Once every six years, the Earth and the Sun cross the equatorial plane of Jupiter (the jovian declinations of the Earth and the Sun become zero). Since the Galilean satellites have their orbital planes very close to the jovian equator, mutual phenomena occur: the satellites occult and eclipse each other during a period of about six months. The 1991 period (Arlot 1990) was very favorable since it happened during the opposition of Jupiter and the Sun (Fig. 1 (click here)). Northern hemisphere observatories were favored since the declination of Jupiter was positive.
Since there is no atmosphere around the Galilean satellites, the photometric observations of such phenomena are very accurate for astrometric purposes. The relative positions of the satellites involved may be obtained with an accuracy better than the one obtained by the other kinds of observation: 200 km for the best photographic observations and 700 km for the eclipses by Jupiter observed visually. Note that CCD photometry of eclipses by Jupiter should reach an accuracy of 40 km according to Mallama et al. (1994).
Such an accuracy is necessary for several reasons. First, the orbital motions of the Galilean satellites have to be very well known in order to prepare the missions of the space probes exploring the jovian system. Second, all the theoretical problems related to the motion of the Galilean satellites are not solved. This motion, which is affected by numerous perturbations, is one of the most complex in celestial mechanics. The theoretical data presented in this paper are calculated using ephemerides G-5 (Arlot 1982) based on the work of Lieske (1977). These ephemerides need improvements. For example, Io is suspected of having a secular acceleration due to energy dissipation (Goldstein & Jacobs 1986; Lieske 1987). Such an effect is very small and observations during several series of mutual events will be needed in order to determine the acceleration.
Coordinated campaigns of observations are very useful for getting a large amount of data. All the events occur in a small interval of time; so, numerous observers observing from several sites are necessary. To get as many events as possible, it is necessary to observe from different longitudes (to get different events) and from several sites (to avoid meteorological problems). In 1979, the conditions were not favorable because of the conjunction of Jupiter with the Sun, so, very few observations were obtained (Arlot et al. 1982). Contrarily, in 1985 the conditions were very favorable and we got 154 lightcurves from 63 sites in Europe and South America. Observations were easier in the Southern hemisphere because of the negative declination of Jupiter (Arlot et al. 1992).
Figure 1: Jovicentric declinations of the Earth and the Sun during
the 1991 occurrence