DSpace Repository

The development of the matrix method for the determination of riometer quiet day curves

Show simple item record

dc.contributor.author Drevin GR en
dc.date.accessioned 2016-09-22T11:14:17Z
dc.date.available 2016-09-22T11:14:17Z
dc.date.submitted 1999 en
dc.identifier.uri http://hdl.handle.net/20.500.11892/106444
dc.description.abstract An important aspect in using riometers to study the ionosphere is the determination of Quiet Day Curves (QDCs). This thesis re-examines the process of determining QDCs by identifying the shortcomings in existing methods, and then proposing a new method for the determining of QDCs which differs fundamentally from existing methods. The existing methods use the values of a given sidereal time interval over a number of consecutive days to determine the QDC value for that time interval. The values in the time intervals preceding and following that interval are not used explicitly and it is only when the QDC values are smoothed that the values outside a given interval have an influence on its smoothed QDC value. With the proposed method both the values in the given time interval, over a number of consecutive days, as well as the values outside that time interval are used to calculate its QDC value. The problems that have to be addressed in the implementation of this method are: (1) The inference of a cut-off frequency directly from a data set: The method proposed in this thesis is based on the filtering of the riometer data in the frequency (Fourier) domain. One of the existing methods, viz. the maximum density method also makes use of filtering in the frequency domain. However, the choice of cut- off frequency in the maximum density method is quite subjective, with the cut-off frequency being the frequency which gives visually acceptable results. Therefore, in this thesis, an attempt is made to determine cut-off frequencies objectively. This is done by determining cut-off frequencies directly from the antenna reception pattern as well as by inference from the data itself. A number of methods which can be used to infer a cut-off frequency from the data are investigated with the most consistent results being obtained with the Lorentz curve and entropy-based methods. (2) The occurrence of Gibbs's phenomenon error at the edges of a filtered data set: An undesirable side effect of filtering in the frequency domain is that errors are introduced at the leading and trailing edges of the data set which is being filtered. This phenomenon is known as the Gibbs phenomenon. Different methods of overcoming this problem are investigated with the pinned Fourier transform being the most effective. (3) The occurrence of localised high frequency content in a data set: It is shown that the riometer data has localised high frequency content. The use of wavelets instead of the Fourier transform is therefore investigated. The error ob- tained with wavelets is much larger than that obtained with the Fourier transform. It is therefore shown that wavelets do not have an advantage over the Fourier transform in this specific application. The use of total least squares regression to improve the accuracy of QDCs, where riometer observation is done at two or more frequencies, is also investigated. This investigation accentuates the need for accurate QDCs as it is shown that inaccurate QDCs could result in error-induced variations in n-values to which one would be tempted to assign physical significance. Total least squares regression is therefore shown to be an effective method of improving the accuracy of QDCs. In conclusion, the QDCs obtained using the proposed method are compared to those obtained with two existing methods, viz. the percentile method and the maximum density method. It is shown that the proposed method is an improvement over these two existing methods. en
dc.language English en
dc.title The development of the matrix method for the determination of riometer quiet day curves en
dc.type Doctoral degree en
dc.description.degree PhD en

Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record