Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://lnfm1.sai.msu.ru/~anna/article/AR01p040.pdf
Äàòà èçìåíåíèÿ: Tue Nov 13 14:09:14 2001
Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 22:50:34 2012
Êîäèðîâêà:

Ïîèñêîâûå ñëîâà: dust
Astronomy Reports, Vol. 45, No. 1, 2001, pp. 34­43. Translated from Astronomicheskioe Zhurnal, Vol. 78, No. 1, 2001, pp. 40­51. Original Russian Text Copyright © 2001 by Sitnik, Mel'nik, Pravdikova.

Streaming Motions of Molecular Clouds, Ionized Hydrogen, and OB Stars in the Cygnus Arm
T. G. Sitnik, A. M. Mel'nik, and V. V. Pravdikova
Sternberg Astronomical Institute, Universitetskioe pr. 13, Moscow, 119899 Russia
Received November 26, 1999

Abstract--The radial velocity fields of molecular clouds, OB stars, and ionized hydrogen in the Cygnus arm (l ~ 72°­85°) are analyzed. A gradient VLSR/l in the mean line-of-sight velocities of molecular clouds and ionized hydrogen due to differential Galactic rotation is detected, and two groups of physically and genetically associated objects moving with different line-of-sight velocities are identified. One of the two molecular-cloud complexes (l ~ 77.3°­80°) is located within 1 kpc of the Sun, closer to the inner edge of the arm, whereas the other complex (l ~ 78.5°­85°) lies 1­1.5 kpc from the Sun and is farther from the inner edge of the arm. The residual azimuthal velocities of the objects in both groups are analyzed. The residual azimuthal velocities of the first molecular-cloud complex are directed opposite to the Galactic rotation (V ~ ­7 km/s), while those of the second complex are near zero or in the direction of Galactic rotation, independent of the distance to the complex (V 1 km/s). Like the molecular clouds, stars of the Cygnus arm form two kinematic groups with similar azimuthal velocities. On the whole, the mean azimuthal velocities V for the ionized hydrogen averaged over large areas agree with the velocities of either the first or second molecular-cloud complex. In terms of density-wave theory, the observed differences between the magnitudes and directions of the azimuthal velocities of the kinematic groups considered could be due to their different locations within the arm. © 2001 MAIK "Nauka / Interperiodica".

1. INTRODUCTION We investigate here the effect of spiral density waves on the interstellar medium of the Cygnus arm. Perturbations of the gravitational potential in a rotating galaxy and the resulting spiral shocks give rise to systematic (streaming) motions of young stars and gas [1, 2]. These streaming motions produce a characteristic residual-velocity field Vres for the stars and gas. The residual velocities are determined relative to a reference frame rotating uniformly with linear velocity Vgal (the mean circular velocity of Galactic rotation at a given Galactocentric distance): Vres = Vobs ­ Vap ­ Vgal, where Vobs and Vap are the observed heliocentric velocity and the velocity of the solar motion toward the apex, respectively. Taking into account the influence of shocks, we expect the following behavior of the azimuthal (V) and radial (VR--along the Galactocentric radius) residualvelocity components for young stars and gas inside the corotation radius [2]. The residual velocities of the stars and gas should be maximum near the inner edge of the arm, which coincides with the shock front. The radial and azimuthal residual velocities of these motions are directed toward the Galactic center and opposite to the Galactic rotation, respectively. The residual velocity decreases in magnitude with distance from the inner edge of the arm. At the outer edge of the arm, the residual velocity VR is close to zero, and the azimuthal velocity V is in the direction of Galactic rotation.

Thus, the residual azimuthal velocity reverses direction across the arm. The perturbation of the gravitational potential also forces the stars and gas to deviate from circular orbits in the inter-arm space. In contrast to the density-wave arms, the inter-arm radial residual velocity is directed away from the Galactic center. The interstellar medium, like stars, must react to the spiral density waves propagating through it. The main problem in analyses of streaming motions of interstellar clouds and rarefied gas is the determination of their heliocentric distances. The distances to interstellar clouds are usually derived from their observed line-of-sight velocities and an adopted Galactic rotation curve. When estimating kinematic distances, it is assumed that line-ofsight velocities are determined solely by Galactic rotation and that the residual velocity is zero. Therefore, the residual gas velocity can be determined only if an independent distance estimate is available. However, we would not attempt even a qualitative analysis without the previous discovery of systematic residual motions of OB associations in the Carina and Cygnus arms (Fig. 1a), predicted by density-wave theory [3, 4]. We have also made use of the special observing conditions in the Cygnus arm (l ~ 70°­90°); namely, it is possible to analyze variations of the azimuthal velocity, which nearly coincides with the radial velocity up to a distance of ~2 kpc. Note that only line-of-sight velocities are known for the gas, whereas our analyses of stellar motions included both line-of-sight velocities and proper motions.

