Astron. Astrophys. 342, 867-880 (1999)

2. Observations

On September 1996, a quiescent filament channel was observed in the decaying remnant of AR 7986. The observations were obtained during a coordinated campaign between space instruments (SOHO and Yohkoh) and ground-based instruments (the German VTT and GCT at the Teide Observatory, Tenerife, and the SVST at La Palma), in the context of the Joint Observing Program JOP 17 concerning the study of the "Dynamics of Solar Active Region". In this paper we only use H observations obtained with the MSDP instrument mounted on the VTT, as well as SOHO/MDI line-of-sight magnetograms taken on September 25.

2.1. MSDP observations in H

The details of MSDP data reduction is described in Mein (1991). Nine images of the same solar area were recorded simultaneously in nine wavelengths of the H profile on a 10241024 CCD camera. By using 200 successive steps across the solar disk, we scanned two strips of 170600 arcsec² every 34 min. The elementary fields of view are overlapping. They are joined by 2-D cross-correlation calculations. The pixel size is 0.43" for the recording, and 0.25" for the output maps. The MSDP also observed simultaneously with H the selected strips in the y line (Ca II, 8542 Å), of which dotted brightenings are relevant of the underlying magnetic network.

The H profile is restored at each point of the field. A special care is taken to correct the scattered light in the spectrograph. Dopplershifts and intensity fluctuations are derived at different distances from line center. The H images of the filament channel presented in this paper are computed at 0.25 Å from line center.

2.2. SOHO/MDI magnetograms

MDI is one of twelve experiments onboard the SOHO spacecraft, which was launched on December 2, 1995. The Solar Oscillations Investigation (SOI) uses the Michelson Doppler Imager (MDI) instrument to probe the interior of the Sun by measuring the photospheric manifestation of solar oscillations (Scherrer et al., 1995). The instrument images the Sun on a 10241024 CCD camera through a series of increasingly narrow filters. The final elements, a pair of tunable Michelson interferometers, enable MDI to record filtergrams with a FWHM bandwidth of 100 mÅ. Several times each day polarizers are inserted to determine the line-of-sight magnetic field. In this paper we are using magnetograms of the full-disc mode with a 96-minute cadence, with a pixel size of 2". Some maps are made using a five-minute average of the polarization signal, while others are merely "snapshots".

The noise in the magnetograms originates from two main components: photon shot noise and leakage from the Doppler signal. Single magnetograms are limited by the shot noise to about 10 Gauss and five-minute average maps to about 3 Gauss. That is because over the five-minute interval most p-modes execute almost one complete cycle, so the P-mode leakage is smaller. There is some noise at about the 2-3 Gauss level from granulation as well (DeForest, private communication).

In Table 1 we list the magnetograms available during the MSDP observing hours on September 25 1996. Unfortunately, during the morning hours, when the seeing is the best on Tenerife, two magnetograms were lost presumably due to telemetry problems. We match the magnetograms with the MSDP image taken at the closest in time.

Table 1. List of the SOHO/MDI magnetograms and MSDP H data used in this paper (observed on September 25 1996). It must be noted that the times of H data and MDI magnetograms correspond to the beginning of the scans (of which duration is a few minutes).

2.3. Co-alignment of the MDI and MSDP data

Since the MSDP instrument has a partial field of view and no pointing information, the co-alignment of the MSDP and MDI images is not trivial, thus we need to use additional observations, such as full-disc H images and y line (Ca II, 8542 Å) MSDP images to place the partial MSDP H image on the solar disc. We rotated a full disc Meudon H spectroheliogram taken at 14:35 UT to the times of each MDI magnetograms and co-aligned the full-disc images. Then we co-aligned the full disc and partial field H observations and "fine-tuned" the co-alignment matching relevant network brightenings observed on CaII MSDP images (taken coincidentally and co-spatially with the H images, see Sect. 2.1) and magnetic polarities from the MDI magnetograms. In order to do so, we performed feature-tracking techniques in a large field of view (about 300600 arcsec²). We estimate that the mis-alignment is not larger than 4" (two MDI pixels).

We then located the observed straight inversion line using a large field of view (about 300600 arcsec²), and for viewing reasons we finally rotated the co-aligned MDI magnetograms and MSDP images to place the inversion line vertically and at the center of the considered fields of view (see Fig. 1d). Thus, in all the figures of this paper, the true North is 72^o to the left from up.

© European Southern Observatory (ESO) 1999

Online publication: February 23, 1999
helpdesk.link@springer.de