------------------------  Technical Note n°3  PHEMU  January 15, 1997  ------------------------

 
 
I - Introduction
 
Before observing the events during an opportunity which is very short, it seems to be necessary to remind the observers to be very careful and to follow a very strict procedure for the photometric observation of the mutual events either with a photoelectric photometer or with a CCD receptor. A technical help in the use of photoelectric photometers or CCD targets will be provided in a next note. The present note would like to state the principles which will lead to useful observations : what to observe, when to observe and how to observe. These observations are easy to make and they are spectacular because of the large magnitude drop to be observed and because of the fastness of these events. The help of previous campaigns of observations will provide us what we need to succeed in the future observations.
 

II - The time-scale for the observations

Since these observations are made in order to improve the models of motion of the objects themselves, all the data must be referred to a well known time-scale to the nearest tenth of a second of time in order to be able to link all the observations together. In fact all the observational data should be referred to UTC (Universal Time). Internal clocks not related to UTC are not confident. If it is not possible to be connected to UTC any time, it is necessary to verify the used clock before and after the observation and to record the difference between the used clock and UTC at that time (never modify the clock during an observation). The sidereal time must never be used in the present work. We learned, through the past experiences that this calibration of the time-scale should never be made after the observation since anything can happen such as a failure of the hardware. Unfortunately, all the observatories do not have a sure clock related to UTC and the best to do is to note the difference of the available time-scale with UTC before and after the observation by calling UTC through the phone network or thanks to a radio-receiver in order to know if a drift occurs in the time-scale. The time accuracy should be better than 0.5 second of time that is to say that each photometric measurement of the recorded light-curve must be dated in UTC with an accuracy better than 0.5 second of time. Note that the satellite Io, for example, has a velocity of 17.2 km/s, so that an accuracy of 0.5 second of time corresponds to an accuracy of 8.6 km in space. Since the internal accuracy of the theory of motion of the satellites is around one kilometer, anyone may understand that an accuracy better than 0.5 second of time is necessary.

At last, be careful to start the observation well in advance. The predictions are not so precise and the events could start a few minutes before the predicted time. The long events may begin 10 or 20 minutes before (or after) the predicted time. Don't forget that predictions are given in Terrestrial Time which differs from the UTC of about one minute of time (TT - UTC ~ 66 seconds in 2002).
 

III- Choosing the events

If you are able to observe only a few events, what events should you choose ? For that, you may follow the following rules:
- choose events when Jupiter is not too low above the horizon and avoid events during twilight, except if you use a CCD with a R or I filter and if a reference satellite is present in the field;
- avoid grazing events and events the magnitude drop of which is less than 20 p. cent; the best events to observe are those the magnitude of which is larger than 20 p. cent and smaller than 90 p. cent;
- choose events occurring at more than 3 jovian radii to Jupiter;
- prefer the occultations to the eclipses and choose the eclipses where the eclipsing satellite is far from the eclipsed one ;
- avoid the too long events if you are not able to observe the entire light curve (Jupiter mat be below the horizon at the end of the event...).
Of course it is better to try to observe the maximum of events.
 

IV - The receptors to be used

The goal is to record the variation of a light signal and to make a photometric measurement depending on the time with a high frequency of acquisition (from 0.5 to 5 points per second of time). For that, several receptors may be used from the more simple recording only one signal to the more elaborated recording simultaneously several signal in several wavelengths. These receptors are:
- single channel photometer ;
- spectral multichannel photometer ;
- spatial multichannel photometer ;
- spectrophotometer ;
- two-dimensional receptor (CCD, video).
We will not provide information on the receptors themselves, but only on the problems related to the recording of the events, the wavelengths to be used, the sampling integrated time and the reduction.

a) the wavelength

The spectral band where the observation will be made depends on several factors : first you have to choose the wavelength where the receptor used is the more efficient; second you may choose a wavelength in which the observation will provide the most interesting information. Note that any wavelength is interesting but we may see what wavelengths are more favorable to reach new information:
- case of a receptor working in only one wavelength : any visible wavelength will provide useful lightcurves. For a CCD receptor, the R and I bands are more efficient, but with a photoelectric photometer you probably will have no choice. Note that it is possible to decrease the light from Jupiter (for events occurring very near the limb of the planet) by using interferential filters such as CH4, (7260 A, 8300 A,...) even it is not very efficient: two-dimensional photometry is more efficient. For observations to be made in a city polluted by light, the 5000 to 5300 A spectral band should be of any interest. The following diagram shows the brightness of the urban sky depending on the wavelength.
- case of a spectral multichannel photometer : if the receptor allows to record the event in several wavelengths, please do it: recording several lightcurves of the same event from the same telescope is highly interesting: it is the only way to decorrelate local photometric accidents depending on the site and on the observational conditions from interesting information on the grounds of the satellites. Note that the time sampling should be sufficient (more than one point every second of time but not more than 10 points per second). As said previously, all the wavelengths are interesting because of the few experience we have on such simultaneous lightcurves.