1063-7729/01/4501-0034$21.00 © 2001 MAIK "Nauka / Interperiodica"


STREAMING MOTIONS OF MOLECULAR CLOUDS

35

Mon OB1 Cer OB3 Cer OB2

(a)
Per OB2

Col 121 Ori OB1 Vel OB2

Y
Sco OB2 Cyg OB7 Cyg OB9 Cyg OB8 Cyg OB1 Cyg OB3

Sgr OB1 Sco OB1 Ser OB1

1 kpc 20 km/s X

(b)
Cá OB3 Cá OB2 Cyg OB7

Y BB
Cyg OB8 1 = 76 Cyg OB3

AA
Cyg OB9 Cyg OB1

1 kpc

X

Fig. 1. (a) The observed residual-velocity field of Cygnus-arm associations and the surrounding interarm space. The show the radial (VR) and azimuthal (V) velocity components. The X axis is directed toward the Galactic center and the coordinate origin. (b) The molecular-cloud complexes AA and BB and Cygnus-arm associations in projection into plane. The boundaries of the Cygnus arm for a pitch angle of i = 10° are shown schematically. The dashed line indicates l = 76°, which is the adopted boundary separating the Cyg OB1 stars into two groups. ASTRONOMY REPORTS Vol. 45 No. 1 2001

dashed lines Sun is at the the Galactic the direction


36

SITNIK et al.

We will investigate the behavior of the interstellar medium in the spiral density wave in the direction of the Cygnus arm (l ~ 72°­87°) by analyzing the residual line-of-sight motions of molecular clouds and ionized hydrogen. Section 2 describes streaming motions in Cygnus-arm associations [4]. Section 3 investigates the residual line-of-sight motions and localization of molecular clouds. Section 4 considers the specifics of stellar kinematics of the Cygnus-arm associations and their relationship to molecular clouds. Section 5 investigates the distribution of ionized hydrogen. We conclude that the resulting kinematic pattern is consistent with the predictions of density-wave theory. 2. RESIDUAL RADIAL VELOCITIES OF CYGNUS-ARM OB ASSOCIATIONS The Cygnus arm contains two star­gas complexes: the distant Cygnus complex (l ~ 70°­81°, b ~ ­1.0°...+5.8°, r ~ 1.0­1.8 kpc), and nearby Cygnus­Cepheus complex (l ~ 81°­122°, b ~ ­5°... +12.5°, r ~ 0.6­0.9 kpc) [5]. The Cyg OB1, OB2, OB3, OB8, and OB9 associations belong to the Cygnus complex and Cyg OB7, OB4, Cep OB2, OB3, OB4, and Cas OB14 belong to the Cygnus­Cepheus complex (Fig. 1b). The Galactic coordinates l and b and the names of associations at Galactic longitudes l ~ 70°­90° are given in the table. A detailed analysis of the residual velocities of Cygnus-arm OB associations was performed by Sitnik and Mel'nik [4]. Here we make note only of the main results of that study. We determined the residual velocities of stars in OB associations and interstellar clouds using the Galactic rotation velocity, Vgal , and the velocity of the Sun toward the apex, Vap, derived from an analysis of Cepheid motions (see version A in [6] and Table 1 in [7]). We adopted a Galactocentric distance for the Sun R0 = 7.1 kpc [6, 8]. We assumed that the heliocentric distances r of associations were 80% of the distances of Blaha and Humphreys [9], based on the fact that this results in an OB-association distance scale that kinematically matches the Cepheid distance scale used in the rotation-curve solution [7]. In other words, we used a so-called short distance scale. The table gives the heliocentric distances r, median line-of-sight velocities (VLSR), residual line-of-sight velocities Vr, res, and residual azimuthal velocities V for the Cygnus-arm associations (see also [4]). Figure 1a shows the distribution of residual azimuthal and radial velocities of the Cygnus-arm associations and in the surrounding inter-arm region [4]. The azimuthal velocities of associations near the inner edge of the Cygnus arm (at Galactic longitudes l ~ 70°­80°) are directed opposite to the Galactic rotation, and have values V ~ ­16...­3 km/s (see also table). The azimuthal velocities of associations at l ~ 84°­122° (closer to the outer edge) are in the direction of Galactic rotation and have values V ~ 0­7 km/s. The residual radial velocities VR of all associations in this region are directed toward the Galactic center and their magnitude

