Could Artificial Intelligence Replace Broadcast Meteorologists?
The world of meteorology is changing. Between the need for 24/7 weather coverage and increasingly severe weather events, broadcast meteorologists are under pressure to deliver accurate, up-to-the-minute forecasts. But could artificial intelligence replace them? The short answer is yes, but it’s not going to happen overnight.
To demonstrate the use of artificial intelligence, this entire blog post, except for this sentence, was written by copy.ai with minimal editing.
Machine learning – the “training” of artificial intelligence to complete a task – has made significant progress in recent years. It’s how Google Translate can translate from one language to another almost instantly, how Netflix recommends movies based on your viewing history, and how Amazon can ship you a product before you know you want it.
A recent study by AccuWeather predicted that 30% of jobs in the weather industry could be replaced by artificial intelligence within a decade. We already have supercomputers that allow us to analyze data from satellites, radar and other sources with extreme speed and accuracy—allowing us to make more accurate predictions about what will happen next over time than ever before in human history.
AI has been used for weather forecasting for some time now, but only on a limited basis and with limited success. Computers were unable to take in all the data necessary for accurate forecasts and make sense of it; they could only analyze localized, short-term information (such as what’s happening right now).
Today’s machine learning algorithms are able to process massive amounts of data at once (in milliseconds), allowing them to look at trends over longer periods of time than ever before – up to ten days into the future! This means that AI is getting better at predicting when rain will fall or snow will fall in any given location across the globe. This is especially true if we’re talking about specific locations: while most meteorologists might be able to predict whether it will rain or snow somewhere within 50 miles of their station, an AI could do so anywhere on Earth. Machine learning algorithms even have the ability now potentially replace human meteorologists altogether!
AI can access weather data and convert that data into conversational language and images. Weather forecasting is a complex process, but AI can use natural language processing to write news stories about the weather and forecast conditions for specific locations. The ability to answer questions about the weather means that AI is already able to provide more information than your average meteorologist.
AI can also tell us how climate change is affecting our planet’s atmosphere on a global scale, while also providing granular details down to a city level (or even smaller). The software can determine when rainstorms will emerge over densely populated areas like New York City or Los Angeles without needing any human input; it can detect anomalies such as unexpected snowfall despite being programmed with normal data; it can even report on changes in wind speed over time so we know whether we should expect gusts strong enough for storm warnings later today or tomorrow morning instead.
Artificial intelligence has been around for a while, but it’s getting better in different ways every day. Meteorologists have been using AI to help them forecast the weather since the 1950s, when IBM built its first supercomputer. It isn’t perfect and sometimes gets it wrong. But then again, so do broadcast meteorologists, if we’re being honest. We make forecasts that are off by a few degrees, or say there will be rain when there isn’t any at all (sometimes called “rain on the parade”). So why not let machines do what they do best—be accurate?
Tim Heller is very real. He’s an AMS Certified Broadcast Meteorologist, Talent Coach, and Weather Content Consultant. He helps local TV stations and broadcast meteorologists implement effective weather coverage strategies on-air, online, and on social media.