THE STELLAR POPULATIONS OF DEEPLY EMBEDDED YOUNG CLUSTERS: NEAR-INFRARED SPECTROSCOPY AND EMERGENT MASS DISTRIBUTIONS

Abstract

ABSTRACT The goal of this thesis is to test the following hypothesis: the initial distribution of stellar masses from a single "episode" of star formation is independent of the local physical conditions of the region. In other words, is the initial mass function (IMF) strictly universal over spatial scales d < 1 \\ pc and over time intervals Delta-tau << 3 x 106 yrs? We discuss the utility of embedded clusters in addressing this question. Using a combination of spectroscopic and photometric techniques, we seek to characterize emergent mass distributions of embedded clusters in order to compare them both with each other and with the field star IMF. Medium resolution (R=1000) near-infrared spectra obtainable with the current generation of NIR grating spectrographs can provide estimates of the photospheric temperatures of optically-invisible stars. Deriving these spectral types requires a three--step process; i) setting up a classification scheme based on near-infrared spectra of spectral standards; ii) understanding the effects of accretion on this classification scheme by studying optically-visible young stellar objects; and iii) applying this classification technique to the deeply embedded clusters. Combining near-infrared photometry with spectral types, accurate stellar luminosities can be derived for heavily reddened young stars thus enabling their placement in the H-R diagram. From their position in the H-R diagram, masses and ages of stars can be estimated from comparison with theoretical pre-main sequence evolutionary models. Because it is not practical to obtain complete spectroscopic samples of embedded cluster members, a technique is developed based solely on near-IR photometry for estimating stellar luminosities from flux--limited surveys. We then describe how spectroscopic surveys of deeply embedded clusters are necessary in order to adopt appropriate mass-luminosity relationships. Stellar luminosity functions constructed from complete extinction-limited samples can then be used to characterize emergent mass distributions of deeply embedded young clusters. Because of systematic uncertainties in these models at the low-mass end, we adopt the ratio of intermediate (10 solar mass > M* > 1.0 solar mass) to low-mass (1.0 solar mass > M * > 0.1 solar mass) stars in order to compare these mass distributions to the Miller-Scalo IMF. As an example of this analysis we present a study of the embedded cluster associated with the NGC2024 nebula. Although this cluster contains an enhanced number of intermediate mass stars, we cannot distinguish the distribution of stellar masses from the field star IMF. A detailed comparison between the stellar luminosity functions of the embedded clusters associated with the NGC2024 cluster and the embedded population found in the Ophiuchus cloud cores suggests that it is unlikely they were drawn from the same parent population. After finding the evolutionary states and accretion properties of both clusters to be similar, we interpret the difference in stellar luminosity functions as a difference in their emergent mass distributions. Synthesizing results for NGC 2024 and Ophiuchus with those from other studies of embedded clusters, we arrive at the following conclusions: i) the emergent mass distributions of most of the embedded young clusters considered are consistent with having been drawn from the Miller--Scalo IMF; and ii) there is a hint that regions of high central stellar density contain a greater proportion of intermediate mass stars.