The seeing effects are caused by atmospheric turbulence through which some of the light arriving from a star is scattered by refractive inhomogeneities. As the light wave propagates through the atmosphere it experiences fluctuations in amplitude and phase. An image formed by focusing this wave exhibits fluctuations in intensity, sharpness and position which are commonly referred to as scintillation, image blurring and image motion.
The seeing observed by a telescope on the ground is contributed by the whole atmosphere crossed by the light wave front and one distinguishes three main contributing causes:
50 m)
and the one generated by the artificial structures of the observatory
itself.
A rigorous quantification of the seeing effect depends on the exposure time. For most astronomical observations, seeing is quantified for the so-called long exposure case, in which the exposure is longer than the time in which the wavefront phase inhomogeneities larger than the telescope pupil pass through it. In practice for a large telescope this is an exposure of a duration of the order of 10 to 30 seconds. As a consequence the image motion effect of seeing will be summed-up in an overall blur effect.
Several figure of merit are used for quantifying seeing:
.
The FWHM 
for a telescope of large diameter (D 
),
expressed
in seconds of arc (arcsec) on the sky, is related to the 
parameter as
where 
is the wavelength.
In the practical case of a limited diameter telescope, expression
()
defines the seeing contribution to the total image FWHM, which results
from the convolution of the diffraction pattern of the telescope
with the optical aberrations and the seeing.
) it is:
In this work one will generally use the FWHM of the seeing disk as a unit of measure for seeing.
Different methods exist to measure or estimate seeing. With a large telescope a good estimate of seeing may be obtained by measuring the object size or by looking at the smallest resolution in the image. In order to have a small transportable instrument for testing the quality of different sites, ESO has developed a 35-cm telescope with a differential image motion monitor (DIMM) based on a method in which the image motion of short exposures is related to the long exposure image size [Sarazin 92].
A main problem for the analysis of seeing effects is due to the fact that measurements of the optical image quality hardly allow separating the different sources. Therefore other instruments are used to characterize and separate the effects from the different layers of the atmosphere such as the scintillometer, used for evaluating high altitude turbulence and the acoustic sounder or SODAR (SOund Detection And Ranging) for measuring turbulence profiles in the atmospheric boundary layer. There exist, however, no direct means to discriminate the different causes of seeing in the immediate environment of the telescope.
The best astronomical site are reported to have a natural seeing varying between 0.3 to 0.6 arcsec (FWHM), which is roughly equally divided between the high atmosphere and the atmospheric boundary layer. Acceptable sites for astronomical research will have natural seeing up to 2 arcsec.
Local seeing, that is the seeing caused by the observatory can represent anything from zero to about 2 arcsec.
