Storms End

Header Image
Storms End: AI enabled solution to visualize and predict Geomagnetic Storms from NASAs' "DISCOVR" spacecraft

Background

Our team, Storm Troopers, from Khulna University, has proudly won the NASA Space Apps Challenge 2023 Hackathon in the Khulna region. We are among the nine teams from Bangladesh selected to compete in the global round of this prestigious competition. The hackathon began in September, with over 250 teams participating in the initial selection round. Each team had to choose from 30 challenges provided by NASA, develop a solution, and present it in a four-minute video. Based on this round, 50 teams were selected for the national-level hackathon, where we competed against the best minds in the country. The national-level hackathon was held non-stop for 36 hours on October 6-7 at Independent University, Bangladesh (IUB).

Our team worked on the challenge "Developing the Oracle of DSCOVR," which focused on utilizing data from NASA's DSCOVR satellite for scientific and technological advancements. We developed a machine learning model that can predict solar storms with 95% accuracy using 1.5 years of satellite data. Following our victory in the Khulna region, we received warm congratulations from Professor Dr. Mahmud Hossain, Vice-Chancellor of Khulna University. He expressed his pride in our achievement, stating that it enhances the university’s reputation and paves the way for future success on both national and global platforms.

What is it

A geomagnetic storm is a major disturbance of Earth's magnetosphere that occurs when there is a very efficient exchange of energy from the solar wind into the space surrounding Earth. While the storms create a beautiful aurora, they also can disrupt navigation systems such as the Global Navigation Satellite System (GNSS) and create harmful geomagnetic induced currents (GICs) in the power grid and pipelines. NASA's "DSCOVR" spacecraft has long been used to collect magnetometer readings from earths magnetic field which allowed scientists to make predictions about geomagnetic storms. But, occasional faulty data which is a result of the environment itself, is making the prediction task tougher. Hence, we; team "Storm Troopers"; are proposing "Storms End"; an AI enabled solution to visualize and predict geomagnetic storm.



Our solution includes a Neural Network consisting of Spatial and Temporal modules to correctly capture the variability in magnetic readings made from "DSCOVR". The model will be able to take previous readings as input and will output the probability for a geomagnetic storm on a spatio-temporal scale. "Storms End" will also have visualization features where users will be able to see geomagnetic patterns in a 3D environment and how each area on earth is getting affected by geomagnetic storm. Similarly, it will also contain visualization on what other measurements made from "DSCOVR" are changing in time and space. Through "Storms End", we wish to build a solution that can accurately predict geomagnetic storms on a finer scale and we hope that our visualization features will motivate others to create solutions in this domain.

Team Background

We are a multidisciplinary team from Khulna University, bringing together expertise from both engineering and communication fields. Our team consists of:

Additional Information

1. নাসার স্পেস অ্যাপস্ চ্যালেঞ্জ হ্যাকাথনে আঞ্চলিক চ্যাম্পিয়ন খুবির দল - The Daily Campus News
2. নাসার স্পেস অ্যাপস্ চ্যালেঞ্জ হ্যাকাথনে আঞ্চলিক চ্যাম্পিয়ন খুবির দল - Channel 24
3. Project Description - NASA Space Apps Challenge Bangladesh 2024
© 2023 Md. Johir Raihan