-------------------------------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------------------------------
Two months after the solar eclipse, here is a first evaluation of the network scientific output.
In the last few weeks before the total solar eclipse (TSE), 50 rolls of Kodak Tmax 400 black & white film from a single batch were pre-exposed in the laboratory, using a reference step-density scale (5 exposures). This pre-calibration allows the compensation of variations in the effective response curve of the films due to the inevitable inaccuracies of the subsequent chemical development process. Those films, together with pieces of linear polarizing film (Edmund Scientific TechSpec cat#D45,204), were sent to all photographic stations, thanks to funding from the Royal Observatory of Belgium.
At the time of the TSE, 28 stations took part in the coordinated observations. They were set up by amateur (17 teams) and professional observers (11 teams) of 16 different nationalities. As several groups organized several observing sites, polarimetric instruments were set up on 30 different observing sites, with notable concentrations in Hungary and France.
By mid-September 1999, most of the photographic films had been collected and developed at the ROB using a uniform standard procedure, and most of the observing logs and reports had been received. Twelve stations were clouded out, mainly because of the unfavorable weather prevailing over Western Europe. Among them, the only station in India was also unable to observe the eclipse because of the bad weather (rain season). Three additional teams did not bring back data for technical reasons (instrument failure during totality, manipulations errors due to a lack of experience). It is interesting to note that the lower success rate of amateurs astronomers, as compared with professional teams, is only due the weather circumstances. Indeed, the amateurs were more concentrated in the West, where the clouds were more abundant, while professional astronomers avoided that unfavorable region.
On the other hand, out of the 29 stations checked so far, 14 stations produced useful data. About half of those data sets can be rated as excellent (good photometric quality, complete image set). Despite of the 50% overall success rate, the successful observations are evenly distributed along the totality band. This well balanced geographical distribution ensures a uniform temporal coverage over a 1h 40min duration, between 10:25 and 12:05 UT.
Graph 1 : statistics of the participating countries and of the observing sites.
Graph 2 : quality distribution of the resulting data
Graph 3 : geographical distribution of the TECONet sites
Obslist : detailed list of observing teams
All photographic films have been developed using the same standard development procedure. In a subsequent step, the films must be scanned to convert them in numerical form in view of further processing. Four stations used CCD cameras and will take care of the proper pre-processing of the digital images. This work is now in progress, but the full analysis process will probably take several months.
In central Romania, the Belgian team set up by the ROB successfully observed the corona using a CCD system. This experiment is based on video CCD camera (740x580 pixels) with 8-bit digitization, placed a the focus of a 200mm, f/2.8 telelens. A motorized rotating filter is placed in front of the optics, and an optoelectronic sensor triggers exposures at 15-degree angular steps. This system collected 120 images in 90 seconds: 5 exposure levels, and 2x12 angles in each polarization sequence. These images were corrected for the dark current, for the flatfield and for non-linearities in the response. They were then co-registered and intensity-rescaled before being composited. The full image set, which includes the whole intensity range between the limb and the sky background at about 3 solar radii, was then processed to extract the Stokes parameters of linear polarization. We used a procedure that was already applied to several previous TSEs (Gabryl et al. 1998, Clette 1994}). The results shown here are preliminary, as several corrections to the raw images must still be refined (detector blemishes, optical artifacts, intensity scale)
Image 1 : greyscale image of the total intensity (logarithmic intensity scale, composite of 120 images)
Image 2 : contour plots of the polarized brightness (K corona, x2 steps)
Image 3 : contour plots of the degree of polarization (5% steps)
(Image credit: JOSO/TECONet99, Royal Observatory of Belgium)
Overall, the coronal intensity distribution is fairly symmetrical, as expected for a period of high solar activity. A lower brightness is present between position angle (PA) 170 deg. and PA 230 deg, apparently marking the position of the South pole of the global heliospheric magnetic field. Some strongly non-radial streamers emerge from the North polar region (PA 30deg. and 320deg.). A comparison with LASCO C2 images shows that these structures extend further out with the same inclination. The deviation, away from bright loop structures around PA 55deg., thus occurs entirely below 2 radii.
