Improved SuperDARN radar signal processing: A first principles statistical approach for reliable measurement uncertainties and enhanced data products

Abstract

Ground-based radar systems are the best way to continuously monitor medium-to-large-scale features of the near-Earth space environment on a global scale. The Super Dual Auroral Radar Network (SuperDARN) radars are used to image the high-latitude ionospheric plasma circulation, which is produced by magnetosphere-ionosphere coupling processes generated by the interaction of both the solar and terrestrial magnetic fields. While investigating ways to expand the usable data products of SuperDARN to include electron density inferred using a multiple-frequency technique, it was determined that SuperDARN error estimates were lacking sufficient rigour.

The method to calculate SuperDARN parameters was developed approximately 25 years ago when available computing resources were significantly less powerful, which required a number of simplifications to ensure both valid data and reasonable processing time. This resulted in very conservative criteria being applied to ensure valid data, but at the expense of both rigorous error analysis and the elimination of some otherwise valid data. With access to modern computing resources, the SuperDARN data processing methodology can be modernized to provide proper error estimates for the SuperDARN parameters (power, drift velocity, width).

This research has resulted in 3 publications, which are presented here as Chapters 5, 6, and 7. The error analysis started with a first principles analysis of the self-clutter generated by the multiple-pulse technique that is used to probe the ionosphere (Chapter 5). Next, the statistical properties of voltage fluctuations as measured by SuperDARN were studied and the variance of these measurements were derived (Chapter 6). Finally, the statistical error analysis was propagated to the standard SuperDARN data products using a new First-Principles Fitting Methodology (Chapter 7). These results can be applied to all previously recorded SuperDARN data and have shown a practical increase in data of >50%. This has significant impact on the SuperDARN and space science communities with respect to, for example, global convection maps and their use in global modelling efforts. These results also enable quantitative experiment design facilitating research into using SuperDARN to provide electron density measurements, with a preliminary investigation using the new SuperDARN fitting methodology presented in Chapter 8.