The Kinematics and Nature of Gould's Belt , a 30 Myr Old

Home , Gould Belt

507

THE KINEMATICS AND NATURE OF GOULD'S BELT A 30 MYR OLD STAR

FORMING REGION

1 2 1 3

P.O. Lindblad ,J.Palous ,K.Lod en , L. Lindegren

Sto ckholm Observatory,SE133 36 Saltsjobaden, Sweden

Astronomical Institute, Academy of Sciences of the Czech Republic,

Bocn I I 1401, 141 31 Prague 4, Czech Republic

Lund Observatory,Box 43, SE221 00 Lund, Sweden

to as Gould's Belt. Of investigations of the kine- ABSTRACT

matics of Gould's Belt we will here refer to those by

Westin 1985 and more recently by Mestres 1996

and byTorra et al. 1997. Lindblad et al. 1973 and

For a sample of 2440 non-sup ergiant Hipparcos stars

Sandqvist et al. 1988 identi ed a lo cal comp onent

b elonging to Stromgren's `early group' luminosities,

of interstellar hydrogen and dust related to Gould's

temp eratures, ages and distances have b een deter-

Belt.

mined. 241 stars younger than 30 Myr with well

determined ages and space velo cities fall within the

lo cal attened system, inclined 20 to the galactic

Very di erent interpretations of the deviating char-

plane, called Gould's Belt. The system is dominated

acter of Gould's Belt have b een suggested in the lit-

byanumb er of nearby prominent asso ciations within

erature.

a distance of 700 p c, foremost the Scorpius-Centaurus

and Orion asso ciations.

We derive galactic rotation parameters for young

2. SELECTION AND CALIBRATIONS

stars of ages less than 30 Myr situated outside

the Gould Belt region. This gives a at rotation

curve with a circular angular velo city at the Sun of

1 1

We selected those stars in the Hipparcos Catalogue

= 25:3 1:5kms kp c .

for which complete Stromgren u; v ; b; y and H pho-

tometry was available and that b elong to Stromgren's

The kinematics of Gould's Belt di ers signi cantly

`early group' Stromgren 1966, closely following the

from that of stars outside the system. Besides a slight

scheme given byWestin 1985. The Stromgren pho-

outwards motion of a few km s it rotates in the

tometry was corrected for interstellar reddening fol-

same direction as the galactic rotation and expands,

lowing mainly the pro cedure of Shobbro ok 1983.

giving apparent values for the galactic rotation pa-

Absolute visual magnitudes M were computed us-

1 1

rameters of B = 21 and K = +12 km s kp c .

ing relations given by Balona & Shobbro ok 1984.

Distances were derived from M and the V magni-

It is suggested that this system was b orn ab out 30

tudes given in the HIP, corrected for reddening. Ef-

40 Myr ago, p ossibly in a spiral arm, with an angular

fective temp eratures and b olometric corrections were

momentum to o large to keep the system gravitation-

computed according to the scheme for early group

ally b ound. The rotation may explain the atness of

stars given by Balona 1994. Ages of the stars were

this inclined system.

then determined byinterp olation in the tables of star

mo dels by Maeder & Meynet 1988.

Key words: lo cal kinematics; Gould's Belt; galactic

rotation; stellar asso ciations.

It should b e stressed that in this sample sup ergiants

were excluded b ecause the reddening corrections ap-

plied will not be accurate for sup ergiants. Also ex-

cluded were stars, where the p ossible variation in the

1. INTRODUCTION

Hipparcos magnitude system Hp > 0:6 mag, and

double stars with a magnitude di erence between

comp onents Hp < 1:23 mag, as well as stars with a

relative error in distance > 0:3 or estimated error in As is well known, a series of investigations, pio-

age determination > 20 Myr. This results in a total neered by Plaskett 1930 and Blaauw 1946, have

sample of 1605 stars b elonging to the `early group'. shown that early typ e stars in the solar neighb our-

In the kinematic analysis we exclude stars with error ho o d, within some 500 p c, show a di erent space dis-

in prop er motion > 2 mas/yr, and error in radial ve- tribution and kinematic b ehaviour from early typ e

lo city > 20 km/s, which brings down the sample with stars at larger distances. This lo cal system of young

relatively accurate space velo cities to 1222 stars. stars, inclined 20 to the galactic plane, is referred

508

Figure1. Spatial distribution of 593 stars in our sample younger than 30 Myr. The Sun is in the centre of the coordinate

system. The X axis points towards the galactic anticentre and the Z axis is normal to the galactic plane.

