Astron. Astrophys. 363, 843-850 (2000)

3. Gas kinematic properties

In the NE part of the galaxy there appears a chain of HII regions which does not match the shape of the spiral arm. This structure, which we call a "spur", is outlined in Fig. 2 by a dash contour. It begins near a bright HII region and extends nearly perpendicular to the spiral arm. The total luminosity of the spur reaches of the total luminosity of the galaxy.

The mean velocity curve used as the reference curve of circular rotation was derived from the velocity measurements across the entire body of the galaxy by applying the custom developed software based on the algorithm described by Begeman (1989) for pure circular rotation. As a first step we find the dynamical center position and the mean value of the systemic velocity . As a second step these parameters are fixed, and the position angle of the kinematical major axis PA and inclination i are estimated in tilted rings of 3" width. Fig. 4 shows the radial dependence for the rotation velocity (Fig. 4a), for PA (Fig. 4b) and for disk inclination i (Fig. 4c). At , velocity data are available only for small emission islands in the WE (part see Fig. 2a) and in this region we fix the mean values for i and PA. The resulting mean disk parameters (, , ) are in good agreement with those found by Afanasiev et al. (1988). The disk orientation parameters being fixed at their mean value and the rotation curve being extracted from the velocity field (Fig. 4a), this figure shows that the circular rotation velocity is approximately constant and does not exhibit any peculiarities for the radius range .

Fig. 4a-c. Analysis of the velocity field of the ionized gas in circular rotation approximation: a  rotation curve, b  position angle of the kinematical major axis, c  disk inclination. Dashed lines indicate the mean values of the orientation parameters

In the most part of the galaxy the profiles of the and [NII] emission lines are quite symmetrical and have a gaussian shape (except the central region where a bar-like structure may be located). But in the "spur" the emission profiles differ from the common picture. In many locations in the "spur" the profiles split into two components: a "normal" component, close to the expected one from the circular rotation velocity field, and an "abnormal" component, shifted by . To study this peculiarity in detail, we have binned resulting in our data cube (by ) an enlarged pixel size of . Double-horned profiles of the spectral lines were fitted by two gaussians, corresponding to the "normal" and "abnormal" velocity components. Both and [NII] emission lines were used. In some regions of the spur the latter appears to be strongly enhanced, almost up to the level of the line intensity.

Fig. 5 reproduces the enlarged -image of the "spur", where different regions are identified by letters A - H. Typical line profiles of and [NII] are also shown for every region in Fig. 5i and Fig. 5j. The vertical arrows in each frame corresponds to the velocity of circular rotation.

Fig. 5a-j. image of the "spur" and examples of emission line profiles from regions around/in the "spur" (see the text for details). In each spectrum the x-axis is in and the y-axis is a intensity in relative units. The arrow in each frame corresponds to the value of the mean circular rotation velocity. "Normal" and "abnormal" components of are marked as "n" and "a", [NII] line is marked as "[NII]". Above each frame the velocity of the "abnormal" component of is given (if present).

The regions with abnormal velocity components have a complex shape and are located mainly between the bright HII regions of the "spur", avoiding the sites of active star formation. Although the brightness of emission lines away from HII regions is rather low, the observed anomalies are reliable features. As an illustration, Fig. 5j shows normal line profiles obtained for a low brightness region. The bright giant HII regions possess quite normal line profiles (see Fig. 5i), and their velocities correspond to the expected ones for the pure rotation.

To obtain a map of residual velocities in the given region of the galaxy, the simulated 2D line-of-sight velocity field corresponding to the mean rotation curve (Fig. 4a), was subtracted from the observed velocity field. The map of the residual velocities overlapped by the isophotes of the image is shown in Fig. 7 - separately for "normal" and "abnormal" components. It shows that local velocities of the "normal" components are not perturbed by HII regions. They are in good agreement with the rotation, and hence are related to the non-disturbed gas. Let us note however that the dispersion of residual velocities is about that exceeds the observational errors and might reflect velocity perturbation by a density wave. On the contrary, velocities of the "abnormal" components differ from circular velocities by about and as mentioned above they are observed mostly between the bright HII regions.

