3 The catalogue of discrete X-ray sources

Table 2 with the full SMC catalogue is available electronically and contains the following information for each X-ray source: (1) source number, (2) likelihood of existence (maximum value from the five detection energy bands) ML $_{\rm exi}$, (3) vignetting corrected net exposure, (4) and (5) source coordinates RA and Dec (J2000.0) derived from observation and energy band with smallest position error, (6) statistical 90% confidence error on the X-ray position with an additioanl systematic uncertainty of about 7 $^{\prime\prime}$, (7) PSPC 0.1 - 2.4 keV count rate and error, (8) and (9) hardness ratios (defined as HR1 = (H - S)/(S+H) and HR2 = (H2 - H1)/(H1+H2) where S, H, H1 and H2 denote count rates in the 0.1-0.4 keV, 0.5-2.0 keV, 0.5-0.9 keV and 0.9-2.0 keV bands, respectively; not calculated in cases where not all required count rates are available), (10) source extent from same detection as source position, (11) likelihood for the extent, (12) ROSAT observation identifier (a "+" at the end marks merged observations), (13) off-axis angle, (14) distance to detector window support structure or detector rim (WSS), (15) number of nearest (within 60 $^{\prime\prime}$) source in KPFH99 (see Sect. 3.1) and (16) remark (cf. Table 1 in HP99). We note here that the count rates given in KPFH99 and here can considerably differ for sources repeatedly covered by PSPC observations. Since the main goal of this work is to identify sources, we selected the detection with smallest error on the X-ray position which may result from a different PSPC observation not used by KPFH99.

The spatial distribution of the 517 PSPC sources in the catalogue is shown in Fig. 3. Haberl & Pietsch (1999) have used the LMC catalogue to identify known sources and classify unknown sources according to the properties of the identified samples. Again we follow the same procedures for the SMC region.

 

 

Table 2: Excerpt of the catalogue of 517 ROSAT PSPC sources from a 6$^\circ$$\times$6$^\circ$ field centered on the SMC

1	2	3	4	5	6	7	8	9	10	11
No	ML $_{\rm exi}$	Exp.	RA	Dec	r90	Count Rate	HR1	HR2	Extent	ML $_{\rm ext}$
		[s]		(J2000.0)	[ $^{\prime\prime}$]	[cts s-1]			[ $^{\prime\prime}$]
49	12.2	15341	01 02 49.5	-71 37 09	42.6	3.41e-03 $\pm$ 4.4e-03	1.00 $\pm$ 0.84	-1.00 $\pm$ 1.69	0.0	0.0
50	47.1	14862	00 55 47.3	-71 38 03	10.0	4.90e-03 $\pm$ 7.4e-04	-0.22 $\pm$ 0.15	-0.21 $\pm$ 0.22	0.0	0.0
51	25.7	15176	00 59 41.7	-71 38 15	14.5	2.94e-03 $\pm$ 6.1e-04	-1.00 $\pm$ 0.12		0.0	0.0
52	326.6	9397	00 36 56.9	-71 38 15	17.5	3.86e-02 $\pm$ 2.8e-03	1.00 $\pm$ 0.25	0.38 $\pm$ 0.06	3.4	19.1
53	21360.4	3470	00 59 11.3	-71 38 45	2.8	8.38 $\pm$ 5.0e-02	0.48 $\pm$ 0.01	0.17 $\pm$ 0.01	1.1	14.5

12	13	14	15	16
ROR	$\delta$	$d_{\rm rib}$	KPFH99	Remark
	[$^\prime$]	[FWHM]
500250p	29	1.3	0	AGN? 13 cm
400300p+	13	6.0	109	fg Star F5/F6V, HD 5572 (SIM)
400300p+	6	20.6	141	SSS? RXJ0059.6-7138, PN LIN357 (KPFH99)
400299p+	36	1.8	19	AGN? [hard] [nonstar] Radio SMC B0034-7155 (FJW97) 13 cm jets
500142p	32	4.3	0	HMXB Be/X RXJ0059.2-7138, soft pulsar (H94, SC96)

Figure 3: Distribution of the 517 X-ray sources detected with the ROSAT PSPC. The source positions are plotted on a weak grey scale image (0.1 - 2.4 keV) similar to Fig. 1 for orientation

