AB Doradus (AB Dor) is the nearest identified moving group. As with other such groups, the age is important for understanding of several key questions. It is important, for example, in establishing the origin of the group and also in comparative studies of the properties of planetary systems, eventually surrounding some of the AB Dor group members, with those existing in other groups. For AB Dor two rather different estimates for its age have been proposed: the first one, of the order of 50 Myr, by Zuckerman and coworkers from a comparison with the Tucana/Horologium moving group and a second one of about 100–125 Myr by Luhman and coworkers from colour–magnitude diagrams. Using this last value and the closeness in velocity space of AB Dor and the Pleiades galactic cluster, Luhman and coworkers suggested coevality for these systems. Because strictly speaking such a closeness does not still guarantee coevality, here we address this problem by computing and comparing the full 3D orbits of AB Dor, Pleiades, α Persei and IC 2602. The latter two open clusters have estimated ages of about 85–90 and 50 Myr. The resulting age 119 ± 20 Myr is consistent with AB Dor and Pleiades being coeval. Our solution and the scenario of open cluster formation proposed by Kroupa and collaborators suggest that the AB Dor moving group may be identified with the expanding subpopulation (Group I) present in this scenario. We also discuss other related aspects as iron and lithium abundances, eventual stellar mass segregation during the formation of the systems and possible fraction of debris discs in the AB Dor group.
In a series of publications, Eggen introduced and discussed the concept of superclusters defined as stellar aggregates in the solar neighbourhood which are characterized by parallel space motions of their members. One of these superclusters is the Local Association (Eggen 1975, 1983a,b) located within a radius of a few hundred parsecs around the Sun and often referred to as the Pleiades supercluster because of the similarity of the space motions of its members and the Pleiades open cluster. Apart from Pleiades, other open star clusters as α Persei, IC 2602, δ Lyrae, NGC 1039, NGC 2516 and also the Scorpio–Centaurus (Sco–Cen) OB association would be contained in the Local Association.
With recent identifications of relatively small nearby groups or associations of post T-Tauri stars with ages from 8 to 50 Myr (Zuckerman & Song 2004b; Torres et al. 2006), located within 100–150 pc of the Sun and with previous knowledge of the Sco–Cen complex (de Zeeuw et al. 1999; Mamajek et al. 2002), it can be said that a new stage of investigation of the structure of the Local Association has begun. We are now even in a position to suggest plausible formation scenarios for some of those groups. For instance, the young kinematic groups TW Hya (Torres et al. 2003; Zuckerman & Song 2004b) with an age of 8.3 Myr (de la Reza, Jilinski & Ortega 2006), β Pictoris (Zuckerman, Song & Webb 2001; Torres et al. 2006) with an age of 11.3 Myr (Ortega et al. 2002, 2004) and η and ε Chamaeleontis groups (Mamajek et al. 2000; Feigelson et al. 2003) with an age of 6.7 Myr (Jilinski, Ortega & de la Reza 2005) may be the result of a sequence of bursts of low-mass star formation related to the older Sco–Cen subgroups with an age of about 18 Myr (Ortega et al. 2006).
In addition to these, two other groups have been identified within 50 pc of the Sun: the Tucana/Horologium (Torres et al. 2000; Zuckerman & Webb 2000) and AB Doradus (AB Dor) (Zuckerman, Song & Bessell 2004a) moving groups. For the AB Dor group, comoving with the star AB Dor only 15 pc from the Sun, Zuckerman et al. (2004a) derived an age of about 50 Myr on the basis of a comparison with the Tucana/Horologium group (Torres et al. 2000; Zuckerman & Webb 2000; Zuckerman & Song 2004b) and from the location of AB Dor M spectral-type stars in the colour–magnitude diagram. However, Luhman et al. (2005) analysing the discrepancy between evolution models for very low mass stars and direct mass measurements of the brown dwarf C component of the triple AB Dor system (Close et al. 2005; Marois et al. 2005; Nielsen et al. 2005), have proposed that the AB Dor moving group may have larger age in the 100–125 Myr range. Moreover, by noting that the mean space motion of the AB Dor group is similar to Pleiades galactic cluster's mean motion, Luhman et al. (2005) suggested that the former is probably coeval with the latter. In this context, it is interesting that Innis, Thompson & Coates (1986) had proposed that the rapid rotators AB Dor and PZ Tel may be part of the group of fast rotating K stars of the Pleiades cluster, what would characterize both of them as young systems. As for PZ Tel, we now know that this star is a member of the β Pictoris moving group (Zuckerman et al. 2001; Torres et al. 2006) for which we have a fairly accurate age estimate. In contrast, for AB Dor we have two conflicting age values.
