Observations and implications of the star formation history of the large magellanic cloud

We present derivations of star formation histories based on color-magnitude diagrams of three fields in the LMC from HST/WFPC2 observations. One held is located in the LMC bar and the other two are in the outer disk. We find that a significant component of stars older than 4 Gyr is required to match the observed color-magnitude diagrams. Models with a dispersion-free age-metallicity relation are unable to reproduce the width of the observed main sequence; models with a range of metallicity at a given age provide a much better fit. Such models allow us to construct complete "population boxes" for the LMC based entirely on color-magnitude diagrams; remarkably, these qualitatively reproduce the age-metallicity relation observed in LMC clusters. We discuss some of the uncertainties in deriving star formation histories by our method and suggest that improvements and confidence in the method will be obtained by independent metallicity determinations. We find, independently of the models, that the LMC bar held has a larger relative component of older stars than the outer fields. The main implications suggested by this study are: (1) The star formation history of field stars appears to differ from the age distribution of clusters. (2) There is no obvious evidence for bursty star formation, but our ability to measure bursts shorter in duration than about 25% of any given age is limited by the statistics of the observed number of stars. (3) There may be some correlation of the star formation rate with the last close passage of the LMC/SMC/Milky Way, but there is no dramatic effect. (4) the derived star formation history is probably consistent with observed abundances, based on recent chemical evolution models.