Matt Lanza

And sometimes Matt's world is kind of a fantasy sometimes.

But...

The reality is that, yes, weather modeling has gotten way better.

We have invested a ton of time, research, energy, funding, et cetera, since Hurricane Katrina to better predict hurricanes.

It remains a very broad bipartisan issue that needs to continue to be addressed.

2024 was the best year on record for the National Hurricane Center in terms of their forecasting.

They knocked it out of the park.

They did a phenomenal job.

Perfect?

No, but way better than it has been.

In fact, like you look today, your five day forecast today is almost as good as like a one or two day forecast was 30, 40 years ago.

So, I mean, think about that for a second.

Think about how much, how many decisions get made in those few days before a hurricane by public officials, by companies, by anyone, you know, people that live on the coast.

And now you have the confidence that you can start making moves four or five days in advance that maybe 30 years ago, you could only do one or two days in advance.

And so that's why we say all these improvements in forecasts have huge economic and sociological benefits to society because of that.

So, well, now with the advent of machine learning and AI, we're starting to see that, okay, well, maybe we don't need to model the physics of the atmosphere.

Maybe we can say, hey, here's 50 years of weather data, what we call reanalysis data.

It's basically all the past weather that's occurred over the last 50 years.

Here it is.

Take a look at this.