The question whether the Milky Way disk has experienced a smooth and constant star formation history (hereafter SFH) or a bursty one has been the subject of a number of studies since the initial suggestions by Scalo (1987) and Barry (1988). Rocha-Pinto et al. (2000a; hereafter RPSMF) present a brief review about this question. There is evidence for three extended periods of enhanced star formation in the disk. The use of the word `burst' for these features (usually lasting 1-3 Gyr) is based on the fact that all methods used to recover the SFH are likely to smear out the original data so that the star formation enhancement features could be narrower than they seem, or be composed by a succession of smaller bursts. In this sense, they were named bursts A, B and C, after Majewski (1993).
The most efficient way to find the SFH is using the stellar age distribution, which can be transformed into a star formation history after various corrections. Twarog (1980) summarized some of these steps. Although his SFH is usually quoted as an evidence for the constancy of the star formation in the disk, he states that during the most recent 4 Gyr, the SFH has been more or less constant, followed by a sharp increase from 4 to 8 Gyr ago, and a slow decline beyond that. His unsmoothed data were also reanalysed by Noh & Scalo (1990) who have found more signs of irregularity.
Barry (1988) has improved this situation substantially by using chromospheric ages. His conclusion was criticized by Soderblom et al. (1991), who showed that the empirical data would be still consistent with a constant SFH if the chromospheric emission-age relation is suitably modified. However, Rocha-Pinto & Maciel (1998) have recently argued that the scatter in Soderblom et al. (1991)'s Fig. 13, which is the main feature that could suggest a non-monotonic age calibration, is probably caused by contamination in the photometric indices due to the chromospheric activity. The chromospheric activity-age relation was also further investigated by Donahue (1993, 1998), and the new proposed calibration still predicts a non-constant SFH if applied to Barry's data.
The SFH derived in this paper is based on a new chromospheric sample compiled by us (Rocha-Pinto et al. 2000b, hereafter Paper I). This paper is organized as follows: In Sect. 2, we address the transformation of the age distribution into SFH. The results are presented in Sect. 3. In Sect. 4, statistical significances for the SFH are provided by means of a number of simulations. The impact of the age errors on the recovered SFH is also studied. Some comparisions with observational constraints are addressed in Sect. 5, and each particular feature of the SFH is discussed in Sect. 6, in view of the results from the simulations and comparisons with other data. The case for a non-monotonic chromospheric activity-age relation is discussed in Sect. 7. Our final conclusions follow in Sect. 8. A summary of this work was presented in RPSMF.
© European Southern Observatory (ESO) 2000
Online publication: June 20, 2000