Local meteorologist warns federal cuts are undermining weather forecasts nationwide

BETHLEHEM, Pa. — A local meteorologist is raising concerns that reductions at the federal level are quietly eroding the reliability of weather forecasts across the country.

It is affecting everything from national outlooks to the local forecasts on which residents rely every day, EPAWA founder and lead forecaster Bobby Martrich said.

In a recent analysis published on EPAWA Weather Consulting’s website, Martrich outlined what he describes as growing deficiencies in the nation’s weather modeling system.

They are driven largely by cuts and constraints affecting the National Weather Service and its parent agency, the National Oceanic and Atmospheric Administration.

EPAWA is based in the Lehigh Valley and provides daily forecasts, long range forecasts, text alert services, a premium forum and private client services.

Ripple effects

Martrich said the ongoing issue is not limited to any single model or forecasting office.

Instead, he argues the problem begins much higher up the chain: with the observational data that feed every major weather model used in the United States.

At the core of the concerns are reductions in routine weather balloon launches at National Weather Service offices nationwide.

Those balloons, launched at least twice daily under normal operations, collect critical data on temperature, humidity, wind speed and direction throughout the atmosphere.

That information is a foundational input for weather models used by federal agencies, television stations, private forecasting firms such as EPAWA and emergency managers alike.

When those data points are missing or delayed, Martrich said, the ripple effects are immediate and widespread.

"Every forecaster in the country is working off the same pool of data," he said Thursday. "When that pool shrinks, forecast confidence shrinks with it."

Martrich specifically pointed to reduced or suspended weather balloon launches at several Midwestern sites — an area he described as "critical" to weather forecasting — going back to last year.

"That's a source region for things, so everything has a downstream effect," he said.

He explained how the data being fed into models being wrong on Day 3 skews what mid-range and long-range forecasts will look like.

"So if the GFS [the American Global Forecast System] says something is going to happen at Day 3 and it's completely wrong, it's going to be off on Day 12, because everything sets in motion from that point downstream and continues to extrapolate from that point," he said.

'Build up of frustration'

To that end, Martrich said posting such a detailed analysis on his website was because of "a long line of buildup of frustration."

There are more eyes on winter forecasting than any other season, he said, and far too many people failing to understand how missing billions of pieces of data typically ingested into these models is throwing everything off.

He said he wants people to understand how it's affecting meteorologists across the country and the day-to-day forecasts on which people rely.

Models have become more volatile from run to run, he said, long-range outlooks have grown less stable, and confidence in storm timing and intensity has decreased, even during otherwise routine weather patterns.

For private forecasters such as Martrich and his team, the issue is particularly frustrating.

EPAWA Weather Consulting relies heavily on National Weather Service data and federally run models to produce local and client-specific forecasts.

While private meteorologists often add interpretation and regional expertise, Martrich emphasized they can't replace missing observational data with analysis alone.

"This is not a public versus private forecasting issue,” Martrich said. “Everyone is affected equally."

Far-reaching consequences

The stakes also extend beyond forecast accuracy.

Emergency management agencies depend on reliable model guidance to plan for flooding, severe storms, winter weather and extreme heat.

Inconsistent or degraded data can complicate decision-making and reduce lead time for public warnings.

Martrich cautioned that the public may not immediately recognize the cause — instead perceiving forecasts as increasingly unreliable or contradictory.

He warned that perception risks undermining trust in weather guidance at a time when extreme weather events are becoming more frequent and more consequential.

Rather than pointing to individual forecasters or models, Martrich argued the solution lies in restoring consistent investment in federal weather infrastructure, including staffing, observational networks and computing power.

"Without robust model guidance, the nation’s ability to anticipate and respond to weather hazards is weakened — with consequences that extend from individual safety to national economic resilience," he wrote.

'Don't know what they're interpreting'

Compounding the problem is the number of publicly available models being shared by people who have no training, nor knowledge in how to interpret what they see, Martrich said.

"It's good to have an interest," he said.

"But if they don't know what they're interpreting, or you know there's a lot of things that get put out on social media, for example, that are just put out there that are looking way too far ahead, then you'll see all these hype posts, this misinformation about it, and people ask me, 'Why do you care?'

"I care because everybody comes to me and has asked me a million questions. So there's a lot of people sharing things, particularly on snowstorms, that probably shouldn't be sharing them.

"When we're already dealing with these bigger issues, we're just not going to talk about something hypothetical or something that's not legit.

"But unfortunately, that's where the attention goes."

Martrich said he'd rather focus on the real problem at hand and what, if anything, is being done.

"I hope there's some kind of congressional legislation, some kind of legislation that can try to get this reversed and get the National Weather Service and these offices fully funded again," he said.

"Because we have to look at observational data and go from there and people think things are off, well this is why."