Chapter 1

Historical introduction
by MARTIN-LUTHER-UNIVERSITAET HALLE-WITTENBERG on 12/23/17. For personal use only.
Gravitational Lensing and Microlensing Downloaded from www.worldscientific.com

El Aleph? repeti. Si, el lugar donde estan, sin confundirse,

todos los lugares del orbe, vistos desde todos los angulos.

The Aleph? I repeated. Yes, the place where, without confusion,

all the places of the world, seen from every angle coexist.

The Aleph, Jorge Luis Borges.

The bending of light in a gravitational field predicted by General Relativ-

ity provided one of the first verifications of Einstein's theory [Dy20]. The
value of the deflection angle which results has now been tested at the 0.1%
level by observing the change in the apparent position of stars whose light
is deflected by the Sun [Le95]. In 1920, Eddington [Ed20] noted that the
light deflection by a stellar object would give rise to secondary images of
background sources appearing just on the opposite side of the deflector.
Chwolson [Ch24] later pointed out that these secondary images could make
foreground stars to look like binaries, and that if the alignment were perfect
the image of the source would be a ring. In 1936, Einstein [Ei36] published
the correct expression for the magnification of the two images of a very
distant star, concluding however that the lensing effect of foreground stars
was of no practical relevance due to the unresolvably small angular sepa-
ration of the images and the low probability for a high amplification event
to take place.
The following year, Zwicky [Zw37a] showed that, if the deflecting object
were a galaxy instead of a single star, the gravitational lensing of the light of
a background galaxy would lead to resolvable images. This 'macrolensing'
effect could provide information about the mass of the intervening galaxy
and could make possible the observation of objects at much larger distances
due to the magnification of their light, with the lens acting as a natural

1
 Historical introduction
by MARTIN-LUTHER-UNIVERSITAET HALLE-WITTENBERG on 12/23/17. For personal use only.
Gravitational Lensing and Microlensing Downloaded from www.worldscientific.com

Fig. 1.1 HST picture of the first binary quasar observed, Q0957+561. The angular
separation between images is 6.1", and the lensing galaxy is close to the image on
the bottom (from http://cfa-www.harvard.edu/glensdata CfA/Arizona Space Telescope
Lens Survey (CASTLES) website, courtesy of C. S. Kochanek, E. E. Falco, C. Impey, J.
Lehar, B. Mc Leod and H.-W. Rix).

telescope. Furthermore, in Zwicky's words, the probability that this effect

be observed 'becomes practically a certainty' [Zw37b].
It was actually the discovery by Walsh, Carswell and Weymann in 1979
[Wa79] of the multiple imaging of the high redshift quasar Q0957+561,
lensed by a foreground galaxy (Figure 1.1), what provided the first obser-
vation of gravitational lensing. This discovery opened a new area of re-
search which has now become an extremely active field in astronomy, with
the potential of giving also crucial information for cosmology. For instance,
the time delay among the multiple images of a quasar allows to measure
the Hubble constant once the lens mass has been reconstructed from the
location and shape of the images, and this procedure is now becoming a
competitive way to determine the expansion rate of the Universe. Also,
lensing statistics are sensitive to cosmological parameters, with the fre-
quency of multiple imaging giving constraints on the cosmological constant
and the distribution of image splittings probing the amount of structures
on galaxy and cluster scales. Gravitational lensing can also affect the ob-
servable properties of active galactic nuclei and quasars, and may have to
by MARTIN-LUTHER-UNIVERSITAET HALLE-WITTENBERG on 12/23/17. For personal use only.
Gravitational Lensing and Microlensing Downloaded from www.worldscientific.com

Fig. 1.2 Giant arcs around the cluster Abell 2218, produced by the lensing effect of
the cluster mass on background galaxies (credit W. J. Couch (University of New South
Wales), R. S. Ellis (Cambridge University) and NASA).

be taken into account when inferring their intrinsic properties.

Another manifestation of gravitational lensing was discovered in 1986
by Lynds and Petrosian [Ly86] and Soucail et al. [So87] in the form of large
luminous arcs. These are due to the lensing effects of rich clusters of galax-
ies on background galaxies, with possible huge tangential magnifications
which distort the galaxy shapes into long (giant) arcs around the cluster's
cores. One of the nicest examples of this phenomenon is shown in Fig-
ure 1.2. Through this kind of observations the central mass of the clusters
can be inferred. On the other hand, outside the central region of a cluster,
and in general also for clusters which are not massive enough to produce
strong lensing effects (i.e. involving multiple image formation), the mass
distribution can be reconstructed from the observed shape of the images
of thousands of background faint blue galaxies around it. These galaxies
look tangentially elongated due to the lensing effect (which in this regime is
referred to as weak lensing), and their average ellipticities form a coherent
pattern around the cluster encoding the information on the lensing matter
distribution [Ty95]. Finally, in empty regions it has been recently possible
to measure galaxy ellipticity correlations up to scales of ~ 10', which are
induced by the weak lensing due to all the large scale structure along the
line of sight to the background galaxies.
Regarding the gravitational lensing by compact objects (e.g. stars), its
systematic theoretical study started in the sixties with the work of Liebes
[Li64] and Refsdal [Re64], who setup the general formalism and discussed
 4 Historical introduction

