By Dr. Kandis Boyd Wyatt
Faculty Member, Transportation and Logistics
Where do people get their weather forecasts from? The answer is the National Weather Service, which provides free forecasts and updates for the United States and its territories; commercial providers offer some tailored services for a fee. In either case, weather forecasts today are more accurate and accessible than ever, with newspapers, television, the internet, and cell phones as the leading sources of meteorological information.
However, your age may determine how you get your weather information. More seasoned adults get their weather forecasts from television or their daily newspaper, while most young people and 20-somethings get their weather forecast from a cell phone application (app).
Likewise, people are more likely to check several times a day when the weather is changing rapidly or is threatening, such as a forecast of heavy rains and flooding or a snowstorm with high accumulations. However, most people ignore the forecast when the weather is favorable.
Who Determines the Weather Forecast?
Shockingly, your forecast may be created using artificial intelligence (AI). AI and weather are nothing new. AI supplements human forecasters and has proven to improve specific forecasts, such as rip currents and wildfires. Mathematical models called algorithms have been used to detect abnormalities and recognize perturbations in the atmosphere. In addition, these algorithms can decrease the standard four- to six-hour time frame to produce a forecast.
How Accurate Is Your Five-Day and Seven-Day Forecast?
Most forecasts are provided in one-day, five-day and seven-day segments. So which is the most reliable weather forecast? The answer may surprise you.
Same-day forecasts are highly accurate in terms of predicting an accurate temperature within three degrees and the timing and amount of expected precipitation. According to Sci Jinks, a five-day forecast is accurate approximately 90% of the time and a seven-day forecast can accurately predict the weather about 80% of the time. However, forecasts more than seven days in advance have an accuracy of less than 50%.
With the advancements in AI, technology and data collection, far more people have access to key weather instruments, models, and satellite data, and thus can become forecasters in their own right. Does this mean your average citizen has the ability to become a professional weather forecaster? Some have done just that, creating and publishing their own forecasts with the help of the internet. As a result, there is now a plethora of websites and cell phone apps all dedicated to predicting the weather.
Will AI Reduce Meteorology Jobs?
People are concerned that AI and forecast automation will reduce the need for trained meteorologists. To a certain degree, this is true.
However, an emerging skill set of programming automated technologies will be needed in the future. In addition, trust in the forecast may be directly related to the experience of the person or technology providing the forecast. Reliability and trust go hand in hand — people will not make informed decisions if they don’t believe the forecast.
For over 100 years, the weather forecast was created by a human/machine mix. Human intervention and supervision were essential to ensure timely forecasts. So it is easy to assume human involvement only adds to the accuracy of the forecast.
Time Will Tell If Weather Forecasts Predicted by Automated Technology Will Be Accepted by the Public
As automated technologies are infused into the process, however, will there be a decrease in acceptance and trust of weather forecasts? Only time will tell. But for now, forecasting skill is basically a product of time, experience, and technology so experienced forecasters can provide the best analyses and predictions possible.