Adobe Stock
Could AI save lives? That’s what meteorologists hope as they bet big on artificial intelligence tools. The National Science Foundation has invested $20 million to launch a new AI inititiate. They hope to create trustworthy forecasting using sophisticated AI that will revolutionizing weather predictions.
The National Science Foundation (NSF) invested $20 million in an initiative that has created a new institute called the “Artificial Intelligence for the Environmental Sciences (AI2ES).”
The $20 million investment by the NSF, as part of a larger program to establish other AI institutes nationwide, will focus on machine learning, molecular science, workforce development, food, and agriculture, the Oklahoman reported. The funds will also pay graduate students, postdoctoral staff, and faculty salaries over five years. The Institute will work with partners in other states, as well as private companies such as NVIDIA, Disaster Tech, Google, and IBM.
The AI2ES is headed by Doctor Amy McGovern, a professor of both meteorology and computer science, who formally worked in robotics. Dr. McGovern is currently with the University of Oklahoma school of Meteorology, the Wall Street Journal reported.
The Institute will study ways in which AI can improve the prediction of severe storms, including hail and tornadoes, coastal oceanography weather prediction, and winter weather forecasting.
“What’s happened is that a number of us, including me, have been demonstrating that AI can really improve weather forecasting,” Dr. McGovern told the Oklahoman. “And so people are starting to see it. They’re starting to see how AI helps across a variety of fields.”
“We don’t want to replace the humans, by the way,” McGovern continued. We want to become a tool that will help the humans improve their forecasts, and help keep them from being overloaded by the tremendous amounts of data that’s available to them.”
“We’ve got results showing that you can use deep learning on the radar data with a really high confidence and really good skill, to forecast tornadoes 30 to 60 minutes out,” McGovern added.
What AI can do is help forecasters crunch through the massive glut of big data. AI can spot patterns quicker and determine what they mean, helping meteorologists make more informed forecasts. After all, and severe weather, second discount.
This could drastically speed up forecasting to better prepare people for weather conditions or looming disasters, and potentially save lives and prevent accidents. However, all this may be bad news for accident lawyers–but good news for the rest of us.
Dr. McGovern is also the lead author of the 2017 study entitled “Using Artificial Intelligence to Improve Real-Time Decision-Making for High-Impact Weather,” which was published in the Journal of the American Meteorological Society.
In the study, Dr. McGovern and her colleagues “demonstrate that applying AI techniques along with a physical understanding of the environment can significantly improve the prediction skill for multiple types of high-impact weather.”
The study advocates that AI provides “big data” of insight through “computational sustainability” in areas such as “the prediction of storm duration, severe wind, severe hail, precipitation classification, forecasting for renewable energy, and aviation turbulence.”
