PG 1115+080 is one of the first case of quadruply imaged quasar (Weymann et al. 1980) and among the best studied lensed quasars, one of the still rare systems with well measured time-delays (Schechter et al. 1997). The tremendous gain in spatial resolution achieved since the eighties (see Fig. 5), has allowed detailed understanding of the system. Since the range of observational data available for PG 1115+080 is so broad, we take this object as an example to show how observations help to pin down the Hubble parameter H0.
PG 1115+080 is a bright quasar at z = 1.722, with a relatively "wide" angular separation between its images, of the order of 2". Two images are isolated (B and C in Fig. 5) and a blend of two brighter images, is located on the other side of the lens in projection on the plane of the sky (A1/A2 on Fig. 5). This blend A1/A2 was not resolved on the discovery images of the object, and PG 1115+080 was subsequently believed to be a triple until more detailed observations revealed it was in fact a quadruple (Young et al. 1981). Applying Refsdal's method (1964) to PG 1115+080 has been a long process. Some of the ingredients necessary to the lens modeling had to wait for years before suitable observations finally became available.
Figure 5. Left: ground based image of PG 1115+080 adapted from Schechter et al (1997), with a field of view of 5" on a side. The data, with a resolution of about 0.8", is of sufficiently good quality for measuring the photometric variations of the quasar images. It is however not good enough to obtain detailed surface photometry of the lensing galaxy. Right: one of the very best images ever obtained of PG 1115+080, with the 8.2m Subaru telescope. The image has a resolution of 0.32", approaching that of the Hubble Space Telescope (Iwamuro et al. 2000), and allowing for precise measurements of the quasar and lens astrometry as well as for detailed surface photometry of the lens.