Astron. Astrophys. 352, 64-84 (1999)

3. Results

In Sect. 3.1 we present sources detected in the NGC 4258 field, in Sect. 3.2, then, we focus on the bulge and inner disk of NGC 4258 where both, point sources and diffuse X-ray emission are discernable (cf. Figs. 1-3). The remaining point sources in the disk of NGC 4258 are subject of Sect. 3.3, and the different emission components of NGC 4258 are summarized in Sect. 3.4.

3.1. HRI and PSPC point sources in the NGC 4258 field

In the field around the center of NGC 4258 30 point sources are detected (cf. Table 2, Figs. 1-3). Of these, 21, are visible with the HRI and PSPC, while 9 are exclusively discernable in the PSPC image. To convert the count rates of the sources (Table 2) to fluxes, energy conversion factors for different spectral models are given in Table 3, and a 5 keV thermal Bremsstrahlung model is used to convert count rates to fluxes for point sources. The ratio of the energy conversion factors depends on the model, and for the given models the ratios vary up to , and this uncertainty applies, if one compares PSPC and HRI fluxes of fainter sources for which, due to a lack of spectral information, a spectral model has to be assumed.

Table 3. Energy conversion factors for the ROSAT HRI and PSPC for different spectral models.
Notes:
) THBR = thermal Bremsstrahlung, THPL = thin thermal plasma (Raymond & Smith 1977 and updates)
) Energy conversion factors in units of  erg cm^{- 2} cts^-1
) Assumed for the conversion from count rates to fluxes for point sources

The light curves of all sources are presented in Fig. 4. If a source was detected during an observation block with a count rate exceeding its error, filled squares are plotted. Vertical bars represent statistical errors, open squares upper limits in the case of non-detections. Horizontal lines give the mean count rates during the whole HRI (blocks 2, 3, 6-9 of Table 1) or PSPC (blocks 1, 4, 5) observations. Solid lines represent sources that were detected, dashed lines represent sources for which only upper limits could be calculated. In general, the mean count rates are in good agreement with the results presented in Table 2. To enable a comparison of the HRI and PSPC count rates, the PSPC rates were divided by 3.2, the ratio between the HRI and PSPC energy conversion factors for the assumed 5 keV thermal Bremsstrahlung model (cf. Table 3).

Fig. 4. Light curves of X-ray sources in the NGC 4258 field. Where the source is detected during an observation block (bottom: number of the block according to Table 1, top: P/H indicating that the observations were performed with the PSPC/HRI) with a count rate exceeding the error filled squares are used as symbols and the errors are indicated as bars. In the case of errors exceeding the count rates upper limits () are plotted as open squares. The horizontal lines give the mean count rate calculated as explained Sect. 2.5. Solid lines represent sources that were detected according to Table 2, dashed lines represent sources for which only upper limits could be calculated. The PSPC count rates were divided by a factor of 3.2 before they were plotted in the diagram (this is the ratio between the energy conversion factors of the HRI and PSPC for a 5 keV thermal Bremsstrahlung spectrum corrected for Galactic absorption, cf. Table 3). If one detects an isolated point source and the spectral model is right, the light curve should be a straight line

To be independent from the assumption of a spectral model, time variability investigations were performed amongst the individual HRI observation blocks, as well as amongst the individual PSPC blocks. The results of a likelihood ratio test were transformed into Gaussian significances for the variability of the sources, and for all sources, for which time variability could be established with a significance , this significance and the maximum flux are given in Table 4. Time variability is established for 7 of the thirty sources by this method.

Table 4. Time variability investigations of the different HRI and PSPC observation blocks.
Notes:
) Gaussian significance for the detection of time variability during the individual HRI (H) or PSPC (P) observation blocks according to Sect. 2.5

One half (15) of the sources in the field are detected within the optical extent of NGC 4258, while two sources (X6 and X26) are located close to the border of the optical extent of NGC 4258. Table 5 gives the luminosities for the sources in NGC 4258, ranging from  erg s^-1 to  erg s^-1 for the central source X18. The ROSAT observations prove time variability in the case of three of the 15 NGC 4258 sources, namely X13, X16, X24. The maximum luminosities are  erg s^-1,  erg s^-1 and  erg s^-1 for X13, X16 and X24, respectively.

