Separating intrinsic and microlensing variability using parallax measurements

In gravitational lens systems with 3 or more resolved images of a quasar, the intrinsic variability may be unambiguously separated from the microlensing variability through parallax measurements from 3 observers when there is no relative motion of the lens masses (Refsdal 1993). In systems with fewer than 3 resolved images, however, this separation is not straightforward. A general approach that may be used for this purpose is presented. For simplicity, only the one-dimensional case is considered in detail: Given a well-sampled time series of the observed flux at two points in space with a known separation, choosing a velocity $v_{\perp}$ of the observers perpendicular to the line of sight determines the microlensing magnification history, and thereby also the intrinsic variability. The velocity is chosen by minimizing some measure ($\chi^2$) of the residual intrinsic variability. In many cases this gives a close approximation to the true magnification. In cases where the relative motion of the lensing point masses is important, only a partial separation will be possible.