A CO J = 2-]1 STUDY OF THE OUTFLOW SOURCES GL-490, GL-2591, M8E-IR, AND W3-IRS-5

We have mapped, at 21" resolution, the CO J = 2 --> 1 emission from the high-velocity sources GL 490, GL 2591, M8E-IR, and W3 IRS 5. New bipolar outflows are detected from GL 2591, M8E-IR, and W3 IRS 5. The new W3 IRS 5 bipolar outflow (with a lobe separation of 30") has the same orientation as a very small-scale bipolar outflow (lobe separation of 2".5) previously found by aperture synthesis, suggesting two episodes of outflow activity from IRS 5. Observations of (CO)-C-13 J = 2 --> 1 are combined with the (CO)-C-12 data to determine the mass and other physical properties of the outflowing gas. We find the mass of the outflowing gas to be 9 M. for GL 490, 5 M. for GL 2591, 0.5 M. for M8E-IR, and 23 M. for W3 IRS 5. The 0.5 M. for M8E-IR will be an underestimate because, in the absence of (CO)-C-13 data, we assume the outflowing gas to be optically thin. Dynamical time scales of the four outflows are similar, varying from almost-equal-to 10,000 to almost-equal-to 20,000 yr. Mass-loss rates, averaged over the dynamical lifetime of the activity, range from almost-equal-to 10(-3) M. yr-1 for W3 IRS 5 to almost-equal-to 5 x 10(-5) M. yr 1 for M8E-IR. A comparison is made between the momentum and mass-loss rates of CO outflows previously found from GL 2591, M8E-IR, and W3 IRS 5 using the fundamental vibrational band of CO in the infrared, and the momentum and mass-loss rates obtained in this paper from CO emission line data. We conclude from the comparison that a suggested direct causal link between the infrared outflows and the extended bipolar flows remains feasible, but is, as yet, unproven.