| dc.description.abstract | The aim of this thesis has been measurement of Stokes profiles and vector magnetic fields in sunspots using the Kodaikanal Tower/Tunnel Telescope and spectrograph. Although this coelostat telescope had been constructed with no polarisation measurements in mind, it has been possible to make such measurements given its limitation of oblique reflections and its advantages of very high spectral and spatial resolution. It is this advantage of high spectral and spatial resolution that made it possible to measure the Stokes profiles and derive both vector magnetic fields as well as physical parameters of the solar photosphere in the region of magnetic concentration.
Starting with an extensive study of the polarisation introduced by the instrument and its effect on the recorded Stokes polarisation profiles, we have attempted the design and construction of a simple polarimeter. This has only a few optical components in the path of a light beam, which minimises the instrumental polarisation. Using this polarimeter, we have photographically recorded the Stokes polarisation profiles in the ?6300 Å region across a few sunspots in an effort to derive both vector magnetic fields as well as the physical parameters across sunspots.
Our next effort has been to establish a regular working procedure for the reduction of the spectral lines into a usable form. One of the major reasons why full?line Stokes polarimetry is not used to produce two?dimensional maps of sunspots and other active regions is the enormous quantity of data to be handled as well as computer time needed. To our knowledge only Kawakami’s work has made any attempt in that direction. The present study has also been motivated to understand the data?handling problems.
From the observed Stokes profiles that have been corrected for the instrumental polarisation, we proceed to recover the vector magnetic fields and other physical parameters through an inversion technique. The inversion technique is still in its infancy, and we are yet to appreciate many of its advantages and understand its limitations. Once again, we have come across only the attempt of Skumanich and Lites (1987) to use this technique and test it on real data. We have attempted to tune the algorithm for better fitting.
On the instrumental front, using photographic emulsion has been a disadvantage. Our test signal?to?noise ratios indicate that the measurement of polarisation is accurate to the order of 2%. One needs to undertake detailed modelling in various situations to translate these quantities into uncertainties of the vector magnetic field and other physical parameters.
The work done here complements the broadband polarisation work done at several observatories, particularly those of the NASA MSFC system as well as the Tokyo Astronomical Observatory.
In principle, crosstalk from V into Q could be made very small in this method and thus make the measurements of umbral fields more reliable compared to the ones obtained using instruments that employ electro?optic modulators. One might expect non?potentiality of sunspot umbrae to have a dominant influence on manifestations of solar activity such as solar flares, eruptive filaments, disparitions brusques. Thus reliable measurement of vector fields in the umbra is crucial for the understanding of these processes. It is for this reason we wish to pursue Stokes polarimetry as a mode of measuring vector fields despite difficulties encountered in both acquisition and interpretation of data.
2. FUTURE IMPROVEMENTS
The background and practical experience gained from this exercise has shown the need for several improvements.
2.1 The polarimetric system:
In order to record polarisation spectral line profiles of Zeeman?sensitive lines in the solar spectrum anywhere between 3000 to 7000 Å, it is necessary to redesign the polarimeter. The major changes envisaged include:
a) replacing the existing ?/4 plate (6302 Å) with an achromatic rotating quarter?wave plate also known as the Pancharatnam plate. The rotation of the achromatic ?/4 plate is intended to chop the polarisation beam and thus to improve the signal?to?noise ratio,
b) replacing the existing polaroid with a Glan–Thompson polariser, as polaroids are not 100% efficient. We also intend to keep the polariser fixed so as to always have linear polarisation of the output beam with the electric vector vibrating along the direction of the grating rule to exploit the full efficiency of the grating.
2.2 The detector system:
In conjunction with what has been mentioned in the previous section, the next obvious goal is to have a better two?dimensional detector system, preferably a charge?coupled device, to improve the quantum efficiency, reduce noise and to increase time resolution. This would help in achieving better polarimetric accuracies.
2.3 Reduction of scattered light and improvements to telescope seeing:
A system of baffles needs to be designed both at the telescope as well as the spectrograph to prevent spatially scattered light reaching the telescope image plane and to prevent spectrally scattered light within the spectrograph contaminating the recorded spectrum.
2.4 A non?linear seeing monitor cum image?guiding system
hooked to the telescope to enable a raster?scanning facility at image plane.
2.5 Auxiliary information in the form of H? spectroheliograms
at the same spatial resolution with near simultaneity is imperative to resolve the 180° ambiguity of the vector field.
2.6 The existing algorithms for the non?linear least?square fitting
need to be modified because of its slow convergence. It may be necessary to search for better inversion techniques compared to the Marquardt algorithm. | |
