Multiply-imaged quasars

4.1 Multiply-imaged quasars

In 1979, gravitational lensing became an observational science when the double quasar Q0957+561 was discovered. This was the first example of a lensed object [193

]. The discovery itself happened rather by accident; the discoverer Dennis Walsh describes in a nice account how this branch of astrophysics came into being [192

It was not entirely clear at the beginning, though, whether the two quasar images really were an illusion provided by curved space-time – or rather physical twins. But intensive observations soon confirmed the almost identical spectra. The intervening “lensing” galaxy was found, and the “supporting” cluster was identified as well. Later very similar lightcurves of the two images (modulo offsets in time and magnitude) confirmed this system beyond any doubt as a bona fide gravitational lens.

By now about two dozen multiply-imaged quasar systems have been found, plus another ten good candidates (updated tables of multiply-imaged quasars and gravitational lens candidates are provided, e.g., by the CASTLE group [56 ]).

This is not really an exceedingly large number, considering a 20 year effort to find lensed quasars. The reasons for this “modest” success rate are:

Quasars are rare and not easy to find (by now roughly 10⁴ are known).
The fraction of quasars that are lensed is small (less than one percent).
It is not trivial at all to identify the lensed (i.e. multiply-imaged) quasars among the known ones.

Gravitationally lensed quasars come in a variety of classes: double, triple and quadruple systems; symmetric and asymmetric image configurations are known.

For an overview of the geometry of multiply-imaged quasar systems, see the collection of images found at [131 ].

A recurring problem connected with double quasars is the question whether they are two images of a single source or rather a physical association of two objects (with three or more images it is more and more likely that it is lensed system). Few systems are as well established as the double quasar Q0957+561; but many are considered “safe” lenses as well. Criteria for “fair”, “good”, or “excellent” lensed quasar candidates comprise the following:

There are two or more point-like images of very similar optical color.
Redshifts (or distances) of both quasar images are identical or very similar.
Spectra of the various images are identical or very similar to each other.
There is a lens (most likely a galaxy) found between the images, with a measured redshift much smaller than the quasar redshift (for a textbook example see Figure 5 ).
If the quasar is intrinsically variable, the fluxes measured from the two (or more) images follow a very similar light curve, except for certain lags – the time delays – and an overall offset in brightness (cf. Figure 8 ).

For most of the known multiple quasar systems, only some of the above criteria are fully confirmed. And there are also good reasons not to require perfect agreement with this list. For example, the lensing galaxy could be superposed to one quasar image and make the quasar appear extended; color/spectra could be affected by dust absorption in the lensing galaxy and appear not identical; the lens could be too faint to be detectable (or even a real dark lens?); the quasar could be variable on time scales shorter than the time delay; microlensing can affect the lightcurves of the images differently. Hence, it is not easy to say how many gravitationally lensed quasar systems exist. The answer depends on the amount of certainty one requires. In a recent compilation, Keeton and Kochanek [92 ] put together 29 quasars as lenses or lens candidates in three probability “classes”.

Figure 5: A recent example for the identification of the lensing galaxy in a double quasar system [43 ]: The left panel shows on infrared (J-band) observation of the two images of double quasar HE 1104-1825 ( $zQ$ = 2.316, $Δ 𝜃$ = 3.2 arcsec). The right panel obtained with some new deconvolution technique nicely reveals the lensing galaxy (at $zG$ = 1.66) between the quasar images. (Credits: European Southern Observatory [55 ].)

Gravitationally lensed quasar systems are studied individually in great detail to get a better understanding of both lens and source (so that, e.g., a measurement of the time delay can be used to determine the Hubble constant). As an ensemble, the lens systems are also analysed statistically in order to get information about the population of lenses (and quasars) in the universe, their distribution in distance (i.e. cosmic time) and mass, and hence about the cosmological model (more about that in Section 4.6). Here we will have a close look on one particularly well investigated system.

4.1.1 The first lens: Double quasar Q0957+561

The quasar Q0957+561 was originally found in a radio survey, subsequently an optical counterpart was identified as well. After the confirmation of its lens nature [193 , 192 ], this quasar attracted quite some attention. Q0957+561 has been looked at in all available wavebands, from X-rays to radio frequencies. More than 100 scientific papers have appeared on Q0957+561 (cf. [140 ]), many more than on any other gravitational lens system. Here we will summarize what is known about this system from optical and radio observations.

