where x is the broad (0.5–8.0 keV) count rate per observation, $\bar{x}$ is the mean count rate, and N is the number of observations. The count rate upper-limit was used for non-detected sources in individual observations. If a source was detected in fewer than 6 of the 11 observations, an upper limit was used to determine the count rate (x) and the mean count rate ($\bar{x}$) for the (s²) lower-limit calculation. The variance values of these objects are hence considered to be lower limits.

The sources in the upper right of the 2MASS color–color plot in Figure 4 ([H − K]  >  1.5) have large infrared excesses and are heavily reddened, and therefore most likely represent highly embedded Class I YSOs (e.g., sources 12, 13, 14, and 15). Source 16 was not detected at J, H, or K wavelengths and therefore we consider it Class 0/I candidate. These inferences are supported by classifications of the same five embedded YSOs based on Spitzer IRAC colors. These five (Class I and candidate Class 0/I) are hereafter referred to, collectively, as the L1630 Class 0/I sample.

Infrared colors can be used in conjunction with median X-ray photon energy to distinguish between highly embedded Class 0/I YSOs and other sources. Soft X-rays (≲ 1 keV) are susceptible to absorption by intervening cloud and/or disk material. This allows estimates of the line-of-sight absorbing column solely on the basis of the median energy of the X-ray source (Feigelson et al. 2005; Getman et al. 2010). Figure 5 makes apparent that, for L1630 YSOs, median energy is correlated with infrared colors; i.e., that redder infrared sources have larger median energies. The relationship from Feigelson et al. (2005) was used to estimate the absorbing columns for X-ray sources in L1630 listed in Table 2. The resulting typical absorbing columns determined for L1630 YSOs lie in the determined range N_H ∼ 10¹⁹–10²² cm⁻² for Class II, Class III, and transition disk YSOs, whereas highly embedded Class 0/I sources have estimated N_H values as high as 10²³ cm⁻². The value of N_H has also been determined via X-ray spectral modeling for strong (>100 count) X-ray sources. The relationship between best-fit column density and median energy for these X-ray-bright YSOs is shown in Figure 6 and is evidently well described by the best-fit curve relating N_H and median energy for ONC YSOs (Feigelson et al. 2005). The spectrally derived N_H for L1630 YSOs are in good agreement with those measured from the COUP survey.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Of the 59 X-ray sources detected in the merged broad band filter, 40 were detected in the hard band filter and all 59 were detected in the soft band filter. The total number of X-ray-emitting YSOs in each observation is listed in Table 3 where S, H, and B correspond to soft, hard, and broad energy filters, respectively. In a few cases, low count rate sources were detected in the broad energy filter but not in the soft or hard filters. In Figure 7, the detection fraction, i.e., the ratio of X-ray-emitting YSOs to non-X-ray-emitting YSOs, for each observation is plotted by YSO classification. Class 0/I sources have the lowest detection fraction, whereas transition disks and Class II YSOs are more readily detected (see Section 4.1). Figures and tables in this paper combine Class 0 and Class I YSOs (i.e., Class 0/I) because we seek to compare the X-ray detectability, variability and spectral characteristics of deeply embedded YSO phases with X-ray emission characteristic at later stages of YSO evolution. The black dashed line in Figure 7 represents the maximum possible detection fraction based on the merged image and accounts for sources included and excluded by the ACIS FOV for each observation. The colored solid lines indicate the detection fraction within specific energy bands.

