IDL Helps South Korean Scientists Predict Space Weather

Background

In 1859, the sun let loose a large coronal ejection that caused a geomagnetic storm that lit up the sky and frazzled communication wires around the world. It caused auroras knows as the “Carrington Event” that were said to have awakened campers in the middle of the night in Colorado and sent thousands into the streets in New York City. If a solar storm of this magnitude were unleashed today, it would be catastrophic. A much smaller solar-induced geomagnetic storm in 1989, left 6 million people in Quebec without power for nine hours. Lloyd’s of London, the world’s oldest insurance market, estimates that losses for an especially severe solar storm could reach $2.6 trillion. Analysts are most concerned about the immediate impact of a large geomagnetic storm on electrical grids, but the bigger worry is the aftermath of having millions of people without electricity for weeks, months, or longer.

Problem

Predicting space weather has become more important as the world has grown increasingly dependent upon systems that are vulnerable to damage from solar storms, like electrical grids and satellites. A coronal mass ejection that reaches the earth’s atmosphere can cause a geomagnetic storm and a temporary loss of electrical power over a large area, but the effects of solar storms could be mitigated with accurate and timely forecasting. Advance warning of an impending solar storm could allow for crucial or sensitive power-grid components to be shut down and protected. There are also about 150 satellites currently orbiting Earth, many for the purpose of relaying television and telephone signals. These satellites are susceptible to disruption from significant modulations in the ionosphere which can occur during solar storms, but having advanced notice could mean putting them in a low-power state to minimize the damage.

Coronal hole detection with Automated Solar Synoptic Analysis (ASSA).

Solution Achieved

Automated Solar Synoptic Analysis (ASSA) is an automated software system that identifies active solar regions, filament channels, and coronal holes -- three major solar sources that cause space weather. ASSA is currently being used by the Korean Space Weather Center (KSWC) of the Radio Research Agency and SELab, Inc. in Korea. Sangwoo Lee, an employee of SELab, Inc. was instrumental in building ASSA along with the ASSA GUI. According to Lee, the R&D team of SELab, Inc. has been successfully using IDL to develop scientific solutions for research as well as for operational uses for many years. “The team uses IDL because of its strong capabilities for image processing and data handling.”

IDL^®, the scientific programming language that extracts meaningful visualizations from complex numerical data, was crucial in developing ASSA and ASSA GUI. Lee said that he and other SELab employees developed ASSA entirely with IDL. “The IDL procedures were installed and are in operation at KSWC,” said Lee. “They are running in real-time with one-hour cadence and the latest results can be viewed on the ASSA website.” The ASSA GUI is a standalone application program developed with IDL. “This lets ordinary users emulate the whole procedures of ASSA in their personal computers, even if they don’t have IDL installed on their computer,” said Lee. The IDL-based GUI is wrapped up as a distributable file package and enables anyone to obtain classification results not just for the latest solar data, but also data for any date. “The entire procedures of data processing for recognition and classification of sunspot groups in ASSA can be checked step by step on the ASSA GUI, which is quite helpful in understanding the results of ASSA,” according to Lee. The ASSA GUI is available for MS Windows and Mac OS, and is being distributed to public through the website.

Detected sunspot groups with ASSA.

ASSA receives necessary solar raw images and processes them in a designated procedure to yield resultant output files. When identifying active solar 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 solar flares. 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.

Detected filaments using ASSA.

The output results of ASSA are routinely checked and validated against NOAA’s daily SRS (Solar Region Summary) and UCOHO (URSIgram code for coronal hole information). Space weather forecasters at NOAA Space Weather Prediction Center produce solar synoptic drawings on a daily basis to predict solar activities like solar flares, filament eruptions, high-speed solar wind streams, and co-rotating interaction regions as well as their possible effects on the Earth. These drawings are done “by hand”, are time and labor intensive, and require programming expertise. The NOAA results are summarized and issued once a day. “Reliable results can only be produced by experts, and even so, consistency in results might vary based on different opinions and judgments of these experts,” said Lee.

ASSA automatically performs detection and classification of sunspot groups quite frequently and in a consistent manner. Space weather forecasters can use ASSA to produce more consistent results and do so more frequently. According to Lee, KSWC and SELab, Inc. has been discussing how ASSA can be improved to help forecasters. In December 2013, the Community Coordinated Modeling Center (CCMA) at NASA installed ASSA and is now offering access to the real-time results produced by ASSA their website. This is the first space weather model provided from Korea to CCMC.