Kinematic and photometric parameters of the regions marked in Fig. 5, are given in Table 2. Column (1) gives the region identification in agreement with Fig. 5. Columns (2) - (3) give the mean velocity residuals (observed velocity minus circular velocity) for "normal" () and "abnormal" () velocity components, measured from profiles. Column (4) gives the residual velocities found for [NII], Columns (5) and (6) provide the intensity ratios of [NII] to lines and the ratio of "normal" to "abnormal" components. The errors in Columns (2)-(6) were obtained by the intensity-weighted averaging of values over the whole region. Column (7) gives the total luminosity (in ). Line intensities were not corrected for internal absorption. Such a correction would increase , but would not change the intensity ratios.

Table 2. Residual velocities and line ratios for different regions of the "spur".

As seen in Table 2, the "abnormal" component is especially strong on the periphery of the "spur" (regions A and D). It is just where the relative intensity of the nitrogen line is observed to be the largest: [NII] in these regions is comparable to and sometimes is larger (see Fig. 5a and 5c).

It should be noted however that there is an uncertainty in the estimates of line ratios due to continuum subtraction the overlapping of two interference orders. In addition, the transparency of the interference filter is different for [NII] and lines, and the velocity variations of these components may also change their observed relative intensity. But it cannot affect the results significantly because within the same region the observed velocity range of any component usually does not exceed . Note also that independent measurements of the line intensity ratios in bright HII regions of the "spur" carried out at the same telescope with the long slit spectrograph UAGS (A.N. Burenkov, private communication) give [NII]/, which is in good agreement with our own measurements.

In the Region B and in the Region D which captures the extension of the bright HII region, the [NII] line intensity is relatively low ([NII]/), and the non-circular component is seen only as an asymmetry in the profile. In Region E, which extends over about 2 kpc between two bright HII regions, the relative intensity of the "abnormal" component is also low, less than 20% of the normal one. Non-circular motions of the gas are traced by an enhanced "red" wing of the profile. A similar asymmetry is typical for [NII] line profiles in this region. In the Region G, neighboring E, the intensity of [NII] becomes comparable to . Non-circular components of are not detected (Fig. 5g), but the [NII] line is redshifted by at least with respect to . The situation is different in the isolated Region F, lying at the inner edge of the "spur". The relative intensity of [NII] looks normal here, but the line possesses a bright blue-shifted non-circular component. Note that this is the same region where the negative relative velocity excess was found earlier by Afanasiev et al. (1988) from long-slit observations of with lower spectral resolution.

Finally, the Region H, lying on the continuation of the "spur" differs from the other regions by an unusually weak line ([NII]/) and by the absence of a noticeable non-circular component.

Let us note that all anomalies in the emission lines profiles cannot result from the errors of the night sky subtraction. In Fig. 6 we present examples of the abnormal emission profiles from Fig. 5 and the night sky spectrum from Fig. 1b on the same intensity scale. Fig. 6a-c show some profiles with double-horned line and/or abnormal [NII] ratio. In contrast, in Fig. 6d we plot a "normal`' profile from the SW side of the galaxy, opposite to the "spur" region. Obviously all lines from the object are more intense than the sky spectrum. Moreover the brightest lines of the sky spectrum are located only near the `normal' component of the line (Fig. 6a and 6b). Therefore the "abnormal" component of the line and the largest [NII] lines are not related with overestimation or underestimation of the sky spectrum contribution.

Fig. 6a-d. The and [NII] profiles (solid lines) in comparison with the night sky spectrum (dashed lines) on the same intensity scale. a  the line profile from the Region A (red "abnormal" component of ), b  the line profile from the Region F (blue "abnormal" component), c  the line profile from the Region H (only the [NII] line without ), and d  the profile from the opposite side of the galaxy.

Fig. 7a and b. Residual velocities in the "spur" after subtraction of a pure circular rotation, a  for the "normal" component and b  for the "abnormal" component. intensity contour is overlapping these maps.

To summarise, the residual velocity distribution looks rather complex. From Table 2 it can be found that the "normally" rotating gas does not show a systematic deviation (within ) from the line-of-sight component of circular rotation. The "abnormal" component of is strongly redshifted everywhere except the isolated region F where the residuals have the same order of magnitude, but are negative. Velocity profiles of [NII] unlike reveal only one component, excluding the region E where there is a hint that some profiles are double-horned. The velocities measured from the [NII] profiles exceed those obtained from the "normal" profile components by in all regions with the exception of the Region H where the sign of the difference is opposite. Finally, the residual velocities found from the "abnormal" components and from the [NII] lines have the same sign in all regions except the region F, which support the hypothesis that these velocity anomalies could be related phenomena.

© European Southern Observatory (ESO) 2000

Online publication: December 5, 2000
helpdesk.link@springer.de