Here we report observations of the two lowest inversion transitions of ammonia with the 70-m Tidbinbilla radio telescope. They were conducted to determine the kinetic temperatures in the dense clumps of the G333 giant molecular cloud associated with
RCW 106 and to examine the effect that accurate temperatures have on the calculation of derived quantities such as mass. This project is part of a larger investigation to understand the timescales and evolutionary sequence associated with high-mass star formation, particularly its earliest stages. Assuming that the initial chemical composition of a giant molecular cloud is uniform, any abundance variations within will be due to evolutionary state. We have identified 63 clumps using SIMBA 1.2-mm dust continuum maps and have calculated gas temperatures for most (78 per cent) of these dense clumps. After using Spitzer GLIMPSE 8.0 $mu$m emission to separate the sample into IR-bright and IR-faint clumps, we use statistical tests to examine whether our classification shows different populations in terms of mass and temperature. We find that clump mass and column density show no significant population difference, and that kinetic temperature is the best parameter to distinguish between the gravitationally bound state of each clump. The kinetic temperature was the only parameter found to have a significantly low probability of being drawn from the same population. This suggests that clump radii does not have a large effect on the temperature of a clump, so clumps of similar radii may have different internal heating mechanisms. We also find that while the IR-bright clumps have a higher median virial mass, both samples have a similar range for both virial mass and FWHM. There are 87 per cent (40 of 46) of the clumps with masses larger than the virial mass, suggesting that they will form stars or are already undergoing star formation.
