Ehgamberdiev et al. (2000), and Boccas et al. (2000)1
as proof
of DIMM’s success and proliferation. I take parameters of the
Cerro Tololo Inter-American Observatory (CTIO) DIMM (aperture diameter 6 cm, baseline 19 cm, CCD pixel size 0.77)
as representative for these instruments. A good theory of
DIMMs was published by Martin (1987). This paper continues
the work of Martin by considering additional effects.
The difference of the tilts over two DIMM subapertures is
proportional to the second derivative of the wave front, or
curvature. Thus, a DIMM is a form of curvature sensor. Another
instrument to deduce seeing from wave-front curvature is
described by Roddier, Graves, & Limburg (1990). On the other
hand, the DIMM can be viewed as a Shack-Hartmann (S-H)
sensor with only two apertures; extensive studies of these sensors are partially applicable to DIMMs (van Dam & Lane 2000;
Irwan & Lane 1999). Other methods to measure such as r
0
absolute image motion or the shearing interferometer (Dainty
& Scaddan 1975; Roddier 1976) are not immune to pointing
errors, making them unsuitable for use as robust site-testing
instruments.
The DIMM’s success and widespread implementation has
led to some confusion. A DIMM is often thought to be a robust
and self-calibrated instrument that always measures the seeing
correctly. However, a number of subtle instrumental effects in
DIMMs do bias the results. As long as site comparison was
based on identical instruments, their systematic errors were of
secondary importance. Now there is a strong need to compare
seeing measures obtained by teams using different equipment.