Advanced Techniques in Spectral Calibration

Abstract

The thesis aimed to develop production-level software to enhance a prototype that processes calibrates and isolates astrophysical maser emissions. The prototype software had various deficiencies that needed to be addressed. The lack of a calibration technique accounting for the Doppler Effect raised questions about the accuracy of the prototype's results. Would implementing a new calibration technique account for the Doppler Effect? The prototype also had fundamental architectural flaws, and by expanding the overall structure and functionality of the prototype with new features, could the scientific capabilities of the software be enhanced? The hardcoded polarization data processed might limit the isolation of astrophysical maser emissions by not considering dual polarization processing. Development, specifically incremental and iterative development, was employed to address the uncovered issues in the prototype, with the ultimate aim being to answer the research questions and improve the core architecture throughout the software. This thesis is practical but includes research-oriented programming, requiring an understanding of concepts like radio astronomy and astrophysics for development. The use of dual polarization is an example of a theoretical idea being implemented in a practical development scenario; as both polarizations contain signal data, combining the data could potentially enhance the isolation of maser emissions. Additionally, including a calibration technique to correct data affected by the Doppler Effect uses theoretical and known concepts. At the same time, general software improvements are subjective; measurable factors like processing speed were used to gauge progress and improvement. Where possible, measurements and comparisons were used to critique and assess the result of the developed production-level software. Testing the production-level software highlighted various improvements and limitations. Through the implementation of a velocity calibration, the Doppler shift affecting frequency data was corrected. Comparing prototype and production software Figures highlighted that dual polarization processing improved the isolation of maser spectral lines. Additionally, directory and file processing speed demonstrated an average increase of 61%. Introducing OptParse options increases the control the user has over the utilization of the software. The sample rate used during observation should be as low as possible, with the most favourable results achieved at sample rates between 512000.0 Hz and 1024000.0 Hz. Objects larger than the bandwidth of the telescopes negatively impact the median calibration due to the oversaturation of data. Additionally, low file counts of astrophysical observations negatively impact the final output of the software. Overall, the production-level software is more equipped to handle the isolation and management of astrophysical maser data.