Gravitationally lensed compact radio sources have many astrophysical and cosmological applications. The foremost being the determination of a time-delay between the individual lensed images in order to constrain the Hubble parameter (e.g. Refsdal 1964). Considerable progress has been made during the last few years in measuring time-delays, both through optical and radio observations (e.g Kundic et al. 1997; Schechter et al. 1997; Lovell et al. 1998; Biggs et al. 1999; Fassnacht et al. 1999; Koopmans et al. 2000). They also allow a detailed study of the mass distribution of the lens galaxy and sometimes the background source, through a large magnification by the lensing potential. Absorption lines in the spectrum of the background source allow the study of the ISM in the lens galaxy and the HI distribution along the lines of sight to the source.
Temporal changes in the brightnesses or spectra of the lensed images also allow the study of uncorrelated external variability. The most important sources of external variability are scintillation at radio wavelengths and microlensing in all wavelength bands. Differencing the image light curves, taking the proper time delay into account, removes intrinsic source variability and leaves only uncorrelated external variability. These difference light curves thus provide valuable information on the compact objects in the lens galaxy and/or on the intervening ionized medium (e.g. Lehár et al. 1992; Haarsma et al. 1997; Schmidt & Wambsganss 1998).
The study of the ionized component of the Galactic interstellar medium (ISM) through scattering of radio emission from pulsars has had a long tradition (e.g. Rickett 1977, 1990). Scattering by the ionized ISM can explain long-term variability at meter wavelengths (e.g. Condon et al. 1979), as well as large-amplitude variability in very compact extra-galactic radio sources (e.g. Rickett, Coles & Bourgois 1984). Low-amplitude variability at shorter wavelengths (about 10 cm), called `flickering', has been observed by Heeschen (1982, 1984) and is probably associated with refractive interstellar scattering of an extended source (e.g. Rickett et al. 1984). Strong intra-day variability of very compact radio sources might result from refractive interstellar scattering as well (e.g. Wagner & Witzel 1995). A power-law model of the plasma-density power spectrum (e.g. Rickett 1977, 1990), combined with some distribution of this plasma in our galaxy (e.g. Taylor & Cordes 1993 [TC93]) is able to explain most of the observed dispersion measures and variability in pulsars at low frequencies, as well as the variability of extra-galactic radio sources at both low and high frequencies. However, especially for compact flat-spectrum radio sources it remains exceedingly difficult to separate intrinsic variability from scintillation by the Galactic ionized ISM.
Gravitationally lensed (i.e. multiply-imaged) flat-spectrum compact radio sources could offer a solution to this problem. As mentioned previously, these systems provide two or more lines-of-sight through the Galactic ionized ISM. For typical image separations of a few arcseconds, one is looking through parts of the Galactic ionized ISM separated by a few hundred AU. One can expect the scattering of radio waves to be independent for the different lines-of-sight. Differencing the image light curves, after a correction for the appropriate time delay and flux-density ratio, produces a difference light curve that only shows uncorrelated external variability. This difference light curve can be studied to obtain information on the Galactic ionized ISM independent from intrinsic source variability.
However, uncorrelated external variability of the lensed images might also originate from microlensing in the lens galaxy (e.g. Chang & Refsdal 1979). This offers the additional opportunity to study the properties of compact objects in the lens galaxy, if microlensing variability dominates or can be separated from scintillation. Optical microlensing in the lens galaxy of Q2237+0305 has unambiguously been shown (e.g. Irwin et al. 1989; Corrigan et al. 1991; Ostensen et al. 1996; Lewis et al. 1998; Wozniak et al. 2000). In the radio, several suggestions of microlensing variability have been made (e.g. Stickel et al. 1988; Quirrenbach et al. 1989; Schramm et al. 1993; Romero et al. 1995; Chu et al. 1996; Wagner et al. 1996; Lewis & Williams 1997; Takalo et al. 1998; Quirrenbach et al. 1998; Kraus et al. 1999; Watson et al. 1999). In none of these cases, however, has one really been able to convincingly distinguish between intrinsic and external variability. Claims of external variability in singly-imaged radio sources through microlensing should therefore be regarded with some caution.
In this paper, we report the first unambiguous case of external variability of a radio gravitational lens, CLASS B1600+434 (Jackson et al. 1995; Jaunsen & Hjorth 1997; Koopmans, de Bruyn & Jackson 1998 [KBJ98]; Koopmans et al. 2000 [KBXF00]). The system consists of two compact flat-spectrum radio images, separated by 1.4 arcsec. The background source, at a redshift of z=1.59, is lensed by an edge-on disk galaxy at a redshift of z=0.41 (Fassnacht & Cohen 1998). A time delay of days (95% statistical confidence) was recently found (KBXF00).
What is furthermore of interest is that this system offers two distinct lines-of-sight through the lens galaxy. Image A passes mainly through the dark-matter halo around the edge-on lens galaxy, whereas image B passes predominantly through its disk and bulge (Koopmans et al. 1998; Maller et al. 2000; CASTLE Survey, Munoz et al. 1999). This makes image A especially sensitive to microlensing by massive compact objects in the halo and image B to microlensing by stars in the disk and bulge. This might even offer an opportunity to study compact objects in the dark-matter halo around the lens galaxy of B1600+434.
The outline of the paper is as follows. In Sect. 2, we present the VLA 8.5-GHz data from KBXF00 in a different way, unambiguously showing the presence of external variability. We also present additional WSRT 1.4 and 5-GHz monitoring data of B1600+434. In Sect. 3, we investigate whether Galactic scintillation can explain the fractional rms variabilities (modulation-indices) and time scales of the short-term variability seen in the VLA 8.5-GHz light curves. Similarly, in Sects. 4 and 5 the possibility of microlensing by compact objects in the lens galaxy is studied. In Sect. 6, we present microlensing simulations of a more complex jet structure and compare the results to B1600+434. In Sect. 7, we discuss a critical test (i.e. the frequency-dependence of the modulation-index) to discriminate between scintillation and microlensing and compare predictions from the VLA 8.5-GHz light curves with the independent multi-frequency WSRT data. In Sect. 8 our results and conclusions are summarized.
© European Southern Observatory (ESO) 2000
Online publication: June 20, 2000