Improving Weather Prediction Accuracy | StarTalk

NEIL DEGRASSE TYSON: You know what we have? We have a video dispatch from an actual local news meteorologist to help us explain how they make their predictions happen. Let's check it out.

NICK GREGORY: Hello, Dr. Tyson. Nick Gregory here at the Fox 5 Weather Center in New York City. Coming up with a weather forecast with all the raw data we get from NOAA and other sources, really is a science—huge amounts of data that get plugged into various computer models, which then present a solution.

We take that solution, apply our own knowledge of past weather events, and predict what we think is going to happen. For example, let's see how we would predict a potential blizzard. So we would start to look for those weather trends. What would they be? Let's say high-pressure ridging in from eastern Canada.

We would then look to the computer models to see what type of jet stream pattern would be in play, and would this trough of low pressure be along the East Coast, potentially with this jet stream riding up the Eastern Seaboard? Pockets of energy then started moving across the country, this upper-level energy, that once it reaches the East Coast, sets the potential for a storm to develop right in this location.

That storm could then intensify and turn into a nor'easter, and eventually track up along that jet stream track, leaving a swath of heavy snow alongside. So there you have it. Perhaps not the most accurate forecast, but we do work with science and technology so that forecast will improve over time.

NEIL DEGRASSE TYSON: Cool. I know this guy. I mean, he's our local weather guy. It's nice you threw him some work. [laughter]

NEIL DEGRASSE TYSON: So to improve accuracy, it's not just an academic exercise. It matters. And what's the biggest reason why we need accurate disaster forecasts from you?

RADLEY HORTON: Well, for one thing, it saves lives, right? I mean, if we can know that a hurricane is going to explosively gain in strength and advance. We can get people out of the way in time, for example, if we can better forecast tornadoes.

It's also big business, right? If we can help inform people on when they should plant crops, inform investments on how much money should be spent on heating fuels, or something like that, for a winter. So there's real economic implications and human lives at stake.