J/ApJ/774/L27 Solar flares predictors (Yang+, 2013)
Magnetic nonpotentiality in photospheric active regions as a predictor of
solar flares.
Yang X., Lin G., Zhang H., Mao X.
<Astrophys. J., 774, L27 (2013)>
=2013ApJ...774L..27Y 2013ApJ...774L..27Y
ADC_Keywords: Sun ; Models
Keywords: methods: statistical; Sun: activity; Sun: flares; Sun: photosphere
Abstract:
Based on several magnetic nonpotentiality parameters obtained from the
vector photospheric active region magnetograms obtained with the Solar
Magnetic Field Telescope at the Huairou Solar Observing Station over
two solar cycles, a machine learning model has been constructed to
predict the occurrence of flares in the corresponding active region
within a certain time window. The Support Vector Classifier, a widely
used general classifier, is applied to build and test the prediction
models. Several classical verification measures are adopted to assess
the quality of the predictions. We investigate different flare levels
within various time windows, and thus it is possible to estimate the
rough classes and erupting times of flares for particular active
regions. Several combinations of predictors have been tested in the
experiments. The True Skill Statistics are higher than 0.36 in 97% of
cases and the Heidke Skill Scores range from 0.23 to 0.48. The
predictors derived from longitudinal magnetic fields do perform well,
however, they are less sensitive in predicting large flares. Employing
the nonpotentiality predictors from vector fields improves the
performance of predicting large flares of magnitude ≥M5.0 and ≥X1.0.
Description:
We use the observational data of photospheric active region vector
magnetograms obtained by the Solar Magnetic Field Telescope (SMFT) at
the Huairou Solar Observing Station, National Astronomical
Observatories of China. The working spectral line for the vector
magnetograms is FeI5324.19. The data employed are selected from all
the vector magnetograms during the period from 1988 to 2008 subject to
criteria discribed in section 2.1.
The records of soft X-ray flares are available from NOAA's National
Geophysical Data Center (ftp://ftp.ngdc.noaa.gov/STP/space-weather/
solar-data/solar-features/solar-flares/x-rays/goes/).
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table4.dat 147 100 Verification results from testing Support Vector
Machine (SVM) Classifier
--------------------------------------------------------------------------------
See also:
J/A+A/574/A37 : Solar flares movies (Dalmasse+, 2015)
J/ApJ/747/L41 : Solar flares probabilities (Bloomfield+, 2012)
J/ApJ/757/94 : Solar flares observed with GOES and AIA (Aschwanden, 2012)
J/ApJ/759/69 : Solar electron events (1995-2005) with WIND/3DP (Wang+, 2012)
J/A+A/304/563 : Cool X-ray flares of Sun with GOES (Phillips+, 1995)
Byte-by-byte Description of file: table4.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 4 A4 --- --- [≤]
5- 8 A4 --- Flev Flare level (minimal)
10- 11 I2 h Twin [6/48] Time window
13- 15 A3 --- Pred Predictor combination (1)
17- 21 F5.3 --- GM [0.2/0.6] The Geometric mean (2)
23- 27 F5.3 --- POD [0.3/0.9] Probability of Detection
29- 33 F5.3 --- e_POD Uncertainty in POD
35- 39 F5.3 --- FOH [0.1/0.8] Frequency of Hits
41- 45 F5.3 --- e_FOH Uncertainty in FOH
47- 51 F5.3 --- FOCN [0.7/1] Frequency of Correct Null forecasts
53- 57 F5.3 --- e_FOCN Uncertainty in FOCN
59- 63 F5.3 --- CSI [0.1/0.6] Critical Success Index
65- 69 F5.3 --- e_CSI Uncertainty in CSI
71- 75 F5.3 --- F1 [0.2/0.7] The F1 measure
77- 81 F5.3 --- e_F1 Uncertainty in F1
83- 87 F5.3 --- TSS [0.3/0.9] True Skill Statistic
89- 93 F5.3 --- e_TSS Uncertainty in TSS
95- 99 F5.3 --- CSS [0.1/0.5] Clayton Skill Score
101-105 F5.3 --- e_CSS Uncertainty in CSS
107-111 F5.3 --- HSS [0.2/0.5] Heidke Skill Score
113-117 F5.3 --- e_HSS Uncertainty in HSS
119-123 F5.3 --- GSS [0.1/0.4] Gilbert Skill Score
125-129 F5.3 --- e_GSS Uncertainty in GSS
131-135 F5.3 --- ACC [0.6/1] Percentage of correct predictions
137-141 F5.3 --- e_ACC Uncertainty in ACC
143-147 F5.3 --- N0/N [0.5/1] Percentage of non-events
--------------------------------------------------------------------------------
Note (1): We have tried five combinations of predictors (see section 2.2),
as follows:
V06 = (<(Δψ)>, <|Jz|>, <|hc|>, |αav|,
<ρfree>, dEm),
V08 = (<(Δφ)>, <(Δψ)>, <|Jz|>, <|hc|>,
|αav|, <ρfree>, dE, dEm),
L05 = (dEm, <(∇hBz)>, (∇hBz)m, Lgnl,
<(ε(Bz))>),
A10 = (<(Δψ)>, <|Jz|>, <|hc|>, |αav|,
<ρfree>, dEm, <(∇hBz)>,
(∇hBz)m, Lgnl, <(ε(Bz))>),
A12 = (<(Δφ)>, <(Δψ)>, <|Jz|>, <|hc|>,
|αav|, <ρfree>, dE, dEm, <(∇hBz)>,
(∇hBz)m, Lgnl, <(ε(Bz))>).
Note (2): Of several verification measures (POD, FOH, TSS, HSS, GSS, CSI, F1,
and (POD*FOH*FOCN)(1/3)), see Section 3.3.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
References:
Yang et al. Paper I. 2012SoPh..280..165Y 2012SoPh..280..165Y
(End) Greg Schwarz [AAS], Emmanuelle Perret [CDS] 06-Mar-2015