ASSA (Automatic Solar Synoptic Analyzer)
1,2
Hong ,
3
Lee ,
3
Oh ,
1
Kim ,
1
Yoon ,
1,2
Lee ,
Sunhak
Sangwoo
Seung Jun
Jae Hun
Gi-Chang
Jae-Hyung
1
1
2
2
Young-Kyu Kim , Jaehyung Lee , Yong Jae Moon and Dong-Hun Lee
1: Korean Space Weather Center, RRA, Jeju, Korea., 2: School of Space Research, KHU, Young-in, Korea., 3: SELab, Inc., Seoul, Korea
Abstract
We have developed an automated software system of identifying solar active regions, filament channels, and coronal holes, those are three major solar sources causing
the space weather. Space weather forecasters of NOAA Space Weather Prediction Center produce the solar synoptic drawings as a daily basis to predict solar activities,
i.e., solar flares, filament eruptions, high speed solar wind streams, and co-rotating interaction regions as well as their possible effects to the Earth. As an attempt to
emulate this process with a fully automated and consistent way, we developed a software application named ASSA(Automatic Solar Synoptic Analyzer). When identifying
solar active regions, ASSA uses high-resolution SDO HMI intensitygram and magnetogram as inputs and providing McIntosh classification and Mt. Wilson magnetic
classification of each active region by applying appropriate image processing techniques such as thresholding, morphology extraction, and region growing. At the same
time, it also extracts morphological and physical properties of active regions in a quantitative way for the short-term prediction of flares and CMEs. When identifying
filament channels and coronal holes, images of global H-alpha network and SDO AIA 193 are used for morphological identification and also SDO HMI magnetograms for
quantitative verification. The output results of ASSA are routinely checked and compared with NOAA's daily SRS (Solar Region Summary) and UCOHO (URSIgram code
for coronal hole information).
The Process Flow in ASSA
Data I/O
Preprocess
Characterize
Detect &
Classify
SDO HMI & AIA,
Ground H-alpha, etc.
(JPEG, JP2)
Masking, Flattening, Exposure
correction, limb darkening
correction… etc
Existence of penumbra, polarity,
neutral line, compactness,
asymmetry, ellipticity, etc
The key parameters for the classification and detection
Table 1.
Table 2.
Parameter
Value
Unit
Parameter
Value
Unit
factor of standard deviation to be subtracted from mean
intensity to determine threshold intensity for sunspot detection
6.0
None
factor of median intensity of solar disk to determine
candidates for coronal hole
0.45
None
factor of standard deviation to be applied to standard-deviationbased region growing
1.5
threshold of skewness of polarity distribution inside
coronal hole area
skewness of polarity
distribution over entire
solar disk
None
factor of median intensity of solar disk to determine
candidates for filament
0.82
None
Compare&
Verify
NOAA SRS report
UCOHO
Archive
McIntosh and Mag class,
Location, Area, Number of spots,
the derived physical parameters, etc.
None
maximum separation between two sunspots to be considered to
belong to the same spot group
6.0
Degree
maximum separation between sunspots to determine unipolar
or bipolar
3.0
Degree
factor of group width to determine threshold of distance
between spots with penumbra
0.5
ratio (penumbral area / entire spot area) to determine threshold
of maturity of penumbra
threshold of asymmetry of outline ellipse of spot
Quantified thresholds for
detection and classification
The Text Output from ASSA
threshold of north-south diameter of spot
0.2
25 if spot area > 100
50 if spot area < 100
2.5
threshold of ratio (sum of spot areas / entire group area) to
determine compactness
0.4
threshold of ratio (plus polarity / minus polarity) to determine
alpha or beta
0.2, 5.0
threshold of number of effective neutral lines separating
opposite polarity to determine beta or beta-gamma
2
None
None
None
Degree
None
factor of median intensity of solar disk to determine
candidates for sunspot
factor of standard deviation to be applied to standarddeviation-based region growing
0.65
2.0
structuring elements for detection of filament-like
elongated feature
None
None
None
Figure 1. The conceptual diagram for the process
flow from the data ingestion to final archive in ASSA.
The examples of the data types, pre-processing,
physical parameters and the comparison data is
shown on the right in the box.
40
Description
Unit
ASSA Number
Four-digit number of each sunspot group
None
Num Spots
Number of sunspots
None
Num Spots (with penumbra)
Number of sunspots with penumbra
None
Location
Heliographic Coordinate
Degree
Area
Area of group region
one millionth of solar disk area
McIntosh Class
Result of McIntosh classification
None
Mag Class
Result of Wilson magnetic classification
None
Total Number of Spots
Total number of sunspots for entire solar disk
None
Wolf Number
10G+S (G: number of groups, S: number of spots)
None
None
None
maximum separation between two fragments to be
considered to belong to the same filament
Item
Pixel
The key parameter and its corresponding threshold
value for the sunspot classification for Macintosh class
and the Wilson magetic Class (Table 1.) and the
detection of Corona Home and filmament channels
(Table 2.)
Figure 2. The text output retrieved from ASSA for active
region classification. The output is created with 1 hour
cadence, and archived in the database in Korea Space
Weather Center, RRA.
The Results from the test-run of ASSA
Figure 3. The results from the test-run of ASSA. The captured image of ASSA Graphic User Interface for interactive running by tuning the threshold and the appropriate input parameters (a), The classification results from ASSA(red) compared with the SRS
(green) by SWPC, NOAA. (b), the magnetic classification results from ASSA (red) compared with the SRS (green) by SWPC, NOAA.(c), the corona hole detection results from ASSA (orange line) compared with the UCOHO results (blue line,(d)), and the
filament channel detection results from ASSA (e)
Acknowledgement to colleagues of NOAA Space Weather Center for sharing their knowledge and experiences on space weather. Special thanks to Mr. Ken Tegnell for his great mentoing to the new space weather forecasters of Korea.
SH 13A-2239, AGU Fall Meeting, 2012