Zoom In Zoom Out Reset image size

With observations of ∼50 L1630 YSOs at up to 11 epochs, we can examine X-ray variability as it relates to YSO infrared class. Of the 52 X-ray-emitting YSOs with infrared classifications, 27 meet our variability requirements (see Section 2.2). X-ray light curves and hardness ratios spanning ∼4 yr were constructed for each YSO meeting these criteria, as shown in Figures 8–11. All YSO classes show significant X-ray variability with at least one X-ray source from each infrared class undergoing an order of magnitude change in count rate. Furthermore, many YSOs show significant variability even on timescales of days (ObsIDs 5384-6413 and 10763-8585; Table 1). X-ray variability with respect to YSO classification is compared to X-ray and IR spectral parameters in Figures 12 and 13(a)–(e) (see Section 4.2). Not all spectrally analyzed sources met the X-ray variability selection requirements (described in Section 2.2) and thus not all X-ray sources are shown in Figure 13. We find that L1630 Class 0/I YSOs tend to be more variable than other YSO classifications. We find that four of the five Class II YSOs that met our variability requirements have very similar levels of variability (s² ≈ 0.4; Figure 13) yet display differing spectral characteristics.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Examples of model fits to X-ray spectra of representative objects from each YSO class are shown in Figure 14. Various X-ray (L_X, T_X, N_H) and infrared (L_IR) properties of L1630 YSOs are plotted in Figures 15(a)–(e). Such parameters are used to investigate trends based on YSO classification (see Section 4). We find that Class III YSOs in L1630 tend to have lower intrinsic L_X than other YSO classifications yet have similar plasma temperatures. We find no clear correlation between YSO classification and plasma temperature and we find three (one Class II and two transition disk) YSOs that display emission from low temperature (∼3 MK) plasma that could be produced via accretion shocks.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Only 5 of the 19 Class 0/I (candidate Class 0/I or Class I) YSOs in L1630 were detected in X-rays, possibly due to factors such as large obscuration, low quiescent X-ray luminosity and, perhaps, less frequent events than Class 0/I YSOs in Serpens or Coronet or Class II/III YSOs. SSV 63NE (ID 16; Figure 16) has previously been discussed as an X-ray-emitting Class 0/I candidate in Simon et al. (2004) due to its spatially correlated X-ray and radio emission and lack of near-infrared counterpart. Three of the five X-ray-emitting YSOs (IDs 12, 13, and 14) are designated as Class I YSOs in Cohen et al. (1984), Simon et al. (2004) and Muzerolle et al. (2005). All five are classified as protostars in Megeath et al. (2012). Furthermore, the detection of IDs 12, 13 and 14 at J, H, and K wavelengths suggest they are indeed Class I (as opposed to Class 0) sources (Figure 4).

Zoom In Zoom Out Reset image size

The X-ray spectrum of SSV 63NE, is shown in Figure 14 and its SED is shown in Figure 17. Model fitting confirms a highly absorbed (N_H ∼ 10²³ cm⁻²) source and yields a plasma temperature of T_X ∼ 7–27 MK (Table 4). Higher temperatures for Class 0 candidates have been found previously: T_X ∼ 50 MK and N_H = 2.47 × 10²³ cm⁻² for source X_E, a sub-region of IRS 7 in the R Corona Australis (R CrA) star-forming core (Hamaguchi et al. 2005) and T_X = 79.4 ± 20 MK and N_H = 1.48 × 10²³ cm⁻² for IRS7E in the Coronet Cluster (FP07). Submillimeter emission has been observed in the region of SSV 63NE with JCMT-SCUBA (Mitchell et al. 2001); however, due to the close proximity of two other X-ray-emitting protostars (Class I YSOs SSN 63E and SSV 63W; Figure 16), it is unclear from which of the three (or a combination thereof) the unresolved submillimeter emission is originating. The large value of N_H we derived from SSV 63NE is consistent with Class 0/I status and—combined with strong submillimeter emission in its vicinity and its lack of a near-IR counterpart–support Class 0 candidacy. However, higher-resolution submillimeter and far-infrared imaging are needed to confirm this source's Class 0 status.

