Astron. Astrophys. 319, 664-668 (1997)

2. Experiment and observation

The observation was carried out on July 11, 1991 at UT in Mexico, Baja California, La Paz ( = - 15' 27", = + 09' 24", h = 10 m) under excellent weather conditions. An achromatic Zeiss refractor, 130 mm in diameter and 1950 mm in focal length was used to realize the experiment. An ultra-narrow-band 0.17 nm tunable image quality interference filter 002FC10-50 with a temperature control (produced by the Andover Corp., USA) was fitted at the eye-piece end of the telescope to isolate radiation of the 530.3 nm spectral line. A rotatable polarizer (fixed successively to four positions differing by in orientation of its plane of polarization) was placed just in front of the interference filter. This design with no really reflecting surfaces, suppressed any instrumental polarization which was then, in fact, not measurable.

A sixty-millimetre Kodak Tri-X Pan roll-film was used to record photographic images of the linearly polarized light of the green-line corona. One set of four images was taken with exposures of 30 s during the first half of totality (the whole eclipse lasted almost 6.5 minutes in La Paz). Besides this, three sets (exposures 1/125 s, 1/15 s and 1 s) of four white-light polarized corona pictures were taken. (Our paper on the white-light corona polarization has been submitted for publication (Badalyan et al. 1996)).

All the original negatives were subjected to microdensitometer image processing at the Ondejov Observatory (Zicha et al. 1992). We chose a 50-micron pixel dimension to record densities. This, about 5 arcsec, resolution was maintained throughout the whole data analysis. Before calculating the intensities, degree and direction of polarization, a crucial requirement had to be fulfilled: as perfect a fitting as possible of the four images of the given set, taken at different positions of the polarizer. We simply had to be sure that physically the same point of the corona was extracted from the corresponding four images for subsequent calculations. This particular procedure was dealt with and realized for us in the Institute of the Problems of Information Transmission in Moscow. The white-light images were relatively easy to identify using well visible prominences at the east and west solar limbs. Subsequently, the green-line images were identified using common characteristic structural details seen both in the green-line and white-light corona pictures. We have currently achieved one pixel accuracy in the process of identification of the images.

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998
helpdesk@link.springer.de