Transcript preliminary version - University of Exeter

Do accretion discs regulate the momentum of young
stars?
S.P. Littlefair1,2, T. Naylor2, R.D. Jeffries3, B. Burningham2, E. Saunders2
1Dept
of Physics & Astronomy, University of Sheffield, 2School of Physics, University of Exeter, 3School of Chemistry and Physics, Keele University
Rotational evolution – high mass stars: Figure 3 shows the
Abstract:
The question of the early evolution of stellar angular momentum is an interesting and essentially
unsolved problem. It is well known that young stars rotate well below their break-up speed[1], and this
has been explained by disc-locking theories, where magnetic field lines connect the star to the disc,
forcing synchronous rotation between the star and disc material at some radius. However,
observational and theoretical support for disc locking is still controversial. In this poster we present
new observations of the rotation distributions in the young associations IC 348 and Cepheus OB3b,
and review the observational support for disc locking.
Periods for CTTs from photometric monitoring:
The most efficient method of determining large numbers of periods for young stars is via photometric
monitoring. Unfortunately, this method has proved insensitive to periods amongst CTTs in the past,
largely due to the irregular variability shown by CTTs[2,3]. This bias against CTTs is a major problem, as
an obvious test of disc locking is to compare the period distribution of CTTs against WTTs. We have
overcome this bias by using a more intensive monitoring strategy than previous studies. For example, in
IC 348 we took data on 15 consecutive nights, and repeated the observations many times within a
single night. By contrast, the survey of Cohen et al. 2004[2] has much lower temporal density, and many
nights can elapse between observations. Because CTTs are erratically variable with time-scales of order
one night, this variability masks the periodic signal. Our data is much less vulnerable to this effect: the
fraction of CTTs to WTTs amongst the periodic stars in our survey is 0.43±0.13, not significantly different
from the cluster as a whole.
period distributions of IC 348 and Cepheus OB3b, along with the distributions for the ONC and
NGC 2264, taken from the literature. Figure 2 shows that Cep OB3b is slightly older than NGC
2264, with a likely age of 4 Myr. With this in mind, it appears that the ONC, NGC 2264 and
ONC
Cep OB3b imply an evolutionary sequence in
which stars are initially disc-locked at 8-day
periods (the ONC) are released at ~ 1 Myr and
gradually spin up, to 2 day periods at ~4 Myr
(Cep OB3b).
A neat evolutionary sequence is spoiled by IC
348, however. Like the ONC, the high mass stars
in IC 348 are rotating slowly, with a peak around
8 days. In fact, the stars in IC 348 are rotating
more slowly than the stars in the ONC, with a
significant absence of rotators below 2 days. The
literature age of IC 348 is circa 2 Myr, similar to
that of NGC 2264. There are two possible
conclusions to be drawn, the most obvious being
that cluster environment can dominate over disc
locking in determining rotation rates. A more
radical suggestion is that the literature age for IC
348 is not correct. The cluster
M > 0.25 M
M < 0.25 M
2 Myr
IC348
M > 0.25 M
M < 0.25 M
Figure 2: De-reddened colour-magnitude
diagrams of the periodic variables in
Cep OB3b and NGC 2264. Isochrones
are from Siess et al 2000[ref]. Credit:
Nathan Mayne, Exeter
photometric sequence in IC 348 is poorly fit by models[7], and there is some evidence that the age might
be nearer 1 Myr[8], in which case it is not surprising that the period distribution is ONC-like.
Correlation of period with disc indicators:
2 Myr
NGC 2264
The low mass stars: The low mass stars present a very confusing picture. It is not
The most obvious test of disc-locking theory is to look for
correlations between rotation period and indicators of disc
presence or accretion. Obvious candidates are infrared
excess or Hα equivalent width. A correlation between rotation
rate and infrared excess in the ONC has previously been
claimed by Herbst et al(2002). However, both their chosen
disc indicator (I-K excess) and rotation rate are strongly
correlated with mass[3,4,5,6]. The observed correlation between
rotation and I-K excess is most likely a secondary correlation,
arising from the primary, underlying correlation between
I-K excess and mass. Likewise, we must be cautious when
interpreting a lack of correlation – figure 1 shows that our IC
348 data shows no correlation between K-L colour or Hα EW.
However, disc holes or inclination effects might hide the
correlation with L-band excess, and there is no reason why
the current Hα EW should represent the time averaged
accretion rate, as the emission strength of Hα is highly
variable.
0.5 Myr
R-I < 1.3
R-I > 1.3
clear why they show a uni-modal distribution, nor why they are rotating more rapidly than the high mass
stars (but see Barnes 2003[9]). Furthermore, the low mass stars do not fit into the evolutionary sequence
above: the distributions in NGC 2264 and IC 348 are very similar, despite the very different high-mass
distributions in these clusters. Worse still, the low mass stars in Cep OB3b (the oldest association
shown here) are rotating much more slowly than those in NGC 2264. Much more additional work is
needed on the low-mass stars, including deeper variability surveys.
Conclusions: The observational evidence for disc locking is mixed. The lack of any clear
correlation with disc or accretion indicators is worrying, but time variability of Hα, variation in disc-holes
or inclination and mass effects make firm conclusions difficult. Likewise, the period distributions do not
fit a clear evolutionary sequence, although uncertainty in cluster ages may explain this. Future work
should include investigating the link between rotation rate and firm disc indicators, such as Spitzer
colours, and investigation into alternative, model independent methods of determining cluster ages.
Finally, the puzzle of the low-mass stars needs further observational and theoretical study.
Figure 1: Rotation period .vs.
disc and accretion indicators
for IC 348. No correlation is
present.
References:
[1] Bouvier et al, 1993, A&A, 272, 176
[2] Cohen et al, 2004, AJ, 127, 1602
[3] Herbst et al, 2002, PASP, 114, 1167
[4] Hillenbrand et al, 1998, AJ, 116, 1816
[5] Lamm et al, 2005, A&A, 430, 1005
[6] Littlefair et al, 2005, MNRAS, 358, 341
[7] Mayne et al, in preparation
[8] Herbig, 1998, ApJ, 497, 736
[9] Barnes, 2003, ApJ, 586, 464
4 Myr
Cep OB3b
V-I < 3.7
V-I > 3.7
Figure 3: Period distributions of IC 348 and Cepheus OB3b compared
to distributions for the ONC[3] and NGC 2264[5]. Also shown are ages
for the clusters commonly quoted in the literature.