Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.naic.edu/~tghosh/thesis-jf/chapter1.txt Дата изменения: Sat Jul 27 06:22:25 2002 Дата индексирования: Sun Apr 10 13:37:14 2016 Кодировка: IBM-866 Поисковые слова: воздушные массы

Chapter I
Introduction
1.1 The Interstellar medium of our Galaxy
The space between the stars in our Galaxy is not devoid of matter but is filled with gas and dust, the interstellar medium (ISM). Most of our knowledge concerning the ISM is based on observations within the disk of our Galaxy, although studies of nearby spiral galaxies suggest that they contain similar interstellar material. The dust component consists of small grains made up of various kinds of ices, graphite, silicates, and possibly metals. The gaseous material is composed of a variety of constituents, ranging from simple particles such as free electrons, protons, neutral H and He atoms, to molecules such as $H_{2}$, CO, and more complex organic and inorganic species. The ISM is also pervaded by a ubiquitous magnetic field with typical strength of order a few microGauss.
The gas reveals its presence through optical and radio absorption lines seen in the spectra of stars and radio continuum sources, and through emission features, such as optical emission lines emitted by HII regions, radio emission in numerous molecular lines, the HI 21-cm line, radio recombination lines, and the radio continuum. Dust appears prominently in dark nebulae and reflection nebulae. The former are dark because they have a relatively dense concentration of dust grains which scatter and absorb light much more efficiently than single atoms. Reflection nebulae are produced by dust grains (in a lower concentration than is found in dark nebulae) which scatter starlight. In broad terms, the material in the ISM comprises at least five components. These are shown in Table 1.1.
1.2 THE ЃTURBULENT’ ISM
An entirely different (yet complimentary) view of the diffuse ISM is obtained from studies of its effects on radio signals as they propagate through it. Various dispersive and scattering effects affect the amplitude and phase of the signals from compact Galactic and extragalatic radio sources and these demonstrate that the diffuse ionized component of the ISM is very irregular in nature. The electron density in this component fluctuates on scale sizes spanning many orders of magnitudes (Armstrong, Cordes, \& Rickett 1981). Which of the constituents listed in Table 1 hosts this turbulent component is still not clear, although there have been suggestions that the WIM is one good candidate. It is this irregular ionized component of the ISM that forms the subject matter of this thesis. Understanding propagation effects is also relevant to interpreting observations of radio sources whose signals become distorted as they travel through the ISM. In fact, many measurements can only be properly interpreted if we can separate extrinsic effects from those intrinsic to the sources.
1.2.1 Historical Background
Studies of radio galaxies by Hewish (for a review, see Hewish 1975) led progressively to the identification of ionospheric scintillation, angular scattering in the outer solar corona, and interplanetary scintillation (IPS). Since only very small diameter sources show IPS, he employed this effect to probe radio source structure in the range 0.1µµ to 1.0µµ at meter wavelengths. Using this technique, the angular sizes of many compact sources were first determined (Cohen et al. 1967; Little \& Hewish 1968). To exploit this approach, Hewish and his colleagues were motivated to build an antenna array for an IPS survey at 81 MHz, and it was with this instrument that pulsars were discovered (Hewish et al. 1968).
The discovery of pulsars opened new ways of probing the ISM. The most direct observables are the arrival times of pulses and the polarization angles, both which are frequency dependent with coefficients equal to the dispersion measure (DM) and rotation measure (RM), respectively (see Appendix A). The ratio RM/DM yields the average line-of-sight component of the magnetic field. Given a model of Galactic rotation and the distribution of neutral hydrogen, measurements of Hi absorption can lead to lower and/or upper bounds on a pulsarµs distance (Weisberg, Rankin, \& Boriakoff 1980). Combining the distance to a pulsar and its DM gives an estimate of the average electron density along the line of sight to the pulsar. The interpretation of these and other observable quantities reveal that the thermal material is not uniformly distributed, but has a complicated distribution on all size scales (see SECTION 2.5).
The extreme regularity of the arrival times of pulsar pulses contrasts with the erratic variations in their intensities. The component of variability having a timescale of minutes at meter wavelengths was soon identified as interstellar scintillation (ISS) produced by electron-density fluctuations in the interstellar medium (Scheuer 1968). Sieber (1982) later re-examined pulsar variability, demonstrating that a clear correlation existed between the timescale of slow pulsar variations and dispersion measure (and therefore distance to the pulsar; see Appendix A). This clearly indicated that the variations were extrinsic to the pulsars. Sieber also found that the timescale increased with the observing wavelength for the slow variations. He noted that the theory of diffractive interstellar scintillation predicted wavelength and dispersion measure dependencies for the timescale that differed radically from these findings. This led to the proposal from Rickett, Coles, \& Bourgois (1984, hereinafter RCB) of refractive interstellar scintillation (RISS) from the same turbulence spectrum that causes diffractive interstellar scintillation (DISS). RCB pointed out that the behaviour described by Sieber was precisely that predicted by the theory of strong scintillation, which predicts that when scintillation becomes strong it exhibits a two-scale character. This theory had already been developed in the optical literature (e.g., Prokhorov et al. 1975) and was confirmed by laser propagation in the atmosphere (Coles \& Frehlich 1982). RISS could also account for the low-frequency variability (LFV) of some radio sources (Shapirovskaya 1978).
