Point of meteorology and atmospheric science at Penn State.

This is weather world.

Every town has one.

And while the one here could apply to a post office or grocery store or gas station, tonight we're talking about a different one, specifically a weather event that has left behind both mental and physical scars, and a fury that has never been matched since.

And for quite a few communities in eastern Pennsylvania, that weather event was a deadly and destructive flood courtesy of a damaging duo of tropical cyclones in August 1955, bearing the names Connie and Diane.

Because, you know, when you're in a terrain which is, you know, steep and wooded, um, it just all flows downhill.

And that's really kind of what happened.

It's it was mainly the Pocono region.

And then kind of down downstream in the Delaware, it went down to east and it got pretty hard hit it really after about Trenton, um, the river gets very much wider and it also becomes a tidal river at that point.

And that really helps spread out all that water.

While Diane often gets credited with the flood, the stage was actually set by Connie several days earlier as the remnants of a once category four hurricane dropped 6 to 12in of rain over much of eastern Pennsylvania on August 13th, 1955, saturating soils and causing some river rises.

On August 18th, Diane practically passed directly over Philadelphia as a weakening tropical storm, with a widespread swath of 4 to 12in of rain falling on the north side of the circulation.

In between the Pocono Plateau and Trenton.

It left roughly 75 miles of real estate for a wall of flood water to cause chaos in the Delaware River basin, the Delaware River crested 11ft above flood stage at new Hope, 16ft above flood stage in the Delaware Water Gap, and an incredible 22ft above flood stage at Easton.

All records that still stand today.

The same is true a short distance west along the Lehigh River, which eventually feeds into the Delaware.

Similar water rises were observed there and have not been matched since.

Uh, just below Laurie Station.

And this is the Lehigh River.

This is River road, and in the flood of 1955, uh, the water level actually got up this high 15, 20ft and flooded this road.

There was there was a dam just above here, uh, which went across the river.

It was actually called initially slate dam built on bedrock in the river.

And it was built by the Lehigh Coal and Navigation Company.

And they put locks in just behind me up here where the dam was.

And they actually built a dam so that their workers could get across to the store in lorries.

There was a flood, uh, I think about 1940 that actually opened up a hole in the dam.

And, uh, the dam was subsequently repaired.

The hole was repaired in 1955 when we had Hurricane Diane.

The hole reopened and it was never repaired.

Once the flood waters broke the dam open, the wall of water got even taller as it moved downstream, setting its sights on the small town of Northampton.

This is, uh, Northampton Borough, and this is Canal Street running along the side in the Newport section of Northampton.

The flood waters here probably would have been at least 30ft.

Actually rose up and covered like a block or two into the houses up on the Newport section.

A lot of people had to relocate, abandon the houses.

So it was devastating down here.

It was definitely more destructive than it was, uh, two miles upstream at Laurie's.

The river was high enough that the, uh, the bridge crossing into Cementon, which is behind us a little ways.

The bridge had to be shut down.

The rise in the river was impacting the the bridge.

Despite the devastation, coverage of the flood has become much more widespread in recent decades and was very limited at the time that it actually occurred in 1955.

You know, we were ten years out of World War Two.

Um, people weren't looking for excitement anymore.

They just wanted to get back to regular life.

And they really even when this happened and as big as it was, it really never got the coverage it deserved because people just wanted it to go away, even though people wanted it to go away.

And the region has not dealt with an equally ferocious flood since, the memories remain etched into the minds of longtime residents, and they quickly rise to the surface anytime the talk of tropical remnants enters the forecast.

For years and years, it was talked about and, uh, a lot of people reminisced about that, even more so in 1972 when Hurricane Agnes came through another devastating flood.

And it just kind of brought back memories of years past.

So, yeah, people never forgot anybody that was around back then, never forgot.

Reporting from Northampton in Northampton County.

I'm Ben Reppert for Weather World.

Nearly 69 million Americans, or 1 in 5, speak a language other than English at home.

Yet English remains the primary language used to communicate weather emergencies here in the United States.

That vulnerability was the catalyst for research and work by Doctor Joseph Trujillo Falcone, an assistant professor at the University of Illinois Department of Climate, Meteorology and Atmospheric Sciences and the Department of Communications.

Joseph, thank you so much for being here tonight.

Excited to be here.

Thank you so much.

Yeah.

So first things first, what inspired your work on that intersection between meteorology and language I would honestly call a lived experience?

I was born in Lima, Peru, but I moved to Dallas Fort Worth in Texas at a very young age and not having experienced thunderstorms, hail, or tornadoes was a really, really wild surprise for me.

I was often scared of these hazards, and I quickly noticed that my own Spanish speaking communities also did not know what was happening.

That was really a true catalyst to inspire me to study the weather more often, and really also be able to connect with communities that we haven't been able to in the past.

Were there some specific cases you studied as part of your work where language barriers led to lives being lost?

Yeah, all the time.

We conduct nationwide surveys and also visit survivors of various extreme weather hazards in the aftermath to really understand what was happening.