Zoom In Zoom Out Reset image size

Table 4. L1630 YSO X-Ray Spectral Analysis

ID	R.A.	Decl.	Count Rate	NH	T1	T2	Norm. 1	Norm. 2	Absorbed Fluxa	Red. χ2	Class
			(ks−1)	(× 1022 cm−2)	(MK)	(MK)			(erg s−1 cm−2)
2	05:46:22.43	−00:08:52.6	5.66	0.54$_{-0.1}^{+0.1}$	26.82$_{-3.4}^{+4.9}$	8.28$_{-0.8}^{+1.0}$	3.97E-5$_{-7.0E-6}^{+5.8E-6}$	2.57E-5$_{-1.2E-5}^{+1.7E-5}$	4.33E-14$_{-1.0E-14}^{+5.1E-15}$	1.25	Transition/ClassII
3	05:46:11.60	−00:02:20.6	11.28	1.23$_{-0.1}^{+0.1}$	31.57$_{-3.3}^{+3.7}$	...	2.15E-4$_{-1.8E-5}^{+2.5E-5}$	...	1.39E-13$_{-1.9E-14}^{+1.5E-14}$	1.22	Transitionb
4	05:46:17.71	−00:00:14.2	1.94	0.98$_{-0.3}^{+0.3}$	35.28$_{-10}^{+19.2}$	...	3.53E-5$_{-9.7E-6}^{+1.3E-5}$	...	2.65E-14$_{-1.1E-14}^{+7.3E-15}$	0.94	Transitionb
5	05:45:38.26	−00:08:11.1	4.54	0.63$_{-0.2}^{+0.3}$	3.16$_{-1.1}^{+3.8}$	17.04$_{-2.9}^{+2.9}$	9.25E-5$_{-5.5E-5}^{+5.9E-4}$	3.75E-5$_{-7.2E-6}^{+1.0E-5}$	3.03E-14$_{-1.5E-14}^{+6.3E-15}$	0.88	Transition
6	05:45:41.67	−00:04:02.1	2.12	1.04$_{-0.2}^{+0.5}$	14.56$_{-3.2}^{+3.9}$	1.06$_{-0.1}^{+0.5}$	2.08E-5$_{-5.1E-6}^{+5.4E-6}$	5.40E-2$_{-9.8E-3}^{+3.0E-1}$	1.45E-14$_{-9.9E-15}^{+1.3E-15}$	0.74	Transition
7	05:45:41.95	−00:12:05.6	5.90	0.17$_{-0.1}^{+0.1}$	29.05$_{-10.9}^{+24.2}$	10.93$_{-1.8}^{+1.0}$	1.76E-5$_{-7.4E-6}^{+7.9E-6}$	1.81E-5$_{-7.4E-6}^{+1.0E-5}$	4.00E-14$_{-1.0E-14}^{+6.1E-15}$	1.07	Transition
8	05:46:07.90	−00:11:56.6	32.32	0.26$_{-0.0}^{+0.0}$	9.48$_{-0.6}^{+1.3}$	30.67$_{-1.6}^{+2.1}$	3.94E-5$_{-1.4E-5}^{+1.4E-5}$	2.44E-4$_{-1.5E-5}^{+1.1E-5}$	2.76E-13$_{-2.0E-14}^{+1.5E-14}$	1.1	Transition
9	05:46:09.86	−00:05:59.4	0.68	4.45$_{-1.5}^{+2.1}$	17.9$_{-6.1}^{+11.2}$	...	6.18E-5$_{-6.2E-5}^{+1.1E-4}$	...	1.09E-14$_{-1.1E-14}^{+2.0E-15}$	0.66	Transition
10	05:46:20.17	−00:10:19.9	0.66	1.02$_{-0.5}^{+0.7}$	21.95$_{-8.0}^{+11.5}$	...	1.78E-5$_{-8.7E-6}^{+1.5E-5}$	...	9.24E-15$_{-9.2E-15}^{+2.9E-15}$	0.51	Transition
12	05:46:07.85	−00:10:01.2	0.90	20.8$_{-13.39}^{+13.49}$	1.72$_{-0.63}^{+4.65}$	...	4.82E-4$_{-4.82E-4}^{+2.65E-3}$	...	3.06E-14$_{-3.06E-14}^{+1.00E-14}$	0.7	Class 0/I
13	05:46:08.43	−00:10:01.0	0.84	30.72$_{-11.5}^{+12.5}$	15.83$_{-5.2}^{+19}$	...	1.41E-3$_{-1.4E-3}^{+7.4E-3}$	...	3.03E-14$_{-3.0E-14}^{+7.0E-15}$	0.92	Class 0/I
14	05:46:13.15	−00:06:04.7	6.24	4.49$_{-0.6}^{+0.5}$	52.13$_{-11.1}^{+27.5}$	...	3.30E-4$_{-7.3E-5}^{+7.6E-5}$	...	2.01E-13$_{-4.4E-14}^{+2.3E-14}$	0.84	Class 0/I
15	05:46:43.12	+00:00:52.4	2.57	9.79$_{-3.3}^{+6.8}$	45.89$_{-25.8}^{+71.0}$	...	2.13E-4$_{-1.0E-4}^{+6.6E-4}$	...	8.32E-14$_{-8.3E-14}^{+3.5E-14}$	0.67	Class 0/I
16	05:46:08.92	−00:09:56.1	0.53	25.76$_{-12.6}^{+15.4}$	16.04$_{-6.9}^{+27.3}$	...	1.42E-3$_{-1.4E-3}^{+1.8E-2}$	...	3.91E-14$_{-3.9E-14}^{+9.2E-15}$	0.64	Class 0/I