 

 

Table 3: PSPC sources detected with the Einstein IPC

1		2	3	4	5	6	7
No		WW92	d	r90	PSPC Count Rate	IPC Rate	Remarks
	No	Name	[ $^{\prime\prime}$]	[ $^{\prime\prime}$]	[cts s-1]	[cts s-1]
8	23	0049.0-7125	15.5	20.2	2.04e-01 $\pm$ 1.1e-02	0.046	fg Star F5V, HD 5028 (WW92)
47	43	0056.8-7152	2.6	1.3	3.68e-01 $\pm$ 4.9e-03	0.047	SSS 1E0056.8-7154, PN LIN333 (KP96)
94	32	0051.3-7213	62.2	30.2	1.21e-02 $\pm$ 2.1e-03	0.006
107	51	0102.3-7218	13.4	0.7	2.24 $\pm$ 2.1e-02	0.807	SNR 0102-72.3 (WW92,FPW99) 13 cm
114	41	0056.2-7219	58.3	7.9	3.03e-03 $\pm$ 5.0e-04	0.002	HMXB? AXJ0058-72.0, pulsar (YK98a)
121	46	0059.8-7220	37.6	7.2	8.93e-03 $\pm$ 8.5e-04	0.003
125	54	0104.5-7222	52.1	3.6	3.36e-02 $\pm$ 1.5e-03	0.008	SNR 0104-72.3 (WW92) 13 cm
143	50	0101.5-7225	17.5	3.4	1.54e-02 $\pm$ 9.6e-04	0.013	HMXB Be/X SAXJ0103.2-7209, pulsar (ISC98, HS94,CO99)
148	44	0057.7-7226	1.4	3.2	5.73e-02 $\pm$ 1.7e-03	0.024	SNR 0057-72.2 (YTK91,WW92) 13 cm
157	36	0053.7-7227	20.9	2.3	2.74e-02 $\pm$ 1.2e-03	0.005	AGN? Radio SMC B0053-7227 (FHW98) 13 cm jets
160	53	0103.8-7227	49.0	7.8	1.63e-02 $\pm$ 1.0e-03	0.011	[hard] associated with DEM S128? (FHP00)
162	45	0059.1-7227	44.5	2.3	2.67e-02 $\pm$ 1.1e-03	0.006	[hard] G star identification in SCC99 unlikely
163	53	0103.8-7227	76.9	6.6	6.86e-03 $\pm$ 6.7e-04	0.011	HMXB Be/X AXJ0105-722, pulsar (YK98b, FHP00)
170	40	0055.8-7229	14.2	3.6	5.48e-03 $\pm$ 5.4e-04	0.006	[hard] [nonstar]
176	13	0035.3-7230	6.1	1.2	4.40e-01 $\pm$ 5.4e-03	0.075	SSS 1E0035.4-7230 (KP96)
185	42	0056.5-7233	80.6	19.0	2.32e-03 $\pm$ 4.7e-04	0.007
194	42	0056.5-7233	26.7	6.5	1.00e-02 $\pm$ 7.2e-04	0.007	SNR 0056-72.5 (WW92) 13 cm
217	52	0103.2-7239	54.2	2.1	3.12e-01 $\pm$ 4.6e-03	0.056	SNR 0103-72.6 (WW92, CSM97) 13 cm
220	47	0100.1-7240	62.1	8.1	9.98e-03 $\pm$ 9.5e-04	0.007
227	55	0105.4-7240	74.9	14.2	9.32e-04 $\pm$ 1.3e-03	0.002
230	55	0105.4-7240	51.4	2.4	1.20e-02 $\pm$ 7.1e-04	0.002	[hard]
232	58	0107.1-7241	23.3	8.6	4.93e-03 $\pm$ 5.6e-04	0.004	[hard]
234	39	0055.6-7241	20.5	3.4	8.73e-03 $\pm$ 6.7e-04	0.003
241	35	0053.2-7242	20.5	3.5	3.68e-02 $\pm$ 1.5e-03	0.009	HMXB Be/X XTEJ0055-724, pulsar (ML98, SCI98, SCB99, CO99)
242	34	0052.1-7242	33.9	25.7	8.50e-02 $\pm$ 4.7e-03	0.008	HMXB? XTEJ0053-724, pulsar (CML98)
252	59	0107.1-7244	67.8	13.7	5.62e-03 $\pm$ 1.1e-03	0.006