in the Cygnus-arm cross section decreases with Galactocentric distance, from 10­23 to 2­8 km/s. Analysis of the stability of the derived residual velocities showed that variations in the rotation-curve parameters and the distance scale used over a broad range do not lead to qualitative changes in the residual velocity field in the Cygnus arm (see Fig. 6 in [4]). The derived variations of the magnitude and direction of the residual velocities VR and V for Cygnusarm associations testify to the density-wave nature of the spiral arm, and its location inside the corotation radius [1, 2, 4]. For convenience, we will refer to regions where the azimuthal motions of associations are opposite to or coinciding with the direction of Galactic rotation as the inner and outer arm regions, respectively. 3. RESIDUAL VELOCITIES OF MOLECULAR CLOUDS IN THE DIRECTION l ~ 73°­87° The Cygnus arm and the Sun are located at approximately the same Galactocentric distance, so that, in the longitude interval l ~ 70°­90°, we are looking at a cross section of the Cygnus arm (Fig. 1b). The line of sight runs along the arm (at least up to heliocentric distances of ~2 kpc) and is almost tangential to circular orbits. The residual velocities of stars and gas Vr, res must therefore nearly coincide with the azimuthal residual velocity V, as can be seen from a comparison of Vr, res and V for OB associations (see table). If streaming motions are induced by density waves, the residual line-of-sight velocities of interstellar clouds and gas Vr, res should change direction across the arm, from opposite of the Galactic rotation near the inner edge of the arm to coinciding with it at the outer edge [1, 2]. Since the inner edge of the Cygnus arm is seen at smaller Galactic longitudes (Fig. 1), these longitudes should be characterized by predominantly Sunward residual motions, whereas the residual motions at larger longitudes should be directed away from the Sun. To look for density-wave effects, we analyzed the distribution of CO in the direction toward the Cygnus arm. 3.1. Analysis of the CO Distribution in the Direction of the Cygnus Arm Leung and Thaddeus [10] have published detailed CO emission maps for the region l ~ 73°­87°, b ~ -4°...+5°. Figure 2 shows the latitude-averaged CO distribution in the (l, VLSR) plane adopted from [10]. Two molecular-cloud ridges can be identified in the velocity interval VLSR ~ ­25...+25 km/s (Fig. 2), which we will refer to as AA (l ~ 77.3°­80°, VLSR ~ ­6...+4 km/s) and BB (l ~ 78.5°­85°, VLSR ~ -1...+13 km/s). We distinguished these clouds--shaded areas in Fig. 2--around the brightest CO emission features at the level 3.8 K deg [10]. The Galactic longitudes, names, and velocity intervals VLSR where a cloud (or part of a cloud) is
ASTRONOMY REPORTS Vol. 45 No. 1 2001


STREAMING MOTIONS OF MOLECULAR CLOUDS Table Associations l, deg 71­74 74­78 74­76 76­78 76.3­79.5 77­80 77.3­80 78­80 80­80.7 80­81.7 81.7­82 82­83 83­84 84­85 84­96 ­ 4.9­9.0 Cyg OB7 b, deg ­0.5­3.6 ­ 0.8­3.0 name Cyg Cyg Cyg Cyg Cyg Cyg OB3 OB1 OB1B OB1A OB8 OB9 r, kpc VLSR, km/s Vr, 1.8 1.5 9 2 13 ­2 ­8 0 ± ± ± ± ± ± 2 2 3 2 3 5
res,

37

CO-clouds km/s V, km/s ­5 ­8 2 ­ 11 ­ 15 ­8 AA BB BB AA­BB ­ 6 ...+ 43 +3...+13 +3...+83 ­ 4...+11 +3...+633 3 ­ 1...+6 3 ­ 4...+5 ­8 1 0 ­1 1 0 ­1 ­1 ­2 name VLSR, km/s Vr, res, km/s ­4 ± 3 ­7 ± 2

2.0­5.8 0.5­2.2

0.7­1.3

Cyg OB2

1.8 1.0 1.0 1.5 1.5 1.5 1.5 1.5 1.5 1.0 1.5 1.0 0.7

­ 15 ± 4 ­7 ± 5

AA­BB 8±2 7±2 7

identified are summarized in the table. Bright CO emission features are projected onto various parts of the plane of the sky (see Fig. 8 in [10]). Therefore, the BB and AA ridges should be viewed as complexes of molecular clouds that could be located at different heliocentric distances. In particular, the features of BB seen at negative velocities in the direction l ~ 81.8°­82° and l ~ 84.5°­85° could be extensions of AA. The most striking feature of Fig. 2 is the systematic decrease of the line-of-sight velocities VLSR of bright CO emission features with increasing Galactic longitude, especially conspicuous for ridges BB. (The radial velocity VLSR is referred to the local standard of rest; i.e., it is corrected for the solar motion toward the standard apex, VLSR = Vobs ­ Vap). This monotonic decrease of VLSR must be due to a decrease of the Galactic rotational velocity Vgal(l, r) with longitude, since Vgal depends only slightly on distance up to r ~ 2 kpc in the direction considered. What we actually observe is the variation of the Galactic rotational velocity with Galactocentric distance. To illustrate this, we show in Fig. 3 the mean line-of-sight velocities VLSR of molecular clouds AA and BB superimposed on a family of curves defining the Galactic rotational velocity V * at each longitude for vargal ious Galactocentric distances. We computed the mean line-of-sight velocities VLSR of the clouds by averaging the corresponding values for each longitude over the shaded areas shown in Fig. 2. We computed the Galactic rotational velocity Vgal using a rotation curve based on the
ASTRONOMY REPORTS Vol. 45 No. 1 2001

motions of Cepheids [6]. To compare VLSR and Vgal, we introduced a correction to allow for the difference between the velocity of the standard solar motion (VX = 10 km/s, VY = 15 km/s, VZ = 7 km/s) [11] and the solar velocity inferred from the Cepheid-motion analysis (VX = 10 km/s, VY = 13 km/s, VZ = 7 km/s) [6]. We added this correction, equal to Vap = 2 sin l ~ 2 km/s, to the velocity of Galactic rotation Vgal . Both velocities * shown in Fig. 3, VLSR and V gal = Vgal + Vap, are thus corrected for the solar motion toward the standard apex. It is evident from Fig. 3 that the mean kinematic distances of the molecular-cloud complexes can be roughly estimated from the relative positions of the observed lineof-sight velocities VLSR of the molecular clouds relative to the family of Galactic-rotation curves. In the longitude interval l ~ 77°­85°, the observed line-of-sight velocities VLSR of molecular clouds AA and BB coincide, on average, with the Galactic rotational velocities for heliocentric distances of 3 and 1.5 kpc, respectively. These are precisely the distances of molecular clouds identified by OH absorption [12], CH emission [13], and H2CO absorption [14] (see references in the above papers). These molecular clouds were observed at the same lineof-sight velocities VLSR as the molecular-cloud complexes AA and BB. However, in view of the distribution of residual velocities of OB associations in the Cygnus arm, we suggest that a different location of the AA clouds is possible.