21	05:46:00.22	−00:08:26.3	1.56	0.94$_{-0.3}^{+0.3}$	23.49$_{-5.1}^{+6.7}$	...	2.51E-5$_{-7.0E-6}^{+9.8E-6}$	...	1.42E-14$_{-5.8E-15}^{+3.6E-15}$	0.55	Class III
24	05:46:04.57	+00:00:38.1	1.07	1.12$_{-0.2}^{+0.2}$	10.89$_{-2.3}^{+2.6}$	...	2.67E-5$_{-8.6E-6}^{+1.4E-5}$	...	8.11E-15$_{-5.2E-15}^{+3.2E-15}$	0.88	Class III
25	05:46:05.74	−00:02:39.7	1.33	1.93$_{-0.4}^{+0.7}$	13.72$_{-3.6}^{+8.1}$	1.77$_{-0.8}^{+2.0}$	2.83E-5$_{-1.5E-5}^{+2.4E-5}$	5.89E-3$_{-2.5E-3}^{+2.5E-1}$	9.32E-15$_{-9.3E-15}^{+1.4E-15}$	0.76	Class III
26	05:46:09.36	−00:06:57.6	0.64	2.17$_{-1.4}^{+1.4}$	14.22$_{-5.0}^{+10.0}$	...	6.06E-5$_{-6.1E-5}^{+1.1E-4}$	...	1.33E-14$_{-1.3E-14}^{+6.4E-15}$	0.86	Class III
29	05:46:08.00	−00:01:52.3	0.97	0.76$_{-0.2}^{+0.2}$	9.22$_{-1.6}^{+6.2}$	...	2.83E-5$_{-1.2E-5}^{+1.6E-5}$	...	1.13E-14$_{-1.0E-14}^{+3.7E-15}$	1.29	Class III
30	05:45:53.11	−00:13:25.0	1.24	0.74$_{-0.4}^{+0.3}$	20.74$_{-7.4}^{+21.5}$	9.58$_{-2.1}^{+2.2}$	7.89E-6$_{-6.0E-6}^{+6.9E-6}$	1.02E-5$_{-8.3E-6}^{+1.4E-5}$	8.88E-15$_{-5.2E-15}^{+3.6E-15}$	0.62	Class III
31	05:46:10.95	−00:17:39.8	3.37	0.1$_{-0.1}^{+0.2}$	41.03$_{-16.2}^{+63.2}$	...	2.15E-5$_{-5.7E-6}^{+7.2E-6}$	...	2.73E-14$_{-1.6E-14}^{+1.2E-14}$	1.02	Class III
40	05:46:19.48	−00:05:20.1	5.71	0.688$_{-0.2}^{+0.3}$	10.05$_{-2.5}^{+2.7}$	32.73$_{-6.2}^{+10.6}$	1.50E-5$_{-1.0E-5}^{+2.2E-5}$	5.38E-5$_{-1.3E-5}^{+6.2E-6}$	4.96E-14$_{-9.6E-15}^{+8.2E-15}$	1.13	Class IIb
42	05:46:00.17	+00:03:07.3	...	0.35$_{-0.4}^{+0.4}$	14.36$_{-3.9}^{+4.6}$	...	4.48E-5$_{-1.7E-5}^{+3.7E-5}$	...	3.53E-14$_{-3.5E-14}^{+1.5E-14}$	0.64	Class II
43	05:46:09.61	−00:03:31.1	0.99	0.9$_{-0.4}^{+0.4}$	37.81$_{-12.5}^{+33.8}$	...	1.79E-5$_{-5.6E-6}^{+9.9E-6}$	...	1.44E-14$_{-7.9E-15}^{+4.3E-15}$	0.54	Class II
45	05:46:11.60	−00:06:27.8	0.97	0.58$_{-0.3}^{+1.1}$	28.7$_{-16.9}^{+13.5}$	...	1.18E-5$_{-3.5E-6}^{+2.5E-5}$	...	9.06E-15$_{-3.9E-15}^{+3.0E-15}$	0.75	Class II
46	05:46:18.89	−00:05:38.2	22.87	1.26$_{-0.1}^{+0.1}$	3.33$_{-0.7}^{+0.8}$	30.18$_{-2.6}^{+3.0}$	1.64E-3$_{-8.9E-4}^{+2.5E-3}$	2.73E-4$_{-2.5E-5}^{+2.7E-5}$	2.10E-13$_{-4.4E-14}^{+2.0E-14}$	1.56	Class II
47	05:46:19.08	+00:03:29.7	26.90	0.96$_{-0.2}^{+0.2}$	118.51$_{-48.2}^{+234.7}$	...	3.65E-4$_{-3.3E-5}^{+5.2E-5}$	...	4.43E-13$_{-7.8E-14}^{+5.6E-14}$	0.88	Class II
51	05:46:35.61	−00:01:37.0	0.74	1.94$_{-0.9}^{+0.8}$	31.62$_{-15.1}^{+36.8}$	...	3.48E-5$_{-1.7E-5}^{+3.1E-5}$	...	1.91E-14$_{-1.9E-14}^{+9.2E-15}$	0.62	Class II
52	05:46:37.06	+00:01:22.0	2.00	0.2$_{-0.2}^{+0.3}$	46.65$_{-19.7}^{+53.7}$	...	1.63E-5$_{-4.1E-6}^{+6.5E-6}$	...	2.00E-14$_{-7.1E-15}^{+7.1E-15}$	0.97	Class II

