The Ratan Active Region Patches (RARPs) Database: A New Database of Solar Active Region Radio Signatures from the RATAN-600 Telescope

TL;DR

Forecasting solar eruptions requires coronal diagnostics beyond photospheric magnetic measurements. The authors build RARPs, a standardized AR-centered archive of RATAN-600 radio spectra (3–18 GHz) and extend RATANSunPy to automated AR extraction, enabling direct ML benchmarking against SHARPs magnetic predictors. They show that radio-derived embeddings offer complementary information and can achieve better probabilistic calibration for M-class flares, highlighting the value of multimodal approaches. Ground-based RATAN-600 data thus provide operational redundancy and broaden the evidence base for space-weather forecasting and solar-activity studies.

Abstract

Solar flares and coronal mass ejections, originating from solar active regions (ARs), are the primary drivers of space weather and can disrupt technological systems. Forecasting efforts heavily rely on photospheric magnetic field data from the Space-weather HMI Active Region Patch (SHARPs) data products. However, the crucial energy release occurs higher in the solar corona. Radio observations from instruments like the RATAN-600 telescope directly probe this region, but their scientific use has been hindered by a lack of standardized and accessible data products. To address this gap, we have developed the Ratan Active Region Patches (RARPs) database, a new public resource of multi-frequency radio spectra for solar ARs. Generated using RATANSunPy software, RARPs provides the first standardized radio counterpart to magnetic field archives. The database contains over 160,000 calibrated AR observations from 2009 to 2025, each including 3-18 GHz spectra and rich metadata. We demonstrate the scientific utility of this database by using machine learning to forecast solar flares. The radio spectra are first compressed into low-dimensional embedded features using an autoencoder, which are then used as predictors in baseline logistic regression classifiers. We compare the predictive power of these embedded RARPs features with that of the 18 SHARPs magnetic field parameters provided in the SHARPs data product headers. Our results show that while SHARPs data provides superior flare discrimination, the radio signatures in RARPs possess clear predictive potential and, for M-class and above flares, yield lower Brier Scores and positive Brier Skill Scores relative to SHARPs, indicating more accurate probabilistic forecasts for these events. This establishes radio data as a valuable and complementary information source.

Figures (9)

• Figure 1: Workflow of the RATANSunPy package. The toolkit supports automatic retrieval of raw RATAN-600 solar patrol data, full-disk calibration and preprocessing, extraction of active-region–centered radio spectra, and derivation of per-AR microwave features in standardized FITS format.
• Figure 2: Schema of AR spectra extraction. The left panel shows how full-disk photospheric data from SDO/HMI are spatially aligned with selected RATAN-600 full-disk radio scans at 3.00 GHz (green), 3.75 GHz (blue), 9.35 GHz (red), and 10.00 GHz (yellow). The right panel illustrates how photospheric active-region patches are matched with their corresponding cutouts extracted from the RATAN-600 full-disk radio spectrum. Image was taken from https://sdo.gsfc.nasa.gov/assets/img/latest/
• Figure 3: Convolutional autoencoder schema with examples of reconstructed data.
• Figure 4: Training and validation loss curves of the convolutional autoencoder over 20 epochs.
• Figure 5: Comparison of ROC-AUC scores for Logistic Regression models on SHARPs and RARPs datasets across flare types and prediction windows.