Relationship of peak fluxes of solar radio bursts and X-ray class of solar flares: Application to early great solar flares

TL;DR

This study establishes a quantitative relationship between solar radio burst fluxes and X-ray flare classes, enabling estimation of flare magnitudes before satellite X-ray data was available, based on historical radio observations.

Contribution

The paper introduces a method to estimate historical solar flare magnitudes from radio flux data, extending flare analysis back to 1956 and filling gaps in early solar flare records.

Findings

Correlation coefficient between radio flux and X-ray flux is approximately 0.7.

X-ray class can be estimated from radio flux with an error factor of 3.

Historical flare light curves can be reconstructed using radio observations.

Abstract

Large solar flares occasionally trigger significant space-weather disturbances that affect the technological infrastructures of modern civilization, and therefore require further investigation. Although these solar flares have been monitored by satellite observations since the 1970s, large solar flares occur only infrequently and restrict systematic statistical research owing to data limitations. However, Toyokawa Observatory has operated solar radio observations at low frequencies (at 3.75 and 9.4 GHz) since 1951 and captured the early great flares as solar radio bursts. To estimate the magnitudes of flares that occurred before the start of solar X-ray (SXR) observations with the Geostationary Operational Environmental Satellite (GOES) satellites, we show the relationship between microwave fluxes at 3.75 and 9.4 GHz and X-ray fluxes of flares that occurred after 1988. In total, we explored 341 solar flares observed with the Nobeyama Radio Polarimeters and Toyokawa Observatory from 1988-2014 and compared them with the SXR observations recorded by the GOES satellites. The correlation coefficient was approximately 0.7. Therefore, the GOES X-ray class can be estimated from the peak flux at 3.75 and 9.4 GHz with a large variance and an error of factor of 3 (1 sigma). Thus, for the first time, we quantitatively estimated the light curves of two early solar flares observed in 1956 February by the Toyokawa solar radio observations using the relationship between SXR thermal radiation and microwave nonthermal radiation (Neupert, 1968, ApJ, 153, 59).