A submillimetre survey of the kinematics of the Perseus molecular cloud - III. Clump kinematics

We explore the kinematic properties of dense continuum clumps in the Perseus molecular cloud, derived from our wide-field (CO)-O-18 J = 3 -> 2 data across four regions - NGC 1333, IC 348/HH 211, L1448 and L1455. Two distinct populations are examined, identified using the automated algorithms clfind (85 clumps) and gaussclumps (122 clumps) on existing SCUBA 850-mu m data. These kinematic signatures are compared to the clumps' dust continuum properties. We calculate each clump's non-thermal linewidth and virial mass from the associated (CO)-O-18 J = 3 -> 2 spectrum. The clumps have supersonic linewidths, <<Sigma(NT)/c(s)>> = 1.76 +/- 0.09 (clfind population) and 1.71 +/- 0.05 (with gaussclumps). The linewidth distributions suggest the (CO)-O-18 line probes a lower density 'envelope' rather than a dense inner core. Similar linewidth distributions for protostellar and starless clumps imply protostars do not have a significant impact on their immediate environment. The proximity to an active young stellar cluster seems to affect the linewidths: those in NGC 1333 are greater than elsewhere. In IC 348 the proximity to the old infrared cluster has little influence, with the linewidths being the smallest of all. The virial analysis suggests that the clumps are bound and close to equipartition, with virial masses similar to the masses derived from the continuum emission. In particular, the starless clumps occupy the same parameter space as the protostars, suggesting they are true stellar precursors and will go on to form stars. We also search for ordered (CO)-O-18 velocity gradients across the face of each core. Approximately one-third have significant detections, which we mainly interpret in terms of rotation. However, we note a correlation between the directions of the identified gradients and outflows across the protostars, indicating we may not have a purely rotational signature. The fitted gradients are in the range G =1 to 16 km s(-1) pc(-1), larger than found in previous work, probably as a result of the higher resolution of our data and/or outflow contamination. These gradients, if interpreted solely in terms of rotation, suggest that the rotation is not dynamically significant: the ratios of clump rotational to gravitational energy are beta(rot) less than or similar to 0.02. Furthermore, derived specific angular momenta are smaller than observed in previous studies, centred around j similar to 10(-3) km s(-1) pc, which indicates we have identified lower levels of rotation, or that the (CO)-O-18 J = 3 -> 2 line probes conditions significantly denser and/or colder than n similar to 10(5) cm(-3) and T similar to 10 K.