... Gaia

For details and recent updates consult the web site http://www.rssd.esa.int/index.php?project=SA

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... asymmetric

In quite different optical configurations, the location of the peak of the diffraction pattern appears to be stable with respect to the spectral distribution of the celestial objects; this is consistent with the fact that, in case of reasonably limited aberrations, this maximum is coincident with the chief ray of geometric optics, which by definition is insensitive to wavelength variations. Therefore, the diffraction peak would appear as a good candidate for the PSF "photo-center'' (location), if only the PSF were sampled with infinite resolution. In practice, it is impossible to measure it directly because of the finite pixel size.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...=0)

As we make reference to a Gaia-like system in this work, we consider the y axis for consistency with most of the literature on Gaia. There the y-axis is the axis of the one-dimensional location process and the axis along which the scanning occurs.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... value

In practice, the optical system would be optimized to reduce coma. Consequently, the actual variation of the COG would, of course, be smaller.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... width

The root mean square wavefront error, rms WFE, is simply the square root of the quantity $\int WFE^2(\rho , \theta)$ taken over the pupil. The image rms width for, e.g. the y axis on the focal plane is computed as $\sqrt{\int (y - \langle y \rangle)^2 L(y) {\rm d}y}$, where $L(y) = \int {\it PSF}(x,y) {\rm d}x$ is the line spread function.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... chromaticity

As for the previous section, chromaticity is evaluated as the difference between the monochromatic PSF at 628 nm and 756 nm, respectively. The chromaticity derived with the fully polychromatic model is about a factor 2 larger.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.