Catálogo de publicaciones

Many open questions in solar physics demand the acquisition of spectropolarimetric measurements of solar light, involving challenging requirements in terms of spatial, spectral and temporal resolution, and polarimetric sensitivity. There are important motives to try acquiring such measurements from the ground instead of using a space-based observatory, e.g. the larger available apertures. However, a major impediment to reach the requirements are the effects of atmospheric seeing. The most important effects are the reduction of spatial resolution, and the introduction of spurious polarimetric signals (denoted as seeing induced crosstalk, SIC) caused by variable seeing induced image aberrations, and the fact that polarimetric measurements are based on time differential photometry. Standard numeric techniques used to correct for seeing induced image aberrations, require short exposure times (<10 ms). In addition, a fundamental solution to reduce SIC is to increase the modulation frequency of the polarimeter beyond the values defined by the typical seeing coherence time (>100 Hz). The two aforementioned properties can be met by a high-cadence polarimeter. This thesis describes the development of such a polarimeter and presents the results obtained with a specific instrument, namely the prototype of the Fast Solar Polarimeter (FSP). FSP is based on a high-cadence (up to 800 fps) camera and a polarization modulator with ferro-electric liquid crystals (FLCs). The camera is custom-made, using a frame-transfer, back-illuminated pn-type CCD (pnCCD) sensor, that has high sensitivity, and provides almost 100% duty cycle. We have characterized and integrated the components of the modulator package, using a model-based optimization procedure to derive the position angles and retardances of the components that minimize the dispersion of the polarimetric efficiencies. We have obtained a variation of 20% of the total polarimetric efficiency, in the 400 to 700 nm wavelength range. In addition, the characterization results of the integrated modulator package allowed us to derive basic calibration and operational requirements. We have also characterized the pnCCD camera and we have proven that it is suitable to be used for high-cadence, high-sensitivity polarimetry. This includes selecting a specific synchronization between camera and modulator ?which allowed us to develop a numeric technique for correcting image smearing post-acquisition? and addressing the effects of non-linearities, residual offsets, common-mode noise and readout noise of the camera. The FSP instrumental concept was validated during five campaigns carried out in the years 2013 to 2015 at the 68 cm German Vacuum Tower Telescope of the El Teide observatory on Tenerife. We have measured the second solar spectrum of Ca I 422.7 nm with a polarimetric sensitivity of 8 105, and we have detected sub-arcsec bipolar transversal Zeeman signals of the quiet Sun at the 1x10e-3 level, using the TESOS filtergraph tuned to the Fe I 630.2 nm spectral line. In addition we have found that the high cadence and duty cycle of FSP allows the restoration of the Stokes images by means of a multi-object multi-frame blind deconvolution, using only the narrow-band science data, consisting of individual short exposure (2.5 ms) frames with low signal to noise ratio. We have studied the residual level of seeing induced crosstalk (SIC) using non-modulated filtergraph measurements. At a modulation frequency of 100 Hz, if no image restoration is used, SIC is below the 7x10e-5 noise level after averaging ~8 min of quiet-Sun Stokes images. On the other hand, after restoring 1.16 min of the same measurements, we found traces of SIC at the noise level of 4x10e-4 only in the edges of the images, where the performance of the adaptive optics and image restoration are reduced due to seeing anisoplanatism. The techniques and methods studied and/or developed in this work are of relevance for any project addressing the development of high-cadence, high- sensitivity imaging polarimetry.