3. SELECTION OF GOULD BELTSTARS

In the present kinematic study we prefer to select the

stars b elonging to Gould's Belt purely on the basis

of their p eculiar spatial distribution. Figure 1 shows

the spatial distribution of the stars in our sample

younger than 30 Myr.

In this gure the inclined Gould's Belt can b e sp otted

as the disturbance in the Z distribution for jX j <

500 p c. After a careful study of the distribution for

anumb er of age and spatial intervals we suggest the

following criteria for memb ership of Gould's Belt:

age < 30 Myr

p osition within the b ox given in Figure 2 de ned

Figure2. Spatial distribution of 560 stars in our sample

450 Y

younger than 30 Myr and with Z < 500 pc. The Sun

450 + Y

is in the centre of the coordinate system. The X axis

points towards the galactic anticentre and the Y axis in

distance from the galactic plane < 500 p c.

the direction of galactic rotation. The rectangular box

shows the adopted limits for the Gould Belt region.

As can b e seen in Figure 2, the system is dominated

byanumb er of nearby prominent asso ciations within

a distance of 700 p c, foremost the Scorpius-Centaurus

eld around the Sun. For space velo cities U; V ; W ,

and Orion asso ciations. The pro jection of the Gould

where the comp onents are directed away from the

Belt stars selected in this way on the sky in galactic

galactic center, in the direction of the galactic rota-

co ordinates is shown in Figure 3.

tion and p erp endicular to the galactic plane, resp ec-

tively, the linear part of the velo city eld may be

describ ed by the solar motion U ;V ;W and the

4. CONDITION EQUATIONS nine gradients @U=@X; @U=@Y ; @U=@Z, etc.

As wehave prop er motions for many more stars than

The radial velo cities and prop er motions, together for whichwehave radial velo cities, it maybeofad-

with the distances, are used to derive the velo city vantage instead to use the expression for the radial

509

Table 1. Galactic rotation parameters for young stars in

the galactic plane outside of Gould's Belt.

Age Myr 0 30 30 60 all

N 142 55 291

U 4:6 1:0 7:4 1:3 7:2 0:6

V 11:8 1:0 10:1 1:3 10:4 0:7

W 7:4 0:6 7:9 0:7 6:3 0:3

A 12:8 1:2 14:7 2:3 13:7 1:0

B 12:5 0:9 16:8 1:6 13:6 0:8

C 0:0 1:4 3:6 2:4 0:8 1:1

K 0:4 1:4 4:1 2:0 1:1 0:8

than 30 Myr, 55 stars 30{60 Myr and 94 stars older

than 60 Myr.

Figure 3. Galactic coordinates for the 249 stars in our

sample selectedtobelong to Gould's Belt.

As is seen, the di erences b etween the three groups

are marginal. The youngest group mayhave a slight

inward motion of a few km s compared to the older

velo cities V and the two galactic comp onents of

stars. The young stars of ages less than 30 Myr show

prop er motion cos b, :

l b

clear evidence of circular motion with a lo cally at

rotation curve and a circular angular velo city at the

V = U cos l cos b V sin l cos b W sin b

1 1

Sun of =25:31:5kms kp c .

2 c

+ r cos b K + A sin 2l + C cos 2l 1

Intro ducing velo city gradients dep ending on Z in

4:74 r cos b = U sin l V cos l

Equations 1 to 3 did not give signi cant improve-

+ r cos b B + A cos 2l C sin 2l 2

ment. A solution adding higher order terms of r to

the same equations and including stars at r> 2kpc

4:74 r =U cos l sin b + V sin l sin b W cos b

did not give signi cant results for the co ecients of

r sin b cos b K + A sin 2l + C cos 2l 3

these higher order terms.

where l; b are the galactic longitude and latitude, r is

the distance from the Sun and:

6. GOULD'S BELT

1 @U @V

A = + 4

2 @Y @X

For stars within the Gould Belt region the corre-

1 @U @V

sp onding solution encounters diculties b ecause of

B = 5

the small extension of the region, the eccentric p osi-

2 @Y @X

tion of the Sun and the very clumpy spatial distribu-

@U 1 @V

tion of the stars.

