A Comparison of Flare Forecasting Methods. I. Results From the “All-Clear” Workshop

Barnes, G.; Leka, K. D.; Schrijver, Carolus J.; Colak, Tufan; Qahwaji, Rami; Ashamari, Omar W; Yuan, Yixuan; Zhang, J.; McAteer, R. T. James; Bloomfield, D. Shaun; Higgins, Paul A. T.; Gallagher, Peter T.; Falconer, D. A.; Georgoulis, Manolis K.; Wheatland, M. S.; Balch, C. C.; Dunn, T. M.; Wagner, Eric

doi:10.3847/0004-637x/829/2/89

Cited by 211 publications

(251 citation statements)

References 67 publications

Supporting

Mentioning

230

Contrasting

Order By: Relevance

“…3F) was empirically established as a powerful eruptive indicator by Schrijver (2007). The parameters H tot c , I tot , / tot , F, and the R value exhibit a very similar evolution, possibly due to an existing correlation between each parameter (Barnes et al 2016). There is no evidence of physical changes prior to the eruptive flare, at about t~120 t 0 (vertical dashed gray line) for the eruptive simulations.…”

Section: Eruptive Indicators Evolution With B Mask = 30 Gmentioning

confidence: 97%

“…Various systems of prediction invoking different categories of models have been developed in the past decades, as e.g., the ''Theophrastus'' tool (McIntosh 1990), the linear-prediction system of Gallagher et al (2002) used by SolarMonitor, the discriminant analysis of Barnes et al (2007), the Automated Solar Activity Prediction (ASAP) of Colak & Qahwaji (2009), based on machine learning, and more recently the statistical learning technique of Yuan et al (2010). A recent comparison between the current forecasting tools using lineof-sight (LOS) magnetograms has been performed by Barnes et al (2016), showing that none of them substantially outperformed all others. All these forecasting techniques, including the future FLARECAST forecasting tool, require scalar quantities derived from photospheric magnetic field to be able to make flare and/or eruption predictions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Testing predictors of eruptivity using parametric flux emergence simulations

Guennou

Pariat

Leake

et al. 2017

J. Space Weather Space Clim.

View full text Add to dashboard Cite

Solar flares and coronal mass ejections (CMEs) are among the most energetic events in the solar system, impacting the near-Earth environment. Flare productivity is empirically known to be correlated with the size and complexity of active regions. Several indicators, based on magnetic field data from active regions, have been tested for flare forecasting in recent years. None of these indicators, or combinations thereof, have yet demonstrated an unambiguous eruption or flare criterion. Furthermore, numerical simulations have been only barely used to test the predictability of these parameters. In this context, we used the 3D parametric magnetohydrodynamic (MHD) numerical simulations of the self-consistent formation of the flux emergence of a twisted flux tube, inducing the formation of stable and unstable magnetic flux ropes of Leake et al. (2013Leake et al. ( , 2014. We use these numerical simulations to investigate the eruptive signatures observable in various magnetic scalar parameters and provide highlights on data analysis processing. Time series of 2D photospheric-like magnetograms are used from parametric simulations of stable and unstable flux emergence, to compute a list of about 100 different indicators. This list includes parameters previously used for operational forecasting, physical parameters used for the first time, as well as new quantities specifically developed for this purpose. Our results indicate that only parameters measuring the total non-potentiality of active regions associated with magnetic inversion line properties, such as the Falconer parameters L ss , WL ss , L sg , and WL sg , as well as the new current integral WL sc and length L sc parameters, present a significant ability to distinguish the eruptive cases of the model from the non-eruptive cases, possibly indicating that they are promising flare and eruption predictors. A preliminary study about the effect of noise on the detection of the eruptive signatures is also proposed.

show abstract

Section: Eruptive Indicators Evolution With B Mask = 30 Gmentioning

confidence: 97%

Section: Introductionmentioning

confidence: 99%

Testing predictors of eruptivity using parametric flux emergence simulations

Guennou

Pariat

Leake

et al. 2017

J. Space Weather Space Clim.

View full text Add to dashboard Cite

show abstract

“…However, since the underlying mechanisms leading to the generation of solar eruptions have not yet been indisputably determined, no sufficient conditions of solar eruptivity have yet been established. Solar flare predictions are thus still strongly driven by empirical methods (e.g., Yu et al 2010;Falconer et al 2011Falconer et al , 2014Barnes et al 2016). Nowadays most predictions rely on the determination and characterization of solar active regions through the use of several parameters and the statistical comparison with the historical eruptivity of past active regions presenting similar values of these parameters.…”

