Implementation of a Near-Realtime Recording and Reporting System of Solar Radio Bursts Facilities: OVRO-LWA astropy, pytorch

Peijin Zhang, Anastasia Kuske, Bin Chen, Mengjia Xu, Gelu Nita, Marin M. Anderson, Judd D. Bowman, Ruby Byrne, Morgan Catha, Xingyao Chen, Sherry Chhabra, Larry D’Addario, Ivey Davis, Jayce Dowell, Katherine Elder, Dale Gary, Gregg Hallinan, Charlie Harnach, Greg Hellbourg, Jack Hickish, Rick Hobbs, David Hodge, Mark Hodges, Yuping Huang, Andrea Isella, Daniel C. Jacobs, Ghislain Kemby, John T. Klinefelter, Matthew Kolopanis, Nikita Kosogorov, James Lamb, Casey Law, Nivedita Mahesh, Surajit Mondal, Brian O’Donnell, Kathryn A. Plant, Corey Posner, Travis Powell, Vinand Prayag, Andres Rizo, Andrew Romero-Wolf, Jun Shi, Greg Taylor, Jordan Trim, Mike Virgin, Akshatha Vydula, Sandy Weinreb, Scott White, David Woody, Sijie Yu, Thomas Zentmeyer · 2026

Read Implementation of a Near-Realtime Recording and Reporting System of Solar Radio Bursts Facilities: OVRO-LWA astropy, pytorch by Peijin Zhang, Anastasia Kuske, Bin Chen, Mengjia Xu, Gelu Nita, Marin M. Anderson, Judd D. Bowman, Ruby Byrne, Morgan Catha, Xingyao Chen, Sherry Chhabra, Larry D’Addario, Ivey Davis, Jayce Dowell, Katherine Elder, Dale Gary, Gregg Hallinan, Charlie Harnach, Greg Hellbourg, Jack Hickish, Rick Hobbs, David Hodge, Mark Hodges, Yuping Huang, Andrea Isella, Daniel C. Jacobs, Ghislain Kemby, John T. Klinefelter, Matthew Kolopanis, Nikita Kosogorov, James Lamb, Casey Law, Nivedita Mahesh, Surajit Mondal, Brian O’Donnell, Kathryn A. Plant, Corey Posner, Travis Powell, Vinand Prayag, Andres Rizo, Andrew Romero-Wolf, Jun Shi, Greg Taylor, Jordan Trim, Mike Virgin, Akshatha Vydula, Sandy Weinreb, Scott White, David Woody, Sijie Yu, Thomas Zentmeyer (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-25446.txt.

Introduction

Zhang and co-authors run a near-live alarm for solar radio bursts at the Owens Valley array. Bursts flag storms in the Sun's corona and feed space weather forecasts. Their system clips data from a live buffer and streams a radio spectrogram. A YOLO-style deep net, trained on physics-made fake bursts, spots type III events in that stream. Reports go out about ten seconds after a burst hits. The setup pairs raw sensitivity with the low delay that forecasters need.

Catalog record

CREATORPeijin Zhang, Anastasia Kuske, Bin Chen, Mengjia Xu, Gelu Nita, Marin M. Anderson, Judd D. Bowman, Ruby Byrne, Morgan Catha, Xingyao Chen, Sherry Chhabra, Larry D’Addario, Ivey Davis, Jayce Dowell, Katherine Elder, Dale Gary, Gregg Hallinan, Charlie Harnach, Greg Hellbourg, Jack Hickish, Rick Hobbs, David Hodge, Mark Hodges, Yuping Huang, Andrea Isella, Daniel C. Jacobs, Ghislain Kemby, John T. Klinefelter, Matthew Kolopanis, Nikita Kosogorov, James Lamb, Casey Law, Nivedita Mahesh, Surajit Mondal, Brian O’Donnell, Kathryn A. Plant, Corey Posner, Travis Powell, Vinand Prayag, Andres Rizo, Andrew Romero-Wolf, Jun Shi, Greg Taylor, Jordan Trim, Mike Virgin, Akshatha Vydula, Sandy Weinreb, Scott White, David Woody, Sijie Yu, Thomas Zentmeyer
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE31879
LICENSEhttp://creativecommons.org/licenses/by-sa/4.0/
RIGHTSCC BY SA
SOURCEarXiv

