Revealing stellar brightness profiles by means of microlensing fold caustics

With a handful of measurements of limb-darkening coefficients, galactic microlensing has already proven to be a powerful technique for studying atmospheres of distant stars. Survey campaigns such as OGLE-III are capable of providing 10 suitable target stars per year that undergo microlensing events involving passages over the caustic created by a binary lens, which last from a few hours to a few days and allow us to resolve the stellar atmosphere by frequent broad-band photometry. For a caustic exit lasting 12 h and a photometric precision of 1.5 per cent, a moderate sampling interval of 30 min (corresponding to similar to25-30 data points) is sufficient for providing a reliable measurement of the linear limb-darkening coefficient with an uncertainty of 8 per cent, which reduces to 3 per cent for a reduced sampling interval of 6 min for the surroundings of the end of the caustic exit. While some additional points over the remaining parts of the light curve are highly valuable, a denser sampling in these regions provides little improvement. Unless an accuracy of less than 5 per cent is desired, limb-darkening coefficients for several filters can be obtained or observing time can be spent on other targets during the same night. The adoption of an inappropriate stellar brightness profile as well as the effect of acceleration between source and caustic yield distinguishable characteristic systematics in the model residuals. Acceleration effects are unlikely to affect the light curve significantly for most events, although a free acceleration parameter blurs the limb-darkening measurement if the passage duration cannot be accurately determined.