To resolve this discrepancy and derive an age estimate for the AB Dor group here we tackle the problem by calculating the Galactic orbits of both the Pleiades cluster and the AB Dor group. We also compare the orbits of AB Dor with those of the α Persei and IC 2602 clusters, which are considered to be younger than Pleiades (Randich et al. 2001a; Randich 2001b; Makarov 2006) with estimated ages of ∼50 Myr comparable to the age of the AB Dor moving group as given by Zuckerman et al. (2004a). For the α Persei cluster Stauffer et al. (1999) and Barrado y Navascués et al. (2004) have estimated ages of 90 ± 10 and 85 ± 20 Myr, respectively, on the basis of lithium depletion.
2 THE ORBITS
As mentioned above, one of the main reasons given by Luhman et al. (2005) for coevality of AB Dor with the Pleiades cluster was the similar space motions of these groups. Nevertheless, this similarity is not a sufficient condition for coevality. The full 3D path for each system has to be calculated.
As in previous work (e.g. Ortega et al. 2002), our method consists in integrating back in time the 3D orbits of the stars and following the evolution to check for the existence of a first minimum of the configuration. When such a minimum exists the orbits are confined to a certain region which is considered their place of birth. Equivalently, when a comparison of the orbits of the centres of two stellar systems is made, the minimum distance gives the radius of the region occupied by those systems. The dynamical age is then determined by the time elapsed since orbital convergence. The resulting age is called dynamical because the orbits are calculated under a (modelled) Galactic potential.
In the case of clusters it is advisable to work with mean values of the distances and velocities instead of using the whole sample of members. This greatly minimizes the errors in the input. In this work, we take the input values for the Pleiades cluster from Luhman et al. (2005). For the AB Dor group, we derive such values using the Hipparcos data for the members proposed by Zuckerman et al. (2004a) and more recent published radial velocities. Finally, the input data for α Persei and IC 2602 were taken from Robichon et al. (1999).
3 DYNAMICAL EVOLUTION RESULTS
The resulting orbits (Figs 1, 2, 3 and 4) are drawn in the Local Standard of Rest (LSR) frame of reference with cartesian axes X, Y and Z positive in the directions of the Galactic Centre, the Galactic rotation and the north Galactic pole, respectively. The evolution of AB Dor with the α Persei and IC 2602 clusters does not attain any minimum. As seen from Figs 3 and 4 the paths of these systems and that of AB Dor group are quite different in both projection planes. Conversely, the calculation for the AB Dor and Pleiades cluster shows (Figs 1 and 2) that a minimum was reached at the age of −119 ± 20 Myr. In Fig. 5, we show the distance between the centres of Pleiades and AB Dor as a function of time. In Table 1, we quote some values of the relative velocity between the centres of these systems. As can be seen from this table, the relative velocity is negligible near the estimated epoch (119 Myr) of Pleiades formation, showing that here we do not have a simple crossing of orbits. The minimum separation of the AB Dor and the Pleiades cluster determines a region of 43 ± 11 pc of radius. As in previous work (Jilinski et al. 2005) the errors in the age and also in the radius were estimated through Monte Carlo tests. Tests with 1000 realizations were performed. Because the dynamical age was computed differentially here, possible systematic errors caused by uncertainties in the potential will mutually compensate each other. Hence the resulting errors will be determined by the uncertainties in the initial data.
|Age||V (pc Myr−1)|
|Age||V (pc Myr−1)|
The region at minimum is quite large. The size of the region of maximum orbits approximation is probably connected with an expansion of the AB Dor group during the formation process. The loss of rotational support of the system resulting from the expansion is reflected in an increased U-velocity, that is, in the velocity component in the direction of the Galactic Centre. However, the orbits of both groups are readily spread out along the positive Y-direction by the action of the Galactic rotation. The groups are moving towards the Galactic plane along the Z-direction.
A scenario for the formation of the Pleiades star cluster involving the generation of an expanding population has been inferred by Kroupa et al. (2001) from their N-body calculations. If this applies also to our case we can propose that the AB Dor moving group might be the expanding population (group I), or at least part thereof, present in that scenario. Our results give a range (100−140 Myr) for the age, somewhat different from that found in Luhman et al. (2005) but not conflicting with their conclusions concerning the age of the AB Dor group.
