Parameters Derived from the SDO/HMI Vector Magnetic Field Data: Potential to Improve Machine-learning-based Solar Flare Prediction Models

Abstract

Citation: J Wang et al. "Parameters derived from the SDO/HMI vector magnetic field data: potential to improve machine-learning-based solar flare prediction models. AbstractIt is well established that solar flares and coronal mass ejections (CMEs) are powered by the free magnetic energy stored in volumetric electric currents in the corona, predominantly in active regions (ARs). Much effort has been made to search for eruption-related signatures from magnetic field observed mostly in the photosphere; and the signatures are further employed for predicting flares and CMEs. The parameters in the Space-weather HMI Active Region Patches (SHARP) data from the Solar Dynamics Observatory/HMI observation of vector magnetic field are designed and generated for this purpose. In this paper, we report research done on modification of these SHARP parameters with an attempt to improve flare prediction. The newly modified parameters are weighed heavily by magnetic polarity inversion lines (PIL) with high magnetic gradient, as suggested by Schrijver, by multiplying the parameters with a PIL mask. We demonstrate that the number of the parameters that can well discriminate erupted and nonerupted ARs increases significantly by a factor of two, in comparison with the original parameters. This improvement suggests that the high-gradient PILs are tightly related with solar eruption that agrees with previous studies. This also provides new data that possess potential to improve the machine-learningbased solar flare prediction models.

The development of PIL extraction application in active region of SDO/HMI magnetogram

“…Furthermore, a supervised random forest model was developed to classify active areas into non-strong flaring and strong flaring groups. Wang et al, (2019) in another study also stated that the properties of PILs in ARs are strongly correlated to solar flares and CME occurrences [13].…”

“…Research on PIL itself and PIL as training data for machine learning to predict the solar flare has been carried out. Wang et al, (2019) extracted the PIL with a high gradient from the modified HARP data as training input for the prediction model [13]. Sadikov & Kosovichev (2017) has analyzed the relationship between X-ray peak flux and PIL characteristics in the active region.…”

Interpreting LSTM Prediction on Solar Flare Eruption with Time-series Clustering

“…Furthermore, Space-weather HMI Active Region Patches, or SHARPs are an additional data product providing physical parameters calculated from the vector field, which are relevant to solar flare production, see Bobra et al (2014) for detailed descriptions of these features. In this paper, we use the SHARP parameters that are calculated from vector magnetogram pixels along the polarity inversion line (PIL), with the detailed procedure of detecting PIL being discussed in Wang et al (2019). This section introduces the PIL-based SHARP parameter dataset, the steps on how we collect the predictors and the overview of M, X, B first flares for training and testing the machine learning model.…”

“…The SDO/HMI vector magnetic field images and SHARP parameters are available for download from Joint Science Operations Center (JSOC). Wang et al (2019) utilizes the method mentioned in Schrijver (2007). With the high-resolution vector magnetic field data of each HARP region, they first took the radial field, B r , and produced two bitmaps, with one labelling all pixels satisfying B r > 200G as 1 and 0 otherwise, and the other labelling all pixels satisfying B r < −200G as 1 and 0 otherwise.…”

“…To calculate all the SHARP parameters from Table 3 in Bobra et al (2014), excluding the ones related to Lorentz force, they multiplied the weighted PIL mask with the HMI vector magnetic field image pixel-wisely, and derived SHARP parameters based on the product image. See Figure 1 in Wang et al (2019) for an example of the weighted PIL mask generated for HARP region 377.…”