One that really stuck out to me in particular was the December 10th to 11th, 2021 tornado outbreak, known as the deadliest December tornado outbreak in US history, passed through four states in the Midwest and also the southeast.

And it was just really, really eye opening to see folks that only spoke Spanish and were receiving these tornado warnings time and time again, but they could not make sense out of them.

And a lot of them were taking additional time to try to understand what it was saying, when they should have been taking that time to take shelter.

So I learned a lot of firsthand accounts of just how important language information can be to ensure safety for all.

Yeah.

For sure.

Speaking of the accounts, how were you able to go about conducting this research?

Did you talk to to folks on the ground impacted?

Yes, 100%.

We got on a rental car and visited impacted communities several months after the disaster happened, and we visited anywhere from local churches, community organizations, local homes and talk to the people that were impacted.

We were really curious to see just how that situation unfolded because unlike big urban areas, especially areas in the Midwest and some areas in the southeast do not have access to Telemundo, Univision or other Spanish language resources.

So you're really just left to everything completely in English.

And so in that sort of scenario where a big tornado is approaching your way, it can be really difficult to receive life saving information.

And so we asked them those questions head on and got a lot of insightful feedback that we then forwarded to NOAA to better enhance their communication products across the board.

Yeah, for sure.

And you worked closely with NOAA on developing some new products, including the use of AI.

Talk a little bit about that.

Yeah.

You know, with artificial intelligence continuing to emerge as next generation technology for us, we're making sure that it also impacts the way you receive weather warnings.

You know, it's very difficult to translate from one language to another.

At the National Weather Service, we have volunteer translation teams across the board.

However, they also have to forecast, and that's their first duty at the end of the day.

So to translate on top of it could be very overwhelming.

And that's how we're using AI.

Over the last four years, we've been training an AI specifically to understand weather and climate terminology, and little by little, we've reached up to 99% accuracy in some hazards and have reduced translation times for the weather service from about an hour to 2 to 3 minutes.

And we've been trying to reimagine how that could look like in the future, not just in Spanish, but other non-English languages, too.

So we've been training that AI to translate into Spanish, Chinese, Vietnamese, French, Samoan, among other languages too.

Yeah.

Is this being applied to all National Weather Service offices or only a few select in trials?

We started off small.

We actually looked at census data and saw where the biggest multilingual communities are across the country, and we started off in those offices.

So far, though, we've reached over 13 million, uh, limited English proficient speakers just by the jurisdictions as to where they're located in.

And so we're slowly starting to scale that effort upwards so much that even this year, the National Hurricane Center announced that as part of their operations, they're going to start integrating some of this for hurricane season.

So I think it's a very huge plus and benefit to so many communities out there.

And so folks have been very resonant with it.

And we've also been testing these products out using nationwide surveys too.

And we gave them a human translation, and then one made by our NOAA based AI tool.

We actually found that especially in empowering people to act, sometimes the AI was doing slightly better than the human translations, and at least for Spanish speakers.

And so it really does go to show that we've been testing this out with the communities themselves.

They've been engaging with it, and thus we feel confident that it'll make a long lasting impact in the years to come.

Yeah, for sure.

The partnerships that you've developed along the way, not just with NOAA, other entities as well.

Yeah, NOAA and then actually really funny and funny enough, here at Penn State University, there was a there is a linguist by the name of John Lipski, and he was actually the very first person that believed in this kind of work.

I remember as a early graduate student, I was reading a book about just the varieties of language and how it can impact how people perceive various things, and I remember emailing the author of the book as a young graduate student like, hey, I really love this book.

I'd love to apply this to weather.

Sure enough, it was a Penn State professor that took me under his wing and played a very pivotal role in bringing this topic up to the nationwide level.

And so, yeah, I've had a lot of collaborators from various universities to even nonprofits and local organizations, because honestly, it's a collective effort.

You know, I only know Spanish, and there's so many other languages here in this country.

Uh, people practice over 350 languages.

So there's a lot to work yet still to be done.

Wow.

Absolutely great work.

And I think it gets to the core of why risk communication is so important.

Talk a little bit about that and kind of next steps in this this work that you're doing 100%.

Just because I always love to emphasize that the way we've been able to forecast and the way we've been able to communicate is really unprecedented.

Just given just a couple of decades ago.

Our technology has reached new levels.

But here's the thing technology can advance as much as it wants to, but if it doesn't reach the communities we are intended for it to reach, what's the point?

And I think this is where risk communication is just so important.

We have to make sure that everything that we're advancing here in the atmospheric science world reaches these communities and makes sure that it results in safety.

And this is why I'm so passionate about this topic, and really elevating the next generation of students and scholars to start integrating into it, too.

You know, I always tell my atmospheric science students, I don't I'm not asking you to be a social scientist, but I'm asking you to be aware of your communities and to always think of them at every single step of the way, whether you make that forecast or whether you advance that next weather model.

Absolutely.

Well, Doctor Joseph Trujillo Falcon from the University of Illinois, thank you so much for being here and talking about your research.

Thank you so