We use the high magnification event seen in the 1999 OGLE campaign light curve of image C of the quadruply imaged gravitational lens Q2237+0305 to study the structure of the quasar engine. We have obtained g- and r-band photometry at the Apache Point Observatory 3.5m telescope where we find that the event has a smaller amplitude in the r-band than in the g- and OGLE V-bands. By comparing the light curves with microlensing simulations we obtain constraints on the sizes of the quasar regions contributing to the g- and r-band flux. Assuming that most of the surface mass density in the central kiloparsec of the lensing galaxy is due to stars and by modeling the source with a Gaussian profile, we obtain for the Gaussian width 1.20 x 10^15 sqrt(M/0.1M_sun)cm < sigma_g < 7.96 x 10^15 sqrt(M/0.1Msun) cm, where M is the mean microlensing mass, and a ratio sigma_r/sigma_g=1.25^{+0.45}_{-0.15}. With the limits on the velocity of the lensing galaxy from Gil-Merino et al. (2005) as our only prior, we obtain 0.60 x 10^15 sqrt(M/0.1Msun) cm < sigma_g < 1.57 x 10^15 sqrt(M/0.1Msun) cm and a ratio sigma_r/sigma_g=1.45^{+0.90}_{-0.25} (all values at 68 percent confidence). Additionally, from our microlensing simulations we find that, during the chromatic microlensing event observed, the continuum emitting region of the quasar crossed a caustic at >72 percent confidence.
Download