Table 1
Evolution of the ESPERTA model in terms of the provided forecast (severity of the SPEs and time when the forecast is issued), validation period and type, and scores (probability of detection-POD, false alarm rate-FAR, median, and average (AVG) warning time-WT).
| References | Forecast | Forecast time | Validation | Median (AVG) WT | POD | FAR |
|---|---|---|---|---|---|---|
| Laurenza et al. (2009) | ≥S1 SPEs | 10 min after ≥ M2 flare peak time | 1995–2005 | 55 min* | 63% | 42% |
| Alberti et al. (2017) | ≥S1 SPEs | 10 min after ≥ M2 flare peak time | 2006–2014 | ~2 h (~7 h) | 59% | 30% |
| Laurenza et al. (2018) | ≥S1 SPEs | 10 min after ≥ M2 flare peak time | 1995–2014 | ~4.8 h (~9 h) | 63% | 38% |
| Laurenza et al. (2018) | ≥S2 SPEs | S1 threshold crossing | 1995–2014 | ~1.7 h (4 h) | 75% | 24% |
| Alberti et al. (2019) | ≥S1 SPEs | 10 min after ≥ M2 flare peak time | 1995–2017 | 6.7 h (–) | 63% | 38% |
| Alberti et al. (2019) | ≥S2 SPEs | S1 threshold crossing | 1995–2017 | – | 75% | 23% |
| Stumpo et al. (2021) | ≥S1 SPEs | 10 min after ≥ M2 flare peak time | 1995–April 2017; stratified cross-validation | – | 63%** | 42% |
Based on the 31–50 MeV protons onset times published by Posner (2007).
In Stumpo et al. (2021) a theoretical POD of 76% was computed without including the 21 missed SPEs due to <M2 flares. Here we considered them to obtain the operational POD, which is comparable with the other ESPERTA ones.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.