No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

In this thesis, the radio spectral properties of the faint sources in two deep extragalactic radio survey areas are investigated to aid in the discrimination between AGN and starbursts. . The detailed 4-point radio spectra of two X-ray bright, optically quiescent sources in the 13H XMM/Newton/ Chandra survey area are measured, and found to be flat in one case, and steep (0: rv -0.6) in another. Deep imaging of both the 13H and 1H XMM/Newton/ Chandra survey fields at 610 MHz with the GMRT are presented. To detection limits of rv320 and 300t.tJy beam-I, 206 and 223 sources are discovered in the 13H and 1H fields respectively. A deep 1.4 GHz survey of the,IH field is performed with the VLA. To a detection limit of 45 p,Jy, 432 soUrces are discovered. . The source counts at both 610 MHz and 1.4 GHz agree with those in other deep fields, and can both be fit with similar models consisting of standard AGN source count models combined with a population of steep spectrum (0: = -0.7) starburst galaxies undergoing power law evolution with an index of 2.5 and 2.6 at 1.4 GHz and 610 MHz respectively. The catalogues of these three surveys are combined with the results of an earlier deep 1.4 GHz survey of the 13H field to calculate useful radio spectral indices for 192 faint radio sources. Assuming that starburst galaxies exhibit spectra within -1 < 0: < -0.4, 83 sources can be identified as AGN on radio spectral grounds.

X-ray binaries are binary star systems in which a compact object (a neutron star or a black hole) and a relatively normal star orbit a common centre of mass. Since the discovery of X-ray binaries with the first X-ray telescopes in the 1960s, astronomers have tried to understand how these bizarre objects behave, and why. Some change in X-ray luminosity by eight orders of magnitude on timescales of days to months due to an increased transfer of mass from the star towards the compact object. Many X-ray binaries are detected at all observable frequencies, from radio to ,-rays. It has been found that many different sources of emission, which peak at different frequencies, are present in X-ray binary spectra and together they produce the observed broadband spectrum. However, disentangling these components has proved challenging. Much of the work in this thesis concerns disentangling the components that occupy the optiCal and near-infrared (NIR) region of the spectrum of X-ray binaries; possibly the region in which the relative contributions of the different components are least certain. In particular one component, the synchrotron emission from jets of outflowing matter, is found in this work to contribute ubiquitously to the optical and NIR light of X-ray binaries with relatively faint stars. These results confirm that the jets are powerful and in some of this work, observations of the jets interacting with the surrounding matter are used to infer their power. An introduction to the subject matter is presented in Chapter 1. Attention is given to the current thinking of the dominating optical and NIR emission processes, and to X-ray binaries that produce jets. In Chapters 2-6 I present a number of investigations into optical and infrared observations of X-ray binaries. Relations, patterns and general trends are discovered that have implications for not only the dominating emission processes but also for .the physical conditions and general behaviour of the inflowing and outflowing matter. In Chapter 7 I summarise the results and discuss follow-up work that could further our understanding of these objects.

The source of the semi-annual variation in geomagnetic indices is investigated with regard to its possible origins in the solar wind and within the magnetosphere/ionosphere system. To further this investigation, the patterns of annual' and diurnal variations of the existing long-running aa and am indices are modelled and the stability of the semi-annual variation over the last 150 years confirmed. The connection between these range indices and the solar wind is evaluated at a wide variety of timescales, varying from 3 hours to 1 year, in order to determine which of the solar wind parameters are responsible for their variations. In this respect, the importance of considering discontinuities in the in-situ data from spacecraft in the solar wind is highlighted. The asymmetry of the two equinox peaks in geomagnetic activity is used to investigate thel axial hypothesis which links the semi-annual variation seen at the Earth to a proposed latitudinally-symmetrical structure of the Sun. It is demonstrated that during the spacecraft era the assumptions underpinning this model are false and that the Earth's sampling of the solar wind has been strongly influenced by solar structure asymmetries and variations on timescales shorter than Earth's annual orbit. The results strongly suggest that these asymmetries arose by chance and so little information about the Sun can be gained from the asymmetries seen in the aa index over the past century and a half. A new geomagnetic index is constructed to characterise more fully the semi-annual variation of the terrestrial current systems. Using this index it is confirmed that the equinoctial hypothesis which best characterises the semi-annual variation of activity is a process internal to the magnetosphere (specifically the substorm current wedge) and not a product of annual and diurnal modulations of energy extraction from the solar wind.

I address the question'of whether active galactic nuclei (AGN) exhibit similar Xray temporal variability properties to black hole X-ray binaries (BHXRBs). I utilise long time-scale X-ray monitoring data to produce broadband power spectral densities (PSDs) for a variety of AGN. Unfortunately X-ray monitoring progranunes of AGN often produce light curves that are irregularly sampled, which causes the observed PSD to become distorted. I therefore further developed a Monte Carlo modelling technique (PSRESP), based upon Uttley et ai, -(2002), to determine the undistorted broadband PSD and associate a robust acceptance probability to the fitted model. Using archival and proprietary Rossi X-ray Timing Explorer (RXTE) data, along with X-ray data from XMM-Newton, I apply PSRESP to determine the undistorted PSD of NGC 3783, which had previously been suggested to be analogous to a 'hard' state BHXRB. I show that a second break is not statistically supported, and I show that this PSD is, in fact, well fitted by a 'soft' state model that has only one break at higher frequencies. These results leave Arakelian 564 as the only AGN which shows a second break at low frequencies. I present a power spectral analysis of a 100 ksec XMM-Newton observation along with RXTE and ASCA data of the narrow line Seyfert 1 galaxy Ark 564. I demonstrate that the PSD of Ark 564 is well fit by the sum of two Lorentzian-shaped components, similar to those seen in BHXRBs. The PSD and lag spectrum of Ark 564 strongly suggests that a two-component Lorentzian interpretation of the origin of most of the variability is correct. Given the super-EddiPgton accretion rate of Ark 564 it is most likely analogous to a very-high state system. These results further confinn the connection between AGN and BHXRBs. I use proprietary X-ray data from our RXTE monitoring progranune, together with other data from the RXTE archive and complementary data from XMM-Newton, ASCA, and EXOSAT to produce broadband PSDs for 32 AGN spanning the frequency range from rv 10-8 to rv 10-3 Hz. The X-ray data for each AGN are reduced and analysed in a consistent manner. In particular, to ensure consistency between different RXTE gain epochs, all RXTE monitoring data is calibrated with respect to public archival Crab PCA data. I apply PSRESP to the resulting PSDs to determine the undistorted shape. All of the 32 AGN are consistent with either an unbroken or single-bend power-law interpretation of their PSDs, with the exception of Ark 564. I thus show that, for an expanded sample, AGN do behave like scaled-up BHXRBs, with the bend timescale scaling as mass divided by accretion rate. Thus, I further endorse the result of McHardy et al. (2006).