Uncertainties in stellar ages provided by grid techniques

The determination of the age of single stars by means of grid-based techniques is a well established method. We discuss the impact on these estimates of the uncertainties in several ingredients routinely adopted in stellar computations. The systematic bias on age determination caused by varying the assumed initial helium abundance, the mixing-length and convective core overshooting parameters, and the microscopic diffusion are quantified and compared with the statistical error owing to the current uncertainty in the observations. The typical uncertainty in the observations accounts for 1ω statistical relative error in age determination ranging on average from about –35 % to +42 %, depending on the mass. However, the age's relative error strongly depends on the evolutionary phase and can be higher than 120 % for stars near the zero-age main-sequence, while it is typically about 20 % or lower in the advanced main-sequence phase. A variation of ±1 in the helium-to-metal enrichment ratio induces a quite modest systematic bias on age estimates. The maximum bias due to the presence of the convective core overshooting is –7% for β = 0.2 and –13% for β = 0.4. The main sources of bias are the uncertainty in the mixing-length value and the neglect of microscopic diffusion, which account each for a bias comparable to the random error uncertainty.