265	27	0050.1-7248	26.4	45.8	5.45e-03 $\pm$ 1.5e-03	0.016	HMXB Be/X RXJ0051.8-7231, pulsar (WW92, ISA97, SCB99)
279	56	0105.6-7251	20.3	11.1	7.44e-04 $\pm$ 1.0e-03	0.007
285	30	0050.9-7253	50.9	12.7	6.86e-03 $\pm$ 8.7e-04	0.012	SNR 0050-72.8 (WW92) 13 cm
295	19	0045.6-7256	54.1	49.8	4.06e-03 $\pm$ 5.1e-03	0.006	AGN? Radio SMC B0045-7255 (FHW98) 13 cm
320	48	0101.3-7300	53.7	17.2	3.58e-03 $\pm$ 7.9e-04	0.027	[fg Star] GSC 9141.7584
333	31	0051.1-7304	79.2	12.5	1.21e-03 $\pm$ 4.0e-04	0.008
361	72	0102.3-7318	8.6	32.4	4.40e-02 $\pm$ 9.6e-03		SSS? RXJ0103.8-7254 (KP96)
379	62	0115.1-7315	21.5	14.0	6.04e-03 $\pm$ 8.2e-04	0.005	AGN z=0.253 (TDZ97)
385	66	0118.1-7316	19.5	13.2	3.13e-02 $\pm$ 1.6e-03	0.013	Cluster z=0.0656 (CGC97)
413	16	0045.5-7325	50.6	6.8	2.98e-02 $\pm$ 1.6e-03	0.020	SNR 0045-73.4 (WW92, RLG94) 13 cm
414	15	0044.7-7324	43.7	8.5	1.73e-02 $\pm$ 1.3e-03	0.007	SNR 0044-7325 (WW92, RLG94) 13 cm
419	16	0045.5-7325	88.2	4.7	3.72e-02 $\pm$ 1.5e-03	0.020	[SNR] 13 cm, part of source 413?
424	25	0050.0-7327	36.8	2.2	1.46e-02 $\pm$ 8.6e-04	0.011	[hard]
431	64	0116.1-7326	67.5	5.9	5.51e-03 $\pm$ 5.5e-04	0.005	[stellar] GSC 9142.1764
434	18	0045.6-7329	37.8	4.0	1.41e-02 $\pm$ 9.6e-04	0.006	[hard]
437	22	0047.3-7330	2.4	4.7	1.77e-02 $\pm$ 1.1e-03	0.007	SNR 0047-73.5 (WW92, FPH00) 13 cm
453	29	0050.3-7335	36.0	1.9	2.15e-02 $\pm$ 1.0e-03	0.009	HMXB Be/X RXJ0052.1-7319, pulsar (LPM99, ISC99)
454	21	0046.6-7335	34.0	10.8	5.15e-03 $\pm$ 7.2e-04	0.004	SNR 0046-73.5 (WW92) 13 cm
461	24	0049.2-7337	26.1	2.0	1.41e-01 $\pm$ 2.7e-03	0.060	SNR 0049-73.6 (WW92, CSM97) 13 cm
478	65	0117.2-7341	16.2	0.8	1.11e-01 $\pm$ 1.8e-03	0.016	fg Star G5V, HD 8191 (WW92, CSM97)
480	14	0043.7-7343	65.9	16.7	1.19e-03 $\pm$ 1.6e-03	0.004
482	63	0115.8-7342	15.9	0.6	5.82 $\pm$ 1.2e-02	0.069	HMXB supergiant SMC X-1, pulsar (WW92)
496	28	0050.1-7346	43.5	10.4	9.39e-03 $\pm$ 2.4e-03	0.010	[nonstar]
623	33	0051.4-7455	2.5	0.7	4.12 $\pm$ 8.4e-03	0.413	fg Star G3:V+, RSCVn type, HD 5303 (WW92), 13 cm

Notes to the remark column to this and following tables:
Abbreviations for references given in parenthesis are described in the literature list.
Unsecure identifications from positional uncertainty begin with ? in the remark.
Candidates from literature are marked with ? after source class name.
Abbreviations: GSC: HST guide star catalogue (Lasker et al. 1990); fg: foreground.

