Technical Article

Using solar ultraviolet irradiance measurements from GOES/EXIS to nowcast flare X-ray properties

¹ University of Johannesburg, Department of Physics, Johannesburg, South Africa
² South African National Space Agency, Hermanus 7200, South Africa
³ University of the Western Cape, Department of Physics and Astronomy, Robert Sobukwe Road, Bellville 7535, South Africa
⁴ University of Colorado Boulder, Laboratory for Atmospheric and Space Physics, 1234 Innovation Dr., Boulder CO 80303, USA

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Received: 23 June 2025
Accepted: 18 March 2026

Abstract

Context. X-rays emitted by solar flares change the properties of the Earth’s ionosphere and can damage to technological systems. Space weather forecasters monitor the X-ray irradiance from the Sun and issue warnings to mitigate these potentially harmful effects. Other wavelengths of solar irradiance are also observed operationally, and might provide advanced information about a flare’s X-ray properties. Aims. We investigate whether extreme ultraviolet (EUV) emissions can provide advance information about the strength and duration of 0.1–0.8 nm soft X-ray (SXR) solar flares. Specifically, we assess the predictive capability of He I (121.6 nm; Lyman-alpha), He II (30.4 nm), and the Mg II index in relation to SXR flare characteristics, with the goal of improving flare nowcasting. Methods. We analyzed operational spectral irradiance measurements from the Extreme Ultraviolet Sensor (EUVS) and the X-ray Sensor (XRS) aboard the Geostationary Operational Environmental Satellite-16 (GOES-16). The dataset includes all M- and X-class flares observed between 2017 and May 2025. We examined correlations between the timing and magnitude of EUV and SXR peaks, including lead times, flare durations, and peak intensities. Results. EUV emissions peak before SXR emissions in approximately 76% of cases. For He II, the average lead time is 4.69 min for X-class flares and 4.09 min for M-class flares. Lyman-alpha leads SXR emissions by approximately 4.98 min for X-class and 4.90 min for M-class flares, while Mg II leads by an average of 5.45 min for X-class and 4.75 min for M-class flares. Conclusions. He II shows the strongest correlation with SXR properties: flare durations correlate at r = 0.63 for M-class and r = 0.72 for X-class flares, while peak strength shows a moderate correlation (r = 0.53). He II enhancements exceeding 20% above the background are strong indicators of X-class flares, with fewer than 1% of M-class flares exhibiting such increases. Lyman-alpha demonstrates moderate predictive value for flare duration (r = 0.49 for M-class and r = 0.63 for X-class) and a weak correlation with flare strength (r = 0.32). The Mg II index similarly correlates moderately with SXR duration (r = 0.48 for M-class and r = 0.67 for X-class) and weakly with flare strength (r = 0.43).