Notes. ^aModel flux including absorption (i.e., observed flux). ^bTentative classification due to excess 8 μm extended emission.

Download table as:  ASCIITypeset image

The strongest X-ray-emitting YSO in our survey is the transition disk/wTTS SSV 61 (ID 8; Figures 14, 16 and 19). Simon et al. (2004) had previously classified SSV 61 as a wTTS based on Hα emission and Li i λ6708 absorption equivalent widths; however, our analysis of IRAC photometry placed SSV 61 on the border between Class III (wTTS) and transition disk YSOs. Our 234 ks merged spectrum of SSV 61 (Figure 14(b)) contains ∼7000 counts, which corresponds to an effective count rate of 32.3 ks⁻¹. Simon et al. (2004) previously published an effective count rate of 251.92 ks⁻¹ for a 62.8 ks observation in 2002 November. Evidently the count rate of SSV 61 had declined during the epochs of our Chandra observations and, with a (relatively small) normalized variance of 0.07 (Figure 11), we conclude that Simon et al. (2004) must have observed SSV 61 in outburst. It is interesting to note that SSV 61 lies close to a submillimeter source (Figure 16) originally reported by Mitchell et al. (2001). Such an association is highly unusual for a wTTS but could be consistent with transition disk status.

Other interesting X-ray/infrared sources are IDs 1, 3, 4, and 38 (Figures 3, 18 and 19). Fang et al. (2009) classified ID 4 as a K3.5 cTTS with an optically thick disk. It was detected as an X-ray source in our observations with an effective count rate of 1.94 ks⁻¹. The other three sources were not included in Fang et al. (2009); however IDs 1 and 38 were classified as pre-MS stars with disks in Megeath et al. (2012). If the bow-shock-like nebulae seen in Spitzer imaging (Figure 3) are attributed to these YSOs then these structures may be wind-collision fronts similar to those associated with other YSOs in the Orion region (Bally et al. 2000).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

When observing YSOs, X-ray detection fraction depends primarily on sensitivity, spectral source hardness, and the degree of intervening obscuration toward the source. Our results further illustrate that, in the case of YSOs characterized by strong variability, the number and/or duration of observation epochs plays a significant role in detection statistics when compared to studies with similar sensitivity. The X-ray detection fractions as a function of time in Table 3 and Figure 7 provide insight into the variable nature of YSOs. Class 0/I YSOs have the lowest overall detection fraction in all bands. These YSOs are rarely detected in the soft energy band in our data, most likely due to their highly embedded environments; whereas in the hard band, photons are more likely to penetrate, thus leading to a higher hard-band detection fraction. The detection fractions for Class II sources are slightly larger than for Class 0/I sources; furthermore, it appears that for several observations, Class II sources are more readily detected in the soft band than in the hard band. This could be an indication either of the relatively steady nature of soft accretion shock emission or of the generally smaller degree of intervening absorption toward Class II sources. The transition disk detection fraction tends to be similar in hard and soft bands, indicative of relatively steady, non-variable emission. Class III YSOs, whose X-ray detection fraction in a given observation is 1.0 by definition, are consistently more detectable in the soft band than the hard band, indicating lower levels of intervening absorption with respect to the other YSO classes. The relatively low detection fractions in individual exposures evident in Figure 7 may reflect the highly variable nature of Class III YSOs.