3.1 Comparison with previous X-ray surveys

In Table 3 we summarize 50 PSPC sources which have an Einstein IPC counterpart (from Tables 2A and B in WW92) within 90 $^{\prime\prime}$. The IPC and PSPC count rates are compared in Fig. 4 where the sources identified with a SNR are marked with a circle. A linear fit to the SNR count rates gives a mean conversion factor of 2.8 for IPC to PSPC count rates, somewhat smaller than found in the LMC which is compatible with a slightly higher foreground absorption. The high mass X-ray binary (HMXB) pulsar SMCX-1 and the RSCVn type foreground star HD5303 were significantly brighter during the ROSAT observations (because the best source detection with smallest position error enters the catalogue, detections during bright source state are preferentially selected) while the pulsar RXJ0051.8-7231 probably associated with the IPC source 0050.1-7248 was bright during the Einstein observations. The unidentified ROSAT source 320 may be associated with 0101.3-7300 (a relatively large correlation distance leaves some doubt) in which case the classification as foreground star suggests a flare during the IPC observation.

A correlation of the PSPC catalogue from the pointed observations with the Bright Source Catalogue (BSC) of the ROSAT all-sky survey (Voges et al. 1996, 1999) yields 16 sources within a distance of 60 $^{\prime\prime}$ which are summarized in Table 4. Strong variability between survey and pointed observations was found for the two brightest sources of this sample (SMCX-1 and HD5303) and for the SSS RXJ0048.4-7332 which was more than a factor 20 fainter during the pointing with identifier 500251p than during survey. However, during other pointings RXJ0048.4-7332 was detected with count rates similar to the survey intensity (see Kahabka et al. 1994).

The first SMC catalogue based on ROSAT PSPC data and published by KPFH99 is less complete than the catalogue presented here mainly due to the limited number of observations and the more restricted field of view (45$^\prime$ vs. 52$^\prime$) used for the analysis. A correlation of the two catalogues shows that 209 sources have a counterpart within 60 $^{\prime\prime}$. For these the source number from KPFH99 is given in Table 2. The 39 sources of KPFH99 without counterpart in the new catalogue are summarized in Table 5. KPFH99 marked 15 of these with "D" as probable artifacts close to the WSS and we confirm that they are not real. Four other sources with relatively high likelihood are probably also false detections: source 223 in KPFH99 is the most extended one and not visible on the PSPC images; sources 75, 165 and 185 are near the WSS. Two sources (25, 186) are also near the WSS but may be real. Another 17 sources have likelihood values smaller than 15, close to our acceptance threshold and a slightly different background map used in the analysis is probably the reason why they are not detected in our analysis. Source 60 may be resolved into two X-ray sources (413 and 419 in the present catalogue) in an observation which was not used by KPFH99 but has source 60 closer to its field center. Source 413 coincides with a known SNR showing radio emission. Inspecting the ATCA $\lambda = 13$ cm image (see below) shows that the radio emission is associated with source 413 suggesting that 419 is a separate source.

   