C = 6

2 @X @Y

Solutions using Equations 1 and 2 separately and 1

1 @U @V

K = + 7

to 3 in combination are given in Table 2. In those

2 @X @Y

solutions stars with an error of the radial velo city >

2kms have b een discarded. Notable is the large

negativevalue of B and the signi cant p ositivevalue

The development in Equations 1 to 3 is restricted to

rst order terms in the distance, and due to the at-

ness of the galactic disc gradients in the Z direction

have b een left out.

Table 2. Galactic rotation parameters for young stars

within the Gould Belt region.

5. YOUNG STARS OUTSIDE GOULD'S BELT

N 144 9 ass.

Eq. 2 1 1{3 4{7

Toinvestigate the velo city eld for young stars in the

U 13:2 0:84 6:9 1:5 9:8 1:0

galactic plane but outside Gould's Belt we select stars

V 13:8 0:9 17:6 1:4 16:8 1:0

situated outside the boxof Figure 2, within 500 p c

W 8:2 3:2 6:2 1:0

of the galactic plane, and within 2000 p c in distance

A 7:8 3:9 11:5 4:1 6:1 4:1 3:9

pro jected on the plane. Table 1 gives the results of an

B 19:8 2:8 20:6 5:2 22:8

analysis using a combination of the Equations 1 to 3.

C 6:1 4:3 6:2 3:9 2:9 3:7 14:5

In this analysis the stars have b een divided into three

K 14:7 2:8 11:0 3:5 11:4

di erent age groups younger than 30 Myr, b etween

30 and 60 Myr and all ages represented within the

sample. In this last group there are 142 stars younger

510

7. DISCUSSION

It is of course a p ossibility that Gould's Belt is just a

random con guration of two or three dominating as-

so ciations concentrated in the third and fourth quad-

rants of galactic longitude, in particular the Scorpius-

Centaurus and the Orion asso ciations. The `Finger of

Go d' e ect could help to give the impression of a at-

tened plane. However, there are arguments to see this

assembly of stars as a coherent system with a com-

mon origin: a the p osition of the Orion asso ciation

b elow the plane and the Scorpius-Centaurus asso ci-

ation ab ove the plane at the opp osite galactic longi-

tude would also have to be at random. Also there

are more stars and asso ciations than these two that

give rise to the inclination; b Lindblad et al. 1973,

Lindblad 1974, Olano 1982, and Sandqvist et al.

1988 show that there exists a lo cal cloud or ring

of H i and dust clouds with velo cities from H CO

which largely coincides with Gould's Belt as outlined

here, shows a similar tilt to the galactic plane, and

shows a similar expanding motion.

A consistentinterpretation of the motions within the

Figure4. Spatial distribution of 9 apparent associations

Gould Belt region is not available yet. The system

within Gould's Belt and their mean position . The

moves away from the galactic center at 5kms ,

1 1

vectors indicate the mean motion and the velocities rela-

it expands at a rate of ab out 12 km s kp c , and it

rotates giving an apparentvalue for the galactic ro-

tive to the mean in a non-rotating coordinate system.

1 1

tation parameter B = 21 km s kp c . However,

there is not an agreement in the values of A and C

from radial velo cities and prop er motions, and the

velo city eld maywell b e non-linear.

of K . These agree very well with those derived by

Selection of stars b elonging to Gould's Belt purely

Westin 1985. Also Mestres 1996 gets a somewhat

on the basis of their spatial distribution should p er-

large negative value for B and a value for K that

haps b e abandoned and the kinematical information

agrees well with ours. However, there are diculties

should also b e used. The plane parallel space velo c-

to get a set of A and C consistent for b oth the prop er

ity comp onents for stars younger than 30 Myr are

motions and the radial velo cities, which is seen also

plotted in Figure 5. Continuous turn of vectors in

in the analysis by Mestres. This may partly b e due to

a stream-likevelo city pattern shows that b esides the

the irregular distribution of stars and stellar groups,

Orion and the Scorpius-Centaurus asso ciations also

but it is also an indication that the velo city eld even

the OB asso ciations in Cygnus should be included

within this small region is non-linear.