The polarization reaches a maximum of 45% in the broad streamer at PA 165deg. At its base, this prominent high-latitude streamer features a bright prominence imbedded in a very dark cavity. A narrow radial strip at PA 325 deg. is marked by a remarkably low polarization, as well as a low brightness. It seems to correspond to a narrow fan of open field lines, which seems to be overlying a long filament channel belonging to the "polar crown". Overall, the measured polarizations are a bit low for a period of strong solar activity. This might indicate that most features seen above the limb were emerging from the front or far sides of the Sun. This would also explain why long streamers extending beyond 3 or 4 solar radii were notably absent from this corona. At this stage however, it is still difficult to tell if such a geometry was responsible for a steeper radial intensity gradient in the corona or if there was an unusually bright sky background. The absolute intensity calibration included in the TECONet observations should help us to conclude on this issue.
Due to the bad weather in Western Europe, the success rate (station count), proved to be a bit lower than expected. Still, the primary goals have been achieved: namely, overcoming the bad weather and breaking the "2-min barrier" of the totality observed from a single site.
After the eclipse, we also realized two other contributions from this network. The project helped to introduce the JOSO to the world, beyond the European context and the professional circles. It attracted the attention of the media in Europe and the USA, and lead to several interviews and to articles in newspapers and magazines.
Moreover, this joint observing project, organized in an international context, proved to be a unique experience for all participants, in particular for amateur astronomers. The observations were often prepared and discussed collectively in amateur associations, and thus involved the participation or interest of many more people than the observers themselves. As a matter of fact, many of them already volunteered to take part in a future eclipse observing network. Therefore, the experience and the personal relations established over that period proves to be an investment for the future, as they provide the base for any future coordinated project of this kind.
Regarding the future of TECONet, a critical phase of the project is now just beginning: namely, the analysis of this fairly large image collection. The first task will consist in digitizing all the negatives produced by TECONet. All numerical images (photo and CCD) will then be submitted to a global analysis, in four successive steps: 1) Pre-processing: cosmetic corrections (blemishes, flatfield, vignetting) and response linearisation. 2) Computation of polarized brightness maps for each station (different procedures will have to be developed to accommodate some differences in the executed sequences) 3) Multi-station intercomparisons: intensity cross-calibration, coronal evolution 4) Determination of electron density distributions (global and local) and of density variations.
As noted already during the TECONet development phase, an analysis team must be constituted to share this extensive work. Steps \#1 and \#2 involve the largest amount of work, and are the ones that require wider support. The final steps are more attractive, as they deliver the scientific results, but are less demanding.
Now that it is available, this original multi-site polarimetric image set will certainly give rise to an additional interest from the solar physics community. We thus hope that the TECONet project will attract new collaborators wanting to contribute to the analysis phase that already looks promising. ALL VOLUNTEERS WANTING TO CONTRIBUTE TO THIS RESEARCH PROJECT ARE WELCOME TO JOIN THE TECONET COMMUNITY.
First of all, we would like to thank all the teams and individuals who took part in this project and invested their time and resources for the preparation and execution of the observing program. The success of TECONet relies entirely on their enthusiasm and energy. We are also grateful to the Royal Observatory of Belgium, in particular to its Director, Professor Paul Paquet, who supported the development of TECONet, materially and financially.
Dr. Frederic CLETTE
TECONet Coordinator
Observatoire Royal de Belgique
Avenue Circulaire, 3
B-1180 Bruxelles
Belgium
tel : ../32/(0)2/373.02.33
fax : ../32/(0)2/373.02.24
e-mail : F.Clette@oma.be
fclette@solar.stanford.edu
--------------------------------------------------------------------------------
IMAGE CREDIT: JOSO/TECONET99, ROYAL OBSERVATORY OF BELGIUM
--------------------------------------------------------------------------------
F. Clette, J.-R. Gabryl
Royal Observatory of Belgium
1999/10/12
--------------------------------------------------------------------------------