Our X-ray detection fraction results are compared with other infrared/X-ray studies of star-forming regions in Table 6. X-ray detection fractions for individual observations were not available in the other studies so we restrict the comparison to the respective merged exposures. The respective sensitivities between 0.5–8.0 keV are L_{x, min} ≈ 4 × 10²⁶ erg s⁻¹, L_{x, min} ≈ 3 × 10²⁷ erg s⁻¹, and L_{x, min} ≈ 2 × 10²⁷ erg s⁻¹ for FP07, W07 and our L1630 observations. This comparison of X-ray luminosity sensitivities is only approximate, given the different source detection thresholds and angular size scales used in the three studies. When comparing YSO detection fraction, it may be important to consider not only the total exposure time and source distances, but also how the component exposures were distributed in time. FP07 used eight exposures of the Coronet Cluster over the course of ∼5 yr; however, five of the exposures were obtained only a day apart. The Serpens Core (W07) was observed for one 90 ks period. L1630 was observed 11 times over the course of four years with only two sets of two exposures obtained less than 2.5 months apart (Table 1).

For Class 0/I YSOs, FP07 and W07 found detection fractions of 90% and 41% in R CrA and Serpens, respectively, whereas for L1630 we obtain a Class 0/I detection fraction of only 26%. The Class II detection fraction for L1630 (58%) is significantly larger than the detection fraction for Class II YSOs in the Serpens Cloud (32%) but again smaller than in the Coronet Cluster (82%). It appears the high detection fraction in R CrA can most likely be attributed to the less deeply embedded nature of its Class 0/I YSOs and the high sensitivity of the R CrA observations. If Class II YSOs were characterized by relatively steady X-ray emission over timescales longer than a typical X-ray observation, punctuated by relatively rare, strong outbursts, then the probability of detecting X-ray-emitting Class II YSOs should increase with the number of X-ray observations. It would therefore appear that, for studies with similar sensitivities—as for W07's Serpens study and our L1630 study—observations spanning longer time scales may yield a higher X-ray detection fraction of Class II sources.

Both coronal activity and star–disk interactions—e.g., magnetically funneled accretion, time-variable absorption from inner disk warps and/or accretion streams, disk instabilities resulting in accretion outbursts, or magnetically confined-plasma suddenly released in star–disk magnetic-reconnection events—are phenomena that can plausibly cause significant X-ray variability. In the case of star–disk interactions, obviously, a circumstellar disk must be present. Class III YSOs have little or no detectable IR excesses and hence their disks have likely already dispersed. Ergo, if star–disk interactions are at least partially responsible for the highly X-ray variable nature in observed YSOs with disks, then Class I, Class II, and transition disk sources should show X-ray variability that is distinct from that of Class III sources. In particular YSOs with disks might be expected to display larger levels of variability than Class III YSOs over a timespan of several years and multiple observations (Bouvier et al. 2007, and references therein). Transition disk YSOs, representative of an evolutionary stage between Class II and Class III, are characterized by disks with large (∼1–10 AU) inner dust clearings (Calvet et al. 2005). If magnetic field lines interacting with gas in the disk are required to provide some mechanism for variability, then transition disks whose "holes" are large enough might extend beyond the immediate influence of the protostellar magnetic field, thus halting such variability. However, even though there is a large gap in the dusty component of a transition disk, this does not necessarily imply a large gap in the gaseous component (Chiang & Murray-Clay 2007; Sacco et al. 2012), and hence star–disk interactions could still be an important driver of YSO X-ray variability.

Regardless of the mechanisms underlying variability, Figures 8–11 show that each class in our variability sample has at least one YSO that underwent an order of magnitude increase in count rate over the ∼4 yr course of our study. Several YSOs also showed significant variability on timescales of days, which is suggestive of magnetic loop flaring (Favata et al. 2005). FP07 displayed similar X-ray duration light curves for YSOs in the Coronet Cluster and suggested that these light curves indicated that Class I YSOs displayed stronger levels of variability than Class II YSOs. Likewise, Figures 8, 9 and 12 suggest that L1630 Class 0/I YSOs are more variable than Class II YSOs, though these figures also suggest that Class 0/I and Class III objects are similarly variable. Similar distinctions in variability are apparent for YSOs in the ρ Ophiuchi dark cloud (Imanishi et al. 2001a, 2001b).

