Forecasting Solar Energetic Particle (SEP) events with Flare X-ray peak ratios

Abstract

Solar flare X-ray peak fluxes and fluences in the 0.1–0.8 nm band are often used in models to forecast solar energetic particle (SEP) events. Garcia (2004) [Forecasting methods for occurrence and magnitude of proton storms with solar soft X rays, Space Weather, 2, S02002, 2004] used ratios of the 0.05–0.4 and 0.1–0.8 nm bands of the X-ray instrument on the GOES spacecraft to plot inferred peak flare temperatures versus peak 0.1–0.8 nm fluxes for flares from 1988 to 2002. Flares associated with E > 10 MeV SEP events of >10 proton flux units (pfu) had statistically lower peak temperatures than those without SEP events and therefore offered a possible empirical forecasting tool for SEP events. We review the soft and hard X-ray flare spectral variations as SEP event forecast tools and repeat Garcia’s work for the period 1998–2016, comparing both the peak ratios and the ratios of the preceding 0.05–0.4 nm peak fluxes to the later 0.1–0.8 nm peak fluxes of flares >M3 to the occurrence of associated SEP events. We divide the events into eastern and western hemisphere sources and compare both small (1.2–10 pfu) and large (≥300 pfu) SEP events with those of >10 pfu. In the western hemisphere X-ray peak ratios are statistically lower for >10 pfu SEP events than for non-SEP events and are even lower for the large (>300 pfu) events. The small SEP events, however, are not distinguished from the non-SEP events. We discuss the possible connections between the flare X-ray peak ratios and associated coronal mass ejections that are presumed to be the sources of the SEPs.

Temperatures of Large Solar X-ray Events and Associated CME Speeds

“…The question now is how lower-temperature X-ray flares are physically connected with the faster CMEs needed for SEP production. [1] pointed out that a global analysis by [9] led to a scaling law of Vcme ∼ T 0.5 e , where Vcme is the CME speed and T e is the X-ray flare temperature, a result seemingly in conflict with our results. The statistical correlation between X-ray flare peak fluxes and associated CME speeds Vcme [10,11,12,13] shows a clear connection between flares and CMEs, but any role for flare temperatures is not included in those correlations.…”

“…Our goal is to determine whether the X-ray ratio plot format of Figure 1 also sorts CME speeds as it does SEP events. We begin with the same [1] list of all GOES > M3 flares with known flare locations from 1998 to 2016. There were 433 flares without and 75 flares with ≥ 10 pfu SEP events in that list.…”

“…We changed the flare locations of four events and added an X3.1 flare on 24 October 2014 to make a total of 509 flare events. As in [1] we consider separately the eastern and western hemisphere flares. The timing and location of each flare was compared with CMEs reported on the SOHO/LASCO web site (https://cdaw.gsfc.nasa.gov/CME_list/index.html) to determine the CME association, if any, for each flare.…”

“…Recently [1] revisited the [2,3] approach using a selection of all > M3 X-ray flares with known source locations from 1998 to 2016. Rather than flare temperatures, [1] used simply the background-corrected peak flare ratios R of short 0.05-0.4 nm to long 0.1-0.8 nm versus the 0.1-0.8 nm peak intensities to sort out flare groups with (1) no associated SEP event; (2) a small (1.2-9.9 pfu) event; (3) a NOAA (≥ 10 pfu) event; or (4) a ≥ 300 pfu event. The flares were divided into eastern and western hemisphere groups and plotted separately as shown in the left panels of Figure 1.…”

“…The flares were divided into eastern and western hemisphere groups and plotted separately as shown in the left panels of Figure 1. [1] used the statistical t-test for three selected bins of peak-flux ratios to show low probabilities P of common populations, implying clear separations between the SEP and non-SEP groups based on flare peak X-ray ratios.…”