38

SITNIK et al.

20

10

B A B A

0

­10

­20 VLSR, km/s

­30

­40

­50

­60

­70

­80
Resolution

­90 87

86

85

84

83

82

81

80

79

78

77

76

75

74

73 l, deg

Fig. 2. The distribution of CO emission in the (l, VLSR) plane summed over the interval b ~ ­4°...+5° [10]. The lower level corresponds to 3 (0.38 K deg) and each subsequent level is 1.33 times the previous one. The clouds AA and BB are indicated by shaded areas.

3.2. Location and Residual Radial Velocities of Molecular-Cloud Complex AA AA (l ~ 77.3°­80°, b ~ ­1.6°...+1.5°, VLSR ~ -6...+4 km/s) is the brightest extended CO-emission feature seen toward Cyg X (Fig. 2). (We set the Galactic latitude boundaries for AA based on the distribution of CO emission in the plane of the sky for the corresponding interval of line-of-sight velocities and Galactic longitudes (see Fig. 8 in [10]).) The AA clouds are projected against an area occupied by associations of the inner part of the cloud, Cyg OB9

and, partially, Cyg OB1, which are located within 1.5 kpc of the Sun. The residual line-of-sight velocities of these associations are directed toward the Sun and have values Vr, res ~ -7 km/s [4]. If the molecular clouds AA, like the associations, are located near the inner edge of the arm, they, too, must have negative residual velocities Vr, res . Allowance for this velocity reduces the inferred distances to the clouds (Fig. 3). In fact, there is certain circumstantial evidence that the AA clouds are nearby. In the direction in which most of the clouds of complex AA are observed, i.e.,
ASTRONOMY REPORTS Vol. 45 No. 1 2001


STREAMING MOTIONS OF MOLECULAR CLOUDS

39

l ~ 77°­82°, b ~ 0°­4°, there is strong optical absorption (AV ~ 3­6m) at heliocentric distances <1 kpc (see Fig. 8a and areas 76/0, 78/1, 80/1, and 81/3 in [15]). In other directions toward the area considered l ~ 70°­100°, b ~ -5°...+5°, the extinction has similar or lower values (AV ~ 3­4m) at larger distances. It is very probable that the brightest cloud of CO emission coincides with a region of isolated interstellar absorption, not only in the plane of the sky, but also in heliocentric distance; i.e., AA lies no further than 1 kpc from the Sun. Based on their comparison of the distributions of CO emission and absorbing material toward l ~ 65°­100°, Dame and Thaddeus [16] also concluded that some of the molecular clouds were close to the Sun, and associated with a well-known region of strong optical absorption--the socalled Cygnus Dust Tongue, at a heliocentric distance of r ~ 0.7 kpc (see also [17] and references therein). There are independent estimates of the distance to the stellar and gas populations in this region. The star WR 143 and association Cyg OB9, with distances of r ~ 0.8 [18] and 1 kpc (see table), respectively, lie in the direction of AA. Since the optical absorption exceeds 3m for 63% of the stars of Cyg OB9 [9], this association is most likely located behind AA. The well-known compact star-forming region ON2 (G75.8+0.4) and the star WR 142 (r ~ 0.9 kpc [18]) are projected against the region occupied by the wedges of the AA ridges. WR 142 and ON2 are associated with the young open cluster Be 87, whose photometric distance is 0.9 kpc [19]. Thus, the molecular-cloud complex AA with its embedded star-forming regions may actually be a nearby object located within 1 kpc of the Sun (Fig. 1b). The velocity of Galactic rotation toward l ~ 77°­80° is V * = 6­7 km/s at a heliocentric distance of r ~ 1 kpc. In gal this case, the residual radial velocity of AA, Vr, res = VLSR ­ V * , falls in the interval ­12...+3 km/s. This is equal, gal on average, to Vr, res ~ ­7 km/s, and is therefore close to the velocities Vr, res of the inner-arm associations Cyg OB9 and OB1 at heliocentric distances of 1­1.5 kpc (see table). Note that, in the direction l ~ 77­85°, the velocity of Galactic rotation V * varies by less than 3 km/s gal over the rather broad distance interval from 0.5 to 2 kpc and, consequently, the residual line-of-sight velocities of clouds are only weakly sensitive to distance errors (Fig. 3). The observed line-of-sight velocities VLSR of molecular clouds AA can therefore be explained by densitywave effects without putting the clouds at a heliocentric distance of 3 kpc. Figure 4 and table give the residual velocities Vr, res of the molecular-cloud complex for an assumed heliocentric distance of r = 1 kpc. It is evident from Figs. 3 and 4 that, if the distance of complex AA is less than 3 kpc, its residual radial velocity Vr, res should always be negative. Therefore, the azimuthal component of the residual velocity of AA is opposite to the Galactic rotation, as must be the case for objects in the inner part of a density-wave arm.
ASTRONOMY REPORTS Vol. 45 No. 1 2001