Figure 12 illustrates the overall variability and spectral hardness (median photon energy) for YSOs of various classes in L1630 over the course of ∼4 yr. Figures 5 and 12 make clear that Class 0/I sources are significantly harder than Class II and III YSOs. This is likely mainly due to their highly embedded nature (i.e., absorption of <1 keV photons by circumstellar material) given that their inferred plasma temperatures are similar to those of Class II and Class III YSOs (Figure 13(a) and Section 4.3). Large (≳ 3 keV) median energies for similarly deeply embedded objects have been inferred previously, including V1647 Ori (ID 14) (Grosso et al. 2005a, 2005b).

Figures 12 and 13(b) illustrate the highly variable nature of Class 0/I YSOs in L1630. It is important to note that when comparing these two figures, one must account for the lower limits of sources that were only detected in 2–5 observations. Many sources in Figure 12 did not have enough counts for spectral fitting and, therefore, were not included in Figure 13. Furthermore, for these sources with non-detections, count rate upper limits (i.e., background count rate at the source position) were included in calculations of the normalized variances (see Section 2.2). Figure 12 suggests both Class 0/I and Class III YSOs are highly variable. However, Figure 13 implies that of the YSOs detected in more than 5 observations, the Class 0/I YSOs are marginally more variable than all other YSO classes, including Class III YSOs. Transition disks in L1630 tend to be less variable than other classes (Figures 12 and 13(a)).

Of the five Class II YSOs that met our variability and spectral fitting requirements, four of them have very similar levels of variance (s² ≈ 0.4; Figure 13). It is interesting to note that all four of these Class II YSOs also show similar plasma temperatures (T_X ∼ 30 MK), yet their other spectral characteristics differ. Flaccomio et al. (2012) found that COUP Class II YSOs were more variable than Class III YSOs and suggested that rotational modulation was the mechanism responsible for high levels of variability. We do not find such a trend for YSOs in L1630. This could be due to the fact that the quasi-continuous 10 day observation of the ONC study was more sensitive to variability due to rotational modulation but was not as sensitive to rare, energetic flaring events as our L1630 study, in which the X-ray observations span four years.

Before considering the X-ray spectral properties of L1630 YSOs obtained from model fitting, it is important to note that the absorbed one-or two-component thermal plasma models typically used to fit X-ray spectra of YSOs are not complete representations of the intrinsic properties of YSOs. Detailed studies of magnetically and accretionally active YSOs show it is likely that their plasmas span a wide range of temperatures (Brickhouse et al. 2010). However, one- or two-component thermal plasma models are sufficient to ascertain global plasma properties, and have been used successfully in numerous previous studies (e.g., Feigelson et al. 2005; Güdel et al. 2007; Forbrich & Preibisch 2007; Winston et al. 2007).

As shown in Figure 6, the line-of-sight absorption (N_H) values derived from spectral analysis of YSOs in L1630 tend to follow the same trend seen in the ONC (Feigelson et al. 2005), the Serpens Cloud (Winston et al. 2007), and the Coronet Cluster (Forbrich & Preibisch 2007). We find a large median X-ray absorbing column density for Class 0/I sources relative to other classifications (Table 5) and other studies (e.g., N_H  ∼ 5 × 10²² cm⁻² for Class I YSOs in Rho Ophiuchi; Ozawa et al. 2005). Winston et al. (2007) found Class 0/I sources in the Serpens Cloud core had average X-ray absorbing column densities ∼5 times that of Class II and Class III YSOs. We have found Class 0/I sources in L1630 have median X-ray absorbing column densities ∼20 times those characteristic of later YSO stages (Table 5). However, note that three of the five X-ray detected Class 0/I sources in L1630 (IDs 12, 13, and 16; Figure 16) reside within ∼10 arc seconds of each other and hence likely reside in the same, highly obscured region of the L1630 molecular cloud. The spatial proximity of these three YSOs with high values of N_H could suggest that for many Class 0/I sources, much of the absorbing material resides in the molecular cloud environment immediately surrounding the protostar, rather than in the collapsing circumstellar envelope of the Class 0/I source. Furthermore, it is interesting to note that these three YSOs have similar intrinsic X-ray luminosities and plasma temperatures (Table 4, Figures 13, and 15). These three highly embedded Class 0/I sources, which have likely formed around the same time and in a similar environment, hence represent an excellent Class 0/I YSO group for further study. Our sample of Class II, transition disk, and Class III L1630 YSOs have very similar values of median N_H (Table 5). This suggests that ambient cloud material, not circumstellar, dominates N_H for these sources and that the Class 0/I YSOs are found near the densest parts of the L1630 cores.