Table 4: PSPC sources detected in the ROSAT all-sky survey

1	2	3	4	5	6	7	8
No	BSC name	d	r90	$r_{\rm BSC}$	PSPC Count Rate	BSC Rate	Remarks
	1RXS	[ $^{\prime\prime}$]	[ $^{\prime\prime}$]	[ $^{\prime\prime}$]	[cts s-1]	[cts s-1]
8	1RXS J005052.4-710914	10.5	20.2	11	2.04e-01 $\pm$ 1.1e-02	1.20e-01 $\pm$ 2.3e-02	fg Star F5V, HD 5028 (WW92)
47	1RXS J005837.9-713542	8.8	1.3	9	3.68e-01 $\pm$ 4.9e-03	2.81e-01 $\pm$ 3.2e-02	SSS 1E0056.8-7154, PN LIN333 (KP96)
52	1RXS J003704.1-713821	34.1	17.5	19	3.86e-02 $\pm$ 2.8e-03	5.71e-02 $\pm$ 1.4e-02	AGN? [hard] [nonstar] Radio SMC B0034-7155 (FJW97) 13 cm jets
107	1RXS J010403.5-720158	11.2	0.7	7	2.24 $\pm$ 2.1e-02	2.02 $\pm$ 7.7e-02	SNR 0102-72.3 (WW92, FPW99) 13 cm
176	1RXS J003723.2-721415	14.0	1.2	8	4.40e-01 $\pm$ 5.4e-03	3.49e-01 $\pm$ 3.0e-02	SSS 1E0035.4-7230 (KP96)
217	1RXS J010500.1-722305	17.5	2.1	11	3.12e-01 $\pm$ 4.6e-03	2.36e-01 $\pm$ 2.8e-02	SNR 0103-72.6 (WW92, CSM97) 13 cm
461	1RXS J005108.0-732134	9.8	2.0	21	1.41e-01 $\pm$ 2.7e-03	8.08e-02 $\pm$ 1.6e-02	SNR 0049-73.6 (WW92, CSM97) 13 cm
478	1RXS J011836.8-732529	6.2	0.8	14	1.11e-01 $\pm$ 1.8e-03	6.45e-02 $\pm$ 1.8e-02	fg Star G5V, HD 8191 (WW92, CSM97)
482	1RXS J011705.9-732632	1.5	0.6	8	5.82 $\pm$ 1.2e-02	7.20e-01 $\pm$ 6.4e-02	HMXB supergiant SMC X-1, pulsar (WW92)
512	1RXS J004820.7-733154	14.9	24.1	8	8.10e-03 $\pm$ 2.5e-03	1.99e-01 $\pm$ 2.2e-02	SSS RXJ0048.4-7332, symbiotic star M0 (KP96, M92)
621	1RXS J004402.7-743737	6.1	5.9	10	8.96e-02 $\pm$ 1.6e-03	6.51e-02 $\pm$ 1.3e-02	[fg Star] GSC 9141.1004
623	1RXS J005307.5-743903	2.4	0.7	6	4.12 $\pm$ 8.4e-03	2.38 $\pm$ 6.3e-02	fg Star G3:V+, RSCVn type, HD 5303 (WW92), 13 cm
683	1RXS J012522.3-750015	2.1	10.2	12	6.26e-02 $\pm$ 5.2e-03	6.42e-02 $\pm$ 1.3e-02	fg Star SAO 255776 (SIM)
709	1RXS J005655.1-751349	2.6	4.3	7	2.13e-01 $\pm$ 2.4e-03	3.00e-01 $\pm$ 2.4e-02	[AGN] GSC 9142.0531
721	1RXS J012254.1-752117	10.2	1.4	9	8.15e-01 $\pm$ 1.2e-02	5.28e-01 $\pm$ 6.7e-02	fg Star WD, FAUST 199 (SIM)
734	1RXS J013114.0-755643	15.0	17.7	7	2.28e-01 $\pm$ 9.7e-03	2.27e-01 $\pm$ 2.0e-02	fg Star K0, CD-76 56 (SIM)

 

 

Table 5: PSPC sources from the catalogue of Kahabka et al. (1999) not detected in the present analysis

