Automated Detection of Solar Radio Bursts using a Statistical Method

Abstract

Radio bursts from the solar corona can provide clues to forecast space weather hazards. After recent technology advancements, regular monitoring of radio bursts has increased and large observational data sets are produced. Hence, manual identification and classification of them is a challenging task. In this paper, we describe an algorithm to automatically identify radio bursts from dynamic solar radio spectrograms using a novel statistical method. We used e-CALLISTO radio spectrometer data observed at Gauribidanur observatory near Bangalore in India during 2013 - 2014. We have studied the classifier performance using the receiver operating characteristics. Further, we studied type III bursts observed in the year 2014 and found that $75\%$ of the observed bursts were below 200 MHz. Our analysis shows that the positions of the flare sites which are associated with the type III bursts with upper-frequency cut-off $\gtrsim 200$ MHz originate close to the solar disk center