- the integrating time should be not too short (it is necessary to have a sufficient signal/noise ratio) and not too long (since the events are very fast, we need to record enough points). The experience shows that, depending on the receptors, the integrating times vary from 0.1 to 2 seconds of time.
- the time sampling depends on the integrating time : after recording the light during the integrating time, it is necessary to store the date and the value before starting a new measurement. The modern receptors allow to make this operation very quickly. The time sampling is commonly included between 10 points per second of time to one point every 2 seconds. More points will provide too much data for an event of a few minutes duration ; less points will not be sufficient to determine the different phases of the events especially in case of short events -. Anyway the time sampling must depend on the duration of the event : for long events (about one hour) one point per second is sufficient. Note that in case of a long event, a photoelectric photometer working successively in several wavelengths will be very efficient; for a very short event, one should be sure to get enough measurement in each wavelength for such a receptor.

c) the diaphragm

The diaphragm is the field, the contents of which are measured together. For example, in the case of an eclipsed satellite, this satellite will be included in the diaphragm and its light flux measured during the event. In the case of an occultation, both the occulted and the occulting satellites will be included in the diaphragm and their light flux measured. This may be complicated by the presence of another satellite in the vicinity of the involved satellite. Several examples will be shown below. In the case of a spatial multichannel photometer or of a two-dimensional receptor the problem is slightly different: we have then the possibility of recording several diaphragms simultaneously.
In the case of a spatial multi-channel photometer, we should choose carefully the different diaphragms before the observation : one for the implied satellite(s), one for the sky background and one (or more) for the reference object. Attention to a satellite (or a star !) arriving in a diaphragm during the observation... or to Jupiter itself (dont forget that Jupiter has a magnitude of 5 per square arcsecond).
In the case of a two-dimensional receptor, a large field is recorded simultaneously : it is not a diaphragm, it is an image. The diaphragms will be designed after the observation on the recorded images themselves. Therefore, it is important to define very well the field of the images before the event to be able to determine efficient diaphragms during the reduction. Be sure that a reference object is available and will not be disturbed during the event by another satellite moving during the event (especially for long events).
The study of the field before each event is now easy thanks to PC- softwares such as SATEL13 available on the ftp server of Bureau des longitudes.
Note that in any case a too large diaphragm will include too much light from the sky background and from Jupiter and that a too small diaphragm will make difficult the stay of the implied satellite in the diaphragm (especially with a bad seeing).
 

V - The observation

The nights are always too short to make all the preparation to the observation of the events and to allow improvisation : therefore, prepare a written procedure with a well-adapted timing in order to be sure to forget nothing (calibrations, ...).

a) the preparation

Even we are doing relative photometry, the photometric calibration using quasi-solar type stars may be interesting. This has to be made well in advance to the event.
The diaphragm(s) must be chosen depending on the configurations of the satellites. Be careful to identify correctly the satellites, especially if you use optical mounting reversing the field. Several cases may occur as shown below.

b) The recording of the event

The measurement of the sky background is necessary in all cases, even if there is no light pollution. In the case of a single channel photometer, this measurement should be made before and after the recording of the event in the case of a short event (less than 20 minutes) and once every 5 minutes in the other cases. The method described in the previous paragraph is applicable for the events near Jupiter: four measures around the implied satellites. If necessary only one measure above and one below will be sufficient. The use of a multi-spatial channel photometer or of a two-dimensional receptor avoid the previous measures.
The atmospheric absorption leads to a magnitude drop proportional to the zenithal distance as determined by the ``droite de Bouguer''. The annex extracted from ``Introduction à l'Astrophysique : les étoiles'' of J. Dufay provides the principles of this mechanism. However one should be careful with the given values (0,165 magn., 0,29 magn., 0,59 magn.) which correspond only to mean values for a given site. In fact these values may vary considerably from one night to another ; they may vary also during the same night. Therefore the ``droite de Bouguer'' will be no more a ``droite'': the points scatter, mainly near the horizon. This, to show that it may be useful to make differential measurements referred to another Galilean satellite or to a standard photometric star whose spectral type is quasi-solar.

c) Observations to be made cautiously

- Observations made low on the horizon: In that case, be careful with the refraction which increases rapidly and which needs to correct the guiding of the telescope. Be also careful with the absorption which increases and may also vary. The measurement of a reference object is highly necessary. The obtained light curve will be as given in fig. 4. Note that for these difficult observations, the best is the used of a two-dimensional receptor or of a multi-spatial channel photometer allowing to measure simultaneously all the interesting objects (and the sky background).

d) Let's remind the errors to be avoided

- to mix satellites (confusing North/South or East/West...);
- to start observing to late and to have not enough time for the calibrations;
- to miss observations because Jupiter is too low on the horizon observations are possible at 10 above the horizon);
- to choose a wrong diaphragm and to need to change it during the event;
- to suppose that the motion of the satellites is linear and uniform;
- to think that we know everything on the galilean satellites (the magnitude may change from one point to another on the orbits;
- to have a wrong time scale and to be not sure of the clock (be sure to have the UTC available);
In brief, prepare carefully the observation and follow minute after minute a procedure written in advance with a precise timing
 

VI - Summary of the most important points to be examined before the observation

1 - be sure to have a time scale in UTC accurate to 0.2 second of time ;
2 - verify that Jupiter and the satellites will be visible during all the observation ;
3 - verify that each point of the lightcurve is correctly referred to the time scale with an accuracy better than 0.2 second ;
4 - think to use the right filter : 5000-5300 A, in an urban site, R or I filter during twilight or near the Moon, but, if possible use the filter designed for the receptor that you use ;
5 - if you are not familiar with the material that you use, take a little more time before the observations to know it
6 - be sure of the identification of the satellites (beware the optical mounting which reverse the field ) ;
7 - determine precisely the size of the diaphragm and what satellites should be in the diaphragm during all the time of the observation (especially for long events) ;
8 - know precisely the motions of the satellites during the events and take into account the refraction when observing low on the horizon ;
9 - take into account the presence of the Moon or of Jupiter to prepare the observation ;
10 - make individual photometric measurements of the satellites before and after the observation ;
11 - measure the sky background in different areas several times during the observation ;
12 - measure the atmospheric absorption thanks to a reference object ;
13 - be careful for the observations during twilight for which a special procedure is necessary.