the lensing of stars in the disk of the Galaxy, in globular clusters and in
the Andromeda galaxy.
The lensing effect of individual stars belonging to a galaxy that is it-
self macrolensing a background source was discussed by Chang and Refsdal
[Ch79] soon after the discovery of the binary quasar Q0957+561. When
both the lensed source and the intervening galaxy are at cosmological dis-
by MARTIN-LUTHER-UNIVERSITAET HALLE-WITTENBERG on 12/23/17. For personal use only.

tances, the passage of one of these stars close to the line of sight to one of
the images further deflects the source light by an angle which is typically
Gravitational Lensing and Microlensing Downloaded from www.worldscientific.com

of 0(/^arcsec). The name 'microlensing' then became associated with this

process, and is now generally applied to any gravitational lensing effect by
a compact object producing unresolvable images of a source and potentially
huge magnifications of its light.
Press and Gunn [Pr73] showed in 1973 that a cosmological density of
massive compact dark objects could manifest through the microlensing of
high redshift sources. In 1981, Gott [Go8l] pointed out the possibility of
detecting the dark halos of remote galaxies by looking for microlensing of
background quasars. It was in 1986 that Paczyriski [Pa86a] gave a new face
to the field when he noted that, by monitoring the light-curves of millions
of stars in the Large Magellanic Cloud (LMC) for more than a year, it
would become possible to test whether the halo of our galaxy consisted of
compact objects with masses between 10~ 6 and 102 M Q , i.e. covering most
of the range where baryonic dark matter in the form of planets, Jupiters,
brown dwarfs or stellar remnants (dead stars, neutron stars or black holes)
could lie. It was later realized [Pa91; Gr91b] that the Galactic bulge stars
also provided an interesting target to look at, since at least the lensing by
faint stars in the disk should grant the observation of microlensing events,
and observations towards the bulge could also allow to probe the dark
constituency of the Galaxy close to the Galactic plane.
Several groups undertook the observations towards the Magellanic Clouds
and the bulge, obtaining the first harvest of microlensing events in 1993
[A193; Au93; Ud93] (see Figure 1.3). Also follow-up telescope networks
have been organized and they are looking for alerted ongoing microlensing
events in the search for planetary signals and providing detailed light curves
which are used to obtain better information about the lenses. Also the ob-
servations of microlensing of unresolved stars in the Andromeda galaxy has
started.

As we have seen in this brief introduction, gravitational lensing is an

 5

"1 1 1 1 1 ] 1 1 1 1 1 I I 1 1

blue I Amax=6.86
t=33.9

<4
a
by MARTIN-LUTHER-UNIVERSITAET HALLE-WITTENBERG on 12/23/17. For personal use only.

o
Gravitational Lensing and Microlensing Downloaded from www.worldscientific.com

o i—|—I—I—I—i-J—I—I—I—|—I—I—l—f-r-t

.1.5

4*1 & * H "^H

jiUlhl^
0.5

_i i i i i i i i i i i i i i i_
400 420 440 460
days from 2 Jan 1992

Fig. 1.3 Light-curve of the first microlensing event of a star in the LMC obtained by
the MACHO Collaboration. The ratio of the light-curves in two colors is shown in the
bottom panel, illustrating the achromaticity of this gravitational effect (courtesy of C.
Alcock, reprinted from [A193] with permission from Nature).

ubiquitous phenomenon in the Universe and it has several attractive fea-

tures. It is a direct manifestation of the deflection of light predicted by
General Relativity, giving rise to many beautiful optical effects, such as the
multiple imaging, the shape distortion and the magnification of faraway
sources. Most importantly, the observation of these effects provides a di-
rect way to measure the amount of matter in the Universe as well as its
spatial distribution at different scales, and allows to obtain fundamental in-
formation on the cosmological parameters and on the large scale structure
distribution in the Universe. On the other hand, microlensing experiments
are a powerful tool to search for compact dark objects and faint stars which
are otherwise unobservable, and the results obtained at present are already
providing crucial insights into the dark matter issue, into the non-luminous
contribution to the mass of the different Galactic components and about the
 6 Historical introduction

morphology of the inner part of the Galaxy. Microlensing has also applica-
tions to stellar astrophysics, allowing to study the binary mass ratios and
distribution of separations, to measure the mass of foreground bright stars
and even in some special circumstances to resolve the size of the sources.
by MARTIN-LUTHER-UNIVERSITAET HALLE-WITTENBERG on 12/23/17. For personal use only.
Gravitational Lensing and Microlensing Downloaded from www.worldscientific.com