4 DISCUSSION AND CONCLUSIONS
In this work, we have considered the question of the age of the AB Dor moving group from the dynamical point of view. As mentioned in the Introduction section, two quite different estimates for the age of this group have been given: ∼50 Myr (Zuckerman et al. 2004a) and ∼125 Myr (Luhman et al. 2005), which is also the age of the Pleiades open cluster. Here we have approached this problem by comparing the 3D orbits of the centres of AB Dor, and of the open clusters Pleiades, α Persei and IC 2602. The results of this analysis show that the AB Dor moving group and the Pleiades cluster are coeval having formed some 119 ± 20 Myr ago. As for Pleiades, the dynamical age is compatible with the evolutionary age determinations for this system. For instance, the upper main-sequence turn-off point gives an age of 100 Myr (e.g. Mermilliod 1981; Meynet, Mermilliod & Maeder 1993) while somewhat more recent determinations using the lithium boundary method yield an age of 125 Myr (Stauffer et al. 1998, 1999; Barrado y Navascués et al. 1999, 2004). As for the AB Dor group, our results are consistent with Luhman et al. (2005) who concluded that an age greater than 50 Myr, of the order of the Pleiades age, would help eliminate the discrepancy between models and measurements of the mass of AB Dor C star. It should be remarked, however, that according to Nielsen et al. (2005) a greater age will be only part of the solution because of problems related to the effective temperature of that brown dwarf.
The size of the birth region of Pleiades and AB Dor moving group yielded by our dynamical solution could, in principle, be explained within the framework of the open cluster formation scenario put forward by Kroupa et al. (2001). In this scenario a bound and an unbound system are born together. We suggest that the unbound system expanding at formation could be identified with the AB Dor group. This overall picture is also in agreement with conclusions of Luhman et al. (2005) who suggested that the AB Dor group could be a remnant of an OB association related to the formation of the Pleiades cluster.
Apart from these dynamical characteristics, the Pleiades and AB Dor group also share some other features that can provide independent supporting evidence of their common origin. For example, this appears to be the case with the iron abundance. In fact, 45 spectroscopic high-resolution determinations of the metallicity (Cayrel de Strobel, Soubiran & Ralite 2001) give a mean value of 0.00 ± 0.01 for Pleiades, whereas for AB Dor a metallicity of −0.02 ± 0.02 was obtained as an average from high-resolution spectroscopic measurements in a sample of 15 member stars, including AB Dor itself (see Table 2). Comparisons of lithium abundance are also interesting and may provide additional support for Pleiades and AB Dor coevality. In this context, da Silva et al. (2006) have shown that lithium abundances data indicate a depletion of this element between 11 Myr and the main sequence. This was done by considering the members of groups β Pictoris, Tucana/Horologium, AB Dor and of the Pleiades cluster. For all stellar temperatures, the lithium abundance of β Pictoris members is larger if compared with those of Tucana/Horologium, whereas stars of AB Dor and Pleiades have similar lithium distributions.
Some possible consequences of the suggested common origin of the AB Dor group and the Pleiades cluster deserve special attention. Two particulary interesting ones are the stellar mass distribution and the fractional distribution of candidates of debris discs in both stellar systems. Concerning the mass distribution, it is interesting to note the apparent absence of A-type stars with masses about 2.5 M⊙ (and also B-type stars) in the proposed list of members of the AB Dor group by Zuckerman et al. (2004a). This, in fact, is notorious if we consider that younger associations such as TW Hya and β Pictoris do contain A-type stars. If this is not an observational selection effect, we can speculate that the absence of A-type stars in the AB Dor group could be related to the fact that they are present in Pleiades. An efficient mass-segregation mechanism should then have been acting during the initial separation phase of these groups.
As for debris discs, a recent 24-micron Spitzer survey (emission on these wavelengths is produced in the planetary internal zone between 1 and 20 au from the central star) in the Pleiades by Gorlova et al. (2006) has shown that 25 per cent of B–A type members and 10 per cent of F–F3 type stars are debris disc candidates due to their excess infrared emission at an age around 100 Myr. It would then be interesting to conduct a similar survey at 24 micron in the AB Dor group and compare its fraction of debris disc candidates with that in Pleiades.
EJ thanks MCT of Brazil for the financial support under the contract 384222/2006−4.