The Kennicutt-Schmidt relation is an empirical power-law linking the surface density of the star formation rate (Σ_SFR) to the surface density of gas (Σ_gas ) averaged over the observed face of a starforming galaxy Kennicutt (1998). The original presentation used observations of CO to measure gas density and Hα emission to measure the population of hot, massive young stars (and infer the star formation rate). Observations of Σ_SFR from a census of young stellar objects in nearby molecular clouds in our Galaxy are up to 17 times higher than the extragalactic relation would predict given their Σ_gas. These clouds primarily form low-mass stars that are essentially invisible to star formation rate tracers. A sample of six giant molecular cloud (GMC) complexes with signposts of massive star formation was identified in our galaxy. The regions selected have a range of total luminosity and morphology.

Deep ground-based observations in the near-infrared with NEWFIRM and IRAC observations with the Spitzer Space Telescope were used to conduct a census of the young stellar content associated with each of these clouds. The star formation rates from the stellar census in each of these regions was compared with the star formation rates measured by extragalactic star formation rate tracers based on monochromatic mid-infrared luminosities. Far-infrared Herschel observations from 160 through 500 μm were used to determine the column density and temperature in each region. The region NGC 6334 served as a test case to compare the Herschel column density measurements with the measurements for near-infrared extinction.

The combination of the column density maps and the stellar census lets us examine Σ_SFR vs. Σ_gas for the massive GMCs. These regions are consistent with the results for the low-mass molecular clouds, indicating Σ_SFR levels that are higher than predicted based on Σ_gas. The overall Σ_gas levels are higher for the massive star forming regions, indicating that they have a higher fraction of dense gas than the clouds that are forming primarily low mass stars. There is still significant spread at a given average gas density, indicating that the star formation history and dense gas fraction play important roles in determining an individual molecular cloud's place in a Σ_SFR vs. Σ_gas diagram.

Zooming in, Σ_SFR vs. Σ_gas was examined within the individual clouds, revealing a decrease relative to the spread that is observed for the average over whole clouds. The dependence of Σ_SFR on Σ_gas increases significantly above A_V∼5-10 which is consistent with previous measurements of a threshold for star formation around A_V=8 or Σ_gas=0.04 g cm^-2. NGC 6334 was found to be consistent with a threshold for massive star formation at Σ_gas=1 g cm^-2.