726 Encyclopedia of the Solar System
FIGURE 7 Schematic of an active optics system. Starlight from
the telescope is sent to a beamsplitter that simultaneously sends
light to the focus and to a wavefront sensor. The computer
analyses the output of the wavefront sensor and sends control
signals to the primary and secondary mirrors to correct any
aberrations in the image. (Courtesy of C. Barbieri.)
limitations arise from the need to be diffraction limited,
the difficulty of building a suitable enclosure, and the cost.
To be competitive with space observatories, all large tele-
scopes must work at the diffraction limit using adaptive
optics. But the need to be diffraction limited will ultimately
cause adaptive optics systems to be too complex on an ex-
tremely large telescope. An enclosure is necessary to keep
the disturbance by wind to acceptable levels, and the cost to
build and operate the telescope will be enormous. At some
point, it may be more cost effective to go into space, where
gravity and the weather are not factors driving the design.
This has been estimated to be at approximately 70-m in di-
ameter. This argument applies to fully steerable telescopes,
not to designs such as the Hobby-Eberly Telescope or the
Large Zenith Telescope.
The drive to build ever-larger telescopes is motivated by
the need to collect as much light as possible and thereby
increase the signal-to-noise (S/N) ratio of observations. One
can derive that for a diffraction-limited telescope and a de-
tector that is background-limited, the S/N in a given inte-
gration time is proportional to:S/N≈(A∗η/ε)^0.^5 /(FWHM), (1)where A is the area of the telescope,ηis the total transmis-
sion of the optics and the detector quantum efficiency,εis
the background emission, and FWHM is the full width at
half maximum of a stellar image.ηtakes into account all of
the light losses that occurs from the reflection of the mir-
rors and transmission losses of lenses as light propagates
from the telescope to the detector. In order to minimize
these losses it is necessary to utilize high reflection coatings
on mirrors and lenses as well as to minimize the number
of lenses. The detector quantum efficiency is the fraction
of light that is absorbed by the detector material. This is
near the theoretical maximum of 1.0 at visual wavelengths
and about 0.8–0.9 for the 1–15μm wavelength range. The
background emission,ε, arises from the sky emission lines at
visual wavelengths and thermal background from the tele-
scope and sky at wavelengths longer than 2μm. To reduce
the thermal emission from the telescope, it is necessary to
have the highest reflectivity mirrors available and to reduce
or eliminate the thermal emission from the secondary mir-
ror. The latter is often accomplished by forming an image
of the secondary within the instrument and then blocking
it with a cooled metal plate. Then the infrared detector will
only sense the thermal emission from the sky and the object
being observed.
After maximizingηand reducingεas much as possible,
one can only increase the telescope area and reduce the
FWHM to further increase the S/N. Reducing the image
FWHM requires decreasing the dome seeing to the abso-
lute minimum, building on sites that have good atmospheric
seeing, and working at the diffraction-limit of the telescope.
Astronomical sites in Hawaii, Chile, and La Palma are prime
locations for large telescopes due to the good seeing they
offer as well as having good weather conditions.
Figure 8 shows the advances in image quality that have
been achieved. The development of adaptive optics has
led to the ability to work at the diffraction limit in the
near-infrared and to achieve improvements in S/N given
by equation 1. Adaptive optics is discussed in Section- The advances in constructing large telescopes coupled
with reducing dome seeing and adaptive optics have pro-
vided the means for studying the surfaces of some KBOs
and larger planetary satellites (see Fig. 1). Ground-based
telescopes provide the discoveries that pose new questions