Table 5. Luminosities of sources located within the area covered by the ellipse of NGC 4258.
Notes:
) Luminosities at the distance of NGC 4258 in units of  erg s^-1, cf. Table 3 for assumed spectral model

3.2. Resolving the central 10 kpc region of NGC 4258

The central 10 kpc (5:05) region of NGC 4258, corresponding to the bulge and inner disk of NGC 4258, is covered by extended emission structures on top of which several point sources are detected (cf. Figs. 1-3). A part of this emission can be spatially resolved with the help of the HRI (Sect. 3.2.1), while the PSPC allows spectral investigations of different regions (Sect. 3.2.2), however having a worse spatial resolution ( PSF FWHM versus for the HRI).

3.2.1. The HRI: spatial resolution

The HRI images show a `ridge' of enhanced diffuse emission along an axis with position angle (p.a.) of , and the count rate in this region is (0.0570.002) cts s^-1 (mask along the arms with a width of 30"). Assuming a thin thermal plasma (keV) for the diffuse emission component, the luminosity is  erg s^-1. The diffuse emission follows the axis (p.a. ) outwards to a distance of to the north west and south east. The south eastern tip of the diffuse emission structure then bends towards the east and becomes invisible east of the bending point. To the north west, the surface brightness of the ridge drops rapidly after reaching a distance of from the center of the galaxy. Further regions of patchy X-ray emission with a lower surface brightness than the described X-ray ridge, are visible south and west/north west of the nucleus of NGC 4258. In these regions, the detection algorithm flags three emission enhancements as point-like sources, X20 (in a region of patchy emission east of the nucleus), X9 and X14 (both west of the center of NGC 4258).

The HRI X-ray morphology of the diffuse emission was reported from the first observation blocks (27.2 ks, cf. CWdP), however the deeper observations with the better photon statistics now allow us to trace the diffuse emission features better and help to separate point source candidates from the diffuse emission features. The point source catalogue of Table 2 contains in total 9 sources located in the bulge/inner disk of NGC 4258 (X7, X9, X13, X14, X16, X17, X18, X19 and X20), two of them (X7 and X19) were contained in the source catalogue of CWdP.

In the central region of NGC 4258, 4 of the 9 sources detected were flagged as extended by the detection algorithm, X9 (FWHM of the extent = 10"), X18 (FWHM = 12"), coinciding with the center of NGC 4258, X19 (FWHM = ) and X20 (FWHM = 12"). The flagged extent might argue (1) for the detection of several unresolved point sources, (2) the detection of an extended source like a giant, X-ray bright star forming region, e.g., corresponding to 300 pc at the distance of NGC 4258, (3) a superposition of (1) and (2), or (4) the detection of local enhancements of the diffuse X-ray emission. X19 is not surrounded by diffuse emission of high surface brightness, and this source most probably cannot be attributed to fluctuations in the diffuse X-ray emission. In opposite, the diffuse emission surrounding X9, X18 and X20 has a higher surface brightness and shows steep brightness gradients on small scales. Since the ROSAT data do not finally prove whether or not X9, X18 and X20 are really point-like sources, these sources are excluded from further discussion of point sources. However, given the positional coincidence with the nucleus, X18 remains a good candidate for a point-like source. To examine the count rate of a possible point source on top of the diffuse emission, the excess count rate in an HRI detect cell of centered on the nucleus was compared to detection cells shifted 10" to the north west and south east along the diffuse X-ray ridge (the sides of all boxes were rotated corresponding to the position angle of the X-ray ridge). The received count rate of  cts s^-1 corresponds to  erg s^-1 for a power law model with photon index 1.8 as reported from ASCA observations (Makashima et al. 1994) and corrected for Galactic foreground only.

To further investigate the ridge of diffuse X-ray emission, a surface brightness profile was calculated along the position angle of . The individual boxes had dimensions of 10" (along the ridge) (perpendicular to the ridge, cf. Fig. 5 for results). The surface brightness peaks at the center of the galaxy, and the brightness along the arms is in clear excess of the background, which is indicated as a dashed line in Fig. 5. The decline from the center region brightness ( cts s^-1 arcmin^{- 2}) to the average values measured for the outer arm region ( cts s^-1 arcmin^{- 2} for distances ) is relatively steep towards the south east (half width at half maximum ) while being less steep towards the north west (), and such a behavior was already indicated qualitatively by the periodic gray-scale representation in Fig. 2.

