Automated Recognition and Characterization of Solar Active Regions 
Based on the SOHO/MDI Images

Principal Investigator / Institution:  Judit M. Pap/UCLA
Co-Investigator:  Dr. Michael Turmon/JPL

The objective of the proposed research is to identify and characterize the 
various surface structures on the Sun which may contribute to the changes 
in solar total and spectral irradiances.  For this purpose we will analyze the 
high precision continuum intensity and magnetic maps produced by the 
Michelson Doppler Imager (MDI) instrument of the Solar Oscillation 
investigation (SOI) experiment on SOHO.  The main goals are to 
(1) characterize the active regions and study the evolution; (2) study the 
energy budget of active regions; (3) identify and characterize the active 
and quiet network; and (4) estimate the contribution of sunspots, faculae, 
network, and large scale structures to the changes in total solar and spectral 
irradiances as measured by the Variability IRradiance Gravity Oscillations 
(VIRGO) experiment on SOHO.  To carry out the analysis of the MDI 
images, we establish state-of-the art image analysis and interpretation 
algorithms, using statistical pattern recognition and machine learning 
techniques, which will guarantee the objective, uniform, and automated 
processing of large numbers of high-quality images.

We will study the changes in the area, position, intensity, and contrast of 
sunspots and faculae during the evolution of active regions and we will 
identify and characterize the active and quite network.  The proposed 
research requires processing of hundreds of images, for which we will 
establish a pattern recognition system adapted from related methods for 
identifying chromospheric features via ground-based Ca(( K intensity maps.  
For identifying and characterizing the active regions and the magnetic network, 
analysis of the full disk, high resolution (1024 x 1024 pixels) continuum 
intensity images and magnetic maps of MDI will be required, which are 
available a few times per day.  The lower spatial resolution (128 x 128 pixels) 
flux budget product with high temporal resolution will be used to study the 
evolutionary effects.  To validate our results, the MDI images will be compared 
to the white-light and Ca(( K images Precision solar Photometric Telescopes 
(PSPT) and the Ca(( K spectroheliograms taken at the National Solar Observatory 
at Sacramento Peak.  Study of the active regions and network 
characteristics will make it possible to study the energy budget
of active regions and the role of various features in irradiance changes.
These studies will improve our understanding of the structural/dynamical
changes in the photosphere and convective zone.

The proposed research meets with several science objectives of the ESA/NASA 
SOI/MDI experiment, such as the "Magnetic Field Measurements," "Radiative 
FluxBudget," and "Limb Figure Measurements." Although the  
analysis of the evolution of active regions and their energy budget 
are important research tasks for both MDI and VIRGO experiments on SOHO, 
no significant progress has been made to study these topics. The 
largest obstacle is the lack of manpower necessary to analyze the 
large number of images produced by MDI.  Within the proposed research 
we will apply the automated image processing system to the MDI  
images and we will provide it to the SOHO/MDI team and the broader 
ESA and NASA community.