Observations of AB Doradus, a nearby, rapidly-rotating K0 dwarf are analysed, and the surface magnetic field is shown to be approximated by a potential field. Evolving the surface magnetic field according to diffusion and the observed differential rotation still yields good correlation between the calculated and observed radial field after 30 days, contradictory to the results of Barnes et al. (1998), leading to the conclusion that there is an additional cause for the evolution of the magnetic field. The chromospheric magnetic field is modelled as a potential field with a source surface. Using the stability criteria g.B = 0 and B.V(g.B) < 0, places where prominences can be stable are investigated. For agreement with the results of Donati et al. (2000)- that prominences form preferentially near the equatorial plane and at and beyond corotation- it is necessary to add a quasidipolar field of maximum strength ~20G.

The search for planets orbiting stars other than the Sun has led to the discovery of over one thousand new worlds. The majority of these planets have been very large, Jupiter sized planets located very close to their host star. Transit surveys such as Kepler and SuperWASP monitor thousands of stars looking for periodic dips in light caused by a planet passing between our view point on Earth and their host star, blocking a fraction of the emitted star light. One of the primary limitations in detecting a small, Earth sized planet comes from stellar activity induced signals within the data collected by exoplanet missions. These signals can, however, be used to our advantage. In this thesis, asymmetries in transit light curves are exploited to reveal properties of both the planet and the host stars themselves. An asymmetry in the near-ultraviolet transit light curve of WASP-12b, one of the largest and hottest planets found to date is thought to be caused by the stellar wind interacting with the magnetic field surrounding the planet. In this thesis, a model for such an interaction is developed and is shown to be consistent with the observations, providing the first potential evidence for the presence of a magnetic field around an exoplanet. The model is then extended to predict the shape of near-ultraviolet light curves around HD 189733b, another hot Jupiter that orbits a very bright star. By looking at the variability in these transit light curves over time, the evolution and structure of the stellar wind is investigated. By tracking the position of bumps in the transit light curve, it is shown here that the data collected by missions such as Kepler has the potential to reveal stellar butterfly patterns. Such patterns are intrinsically linked with the stellar dynamo which governs the properties of the stellar magnetic field. Finally, the support of large-scale magnetic loops on young stars is investigated. These loops trap large amounts of hot, dense material and so a rapid destabilisation could lead to a flaring event, which could have devastating consequences for a nearby exoplanet.

In low-mass (≲1.2 M⊙) main-sequence stars, the combination of differential rotation and turbulent flows in the outer convective region generates strong magnetic fields. It has been observed that in these stars, the rotation rate and the strength of the magnetic field decrease over time. This is thought to result from a feedback loop in which magnetized winds carry angular momentum away from the star, braking its rotation and weakening the magnetic dynamo. A well-calibrated age-rotation-activity relation (ARAR) would be particularly valuable for low-mass stars. If we knew the dependence of rotation or magnetic activity on age, a measurement of one of these quantities could be used to determine an accurate age for any isolated field star. Empirical calibrations of the ARAR rely on observations of the co-eval populations of stars in open clusters. In this work, I characterize rotation and magnetic activity, using light curves for the former and X-ray and Hα emission for the latter, in two open clusters of different ages (Alpha Persei, ≈60 Myr, and Messier 37, ≈500 Myr) to analyze the relation between rotation and activity across the low-mass stellar range. I also compare coronal (X-rays) and chromospheric (Hα) activity to understand how magnetic heating varies across stellar atmospheric layers. My results inform models of angular momentum evolution in low-mass stars.

The oldest open clusters within 250 pc of the Sun, the Hyades and Praesepe, are important benchmarks for calibrating stellar properties such as rotation and magnetic activity. As they have the same age and roughly solar metallicity, these clusters serve as an ideal laboratory for testing the agreement between theoretical and empirical rotation-activity relations at ~650 Myr. The re-purposed Kepler mission, K2, has allowed me to measure rotation periods for dozens of Hyads and hundreds of Praesepe members, including the first periods measured for fully convective Hyads. These data have enabled new tests of models describing the evolution of stellar rotation; discrepancies with these models imply that we still do not fully understand how magnetic fields affect stellar spin-down. I show how we can compare the dependence of H-alpha and X-ray emission on rotation in order to test theories of magnetic field topology and stellar dynamos. These tests inform models of stellar wind-driven angular momentum loss and the age-rotation-activity relation. I also present rotation periods measured for 48 Hyads and 677 Praesepe members with K2, and discuss the impact of unresolved binaries on the study of rotational evolution.

Mercury-manganese (HgMn) stars belong to the class of chemically peculiar (CP) stars. It was recently discovered that some HgMn stars have spots of chemical elements on their surfaces. According to conventional picture of CP stars, magnetic field facilitates the formation and long term stability of chemical spots by controlling stratification of elements in stellar atmosphere. However, previous attempts to find magnetic field in HgMn stars set an upper limit on its strength at the level of about 20-100 Gauss. Observational evidence suggested that even weaker magnetic fields can be responsible for the formation of chemical spots. The main goal of our work was to verify this possibility. The search for weak magnetic fields requires the use of least-squares deconvolution (LSD) technique.  This method combines information from many spectral lines providing a mean line profile with increased signal-to-noise ratio. Up to now there was no extensive comparison of the LSD profile with real spectral lines. We showed that the LSD profile of the intensity spectrum does not behave like a real spectral line as a function of chemical composition. However, for circular polarization, LSD profile resembles the profile of a spectral line with mean atomic parameters. We performed a comprehensive search for magnetic field in 47 HgMn stars and their companions, based on high-quality spectropolarimetric data obtained with the HARPSpol polarimeter at the ESO 3.6-m telescope. With the help of LSD technique, an upper limit on the mean longitudinal magnetic field was brought down to 2-10 G for most stars. We concluded that magnetic field is not responsible for the spot formation in HgMn stars. We obtained full rotational phase coverage for the HgMn stars φ Phe and 66 Eri. This enabled us to investigate line profile variability, reconstruct surface maps of chemical elements, and perform a search for magnetic field with very high sensitivity. For φ Phe we derived surface maps of Y, Sr, Ti, Cr, and obtained an upper limit of 4 G on the field strength. We also found marginal indication of vertical stratification of Y and Ti. No magnetic field was detected in both components of 66 Eri, with an upper limit of 10-24 G. We discovered chemical spots of Y, Sr, Ba, and Ti, in the primary star. We demonstrated a relation between the binary orbit and the morphology of these spots.