Fig. 5. Surface brightness profile and errors of the ROSAT HRI X-ray emission along a position angle of and centered on the nucleus of the galaxy. The individual boxes had dimensions of , and the dashed line indicates the field background.

3.2.2. The PSPC: spectral capabilities

Due to its spectral capabilities and higher sensitivity for low surface brightness emission, the PSPC detector provides important complementary information to the HRI instrument. However, the wider PSF of the PSPC compared to the HRI does not allow spatial resolution of different emission components to the same detail as with the HRI.

The PSPC broad band image of the inner region of NGC 4258 (cf. Fig. 3) shows diffuse X-ray emission with an extent of . In the direction of the center, the surface brightness of the X-ray emission increases steeply. Along the minor axis in north west direction, a smooth decay, similar to that along the major axis, is detected, while the diffuse emission is more clearly structured towards the south west. Fig. 6 shows the PSPC images of the central region of NGC 4258 in the energy sub-bands soft, hard1 and hard2. The soft band shows diffuse X-ray emission, and the soft band emission maximum, which does not coincide with the center of the galaxy, is shifted towards the east. The offset of the emission can be explained by a ballistic model for the S band emission, which is assumed to emerge from hot gas filling the south-eastern halo hemisphere of the inclined galaxy (cf. the models presented in Paper I). The extended emission appears to have more structure in the hard bands, and the smooth distribution of the S band emission might partially be due to the wider PSF of the PSPC S band. At the spatial resolution of the hard bands, point source contributions on top of the diffuse emission become visible. The emission maxima of the hard bands coincide with the center of NGC 4258. In the H2 band a ridge-like region of high surface brightness, diffuse X-ray emission is discernable at a p.a. of , and a bending of the ridge is indicated at a distance north west and south east of the nucleus.

Fig. 6. Contour plots of the central emission region of NGC 4258 for soft (S), hard1 (H1), and hard2 (H2) ROSAT PSPC bands. The images were smoothed with a Gaussian filter with FWHM of the average PSPC point response function (47", 26", and 25", respectively) and exposure corrected for the individual energy bands. Soft band contours are given in units of 3, 4, 6, 9, 15 and 25 (1 cts s-1 arcmin- 2) above the background ( cts s-1 arcmin- 2). The hard band contours (background negligible) are 3, 5, 8, 12, 17, 25, 35, 50 and  cts s- 1 arcmin-2. Sources are marked according to Fig. 1 and Table 2

The spectral capabilities of the PSPC were used to further investigate the diffuse X-ray emission. In a first approach, all photons detected in the bulge and inner disk region (see Table 6 for the extracted area, the counts and hardness ratios) were extracted in one spectral file (All ). A field background was calculated from two source-free regions, one region encircled by the sources X23, X28 and X30 and another north of X2. Simple spectral models, thermal Bremsstrahlung (THBR), power law (POWL) and a thin thermal plasma (THPL, Raymond & Smith 1977 and updates), were fitted to the data (see Table 7 for results). As expected, non of the one component models is able to describe the spectral behavior of the file All , for which one expects contributions from point sources, the anomalous spiral arms, and the halo component. To separate the different emission components, we made use of the spatial resolution of the PSPC, and subdivided All into three different regions, the Nucleus (cut diameter of 24" corresponding to the on-axis FWHM of the PSPC PSF at 1.0 keV), the Point sources representing the integrated emission from the non nuclear point sources in the bulge and inner disk, and Diffuse representing the ROSAT unresolved X-ray emission (all point sources of Table 2 screened). Similar to All , the field background was subtracted from the component Nucleus , Point sources and Diffuse . To search for possible contributions of the active nucleus on top of the extended X-ray emission along the X-ray ridge, the spectral file Nucleus local , containing the same source photons as Nucleus , was constructed by subtraction of a `local' background from two detection cells of radius, shifted 24" towards the north-north-west and south-south-east in the direction of the X-ray ridge.

