Automated detection of Pi 2 pulsations using wavelet analysis: 1. Method and an application for substorm monitoring

• Published in: Earth, planets, and space Vol. 50; no. 9; pp. 773 - 783
• Main Authors: Nosé, M., Iyemori, T., Takeda, M., Kamei, T., Milling, D. K., Orr, D., Singer, H. J., Worthington, E. W., Sumitomo, N.
• Format: Journal Article
• Language: English
• Published: Tokyo Terra 01.01.1998; Springer Nature B.V
• Subjects: Algorithms; Earth, ocean, space; Exact sciences and technology; External geophysics; Geomagnetic field; Magnetic storms, substorms; Nowcasting; Physics of the magnetosphere; Real time; Wavelet analysis; Wavelet transformation; Detection; Pi 2 micropulsation; Algorithm; Forecasting
• ISSN: 1880-5981; 1343-8832; 1880-5981
• DOI: 10.1186/BF03352169

Wavelet analysis is suitable for investigating waves, such as Pi 2 pulsations, which are limited in both time and frequency. We have developed an algorithm to detect Pi 2 pulsations by wavelet analysis. We tested the algorithm and found that the results of Pi 2 detection are consistent with those obtained by visual inspection. The algorithm is applied in a project which aims at the nowcasting of substorm onsets. In this project we use real-time geomagnetic field data, with a sampling rate of 1 second, obtained at mid- and low-latitude stations (Mineyama in Japan, the York SAMNET station in the U.K., and Boulder in the U.S.). These stations are each separated by about 120° in longitude, so at least one station is on the nightside at all times. We plan to analyze the real-time data at each station using the Pi 2 detection algorithm, and to exchange the detection results among these stations via the Internet. Therefore we can obtain information about substorm onsets in real-time, even if we are on the dayside. We have constructed a system to detect Pi 2 pulsations automatically at Mineyama observatory. The detection results for the period of February to August 1996 showed that the rate of successful detection of Pi 2 pulsations was 83.4% for the nightside (18-06MLT) and 26.5% for the dayside (06-18MLT). The detection results near local midnight (20-02MLT) give the rate of successful detection of 93.2%.