In the optical light, Q0957+561 appears as two point images of roughly 17 mag (R band) separated by 6.1 arcseconds (see Figure 6 ). The spectra of the two quasars reveal both redshifts to be $zQ = 1.41$ . Between the two images, not quite on the connecting line, the lensing galaxy (with redshift $zG = 0.36$ ) appears as a fuzzy patch close to the component B. This galaxy is part of a cluster of galaxies at about the same redshift. This is the reason for the relatively large separation for a galaxy-type lens (typical galaxies with masses of $1011−12M ⊙$ produce splitting angles of only about one arcsecond, see Equation (10 )). In this lens system, the mass in the galaxy cluster helps to increase the deflection angles to this large separation.

Figure 6: In this false color Hubble Space Telescope image of the double quasar Q0957+561A,B. The two images A (bottom) and B (top) are separated by 6.1 arcseconds. Image B is about 1 arcsecond away from the core of the galaxy, and hence seen “through” the halo of the galaxy. (Credits: E.E. Falco et al. (CASTLE collaboration [56

]) and NASA.)

A recent image of Q0957+561 taken with the MERLIN radio telescope is shown in Figure 7 . The positions of the two point-like objects in this radio observation coincide with the optical sources. There is no radio emission detected at the position of the galaxy center, i.e. the lensing galaxy is radio-quiet. But this also tells us that a possible third image of the quasar must be very faint, below the detection limit of all the radio observations⁶ . In Figure 7 , a “jet” can be seen emerging from image A (at the top). It is not unusual for radio quasars to have such a “jet” feature. This is most likely matter that is ejected from the central engine of the quasar with very high speed along the polar axis of the central black hole. The reason that this jet is seen only around one image is that it lies outside the caustic region in the source plane, which marks the part that is multiply imaged. Only the compact core of the quasar lies inside the caustic and is doubly imaged.

Figure 7: Radio image of Q0957+561 from MERLIN telescope. It clearly shows the two point like images of the quasar core and the jet emanating only from the Northern part. (Credits: N. Jackson, Jodrell Bank.)

As stated above, a virtual “proof” of a gravitational lens system is a measurement of the “time delay” $Δt$ , the relative shift of the light curves of the two or more images, $IA (t)$ and $IB(t)$ , so that $IB (t) = const × IA(t + Δt)$ . Any intrinsic fluctuation of the quasar shows up in both images, in general with an overall offset in apparent magnitude and an offset in time.

Q0957+561 is the first lens system in which the time delay was firmly established. After a decade long attempt and various groups claiming either of two favorable values [138 , 144 , 161 , 190 ], Kundić et al. [105 ] confirmed the shorter of the two (cf. Figure 8 ; see also Oscoz et al. [133 ] and Schild & Thomson [162 ]):

Figure 8: Optical Lightcurves of images Q0957+561 A and B (top panel: g-band; bottom panel: r-band). The blue curve is the one of leading image A, the red one the trailing image B. Note the steep drop that occured in December 1994 in image A and was seen in February 1996 in image B. The light curves are shifted in time by about 417 days relative to each other. (Credits: Tomislav Kundić; see also [105

])

With a model of the lens system, the time delay can be used to determine the Hubble constant⁷ . In Q0957+561, the lensing action is done by an individual galaxy plus an associated galaxy cluster (to which the galaxy belongs). This provides some additional problems, a degeneracy in the determination of the Hubble constant [65 ]. The appearance of the double quasar system including the time delay could be identical for different partitions of the matter between galaxy and cluster, but the derived value of the Hubble constant could be quite different. However, this degeneracy can be “broken”, once the focussing contribution of the galaxy cluster can be determined independently. And the latter has been attempted recently [58 ]. The resulting value for the Hubble constant [105 ] obtained by employing a detailed lens model by Grogin and Narayan [70 ] and the measured velocity dispersion of the lensing galaxy [57 ] is

where the uncertainty comprises the 95% confidence level.

Going further. In addition to the above mentioned CASTLES list of multiple quasars with detailed information on observations, models and references [56 ], a similar compilation of lens candidates subdivided into “presently accepted” and “additional proposed” multiply imaged objects can be found at [131 ]. More time delays are becoming available for other lens systems (e.g., [22 , 160 ]). Blandford & Kundić [25 ] provide a nice review in which they explore the potential to get a good determination of the extragalactic distance scale by combining measured time delays with good models; see also [159 ] and [205 ] for very recent summaries of the current situation on time delays and determination of the Hubble constant from lensing.