Source: https://arxiv.org/abs/2603.25446

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Implementation of a Near-Realtime Recording and Reporting System of Solar Radio Bursts Facilities: OVRO-LWA astropy, pytorch Peijin Zhang, Anastasia Kuske, Bin Chen, Mengjia Xu, Gelu Nita, Marin M. Anderson, Judd D. Bowman, Ruby Byrne, Morgan Catha, Xingyao Chen, Sherry Chhabra, Larry D’Addario, Ivey Davis, Jayce Dowell, Katherine Elder, Dale Gary, Gregg Hallinan, Charlie Harnach, Greg Hellbourg, Jack Hickish, Rick Hobbs, David Hodge, Mark Hodges, Yuping Huang, Andrea Isella, Daniel C. Jacobs, Ghislain Kemby, John T. Klinefelter, Matthew Kolopanis, Nikita Kosogorov, James Lamb, Casey Law, Nivedita Mahesh, Surajit Mondal, Brian O’Donnell, Kathryn A. Plant, Corey Posner, Travis Powell, Vinand Prayag, Andres Rizo, Andrew Romero-Wolf, Jun Shi, Greg Taylor, Jordan Trim, Mike Virgin, Akshatha Vydula, Sandy Weinreb, Scott White, David Woody, Sijie Yu, Thomas Zentmeyer Abstract Strong solar activities are often accompanied by a variety of radio bursts. These radio bursts not only serve as valuable diagnostics of coronal and heliospheric processes but also as potential tools in space weather monitoring and forecasting. However, space weather applications call for the capability for low-latency and high-sensitivity radio burst recording and reporting, which has remained lacking. In this work, we present the development of a near-realtime radio burst recording and reporting system with the Owens Valley Radio Observatory’s Long Wavelength Array. The system directly clips data from the realtime buffer and streams it as a live realtime radio dynamic spectrogram. The spectrograms are then fed to a deep-learning–based burst identification module for type III radio bursts. The identifier is built on a YOLO (You Only Look Once) architecture, trained by synthetic type III radio bursts generated by using a physic-based model to achieve accurate and robust detection. This system enables continuous real-time radio spectrum streaming and the automatic reporting of type III radio bursts within ∼ \sim 10 seconds of their occurrence. I Introduction Radio observations are an important tool for space weather monitoring and forecasting. Solar radio bursts are among the earliest electromagnetic signatures of energetic particle acceleration and shock formation in the solar corona and heliosphere. Because many radio bursts are closely associated with solar flares, coronal mass ejections (CMEs), and the release of solar energetic particles (SEPs), they carry strong potential in space-weather monitoring and forecasting ( 2 ; 8 ; 23 ; 30 ; 1 ) . In particular, rapidly drifting type III bursts trace energetic electron beams escaping along open magnetic field lines, while type II bursts reveal CME-driven shocks that are primary accelerators of large SEP events ( 15 ; 19 ) . These strong physical connections make solar radio observations particularly valuable for both scientific diagnostics and operational space-weather applications. To fully exploit this potential, however, radio burst detection must be performed automatically with very low latency and high accuracy, enabling timely identification of eruptive activity without reliance on manual inspection. Over the past decade, major advances in low-frequency radio astronomy have been driven by new-generation interferometric arrays such as the Low Frequency Array (LOFAR; 29 ; 4 ) and the Murchison Widefield Array (MWA; 28 ). These instruments provide high sensitivity, broad bandwidth, and high resolution in both the time–frequency and spatial domains, enabling detailed studies of solar radio bursts across a wide range of intensities and morphologies ( 17 ; 18 ; 15 , e.g.,) . Results from these facilities have demonstrated strong connections between radio burst properties and solar eruptive phenomena, including CME kinematics, shock evolution, and coronal density structure ( 20 ) . Despite their scientific power, however, these arrays are primarily research-oriented. Their data products are typically processed offline, with substantial latency between observation and analysis, and they do not provide