On the baseline flux determination of microlensing events detectable with the difference image analysis method

To improve photometric precision by removing the blending effect, a newly developed technique of difference image analysis (DIA) has been adopted by several gravitational microlensing experiment groups. However, the principal problem of the DIA method is that, by its nature, it has difficulties in measuring the baseline flux F0 of a source star, causing a degeneracy problem in determining the lensing parameters of an event. Therefore, it is often believed that the DIA method is not as powerful as the classical method based on PSF photometry for determining the Einstein time-scales tE of events. In this paper, we demonstrate that the degeneracy problem in microlensing events, detectable from searches using the DIA method, is not as serious as is often thought. This is because a substantial fraction of events will be high amplification events for which the deviations of the amplification curves, constructed with the wrong baseline fluxes from their corresponding best-fit standard amplification curves, will be considerable, even for a small amount of the fractional baseline flux deviation ΔF0F0. With a model luminosity function of source stars and under realistic observational conditions, we find that ∼30 per cent of detectable Galactic bulge events are expected to have high amplifications and their baseline fluxes can be determined with uncertainties ΔF0F0≤0.5.