Section: Introductionmentioning

confidence: 99%

Relative magnetic helicity as a diagnostic of solar eruptivity

Pariat

Leake

Valori³

et al. 2017

A&A

103

153

View full text Add to dashboard Cite

Context. The discovery of clear criteria that can deterministically describe the eruptive state of a solar active region would lead to major improvements on space weather predictions. Aims. Using series of numerical simulations of the emergence of a magnetic flux rope in a magnetized coronal, leading either to eruptions or to stable configurations, we test several global scalar quantities for the ability to discriminate between the eruptive and the non-eruptive simulations. Methods. From the magnetic field generated by the three-dimensional magnetohydrodynamical simulations, we compute and analyze the evolution of the magnetic flux, of the magnetic energy and its decomposition into potential and free energies, and of the relative magnetic helicity and its decomposition. Results. Unlike the magnetic flux and magnetic energies, magnetic helicities are able to markedly distinguish the eruptive from the non-eruptive simulations. We find that the ratio of the magnetic helicity of the current-carrying magnetic field to the total relative helicity presents the highest values for the eruptive simulations, in the pre-eruptive phase only. We observe that the eruptive simulations do not possess the highest value of total magnetic helicity. Conclusions. In the framework of our numerical study, the magnetic energies and the total relative helicity do not correspond to good eruptivity proxies. Our study highlights that the ratio of magnetic helicities diagnoses very clearly the eruptive potential of our parametric simulations. Our study shows that magnetic-helicity-based quantities may be very efficient for the prediction of solar eruptions.

show abstract

“…Nevertheless, the fundamental magnetic coupling between the photosphere and the corona has motivated the use of photospheric field parameters for predicting flares, not based on physical flare models but on various approaches of statistics and machine learning (see Bloomfield et al 2012;Barnes et al 2016, and references therein). In particular, machine learning is a subfield of computer science that enables algorithms to learn from the input (training) data and make data-driven predictions.…”

Section: Introductionmentioning

confidence: 99%

Predicting Solar Flares Using SDO/HMI Vector Magnetic Data Products and the Random Forest Algorithm

Liu

Deng

Wang

et al. 2017

ApJ

124

122

View full text Add to dashboard Cite

Adverse space weather effects can often be traced to solar flares, prediction of which has drawn significant research interests. The Helioseismic and Magnetic Imager (HMI) produces full-disk vector magnetograms with continuous high cadence, while flare prediction efforts utilizing this unprecedented data source are still limited. Here we report results of flare prediction using physical parameters provided by the Space-weather HMI Active Region Patches (SHARP) and related data products. We survey X-ray flares occurred from 2010 May to 2016 December, and categorize their source regions into four classes (B, C, M, and X) according to the maximum GOES magnitude of flares they generated. We then retrieve SHARP related parameters for each selected region at the beginning of its flare date to build a database. Finally, we train a machine-learning algorithm, called random forest (RF), to predict the occurrence of a certain class of flares in a given active region within 24 hours, evaluate the classifier performance using the 10-fold cross validation scheme, and characterize the results using standard performance metrics. Compared to previous works, our experiments indicate that using the HMI parameters and RF is a valid method for flare forecasting with fairly reasonable prediction performance. To our knowledge, this is the first time that RF is used to make multi-class predictions of solar flares. We also find that the total unsigned quantities of vertical current, current helicity, and flux near polarity inversion line are among the most important parameters for classifying flaring regions into different classes.

show abstract

A Comparison of Flare Forecasting Methods. I. Results From the “All-Clear” Workshop

Cited by 211 publications

References 67 publications

Testing predictors of eruptivity using parametric flux emergence simulations

Testing predictors of eruptivity using parametric flux emergence simulations

Relative magnetic helicity as a diagnostic of solar eruptivity

Predicting Solar Flares Using SDO/HMI Vector Magnetic Data Products and the Random Forest Algorithm

Contact Info

Product

Resources

About