J/ApJ/786/19 Statistical analysis of solar active regions (Barnes+, 2014)
Helioseismology of pre-emerging active regions.
III. Statistical analysis.
Barnes G., Birch A.C., Leka K.D., Braun D.C.
<Astrophys. J., 786, 19 (2014)>
=2014ApJ...786...19B 2014ApJ...786...19B (SIMBAD/NED BibCode)
ADC_Keywords: Sun ; Models
Keywords: methods: statistical - Sun: helioseismology - Sun: interior -
Sun: magnetic fields - Sun: oscillations - sunspots
Abstract:
The subsurface properties of active regions (ARs) prior to their
appearance at the solar surface may shed light on the process of AR
formation. Helioseismic holography has been applied to samples taken
from two populations of regions on the Sun (pre-emergence and without
emergence), each sample having over 100 members, that were selected to
minimize systematic bias, as described in Paper I (Leka et al.
2013ApJ...762..130L 2013ApJ...762..130L). Paper II (Birch et al. 2013ApJ...762..131B 2013ApJ...762..131B)
showed that there are statistically significant signatures in the
average helioseismic properties that precede the formation of an AR.
This paper describes a more detailed analysis of the samples of
pre-emergence regions and regions without emergence based on
discriminant analysis. The property that is best able to distinguish
the populations is found to be the surface magnetic field, even a day
before the emergence time. However, after accounting for the
correlations between the surface field and the quantities derived from
helioseismology, there is still evidence of a helioseismic precursor
to AR emergence that is present for at least a day prior to emergence,
although the analysis presented cannot definitively determine the
subsurface properties prior to emergence due to the small sample sizes.
Description:
In brief, samples from two populations are considered: "pre-emergence"
targets (PE) that track a 32°x32° patch of the Sun prior to the
emergence of a NOAA-numbered AR and "non-emergence" targets (NE)
selected for lack of emergence and lack of strong fields in the
central portions of the tracked patch. The PE sample size comprises
107 targets obtained between 2001 and 2007, matched to 107 NE targets
drawn from an initially larger sample and selected further to match
the PE distributions in time and observing location on the disk.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table4.dat 29 1325 Best Performing Variables with Skill Score>0.27
table5.dat 29 1335 Best Performing Variables with Skill Score>0.24
--------------------------------------------------------------------------------
See also:
J/ApJ/759/141 : Emission measures in solar active regions (Warren+, 2012)
Byte-by-byte Description of file: table4.dat table5.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 7 A7 --- Variable Variable name
9 I1 --- Weight [1/3] Weight factor used in spatial average (1)
11- 14 A4 --- Filter Filter number
16 I1 --- TI [0/4] Time Interval
18- 23 F6.3 --- PSS Peirce Skill Score (2)
25- 29 F5.3 --- e_PSS Uncertainty in PSS
--------------------------------------------------------------------------------
Note (1): Weight factor as follows:
1 = uniform;
2 = cos θ;
3 = sin θ.
Note (2): We used the Peirce skill score (Peirce, 1884Sci.....4..453P 1884Sci.....4..453P). It is
given by PSS=npp/np-nnp/nn, where where npp is the number of regions
that were classified by the discriminant analysis to be emergences and did
emerge, np is the number of PE regions, nnp is the number of regions that
were classified by the discriminant analysis to be non-emergences but did
emerge, and nn is the number of NE regions.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
References:
Leka et al. Paper I 2013ApJ...762..130L 2013ApJ...762..130L
Birch et al. Paper II 2013ApJ...762..131B 2013ApJ...762..131B
(End) Prepared by [AAS], Tiphaine Pouvreau [CDS] 20-Jul-2017