15 10 VLSR, km/s 5

0.5 1.0 0 1.5 2.0 ­5 2.5 ­10 3.0 ­15 3.5 87 85 83 81 79 77 75 73 l, deg

Fig. 3. Mean line-of-sight velocities VLSR of molecular clouds AA and BB (filled circles), the ionized hydrogen (open circles), and stars of the Cyg OB1, OB3, OB9, and OB8 associations (dots), together with a family of curves defining the velocity of Galactic rotation V * at each longal gitude for various heliocentric distances. Distances (in kpc) are given to the left of the curves. Both VLSR and V * are gal corrected for the solar motion toward the standard apex. Only stars with VLSR errors of less than 15 km/s are shown.

It is quite possible that complex AA, which is a relatively compact object in the plane of the sky, is actually extended along the line of sight and, consequently, along the edge of the arm (Fig. 1b). This could explain the unique optical absorption in this direction. 3.3. Location and Residual Radial Velocities of Molecular-Cloud Complex BB The extended molecular-cloud complex BB (l ~ 78.5°­85°, VLSR ~ ­1...+13 km/s; see Fig. 2) is seen projected against a region occupied by associations of both the inner (Cyg OB9 and Cyg OB8; l < 80°) and outer (part of Cyg OB7, l > 84°) arm (see table). No associations are observed at longitudes l ~ 80°­84°, i.e., toward most of BB. It is clear from Fig. 3 that, at each longitude except l > 83°, the mean line-of-sight velocities VLSR of bright features of the molecular clouds BB are greater than or equal to the maximum possible velocities of Galactic rotation V * . (In this direction, the Galactic rotational gal velocities peak within 2 kpc). Therefore, most of the BB clouds have positive or almost zero residual line-ofsight velocities Vr, res, independent of their heliocentric distances (Fig. 4), as is characteristic of the central and outer arm regions. It is precisely such line-of-sight stel-


40 Vr, 10
res,

SITNIK et al. km/s
r = 1.0 kpc

5

Cyg OB1B

0
Cyg OB9

Cyg OB3

­5

Cyg OB1A

­10
Cyg OB8

­15

87

85

83

81

79

77

75

73

l, deg

Fig. 4. Residual velocities V of molecular clouds AA and BB (filled circles), the ionized hydrogen (open circles), and Cygnus-arm associations (rectangles) determined for heliocentric distance 1 kpc. Two groups, A and B, are identified in the Cyg OB1 association.

lar stream patterns with positive Vr, res that Sitnik and Mel'nik [4] found in associations of the outer part of the Cygnus arm, including Cyg OB7 (Fig. 1a). We can say the following about distances to the molecular-cloud complex BB. (1) Numerous analyses of extinction in the direction of the Cygnus arm indicate that the dust is concentrated in the same regions as the OB stars (see, e.g., [20]), and that the photometric distances to Cygnus OB associations are 0.7­1.8 kpc (see table). (2) The relative positions of the complexes AA and BB in the plane of the sky and their kinematics suggest that BB is located near the middle of the arm, implying a minimum residual azimuthal velocity. The kinematic distance of BB corresponding to the minimum residual velocity Vr, res is 1­1.5 kpc (Fig. 3). (3) Dame and Thaddeus [16] believe that some molecular clouds at l ~ 65°­100° are associated with the radio source Cyg X, which is extended along the arm and has (according to [16]) a heliocentric distance of r ~ 1.7 kpc (the molecular-cloud complex BB, in our case). Apparently, BB consists of molecular clouds at distances of ~1 to ~1.5 kpc (Fig. 1b). To summarize, we have direct evidence for variations of the residual line-of-sight velocities Vr, res (or V) of molecular clouds across the Cygnus arm. Given the relative positions of molecular clouds AA and BB, their residual velocities, and the fact that the Galactic spiral arms are trailing, we conclude that complexes AA and BB are located in different parts of the arm cross section.

Figure 1b shows schematically the positions of the molecular clouds and Cygnus-arm associations. Molecularcloud complex AA (l ~ 77.3°­80°), which is located within 1 kpc of the Sun, lies near the middle-arm line and has zero or positive residual velocities Vr, res. Although AA and BB are located at different heliocentric distances, they represent a cross section of the arm in the plane of the sky. Therefore, the change in the direction of V for the molecular clouds observed at l ~ 78°­79°, which is opposite to and coincides with the direction of Galactic rotation in complexes AA (V ~ Vr, res ~ ­7 km/s) and BB (V ~ Vr, res +1 km/s), respectively, is characteristic of the velocity-variation pattern predicted by density-wave theory. 4. LINE-OF-SIGHT VELOCITIES OF OB STARS IN CYGNUS-ARM ASSOCIATIONS Figure 3 also shows the line-of-sight velocities VLSR of stars in the Cyg OB1, OB3, OB8, OB9, and OB7 associations, which are located in the region studied and have heliocentric distances of 0.7­1.8 kpc. We adopted the lists of association stars from the catalog of Blaha and Humphreys [9] and their radial velocities from the WEB catalog [21]. The distribution of line-of-sight velocities VLSR of stars in the interval l ~ 72°­80° proved surprising. The stars in Cyg OB1, Cyg OB3, Cyg OB8, and Cyg OB9 separated into two groups with line-of-sight velocities VLSR < 5 km/s and VLSR > 7 km/s (Fig. 3). The histogram of stellar velocities VLSR has two pronounced maxima (Fig. 5). This bimodal distribution indicating two
ASTRONOMY REPORTS Vol. 45 No. 1 2001