	KPFH99	Comment1
No	Name
6	0035.0-7354	D
14	0036.0-7210	ML $_{\rm exi}$ = 11
24	0038.0-7344	ML $_{\rm exi}$ = 14
25	0038.0-7327	near WSS, may be real
38	0042.2-7338	D
60	0047.5-7308	two catalogue sources (413/419)
73	0050.7-7226	D
75	0050.8-7341	WSS, not real
81	0051.7-7341	ML $_{\rm exi}$ = 10
90	0053.0-7239	ML $_{\rm exi}$ = 10
91	0053.1-7311	ML $_{\rm exi}$ = 11
97	0053.8-7252	ML $_{\rm exi}$ = 11
99	0054.3-7330	ML $_{\rm exi}$ = 11
112	0055.9-7257	D
113	0056.4-7159	D
119	0057.2-7156	ML $_{\rm exi}$ = 13
132	0058.5-7249	D
136	0058.9-7255	D
142	0059.7-7148	ML $_{\rm exi}$ = 12
156	0101.0-7151	ML $_{\rm exi}$ = 11
157	0101.0-7211	D
161	0101.3-7118	D
165	0101.8-7249	WSS, not real
167	0101.8-7233	ML $_{\rm exi}$ = 14
168	0102.1-7236	D
172	0102.8-7216	D
173	0102.9-7111	D
185	0104.2-7102	WSS, not real
186	0104.7-7203	near WSS, may be real
190	0105.3-7126	ML $_{\rm exi}$ = 14
192	0105.4-7219	ML $_{\rm exi}$ = 11
197	0106.3-7125	D
207	0108.5-7247	ML $_{\rm exi}$ = 12
210	0109.0-7243	ML $_{\rm exi}$ = 13
213	0109.5-7226	ML $_{\rm exi}$ = 10
215	0110.6-7217	ML $_{\rm exi}$ = 12
223	0112.7-7207	not real, largest extent in KPFH99
241	0116.8-7314	D
243	0117.2-7238	D

¹ Artifacts marked as D by KPFH99; likelihood of existence ML $_{\rm exi}$ from KPFH99.

3.2 Source identification and classification

Source identifications were mainly obtained by cross-correlation of the SMC catalogue with the SIMBAD data base operated at CDS. The derived samples of known X-ray sources show identical properties to those of the LMC and therefore we used similar classification criteria as for the LMC sources (Table 6; cf. Table 3 of HP99). As only exception, the selection for SSS was expanded to HR1+EHR1 < -0.70 which includes all sources with HR2+EHR2 < -0.70.

To select objects with high X-ray to optical flux ratio we enlarged the $f_{\rm x}$/ $f_{\rm opt}$ classification used for the LMC. By using the optically brightest object in the error circle (to be conservative the 90% confidence error including the systematic error was multiplied by 1.5) a lower limit for the X-ray to optical flux ratio is derived. For this purpose optical Rand B magnitudes were taken from the USNO-A1.0 catalogue produced by the US Naval Observatory. Finally selecting objects with log($f_{\rm x}$/$f_{\rm B}$) greater than 0 excludes foreground stars and we classified these as $\left[{\rm nonstar.}\right]$Sources of this type in the observed regions outside the main regions of the SMC are candidates for background sources and if south of -74$^\circ$ 30$^\prime$ were in addition classified as [AGN]. However, we note that the H I emission of the SMC (Stanimirovic et al. 1999) extends that far south and finding X-ray binaries there is not completely ruled out.

Figure 4: Einstein IPC vs. ROSAT PSPC count rates. Sources identified with SNRs are marked with circles. The line is a linear fit to the SNR data points

Figures 5, 6 and 7 illustrate the selection criteria used for source extent, hardness ratios and X-ray to optical flux ratio. As in the LMC, the identified well known SNRs show the highest extent likelihoods and the two classified SNR candidates (mainly from hardness ratio selection) are found with significant X-ray extent. Also similar to the LMC the different source types separate in different regions of the HR diagram (Fig. 6) which can be divided into the same areas as in the LMC case. There is one object (source 380) with hardness ratios compatible to that of foreground stars which was classified as [nonstar] due to its high X-ray to optical flux ratio. A nearby radio source suggests the source as AGN candidate. The X-ray binary with HR1 near 0.5 marks the peculiar soft X-ray pulsar (source 53, H94). Foreground stars are well recognized from their hardness ratios and also from their low X-ray to optical flux ratio, i.e. typically log($f_{\rm x}$/ $f_{\rm opt}$)< -1(Fig. 7). Most of the objects with high $f_{\rm x}$/ $f_{\rm opt}$ have errors on HR1 larger than 0.25 and are not shown in (Fig. 7).