The X-ray detected sample of YSOs in L1630 range over three orders of magnitude in L_X (from ∼10²⁹–10³² erg s⁻¹), suggesting a large range of stellar masses (Preibisch & Feigelson 2005). In Figure 15(a), many YSOs in L1630 to lie above the level L_X = 10⁻³ L_IR. Unless L_IR is significantly larger than L_bol—which is more likely for Class II and III YSOs than for Class I—this suggests that many L1630 YSOs lie above the "saturation level" (e.g., Preibisch & Feigelson 2005). However, it is important to note that the YSO X-ray luminosities (along with other spectral properties) were calculated using merged spectra obtained from multiple observations. Many YSOs in L1630 undergo flaring events that sometimes increase flux by an order of magnitude (Figures 8–11), possibly resulting in larger values of time-averaged L_X than in previous studies.

In Figure 15(b) and Table 5, we see that the sample of Class III YSOs in L1630 tend to have lower X-ray luminosities than Class 0/I, Class II and transition disk objects, yet these L1630 YSO classifications have similar plasma temperatures. The similar temperatures found in the one-temperature fits and the "hot" components of the two-temperature fits for YSOs in L1630 indicate that magnetic reconnection dominates the plasma heating for all classes. FP07 and Winston et al. (2007) also found no clear correlation between plasma temperature and YSO classification; however, FP07 found Class I and Class II objects tend to show hotter plasma temperatures than Class III YSOs. In most cases, accretion shocks should not be contributing to L_X or T_X in non-accreting Class III YSOs. Hence, either accretion does not play a significant role in increasing plasma temperature and luminosity, or Class III YSOs have another mechanism to heat plasma to similar temperatures.

Accretion shocks are expected to produce X-ray emission at characteristic temperatures of ∼3 MK and cooler (Kastner et al. 2002; Huenemoerder et al. 2007; Grosso et al. 2010; Teets et al. 2012). FP07 concluded that Class I and II sources in their study show no such excess soft emission indicative of accretion shocks. As shown in Figures 15(b)–(e), two transition disks and one Class II YSOs do display a cool plasma component indicative of accretion shocks. These sources also tend to be more X-ray luminous than other YSOs in L1630; however, this could merely reflect the fact that high count-rate sources are easier to fit with two-temperature models. In the case of Class 0/I YSOs (which are actively accreting), non-detection of a cool plasma component does not necessarily imply that accretion shocks are not present. Class 0/I YSOs (particularly those in L1630) are heavily absorbed (Table 5; Figures 15(c) and 13(c)) and any unabsorbed soft X-ray emission from these sources indicative of accretion shocks would be difficult to detect with X-ray CCD spectro-imaging. Furthermore, observations to include only those photons with energies ⩾ 0.5 keV may have removed some of the soft excess detectable in L1630 YSOs by Chandra. We also note that one Class III YSO was found to have a ∼1 MK plasma component, demonstrating that processes other than accretion (e.g., coronal activity) may also contribute to soft X-ray fluxes from YSOs. However, in the absence of Hα spectroscopy, we cannot exclude the possibility that this system is actively accreting gas, despite a lack of thermal IR excess due to circumstellar dust (Günther et al. 2013).

A weak trend of increasing plasma temperature with infrared luminosity can be seen for YSOs with log (L_IR) ≳ 33 in Figure 14(d). The source of infrared luminosity in Class III sources should be mostly photospheric, as they are non-accreting and have little or no disk material, whereas in earlier stages of star formation (e.g., Class 0/I, Class II, and transition disks), the disk (and for Class 0/I, the infalling envelope) contributes significantly to infrared emission. If T_X is an indication of magnetic activity, then Figure 15(d) would imply YSOs with disks tend to have more magnetic activity as L_IR increases. Figure 15(e) shows a general decrease in plasma temperature with increasing L_X/L_IR. More interestingly, the Class III sources in L1630, which as a class have relatively low X-ray luminosities but a large spread of infrared luminosities, tend to decrease in plasma temperature with increasing L_X/L_IR.