Real-time flare detection in ground-based H$\alpha$ imaging at Kanzelh\"ohe Observatory

TL;DR

This paper presents a real-time automated flare detection system using H-alpha imaging at Kanzelhöhe Observatory, achieving high detection accuracy and rapid data provision for space weather monitoring.

Contribution

The paper introduces a new real-time flare detection system integrated into KSO's H-alpha imaging, with algorithms validated over five months showing high accuracy and fast data dissemination.

Findings

99.96% image data provision success rate

All flares with area ≥50 micro-hemispheres detected

Automatic flare timing accuracy within 3 minutes

Abstract

Kanzelh\"ohe Observatory (KSO) regularly performs high-cadence full-disk imaging of the solar chromosphere in the H$\alpha$ and CaIIK spectrallines as well as the solar photosphere in white-light. In the frame of ESA's Space Situational Awareness (SSA) programme, a new system for real-time H$\alpha$ data provision and automatic flare detection was developed at KSO. The data and events detected are published in near real-time at ESA's SSA Space Weather portal (http://swe.ssa.esa.int/web/guest/kso-federated). In this paper, we describe the H$\alpha$ instrument, the image recognition algorithms developed, the implementation into the KSO H$\alpha$ observing system and present the evaluation results of the real-time data provision and flare detection for a period of five months. The H$\alpha$ data provision worked in $99.96$% of the images, with a mean time lag between image recording and online provision of 4s. Within the given criteria for the automatic image recognition system (at least three H$\alpha$ images are needed for a positive detection), all flares with an area $\ge$50 micro-hemispheres and located within $60^\circ$ of the Sun's center that occurred during the KSO observing times were detected, in total a number of 87 events. The automatically determined flare importance and brightness classes were correct in $\sim$85%. The mean flare positions in heliographic longitude and latitude were correct within $\sim$1$^\circ$. The median of the absolute differences for the flare start times and peak times from the automatic detections in comparison to the official NOAA (and KSO) visual flare reports were 3 min (1 min).