Type II and Type III Radio Bursts and their Correlation with Solar Energetic Proton Events

TL;DR

This study analyzes radio burst data from 2010-2013 to develop a statistical model for predicting solar energetic proton events, highlighting the significance of type III burst properties in forecasting.

Contribution

It introduces a new catalog linking radio burst properties with SEP events and develops a predictive model with accuracy comparable to human forecasters.

Findings

All >10 MeV SEP events are associated with type II bursts.

92% of SEP events are associated with type III bursts.

Radio burst properties, especially type III intensity and duration, are highly predictive of SEP occurrence.

Abstract

Using the Wind/WAVES radio observations from 2010-2013, we present an analysis of the 123 decametric-hectometric (DH) type II solar radio bursts during this period, the associated type III burst properties, and their correlation with solar energetic proton (SEP) properties determined from analysis of the Geostationary Operational Environmental Satellite (GOES) observations. We present a useful catalog of the type II burst, type III burst, Langmuir wave, and proton flux properties for these 123 events, which we employ to develop a statistical relationship between the radio properties and peak proton flux that can be used to forecast SEP events. We find that all SEP events with a peak > 10 MeV flux above 15 pfu are associated with a type II burst and virtually all SEP events, 92%, are also associated with a type III radio burst. Based on a principal component analysis, the radio burst properties that are most highly correlated with the occurrence of gradual SEP events and account for the most variance in the radio properties are the type III burst intensity and duration. Further, a logistic regression analysis with the radio-derived principal component (dominated by the type III and type II radio burst intensity and type III duration) obtains SEP predictions approaching the human forecaster rates, with a false alarm rate of 22%, a probability of detection of 62%, and with 85% of the classifications correct. Therefore, type III radio bursts that occur along with a DH type II burst are shown to be an important diagnostic that can be used to forecast SEP events.