continuous real-time data streams optimized for rapid event reporting. As a result, their exceptional data quality has not yet translated into operational real-time space weather capability. Real-time monitoring of solar radio emission has long been recognized as a critical capability for operational space-weather nowcasting and forecasting. Several established systems, such as the Radio Solar Telescope Network (RSTN) ( 21 ; 9 ) , e-Callisto ( 3 ) can provide full-day coverage of dynamic spectra that have enabled decades of research on burst occurrence and flare–CME associations. However, these systems typically exhibit latency at the minute scale between realtime acquisition and operation/public availability, and their sensitivity and spectral resolution are often insufficient to capture the faint, rapidly evolving radio bursts now known to accompany many weak events or as precursors of major events ( 13 ) . Currently, despite the wealth of broadband instrumentation currently available, there remains a substantial gap between data acquisition at the telescope and low-latency, high-sensitivity burst reporting suitable for immediate scientific response or operational space-weather use. The Owens Valley Radio Observatory’s Long Wavelength Array (OVRO-LWA) ( 10 ) occupies a unique position between these two regimes. OVRO-LWA has 352 cross-dipole antennas, observing over a frequency range of 13.4–86.9 MHz (instantaneous bandwidth 73.5 MHz). In its beamformer (total-power) mode, the raw frequency resolution is 24 kHz across 3072 channels, and the Level-1 spectrogram cadence is 256 ms (fast visibility at 0.1 s). In addition, time resolution in the beamformer chain can reach 1 ms (typically 64 ms), while slow visibility products provide 10 s cadence for longer-timescale monitoring. As a wide-field, low-frequency interferometric array, OVRO-LWA delivers high sensitivity and broadband coverage, while its beamformed total-power data stream enables continuous monitoring of solar radio emission. Crucially, the OVRO-LWA recording architecture makes it possible to access data with very low latency, opening a pathway to near-real-time spectral and imaging monitoring. When combined with reliable automatic detection, such a system has the potential to detect and report solar radio bursts within seconds of their onset, providing a powerful new capability for both scientific response and operational space-weather nowcasting and forecasting. With the development of large low-frequency array infrastructures, the data volume and rate of production of observation data has grown to the point where manual inspection is infeasible. There are early attempts of using thresholding algorithms to identify solar radio bursts ( 16 ; 31 ) . Machine-learning methods—especially deep object-detection networks—have recently shown superior performance in identifying diverse solar radio burst morphologies ( 27 ; 11 ; 7 ) . Yet a major remaining challenge is the development of a complete end-to-end system capable not only of accurate detection but also of robust real-time ingestion, processing, and reporting. In this paper, we introduce a real-time solar radio burst reporting system developed for OVRO-LWA. The system streams spectroscopic data directly from the instrument’s live buffer to a burst identification module based on YOLO (You Only Look Once 22 ) trained using a physics-based type III radio burst generation model. This integrated pipeline achieves high sensitivity while maintaining low latency, enabling automatic identification and reporting of type III radio bursts within 10 seconds of onset. Section 2 presents the real-time streaming of the dynamic spectrum, and Section 3 presents the event detection method. Section 4 is the conclusion and discussion. II Realtime Spectrum System Framework The real-time spectrum system is designed to sit non-invasively on top of the existing OVRO–LWA recording infrastructure while providing a low-latency, data stream for monitoring and automatic burst detection. Figure 1 summarizes the overall workflow. At the front end, the ovro_lwa_recorder process receives beamformed total-power data and

Read the complete text