STREAMING MOTIONS OF MOLECULAR CLOUDS

41

velocity groups is due primarily to stars of the richest association, Cyg OB1. The kinematically identified stellar groups in Cyg OB1 are shifted relative to each other in Galactic longitude (Fig. 6a). The stars of Cyg OB1 belonging to the first group (A), have l ~ 76°­78° and move with a median velocity of VLSR = ­2 ± 1 km/s, whereas stars belonging to the second group (B) have l ~ 74°­76° and move with a median velocity of VLSR = 13 ± 3 km/s (Figs. 3, 6a). The stars of the first group are concentrated in the (l, VLSR) plane near the molecularcloud complex AA (Fig. 3); i.e., they move with the same radial velocities as the AA clouds. However, in the plane of the sky, these stars lie outside regions of bright CO emission. The radial velocities of stars of the second group are in good agreement with the longitude dependence of the mean VLSR values for molecular clouds BB (Fig. 3), and continue the linear VLSR(l) relation toward lower Galactic longitudes (l ~ 73°­76°). The agreement of the line-of-sight velocities of stars of the first and second groups with those of molecular clouds AA and BB (Fig. 3), respectively, can be explained as follows. OB stars of Cyg OB1, like those of other Cygnus-arm associations, formed from clouds in extended molecular­dust complexes. AA and BB could be the remnants of such complexes. The stars have the same velocities as the remnants of their parent cloud complexes. It is possible that stars of the first and second groups in Cyg OB1 do not only have different line-of-sight velocities, but also, like the clouds, have different heliocentric distances. Figure 1b shows how Cyg OB1 stars with positive and negative residual lineof-sight velocities could be observed along the same line of sight in a spiral arm with pitch angle i = 10° [22, 23]. However, we were not able to separate these kinematically distinct groups of the Cyg OB1 association into radial subgroups. This might suggest that the separation between the two groups is less than the standard errors in their heliocentric distances, 0.3 kpc. Differences in distance estimates for Cyg OB1, 1.8 kpc [9] and 1.2 kpc [24], may be partially due to the fact that the two stellar groups have been combined into a single association. Since photometric distances to the associations are known (see table), as we can see from Fig. 3, the residual line-of-sight velocities of stars of one group at l < 80° are directed opposite to the Galactic rotation, while those of the other group are either close to zero or are directed along the Galactic rotation (Fig. 4). Figure 6b shows how the residual line-of-sight velocities Vr, res of stars in the Cyg OB1, OB3, OB8, and OB9 associations depend on the absolute bolometric magnitude Mbol . (See [3] for a description of corrections applied to the MV values adopted from Blaha and Humphreys [9].) We can see from this figure that stars with streaming motions opposite of the Galactic rotation are, on average, more luminous and possibly younger than stars with streaming motions along the direction of Galactic rotation. Consequently, we expect the stars of different kinematic groups in the Cygnus arm to have different ages.
ASTRONOMY REPORTS Vol. 45 No. 1 2001

N 15

10

5

­ 40

­30 ­20

­10 0 10 VLSR, km/s

20

30

40

Fig. 5. Histograms of VLSR for stars of the Cyg OB1, OB3, OB9, and OB8 associations. The shaded area is the velocity histogram for the stars of Cyg OB1.

Thus, the stars of Cygnus-arm associations and the remnants of the parent molecular clouds form two kinematic groups, with the residual azimuthal velocities of one group directed opposite to the Galactic rotation and those of the other group being close to zero or directed along the Galactic rotation. Since stars are genetically and physically associated with molecular clouds, it is possible that stars with Vr, res directed opposite to the Galactic rotation have, like AA and BB, different locations in the arm cross section. Should this be the case, the direction of the residual velocities of young stars and molecular clouds could be interpreted in terms of density-wave effects in the gaseous subsystem of the Galaxy. 5. IONIZED HYDROGEN TOWARD GALACTIC LONGITUDES l ~ 72.5°­85° We used the database of Lozinskaya et al. [17] to analyze the radial velocity field of ionized hydrogen in the direction of the Cygnus arm (l ~ 72.5°­85°, b ~ -1°...+4°). The ionized hydrogen line-of-sight velocities were determined at the positions of intensity peaks for each feature of the H profile. (See [17] for a detailed description of the observations and their reduction.) The bulk of the ionized gas, which is described by the bright component in the line profile, emits in the radial velocity interval from ­20 to +25 km/s. This nearly coincides with the radial velocity interval for bright CO emission associated with the Cygnus arm. To compare the line-of-sight velocities VLSR of the ionized hydrogen and molecular clouds, we averaged the velocities of the main component of the H line over areas with l = 0.5° and b = 5° [17]. Figure 3 shows the resulting velocities. Note that the number of main-line velocity measurements per area varies from 100­970 VLSR values


42 VLSR, km/s (a)

SITNIK et al.