Supergiant stars of spectral types B and A are characterized by variable and structured winds, as revealed by variability of optical and ultraviolet spectral lines. Non- radial pulsations and magnetically supported loops have been proposed as explanations for these phenomena. The latter hypothesis is tested using a time series of 65 high-resolution (λ/∆λ ∼ 65, 000) circular polarization (Stokes I and V ) spectra of the late B type supergiant Rigel (β Ori, B8 Iae), obtained with the instruments ESPaDOnS and Narval at the Canada-France-Hawaii Telescope and the Bernard Lyot Telescope, respectively. Examination of the unpolarized (Stokes I) spectra using standard spectral analysis tools confirms complex line profile variability during the 5 month period of observations; the high spectral resolution allows the identification of a weak, transient Hα feature similar in behaviour to a High Velocity Absorption event. Analysis of the Stokes V spectra using the cross-correlation technique Least Squares Deconvolution (LSD) yields no evidence of a magnetic field in either LSD Stokes V profiles or longitudinal field measurements, with longitudinal field 1σ error bars of ∼ 12 G for individual observations, and a mean field in the best observed period of 3 ± 2 G. Synthetic LSD profiles fit to the observations using a Monte Carlo approach yield an upper limit on the surface dipolar field strength of Bdip ≤ 50 G for most orientations of the rotational and magnetic axes, lowered to Bdip ≤ 35 G if the mean LSD profile from the most densely time-sampled epoch (with an LSD SNR of ∼80,000) is used. A simple two-spot geometry representing the footpoints of a magnetic loop emerging from the photosphere yields upper limits on the spot magnetic fields of 60–600 G, depending on the filling factor of the spots. Given existing measurements of the mass loss rate and the wind terminal velocity, these results cannot rule out a magnetically confined wind as, for Bdip ≤ 15 G, η∗ ≥ 1. However, the detailed pattern of line profile variability seems inconsistent with the periodic wind modulation characteristic of known magnetic early-type stars, suggesting that magnetic fields do not play a dominant role in Rigel’s variable winds. / Thesis (Master, Physics, Engineering Physics and Astronomy) -- Queen's University, 2012-04-29 02:10:41.308

No description available.

The Hanle effect has been proposed as a new diagnostic of circumstellar magnetic fields for early-type stars, for which it is sensitive to field strengths in the 1-300 G range. In this paper we compute the polarized P Cygni line profiles that result from the Hanle effect. For modeling the polarization, we employ a variant of the "last scattering approximation." For cases in which the Sobolev optical depths are greater than unity, the emergent line intensity is assumed to be unpolarized, while for smaller optical depths, the Stokes source functions for the Hanle effect with optically thin line scattering are used. For a typical P Cygni line, the polarized emission forms in the outer wind, because the Sobolev optical depth is large at the inner wind. For low surface field strengths, weak P Cygni lines are needed to measure the circumstellar field. For high values of the surface fields, both the Zeeman and Hanle diagnostics can be used, with the Zeeman effect probing the photospheric magnetic fields and the Hanle effect measuring the magnetic field in the wind flow. Polarized line profiles are calculated for a self-consistent structure of the flow and the magnetic geometry based on the WCFields model, which is applicable to slowly rotating stellar winds with magnetic fields drawn out by the gas flow. For surface fields of a few hundred gauss, we find that the Hanle effect can produce line polarizations in the range of a few tenths of a percent up to about 2%.

In this thesis I study the magnetic activity of young stars via observations of stellar prominences on single stars and by applying the Zeeman Doppler imaging (ZDI) technique to map the magnetic fields and measure differential rotation of a young binary system. Stellar prominences can be observed as absorption transients in the rotationally broadened chromospheric lines of rapidly rotating stars. Observations of Speedy Mic(K3V) reveal a densely packed prominence system at heights far above the stellar co-rotation radius. Further observations were used to estimate prominence column densities and masses. From very high signal-to-noise observations, loops of emission are found that trace the path of prominences seen transiting the stellar disc. I also present what appears to be the first observation of an erupting stellar prominence on AB Doradus (K0V). I modify an existing ZDI code so that it can recover the magnetic maps of a binary system. The new code is applied to observations of the pre-main sequence binary system HD 155555 (G5IV+K0IV). The radial magnetic maps reveal a complex surface magnetic topology with mixed polarities at all latitudes and rings of azimuthal field present on both stars. The evolution of the relative field strengths between observations in 2004 and 2007 could be indicative of a magnetic activity cycle. I adapt the sheared image technique for measuring differential rotation parameters to the binary case. Both stellar components of HD 155555 are found to have rates of differential rotation similar to those of the same spectral type main sequence single stars. This is in apparent conflict with previous work on evolved binary systems where low rates of differential rotation were found, leading to the suggestion of suppression by binary tidal forces. I find that the depth of convection zone alone can likely explain the differential rotation results without invoking tidal forces.

No description available.