Table 6. Extracted spectral components.
Notes:
) Without sources X14, X18 and X20, cf. Sect. 3.2.1

Table 7. Spectral investigations of the extracted components: Results and errors calculated from error ellipses in the case of
Notes:
) THBR = thermal Bremsstrahlung, POWL = power law, THPL = thin thermal plasma (Raymond & Smith 1977 and updates)
) Luminosity corrected for Galactic foreground absorption in the 0.1-2.4 keV band
) Predicted luminosity corrected for total absorption
) Lower boundary fixed to Galactic foreground absorption
) Value fixed

From the hardness ratios, there is a clear difference between the different emission components. The highest absorption and the hardest spectral behavior is indicated for the file Nucleus local . The integrated point source emission shows a slightly lower HR1 value and a clearly lower HR2 value, arguing for lower absorption and a less hard spectrum. While the lower absorption can be explained in terms of the viewing geometry of the sources, the softer overall spectrum of the point sources might indicate that the detected point sources are not all X-ray binaries, for which a very hard X-ray spectrum is expected, but some might be superbubbles of hot gas or supernova remnants, for which one expects on average softer X-ray spectra.

Similar to the file All , simple spectral models were fitted to the other spectral components (see Table 7 for results), and errors of the spectral parameters were calculated in the case of values below 2.5. A THBR model gives the best description of the file Nucleus . The predicted temperature is 0.63 keV, the absorption  cm^-2. The value of 1.9 indicates that more component models might be necessary for a better description. However, we were not able to improve the value by physically motivated two component models, namely a combination of the best fit THBR model with either a THPL model describing a possible halo component or a POWL model describing possible contributions of the AGN. A POWL model with (according to Makashima et al. 1994) enables the best fit of Nucleus local , and this argues for the detection of an AGN component in the ROSAT band. As expected from the hardness ratios, a relatively high absorption ( cm^-2) is fitted. If indeed due to the nucleus, the luminosity was  erg s^-1, in clear excess of the HRI prediction for such a component (Sect. 3.2.1). Together with bad value of 2.4 the necessity of multi components again is obvious, but we failed in establishing such models due to the low photon statistics of Nucleus local .

For the diffuse emission, all one component models failed, the best value being 8.1 for the THBR model. A physically motivated two component model, describing thermal emission from the anomalous spiral arms and the outer disk/halo, allows a good description of the X-ray emission (). With the exception of the absorption of the outer disk/halo component (fixed according to the Galactic foreground), all parameters were left free. The fit and the residuals are shown in Fig. 7. The spectral results yield a temperature of 0.19 keV ( error taken from error ellipse for temperature and normalisation of first component) and a luminosity of  erg s^-1 for the outer disk/halo component, while the component describing the anomalous arms is higher absorbed ( cm^-2) and of higher temperature (0.50 keV). The predicted luminosity of the arms is  erg s^-1 corrected for Galactic foreground only or  erg s^-1 corrected for the total fitted absorption. The relatively small errors of the temperatures as well as the great difference of the absorption of the two thermal components support our assumption of different origin in favour of a single component with a broad temperature distribution. However, a two component model still has to be an oversimplification since one expects a temperature distribution instead of a fixed temperature for both components. In addition, the thermal model used (Raymond & Smith 1977 and updates) assumes an equilibrium state of the hot gas which seems unrealistic in the case of shock heated or outflowing gas.

Fig. 7. Results of a two component fit of the diffuse emission, cf. Sect. 3.2.2. Top panel: Count rate at the detector normalized to the energy, in counts s-1 cm-2, the crosses represent the measured count rates, the solid curve gives the best fit, the dotted curves represent the individual models. Middle panel: Flux of the observed X-ray emission normalized to the energy, in photons ), the crosses represent the observed flux (count rates defolded by the spectral model), the solid curve gives the best fit, the dotted curves represent the individual models. Bottom panel: Residuals of the fit in units of Gaussian significance.

The integrated point source emission can be described relatively well with the help of a THBR () or POWL () model, while a thin thermal plasma (THPL) failed (). Predicted absorptions are around  cm^-2, as expected for sources in the disk of the inclined galaxy, the average temperature or power law index are 1.1 keV and 2.3, respectively.

3.3. Point sources in the outer disk of NGC 4258

Besides the sources in the bulge/inner disk of NGC 4258, a couple of point sources were detected in the outer disk of NGC 4258, namely, X8, X10, X11, X12, X24 and X25. The sources in the outer disk have luminosities from  erg s^-1 (HRI measurement of X10) to  erg s^-1 (PSPC measurement of X24). The latter source is time variable and reaches a luminosity maximum of  erg s^-1 during observation block 5 (PSPC). For this bright ( PSPC net counts) source as well as for source X8 ( PSPC net counts) simple spectral models (POWL, THBR and THPL) were fitted to the PSPC data (see Table 8 for results). Independent of the models, the predicted averaged luminosities are  erg s^-1 and  erg s^-1 for X8 and X24, respectively, and these values are in good agreement with the results presented in Table 5.