30 20 10 0 ­10 ­20 ­30

78 77 VLSR, km/s 30 20 10 0 ­10 ­20 ­30 ­3 ­4 ­5 (b)

76 l, deg

75 Cyg Cyg Cyg Cyg OB1 OB3 OB9 OB8

74

clouds of ridges BB (Fig. 3). In this region, ionized hydrogen is observed at positive line-of-sight velocities that are close to the corresponding velocities for the second group of Cyg OB1 stars. Both in the interval l ~ 77°­80°, where both molecular clouds are observed, and in the interval l ~ 80°­85°, the ionized hydrogen has line-of-sight velocities that are either equal to those of the BB or AA molecular-cloud complexes, or fall in the interval between the mean velocities of the two clouds. This could be partially due to the fact that the mean velocities of the ionized gas were derived for large areas that overlap with both clouds. (The mean ionized-hydrogen velocities derived for a finer partition-- e.g., into 0.5 â 0.5 deg2 areas--are, indeed, close to the mean velocities of the clouds observed in the corresponding directions). Furthermore, we cannot rule out the possibility of effects due to peculiar gas motions driven by ionizing radiation and stellar winds from young stars in such a dynamically active region as Cygnus [17], especially since the observing coverage is higher toward HII regions than between them. Since the ionizing radiation is produced by stars of the Cyg OB1, OB2, OB3, OB8, OB9, and OB7 associations, most of the ionized hydrogen is located in the regions occupied by these associations; i.e., at photometric distances of 0.7­1.8 kpc. OB stars and the hydrogen they have ionized are primarily observed outside molecular clouds, although the Galactic longitude distribution of the radial velocities of the bulk of ionized hydrogen and OB stars agrees with the corresponding distribution for CO emission. Figure 4 shows the residual azimuthal velocities of the ionized hydrogen Vr, res for a heliocentric distance of 1 kpc. On the whole, the ionized-hydrogen streaming motions are in agreement with those for either AA or BB. 6. CONCLUSIONS We have analyzed the distribution of young stars, molecular clouds, and ionized hydrogen in the direction of the Cygnus arm (l ~ 72°­85°) and discovered the following features. (1) There is a considerable radial velocity gradient VLSR/l for molecular clouds and ionized hydrogen due to differential Galactic rotation (Fig. 3). (2) We have identified two kinematically distinct groups of genetically and physically associated objects. (a) Two molecular-cloud ridges have mean line-ofsight velocities VLSR ~ ­2...+1 km/s (AA) and VLSR ~ 0...+9 km/s (BB) (Figs. 1b, 2, 3). (b) Of two OB-star groups in Cygnus-arm associations, the first has VLSR that are negative or close to zero and is located near the clouds AA, and the second is located outside the clouds BB but has the positive velocities expected for clouds BB at these Galactic longitudes (Figs. 3, 5). In the case of Cyg OB1, the mean velocities of the OB-star groups are ­2 ± 1 km/s and +13 ± 3 km/s, respectively.
ASTRONOMY REPORTS Vol. 45 No. 1 2001

­6

­7 Mbol

­8

­9

­10

Fig. 6. (a) Dependence of VLSR for stars in the Cyg OB1 association on Galactic longitude and (b) dependence of Vr, res for stars in Cygnus-arm associations on Mbol .

in the interval l ~ 73°­82.5° to 20­40 in other regions. Thus, the mean ionized-hydrogen velocity has the highest frequency of occurrence in each area. The distributions of the mean velocities VLSR of the main H component and of molecular clouds are in good agreement (Fig. 3). The gradient of line-of-sight velocities VLSR of the bulk of ionized hydrogen in Galactic longitude is, as for the molecular clouds, due to the fact that the Galactic rotational velocity decreases with longitude (see also [17]). The linear dependence of the HII radial velocity VLSR at Galactic longitudes l ~ 72.5°­77° continues the similar VLSR(l) dependence for the molecular


