In the
Pictoris observations,
no rapid light variations, neither in V magnitude, in the b-y color index nor
in the m1 and c1 parameters were detectable during the three campaigns
(1995-1996, 1996-1997 and 1997-1998). This is nevertheless interesting and
significant. It is now clear that an
event like the one, which occurred in November 10, 1981, is very rare.
Assuming that the event of November 1981 is periodic, and using the available data with
the Geneva and Strömgren observations, it is now possible to eliminate a large range of
value for the putative period.
Each
given measurement at time t which does not show this variation
allows to eliminate the period P1 given by the time
difference between the measurement and the date
November 10:
.
In fact, considering that if this event had occurred again, it would have been detectable during one day (which is a very conservative assumption), each measurement allows to eliminate the range of period P1 plus or minus half a day. Moreover, each period Pn resulting from the division of P1 by an integer n can also be eliminated. In other words, for each measurement at time t, the ranges of period [(P1-0.5)/n; (P1+0.5)/n], where n is an integer, are not possible. The resulting set of impossible periods is plotted in Fig. 7 for the Geneva and Strömgren data. From this plot we see that 90% of periods between 0 and 1000 days can be eliminated. 49% of periods between 1000 and 2000 days, 29% of periods between 2000 and 3000 days and 57% of periods between 5000 and 6000 days are also eliminated. To cover the periods between 3000 and 5000 days, new observations are required from nowadays to the year 2009.
Copyright The European Southern Observatory (ESO)