Table 8. Spectral investigations of the point sources X8 and X24: Results and errors calculated from error ellipses.
Notes:
) THBR = thermal Bremsstrahlung, POWL = power law, THPL = tin thermal plasma (Raymond & Smith 1977 and updates)
) In excess of the Galactic foreground
) Luminosity corrected for Galactic foreground absorption in the 0.1-2.4 keV band
) Predicted luminosity corrected for total absorption

In the case of X8 all models achieve formally values of . The predicted absorption of the THPL model ( cm^-2) is clearly higher than the ones predicted for the POWL ( cm^-2) and THBR ( cm^-2) model, and in opposite to the THPL spectrum with  keV, the POWL () and THBR ( keV) predict an extremely hard source spectrum. Neither the spectra nor the time variability investigations (no time variability could be established for X8) allow us to finally arrive at a conclusion concerning the nature of the source. It may either be a real point like source with a very hard spectrum, or, as suggested by the THPL model, due to X-ray emission from hot gas suffering high absorption, as was expected for gas in a superbubble surrounded by a dense shell of cold gas.

While the THPL model shows a higher value in the case of X24, the POWL and THBR models fit the data well. The latter models suggest  keV or , and absorption in excess of the Galactic foreground is  cm^-2 and  cm^-2 for the POWL and THBR model, respectively. Even if the spectrum might not be as hard as expected for an X-ray binary system or an AGN, the time variability of X24 argues for the detection of an X-ray binary system or a background AGN.

3.4. The emission components of NGC 4258

The X-ray emission components of NGC 4258 for different regions of the galaxy are summarized in Table 9. The integrated point source luminosity of NGC 4258 as measured with the HRI and PSPC is  erg s^-1 and this value is in good agreement with the results of Sect 3.2.2. Diffuse emission from the bulge and inner disk (a region embraced by a circle with a diameter of 1.25  the diameter of the minor axes) contributes  erg s^-1. The PSPC and HRI measurements slightly differ, most probably due to uncertainties in the spectral model of the diffuse emission. A thin thermal plasma ( keV) is used to describe the diffuse emission components as given in Table 9, however, as shown in Sect. 3.2.2, one has to use different models for the different emission regions. Diffuse emission along the X-ray ridge contributes nearly one half of the entire diffuse emission of the bulge/inner disk region. With both detectors, no diffuse emission from the outer disk is detected. For the PSPC, even a depression ( cts s^-1) is measured, and this can be explained by a shadowing of the very soft 0.1-0.4 keV X-ray background. This background emission originates from sources `behind' the disk of NGC 4258, and it is absorbed (partly) by cold gas in the NGC 4258 disk (PSPC S band count rate ( cts s^-1). Contrary, the PSPC H band count rate ( cts s^-1) might be caused by emission from unresolved point sources in the outer disk of NGC 4258.

Table 9. Count rates, hardness ratios and luminosities for different emission components of NGC 4258. For point source components a thermal Bremsstrahlung spectrum (keV) was assumed, for for diffuse emission a thin thermal plasma spectrum ( keV). In the cvase of non-detrections 2 upper limits are given.
Notes:
) Within an area encircled by 1.25 the ellipse of NGC 4258
Without sources X14, X18 and X20, cf. Sect. 3.2.1
) Diffuse emission in bulge and inner disk within a circle of diameter 1.25 the of the minor axis around the center of the galaxy
) Within the area encircled by 1.25 the ellipse of NGC 4258 but outside bulge + inner disk, point sources excluded

The hardness ratios HR1 of all emission components of NGC 4258 indicate absorption in excess of the galactic foreground. The HR2 value of the entire bulge region is in agreement with the one of the X-ray ridge, and this might indicate a similar spectral behavior, however, as already pointed out, these values are averaged over regions in which the spectral behavior of the diffuse emission cannot be viewed as constant. For both, the bulge as well as the X-ray ridge, relatively soft spectra are suggested (HR2), while the point source content exhibits a harder spectral behavior (HR2). These findings were supported by the spectral investigations of different spectral regions in Sect. 3.2.2 and the bright point sources X8 and X24 (Sect. 3.3).

© European Southern Observatory (ESO) 1999

Online publication: November 23, 1999
helpdesk.link@springer.de