Figure 5: Source extent and extent likelihood for the SMC PSPC sources with off-axis angle less than 18$^\prime$ (plus signs). SNRs are marked with squares (filled: secure, open: candidates, crossed: classified in this work) and known point sources (X-ray binaries, SSSs, stars and AGN) with hexagons

Figure 6: Hardness ratios of identified and classified PSPC sources in the SMC area. X-ray binaries are marked with a hexagon, SSSs with crossed square, SNRs with square, stars with $\times$, and AGN with triangle. The thick lines separate areas where only members of a single class were found in the LMC (HP99). Only sources with error on HR1 and HR2 less than 0.25 are shown

Figure 7: Flux ratio $f_{\rm x}$/ $f_{\rm opt}$ as function of hardness ratio 1 (HR1). The optically identified sources are marked with different symbols (X-ray binaries: hexagon, SSSs: crossed square, foreground stars: $\times$ and AGN: triangle). Unidentified sources with error on HR1 less than 0.25 and nearby GSC entry are shown with HR1 error bars. Some lower limits derived from the optically brightest objects in the error circle are indicated by arrows

Table 7 summarizes the properties of 158 identified and classified sources sorted by source class. The classification (including candidates) is complete for the 44 brightest sources down to a PSPC count rate of 0.02 cts s^-1. This sample consists of 27% foreground stars, 16% background objects, 20% high mass X-ray binaries, 16% SNRs, 9% SSSs in the SMC and 9% sources which are probably either AGN or HMXBs. We expect at lower intensity levels less foreground stars but more background AGN.

The majority of classified sources (Table 7) consists of AGN and AGN candidates (46). Six AGN were optically identified by Tinney et al. (1997) and we suggest 22 new candidates which have an unresolved radio counterpart (mainly at $\lambda = 13$ cm). We classify as AGN additional 12 sources with high X-ray to optical flux ratio which are located south of the main body of the SMC. Table 7 contains further six uncertain AGN candidates due to large position uncertainties. The highly extended source 385 is identified with a cluster of galaxies. Another 24 sources are identified (12) and classified (12) as foreground stars.

X-ray sources intrinsic to the SMC are HMXBs, SSSs and SNRs. Fifteen HMXB and candidates, mainly X-ray pulsars discovered with instruments sensitive at energies of 1 - 10 keV are included. Only three pulsars compiled in Table 4 of KPFH99 (namely AXJ0051-722, XTEJ0054-720 and XTEJ0111.2-7317) were not detected in PSPC data. We note that source 79 in KPFH99 (source 188 in Table 7) is probably not associated with AXJ0051-722 because of the large (3$^\prime$) offset in position. Most of the X-ray binary systems were detected with PSPC count rates between 2.7 10^-3 cts s^-1 and 8.9 10^-2 cts s^-1 which converts to X-ray luminosities between 4.5 10³⁴ erg s^-1 and 1.5 10³⁶ erg s^-1 (using a typical conversion factor of 1.67 10³⁷ erg s^-1/cts s^-1; Kahabka & Pietsch 1996, hereafter KP96). This "quiescent" level is similar to that seen from persistent galactic high mass X-ray binaries like e.g. XPersei, RXJ0146.9-6121 or RXJ1037.5-5647 (Haberl et al. 1998; Reig & Roche 1999). With the exception of SMCX-1 and the peculiar soft X-ray pulsar RXJ0059.2-7138 no other X-ray binary was found with a luminosity in excess of 10³⁸ erg s^-1 in the ROSAT band (0.1 - 2.4 keV). This casts doubt on earlier suggestions made before the launch of ROSAT (when only few X-ray binary systems were known in the MCs) that the fraction of high luminosity sources emitting near the Eddington limit in the MCs is higher than in the galactic population (Pakull 1989).

Four sources were earlier identified as SSSs, two were suggested by KP96 and three more are classified here as SSSs. Source 691 is likely identified with the SMC symbiotic star LIN 358 and is located together with source 664 remarkably far south of the SMC main body. Source 16 as third classified SSS is found north of the main body of the SMC.

SNRs form the largest group of the SMC sources in Table 7. However, it should be noted that the class of SNRs is more complete than the class of X-ray binaries because the latter can not be uniquely separated from the hard sources. Fourteen clearly extended X-ray sources are identified with known SNRs and three with new candidates from radio investigations. In the work here another two sources are classified as SNR, both significantly extended X-ray sources. SNR0102-72.3 (source 107) is by far the brightest in this sample of SNRs and amounts to about 75% of the total known SNR X-ray luminosity in the SMC of $\sim$5 10³⁷ erg s^-1 (using the same count rate conversion factor from above).

