Microburst Identification: An Initial Algorithm

Relativistic Microburst Sorting, Omar Alamin:

This project concerns detection and sorting of relativistic microbursts in the earth’s magnetosphere. The satellite Firebird II has the partial purpose of recording when electron precipitation from the magnetosphere occurs within a few chosen regions of the atmosphere. Particularly localized and fast moments of energetic precipitation which have come to be known as microbursts can reach up to relativistic scales in energy. The end goal is to determine if there are correlations between microbursts and chorus waves. The first step to identifying correlation would be to create an algorithm to catalogue microbursts from the signal received by Firebird. Thus, this poster shows an initial form of an algorithm to automatically identify microbursts. To remove the general upward and downward trends of the original signal, a moving median technique was used for a smoothed, then detrended signal. To begin identifying outliers and collecting what was believed to be true microbursts, the algorithm picked out any part of the signal that was not within three standard deviations of the average of its neighbors. Future researchers may tweak the threshold used here to satisfaction. Further, if the value did not exceed the average by some threshold it was to be considered part of the baseline noise of the instrument and not counted toward microburst detection. What follows are the results of this algorithm as well as statistical analysis of an example signal.