Geomagnetic storm catalog with solar wind drivers – Updated to July 2026

Geomagnetic storms and responsible solar wind drivers lists Detailed description of the derivation of the geomagnetic storm list is given in Section 2 of "Rapid Geomagnetic Variations During High-Speed Stream, Sheath and Magnetic Cloud-Driven Geomagnetic Storms From 1996 to 2023" by Pedersen et al (2024) and the software to generate the geomagnetic storm list is provided in Zenodo - Software for detection of geomagnetic storms and storm phases in MATLAB. To be included, the storms must reach a minimum SYM-H ≤ -50 nT. The start and end times for the initial, main and recovery phases of each storm are given in Universal Time (UT). This repository contains two geomagnetic storm catalogs: Geomagnetic_storms_detailed_1996-2026.txt: Catalog of geomagnetic storms and detailed time intervals of each solar wind driver passing Earth during the storms. Geomagnetic_storms_extended_1981-2026.txt: Catalog of all geomagnetic storms since the beginning of the SYM-H index in 1981. Only the primary solar wind driver (PD) active for the longest time during that storm's main phase is listed for storms from 1996 onwards. Solar wind and interplanetary drivers The Interplanetary Coronal Mass Ejections (ICME) are based on the Richardson and Cane ICME list. The sheaths are the time from the disturbance ahead of the ICME (the time interval from the disturbance until the main ejecta arrived). The main ejecta of the ICME is categorized as an ejecta without a magnetic cloud if the MC indicator of Richardson and Cane is 0, otherwise it is reported as a magnetic cloud when the MC indicator is either 2 or 1, which means that these ICMEs contain some characteristics resembling MCs. The High-speed stream and their associated Stream Interaction Regions (HSS/SIR) are classified based on an updated version of the Grandin et al. (2019) list of HSSs/SIRs. For any time interval where there are conflicts between the HSS/SIR and ICME categories, the ICME category is set to take precedence as the Richardson and Cane list uses multiple detection methods, while the HSS/SIR list of Grandin et al (2019) is based on an automatic procedure to detect these stuctures using only the solar wind and interplanetary magnetic field data. Frequently asked questions About the Geomagnetic_storms_detailed_1996-2026.txt list Why are there sometimes many rows for the same geomagnetic storm? For storms with several solar wind drivers, each solar wind driver has their distinct row. This is so that if a storm has multiple of the same driver (e.g. sheaths or MCs), the times of each one of them can be accurately displayed. What is the resolution of the solar wind drivers? The start and end times have a 1 hour resolution, with the exception of sheath start times which have 1 minute resolution. Why do some HSS/SIR, magnetic clouds, and ejecta end one minute before the hour (e.g. 11:59) or at non-hourly timestamps (e.g., 16:11) instead of on the full hour? To report only one solar wind driver active at a time, when a new interplanetary structure (such as an ICME or a subsequent sheath) arrives while an earlier HSS/SIR, MC, or ejecta is still ongoing, the overlapping interval is removed by prioritizing the newly arrived structure. The end time of the preceding structure is therefore truncated to one minute before the new structure's start time (e.g., ending at 11:59 if the new structure begins at 12:00, or ending at 16:11 if a new sheath begins at 16:12). Why is the same solar wind driver sometimes reported multiple times? If a solar wind driver occurs during two or more geomagnetic storms then that driver will be reported separately as a row for each of the geomagnetic storms. Occasionally also HSS/SIR, magnetic clouds and ejectas do not start exactly on the hour, why? The resolution for these drivers are one hour, but because of small inaccuracies in the arrival times of these structures, in the cases where a driver arrives very shortly after an SSC, it has been stretched so that its beginning is set to the time of the SSC to accurately capture the driver responsible for causing the SSC. This is done in accordance with Pedersen et al. (2024) "To account for slight uncertainty in the arrival times of new interplanetary structures, the driver causing the SSC was set as the interplanetary structure present for the majority of the time from the beginning of the SSC to 20 min after." About the Geomagnetic_storms_extended_1981-2026.txt list Why does the primary solar wind drivers not extend all the way back to 1981? The Richardson and Cane's ICME list began in 1996, therefore we do not have solar wind drivers before that time. The full list since 1981 is given in case anyone is interested in the storms and storm phases before 1996. How exactly is the primary driver determined for each storm? The primary driver (PD) is set as the solar wind driver (HSS/SIR, sheath, magnetic cloud or ejecta without magnetic cloud) that was active for the largest fraction of that storm's main phase. This is also the categorization that was used in Table 1 of Pedersen et al. (2024). For any questions feel free to contact Marcus Pedersen (email [e-mail]). This repository will be kept regularly updated; however, a newer version of the storm and driver lists may be available upon request.