The classification of sources as [hard], [nonstar] and [stellar] does not allow a unique identification of the origin of the X-ray emission but restricts the number of possibilities. [Hard] and [nonstar] objects are likely candidates for X-ray binaries or background AGN while the [stellar] classification most likely selects foreground stars, but bright SMC stars can not be excluded as e.g. the Be X-ray binaries show typical optical magnitudes of 15 (cf. HP99).

Figure 8: Distribution of foreground and background objects in the SMC region detected with the ROSAT PSPC. Foreground stars and candidates are marked with circle, AGN and AGN candidates with squares. The only source (385) identified with a cluster of galaxies is marked by a plus. Source numbers refer to Table 7

Figure 9: Distribution of SSS (plus) and X-ray binaries (circle) including new candidates from this work

Figure 10: Distribution of SNRs and SNR candidates in the SMC

 

 

Table 6: Classification criteria

Source class	selection	unidentified	classified
			sources
SSS	HR1+EHR1 < -0.70	3	3
fg star	HR1+EHR1 <0.25, HR2 - EHR2 > -0.70	7
	HR1 <0.25, HR2 > -0.70, EHR1 <0.25, EHR2 <0.25	1
	EHR1 <0.25, -0.75 <HR1 <0.25, log($f_{\rm x}$/ $f_{\rm opt}$) < -0.5	2
	EHR1 <0.25, 0.25 <HR1 <0.75, log($f_{\rm x}$/ $f_{\rm opt}$) < -1.0	3	12
SNR	ML $_{\rm ext}$ >50, offaxis angle <18$^\prime$	1
	HR1 - R1 >0.25, HR2 - EHR2 > -0.70, HR2+EHR2 < -0.10	1
	HR1 >0.25, -0.70 <HR2 < -0.10, EHR1 <0.25, EHR2 <0.25	0	2
hard	HR1 - EHR1>0.75, HR2 - EHR2 > -0.10	25
	HR1 >0.75, HR2 > - 0.10, EHR1 <0.25, EHR2 <0.25	1	26
AGN	EHR1 <0.25, HR1 >0.25, log($f_{\rm x}$/ $f_{\rm opt}$) > -1.0	1
	nonstar south of -74$^\circ$ 30$^\prime$	12	13
stellar	r90 <20 $^{\prime\prime}$, distance to GSC entry <r90	16	9
nonstar	lower limit of log($f_{\rm x}$/ $f_{\rm opt}$)>0.0	31	31

3.3 Radio sources

Besides the radio catalogues covered by SIMBAD we correlated the PSPC catalogue with a preliminary list of radio sources resulting from the ATCA mosaic high resolution survey (Filipovic & Stavely-Smith, in preparation). The PSPC sources detected at $\lambda = 13$ cm ( $\nu = 2.37$ GHz) are indicated by "13 cm" in the comment column of Table 7. An inspection of the source morphologies on the 13 cm image revealed many point sources which are candidates for background AGN. In three cases jet like structures indicate radio lobes (sources 52, 157 and 390). Most known SNRs and SNR candidates are detected in the ATCA $\lambda = 13$ cm image and a new candidate (source 448) is found in the Eastern Wing of the SMC.

3.4 Spatial distribution of source populations

The spatial distribution of the different identified (including elsewhere published candidates) and classified PSPC sources from our SMC catalogue is shown in Figs. 8-10. Foreground stars and AGN are found in the whole field covered by PSPC observations (Fig. 8). Although, a lower density of detected AGN is expected along the main body of the SMC due to absorption of the X-ray emission, the small number of background objects in that area seen in Fig. 8 is mainly caused by incomplete classification of hard X-ray sources. The SMC intrinsic sources like HMXBs and SNRs are preferentially distributed along the main body of the SMC (Figs. 9 and 10). The high density might be caused by projection of a more extended distribution as Cepheid distances indicate a depth of the SMC of about 20 kpc (Mathewson et al. 1988). SSSs are detected around the main body (Fig. 9), very similar to the LMC where the SSSs are only found around the optical bar (HP99). SSSs may also exist inside the main SMC body but their supersoft X-ray emission is absorbed by the dense gas and dust.