*Toshinori MATSUSHITA1, Kanako SEKI1, Nozomu NISHITANI1, Tomoaki HORI1
(1.STEL, Nagoya University)
Keywords:SuperDARN, Pc5 waves
Ultra-low-frequency Pc5 waves have been observed by many methods such as ground-based magnetometers, HF radars and satellites. It has been demonstrated by numerical experiments that magnetospheric Pc5 waves are globally and directly generated on the dayside by solar wind dynamic pressure variations and/or on the dawn/dusk flank by Kelvin-Helmholtz surface waves. In addition, there are storm-time Pc5 waves on the dusk side magnetosphere that are associated with instabilities in the storm time ring current caused by the particle injection. The Pc5 waves can play an important role in mass and energy transport within the inner magnetosphere such as the radial diffusion of outer radiation belt electrons, as suggested by previous studies. Outstanding problems in Pc5 studies include clarification of their global characteristics and distribution, generation mechanisms, and especially their dependence on the solar wind parameters.In this study, we try to develop a new automatic identification method of Pc5 waves using ~20-sec time resolution data obtained by the SuperDARN Hokkaido HF radar operated in the RBSP mode. In this method, we use the Doppler velocity data and the power spectrum density calculated by the wavelet transformation. We set criteria which can detect Pc5 waves even when harmonic oscillations coexist. We show an example for the identification method using the Doppler velocity data obtained by the SuperDARN Hokkaido HF radar in details. Then, the candidates of Pc5 event are verified by inspection. From the rate of error identification, we evaluate the accuracy of the automatic identification method statistically. In the presentation we will also report on the preliminary results of mid-latitude Pc5 characteristics such as frequency distribution and MLT dependence.