1.3 The Effects of Propagation through the ISM
The effects introduced during propagation on the signals from radio sources by the ISM of our Galaxy will now be summarized.
1.3.1 Effects from the homogeneous ISM
1. Line absorption by atomic and molecular gas within the ISM
2. Continuum free-free absorption by ionized material within the ISM
3. Frequency dependent delay of pulsar signals due to dispersion in the ionized material within the ISM
4. Rotation of the plane of linear polarization for polarized radio sources due to Faraday rotation in magnetoionic material within the ISM
1.3.2 Effects from the turbulent ISM
The propagation effects from Galactic electron-density inhomogeneities can be divided into diffractive and refractive phenomena, due respectively to small ($10^{6} н 10^{8}$ m) and large ($10^{10} н 10^{12}$ m) scales in the scattering medium. The dividing scale here is approximately that of the radius of the first Fresnel zone at a typical distance to the scattering medium (for reviews, see Spangler 1988 and Rickett 1990).
[italics]Diffractive phenomena
1. Angular broadening of sources
2. Temporal broadening of pulsar pulses
3. Fast pulsar intensity fluctuations
[italics]Refractive phenomena
4. Low-frequency variability of compact radio sources ($\nu \lt 1$ GHz)
5. Extreme scattering events (ESE) in source intensity
6. Slow pulsar intensity variations
7. Pulsar pulse arrival time fluctuations
8. Apparent organized patterns in pulsar dynamic spectra such as drifting bands and fringing
9. Fluctuations in the apparent position of sources (ММimage wanderµµ)
1.4 Scope of this thesis
The research described in this thesis focusses on two of the phenomena associated with the propagation of radio signal from extragalactic sources through the turbulent ISM. An extended discussion of these is given in Chapter 2, focusing on the propagation effects most relevant to our investigation, and introducing the related observable quantities. These quantities help us model the distribution of electron-density fluctuations in the Galaxy.
When, via the Research Experience for Undergraduates (REU) Program of NSF, I came to Arecibo Observatory as a summer student in 1990, it was to work on an experiment to study Low Frequency Variability (LFV) using the NAIC Arecibo 305-m radio telescope (see Appendix B) to monitor the low-frequency flux density of 33 extragalactic radio sources. The measurements were made at approximately bimonthly intervals between 1980 January and 1989 February, and at less regular intervals between 1989 October and 1993 October, for a total of 64 observing sessions. The principal investigator of the project was the observatory director, Dr. Daniel Altschuler, with Dr. Tapasi Ghosh also being involved after joining the Arecibo staff in 1992. I returned to Arecibo on a number of occasions subsequent to 1990 to reduce the observations made for this project since my previous visit, and to complete analysis of the accumulated earlier data. After completion of the full reduction and standardization of the data, I was fully involved in the error analysis and interpretation of the results, as described in Chapter 3. The discovery of the first Extreme Scattering Event ever recognized below 1 GHz (see SECTION3.5) was entirely my own discovery.
Dr. Ghosh had been involved earlier in a LFV study made with the Ooty Radio Telescope in India. This had confirmed the cause of these variations to be refractive interstellar scintillations, and she had tried to model her results in terms of their dependence on line of sight direction through the Galaxy. This proved to only be possible if the electron density fluctuations in the ISM possess a spiral structure, similar to other major interstellar components. Consequently, Dr. Ghosh, Dr. Chris Salter of Arecibo, and myself discussed the possibility of investigating the distribution of fluctuations in the electron density distribution via direct Very Long Baseline Interferometry (VLBI) measurement of the scatter broadening of the images of compact extragalactic sources. This seemed very promising for source directions at low galactic latitude, which principally sampled lines of sight towards a number of different environments such as tangential points to inner spiral arms, interarm regions, and along the local Orion Arm. To this end, a three-frequency Very Long Baseline Array (VLBA; see Appendix C) proposal was submitted to the National Radio Astronomy Observatory (NRAO). This was allocated full requested observing time (albeit without the requested inclusion of a single VLA antenna in the array). I helped to schedule this series of observations and imaged the full $\lambda$18-cm data set from this project. When it became clear that shorter baselines were needed to map some of the targets satisfactorily, and produce a complete ММatlasµµ of the targets at this wavelength, a proposal to the Jodrell Bank MERLIN (see Appendix D) array in England was prepared by myself and my collaborators, this again receiving the full requested observing time (Chapter 4). Although the VLBA data at the two longer wavelengths ($\lambda$49 and 92 cm) are not yet fully reduced, I have proceeded with the interpretation of the complete $\lambda$18-cm sample. The background to this is described in Chapter 5, with the actual interpretation of the new results being given in detail in Chapter 6.
Conclusions drawn from the work described in this thesis are summarized in Chapter 7, along with suggestions for further work.