into the kinematical analysis see also Ho ogerwerf et

al. 1997. This will be considered in future discus-

sions. The gure con rms the impression of a non-

As p ointed out ab ove, Gould's Belt is dominated by

linear velo city eld.

a few asso ciations, predominantly in the third and

fourth quadrants of galactic longitude. In order to

It has b een suggested Lindblad et al. 1973 that

decrease the errors and get a b etter impression of the

this system was b orn after a compression of a gi-

velo city eld within Gould's Belt wehave identi ed,

ant molecular cloud of some 10 M in a spiral arm

within di erentintervals of galactic longitude, dom-

30 40 Myr ago. After the rst generation of stars

inating groups of stars with rather constant prop er

was formed these stars disp ersed while the interstel-

motion within a limited distance interval. These stel-

lar matter was driven out by sup ernovae and stellar

lar groups and their average space motions pro jected

winds giving birth to later generations of stars in the

on the galactic plane are shown in Figure 4.

pro cess. The stars b orn, however, could not just have

expanded from a small area in the eld of di eren-

tial rotation b ecause this would have resulted in a

A solution for the galactic rotation constants using

value of B = 0, which is certainly contradicted by

Equations 4 to 7 for these nine groups gives the result

the observations. Thus, we suggest that the original

in the last column of Table 2. Figure 4 indicates a

cloud had an angular momentum whichwas to o large

clo ckwise rotation of the system, which is the cause

to make the cloud gravitationally b ound and causing

of the large negativevalue of B . Also an expansion is

the expansion.

seen which causes the p ositive non-zero value of K .

It should b e p ointed out again that the dominating

The rotation of the system might b e an explanation

asso ciations are the Scorpius-Centaurus asso ciation

why it has b een able to keep its atness and appre-

at X = 170 p c, Y = 30 p c and X = 100 p c, Y =

ciable inclination to the galactic plane over several

110 p c and the Orion asso ciation at X = 380 p c,

Y = 180 pc plus an asso ciation of stars at X = times 10 years. This mo del is now b eing tested with

60 p c, Y = 460 p c. numerical N-b o dy calculations.

511

Figure5. Space velocity vectors relative to the Sun projected to the galactic plane for 361 stars younger than 30 Myr.

The arrow in the upper right corner is 20 km s long.

REFERENCES Torra, J., G omez, A.E., Figueras, F., Comer on,

Grenier, S., Mennessier, M.O., Mestres, M. &

Fern andez, D. 1997, ESA SP{402, this volume

Balona, L.A. 1994, MNRAS 268, 119

Westin, T.N.G. 1985, A&AS 60, 99

Balona, L.A., Shobbro ok, R.R. 1984, MNRAS 211,

375

Blaauw, A. 1946, Publ Kapteyn Lab. No.52, 88

Ho ogerwerf, R., Lub, J., de Bruijne, J.H.J., Blaauw,

A., Brown, A.G.A. & de Zeeuw, P.T., 1997, ESA

SP{402, this volume

Lindblad, P.O. 1974, Gould's Belt. In Contop oulos,

G. ed. Highlights of Astronomy,IAU, p. 381

Lindblad, P.O., Grap e, K., Sandqvist, Aa., Schob er,

J. 1973, A&A 24, 309

Maeder, A., Meynet, G. 1988, A&AS 76, 411

Mestres, M. 1996, Lic. Thesis, Univ. Barcelona

Olano, C.A. 1982, A&A 112,195

Plaskett, J.S. 1930, MNRAS 90, 616

Sandqvist, Aa., Tomb oulides, H., Lindblad, P.O.

1988, A&A 205, 225

Shobbro ok, R.R. 1983, MNRAS 205, 1215

Stromgren, B. 1966, ARA&A 4, 433