STREAMING MOTIONS OF MOLECULAR CLOUDS

43

(c) The hydrogen ionized by Cygnus-arm stars and emitting in the interval l ~ 72°­85° has mean radial velocities corresponding to both AA and BB (Fig. 3). (3) The molecular clouds are situated at different heliocentric distances and at different locations in the Cygnus-arm cross section. (a) Clouds AA (l ~ 77.3°­80°) are within 1 kpc of the Sun and are closer to the inner edge of the arm. (b) Clouds BB (l ~ 78.5°­85°) are located at heliocentric distances of 1­1.5 kpc and are further from the inner edge of the arm. (4) The residual azimuthal velocities V of the two groups of objects have opposite directions. (a) The nearby clouds AA, most stars in Cyg OB1 (Cyg OB1 A group), OB3, OB9, and OB8, and a small fraction of the ionized hydrogen move opposite to the Galactic rotation, with V ~ ­13...­2 km/s (Fig. 4). (b) The more distant BB clouds, the remaining stars of Cyg OB1 (Cyg OB1 B), OB3, OB9, and OB8, and most of the ionized hydrogen have velocities V ~ -2...+5 km/s; i.e., they are close to zero or are directed along the Galactic rotation (Fig. 4). The large-scale motions of molecular clouds, OB stars, and ionized hydrogen are dominated by Galactic rotation and residual motions due to spiral densitywave effects. The role of Galactic rotation is clearly illustrated by the gradient of the mean line-of-sight velocities of the molecular clouds and ionized hydrogen observed along a 13° interval of Galactic longitude (corresponding to more than 200 pc at an average distance of 1 kpc). The role of density-wave effects is demonstrated by the kinematic signatures of the residual azimuthal velocities of molecular clouds, ionized hydrogen, and OB stars in the observed field. Analysis of the relative positions and velocity fields of molecular clouds indicates that the azimuthal component of the residual velocity reverses direction across the Cygnus arm: from being opposite of the Galactic rotation (clouds AA) to coinciding with it (clouds BB). Since the stellar and gaseous populations of each group are genetically and physically related, the OB stars of the Cygnus-arm associations and ionized hydrogen in each of the two kinematical groups should, like the molecular clouds, have different locations in the arm cross section. Thus, in spite of problems with localizing the clouds and ionized hydrogen, we were able to detect the azimuthal-velocity variations characteristic of densitywave arms: a decrease of the magnitude of V, with a possible reversal of its direction. The most common residual motions in the interstellar medium of the Cygnus arm, like the residual motions of stars in associations in this arm, reflect its density-wave nature. ACKNOWLEDGMENTS This work was supported by the Russian Foundation for Basic Research (project codes 98-02-16 032a and 99-02-17 842), the "Program of the Support of Leading
ASTRONOMY REPORTS Vol. 45 No. 1 2001

Scientific Schools" (grant no. 96-15-96 656), and the "Astronomy State Science and Technology Program" (grant no. 1.3.1.2). We are grateful to K.V. Bychkov, Yu.N. Efremov, A.V. Zasov, T.A. Lozinskaya, and A.S. Rastorguev for discussions and valuable comments. REFERENCES
1. C. C. Lin, C. Yuan, and F. H. Shu, Astrophys. J. 155, 721 (1969). 2. W. W. Roberts, Astrophys. J. 158, 123 (1969). 3. A. M. Mel'nik, T. G. Sitnik, A. K. Dambis, et al., Pis'ma Astron. Zh. 24, 689 (1998) [Astron. Lett. 24, 594 (1998)]. 4. T. G. Sitnik and A. M. Mel'nik, Pis'ma Astron. Zh. 25, 194 (1999) [Astron. Lett. 25, 156 (1999)]. 5. Yu. N. Efremov and T. G. Sitnik, Pis'ma Astron. Zh. 14, 817 (1988) [Sov. Astron. Lett. 14, 347 (1988)]. 6. A. K. Dambis, A. M. Mel'nik, and A. S. Rastorguev, Pis'ma Astron. Zh. 21, 331 (1995) [Astron. Lett. 21, 291 (1995)]. 7. T. G. Sitnik and A. M. Mel'nik, Pis'ma Astron. Zh. 22, 471 (1996) [Astron. Lett. 22, 422 (1996)]. 8. E. V. Glushkova, A. K. Dambis, A. M. Melnik, and A. S. Rastorguev, Astron. Astrophys. 329, 514 (1998). 9. C. Blaha and R. M. Humphreys, Astron. J. 98, 1598 (1989). 10. H. O. Leung and P. Thaddeus, Astrophys. J., Suppl. Ser. 81, 267 (1992). 11. P. G. Kulikovskioe, Stellar Astronomy (Nauka, Moscow, 1985). 12. M. I. Pashchenko, Astron. Zh. 50, 685 (1973) [Sov. Astron. 17, 438 (1973)]. 13. O. E. H. Rydbeck, E. Kollberg, A. Hjalmarson, et al., Astron. Astrophys., Suppl. Ser. 31, 333 (1976). 14. A. Piepenbrink and H. J. Wendker, Astron. Astrophys. 191, 313 (1988). 15. Th. Neckel and G. Klare, Astron. Astrophys., Suppl. Ser. 42, 251 (1980). 16. T. M. Dame and P. Thaddeus, Astrophys. J. 297, 751 (1985). 17. T. A. Lozinskaya, V. V. Pravdikova, T. G. Sitnik, et al., Astron. Zh. 75, 514 (1998) [Astron. Rep. 42, 453 (1998)]. 18. I. Lundstrom and B. Stenholm, Astron. Astrophys., Suppl. Ser. 58, 163 (1984). 19. D. G. Turner and D. Forbes, Publ. Astron. Soc. Pac. 94, 789 (1982). 20. N. B. Kalandadze and L. P. Kolesnik, Astron. Astrophys. 32, 57 (1977). 21. M. Duflot, P. Figon, and N. Meyssonier, Astron. Astrophys., Suppl. Ser. 114, 269 (1995). 22. Ju. N. Efremov, Astron. Astrophys. Trans. 15, 3 (1998). 23. A. M. Mel'nik, A. K. Dambis, and A. S. Rastorguev, Pis'ma Astron. Zh. 25, 602 (1999) [Astron. Lett. 25, 518 (1999)]. 24. C. D. Garmany and R. E. Stencel, Astron. Astrophys., Suppl. Ser. 94, 214 (1992).

Translated by A. Dambis