3. Data analysis and results
For the first time, Jupiter radio emissions in the frequency range 1-40 MHz have been observed from ground and space during the same period with comparable geometry. The NDA covers the range 10-40 MHz and the Wind/WAVES experiment records the lower part of the spectrum up to 13.8 MHz. For our study we have used the data from March 21 to July 08 in 1995 and from April 24 to August 12 in 1996. During those two periods the ground-based observations are performed mainly during the night and thus are less contaminated by man-made interference. The Jovicentric declination of the Earth and Wind spacecraft as seen from Jupiter lies between -3.4o and -2o for 1995 and 1996, respectively. Those values of DE are quite different from those recorded before and after Voyager encounter where DE was about +3o. Due to the fact that Wind spacecraft is at a distance from Earth smaller than 1.5 106 km the observation conditions (i.e. the observation time and geometrical corrections) are similar in the space and on the ground.
The observations we report in our analysis are performed during the two quoted periods which correspond to a total of 221 days. We selected these periods because the meridian transit of Jupiter at Nançay (05:00 to 22:00 UT) is mainly during the night when the observation conditions are better. The NDA tracks Jupiter during 8 hours around the meridian time transit while Wind/WAVES observes continuously. In order to compare the two sets of data, we restricted Wind/WAVES's records to the interval of time when the NDA was observing which results in 1760 common observational hours. We establish a database which allows to characterize the observational parameters of each event: the beginning and the end of emission time, the frequency range, the main polarization, the arc curvature and a label from 0 to 2 for increasing intensity. A list of events had been first established and used by Kaiser & Garcia (1997) for each day associated to nearly the same period. The Wind/WAVES data had been reinvestigated to give the same informations than the NDA list during the same eight hours every day for all the events, even the faint ones, which were neglected by Kaiser & Garcia (1997).
3.2. Occurrence probability along the day
From the daily-8 hour observations performed by the NDA, in 1995 and 1996, we selected only 221 days. We rejected the data acquired when the meridian transit of Jupiter occurred during the day between 5UT and 21 UT. Thus we kept two periods (March 21, 1995 to July 8, 1995) and (April 24, 1996 to August 12, 1996) which correspond to 1760 hours of observations. During those periods, the meridian transit of Jupiter shifts each day by about 4 minutes. Fig. 1 shows the number of observations versus the universal time (UT) as covered by the NDA radiotelescope, each bin corresponding to half an hour. For the period considered in our study the number of observations occurrence per half-hour is greater than 100 when Jupiter is tracked between 21-05 UT. During this daily observation we recorded 221 events with the NDA corresponding to 216 hours. The duration of each event is highly variable from half an hour to a few hours with the NDA; the average duration of each event is about one hour. The Wind spacecraft recorded 252 events with a variable duration as in the previous case. The total duration of these jovian emissions recorded by Wind spacecraft below 13.8 MHz is 410 hours, leading to an average event duration of 1h 40 min. In the following we define the occurrence probability as the ratio between the emission duration and the observing duration for each half hour.
The occurrence probability is found to be about 38 for the NDA (Fig. 2a) and Wind/WAVES (Fig. 2b) at the maximum of detection. This maximum occurs between 3 and 5 UT for the ground-based antennas whereas the occurrence evolves only slowly along the morning for Wind/WAVES observations. In the afternoon the occurrence probabilities are very different due to the great importance of man-made interference at Nançay station. Even if small percentage of detection is still possible, the efficiency of the NDA is smaller in the afternoon. However the NDA is more sensitive than the WAVES/Wind experiment but several facts diminish its capability. The attenuation produced by the ionosphere in the lower part of the spectrum, and the difference in the jovian emissions in the lower and higher parts of the spectrum as observed by each receiver. The result is an apparent equivalent capability in the early morning when the lowest frequencies are partly detectable at Nançay station.
3.3. Occurrence probability in the (, CML) plane for all events
We organised the data in the usual coordinates defined as function of the Io phase () and the central meridian longitude (CML). Fig. 3 represents the sampling of the jovian plane (, CML) with NDA observations. The case of Wind/WAVES would lead to a better coverage of the plane but since we restrict the data to the same daily eight hours, as for the NDA, the same slight modulation exists. However, the projection along the two axes is nearly uniform and thus it does not sidestep the issue. The diagram indicates a quite regular sampling of the plane. One can see that the A region as well as a part of the B region have been less sampled. We have to keep this in mind when considering the plane (, CML) as shown in Fig. 3. The (, CML) diagram of the emission recorded by Wind/WAVES and by the NDA are shown in Fig. 4 and Fig. 5, respectively. One must note that the appearance of the two diagrams are very different because the two receivers record events in two complementary frequency ranges. Even if the two bands superpose on 4 MHz, the occurrence in the (, CML) diagram shows that we are partly dealing with different components of the jovian emission.
In Fig. 4 one can see that it is not possible to separate between the occurrence areas of controlled and not controlled sources observed by Wind/WAVES. However the Io-C and Io-B-D areas are slightly visible whereas the AA' sources are not visible. The occurrence probability as a function of Io-phase shows two major peaks recorded when Io-phase is equal to 95o and 250o. Each peak is the contribution of controlled and not-controlled emissions. The non-Io controlled emissions have a probability of occurrence of about 20 for all the values of . For two narrow ranges of values the occurrence increases up to 85 () and to 72 () corresponding mainly to the Io-C and Io-D/B, respectively. The occurrence probability evolves slowly in CML. It is found about 20 for all CML values and exceeds 30 when 100o CML 185o or 320o CML 10o. The maximum value is obtained when CML.
The occurrence in the NDA data (Fig. 5) is about 15 for all the values of . This occurrence reaches 44 when and 46 when . Two small maxima are visible: 28 () and 22 (). On the diagram one can localise the Io-A/A', C and B/D sources. The maximum around is unusual whereas the other bumps results from D, B, A' and A/C sources. The NDA occurrence in CML is in average 10. It reaches 32 when CML due to the contribution of non-Io-AA' emissions. It is around 20 in the CML range 120o-180o. The polarization occurrence probabilities obtained with the NDA are shown in Fig. 6a and 6b for the right-hand and left-hand polarizations, respectively. The main emissions with right-hand polarizations are associated to Io-B region and to another area with a CML between 200o and 270o corresponding to Io-A/A' and non-Io-A. The left-hand emissions have lower occurrence probability and they appear in Io-controlled (Io-C and Io-D) and non-Io-controlled (non-Io-C) regions. It is important to note the absence of Io-C emission with right-hand polarization (see Fig. 6a), only the left-hand component is observed (see Fig. 6b) contrary to previous studies (e.g. Boudjada & Genova 1991). The slight enhanced occurrence around appears on both polarization diagrams.
From the analysis of the occurrence probabilities in the (, CML) diagram for the NDA and Wind/WAVES it appears that the two diagrams present some differences. The non-Io controlled emissions observed by Wind/WAVES are observed less often with the NDA. The occurrence areas of controlled emissions (Io-A/A' and Io-B) appear more clearly from ground observations than from space.
© European Southern Observatory (ESO) 2000
Online publication: February 25, 2000