Weather forecasts are imperfect. Under Trump's budget, they'll be nearly impossible.

If President Donald Trump gets his way, weather forecasts in America will become less accurate over time, lagging behind those produced in Europe, Japan and other parts of the world. 

Trump’s budget contains a head-scratching provision that would cancel key upgrades to the National Weather Service’s (NWS) computer modeling program, forcing an agency that has already fallen behind its peers abroad to fall even further behind. 

If approved by Congress in its current form, the National Weather Service would fail to catch up in both computing power and all-around reliability, leaving American forecasters in the dark about potentially dangerous weather events like blizzards, hurricanes, and tornado outbreaks. 

As we reported on Tuesday, the Trump administration’s budget request includes language that calls for the National Oceanic and Atmospheric Administration (NOAA), which houses the NWS, to slow its implementation of more accurate computer models. The agency is already well behind Europe when it comes to the accuracy of its main forecasting model, known as the Global Forecast System or GFS, and Trump would simply let the accuracy gap grow. 

The NOAA budget, if implemented, would also eliminate the array of Pacific Ocean buoys that enable forecasters to detect and predict the evolution of El Niño events, which can alter worldwide weather patterns. That part of the Trump budget would also do away with a network of specially-designed ocean instruments to detect destructive tsunamis in the Pacific Ocean before they hit land. 

Here’s the language alarming meteorologists and some lawmakers, who say it would significantly set back weather forecasting improvements: 

Science and Technology Integration: Reduce Investment in Numerical Weather Prediction Modeling: NOAA requests a reduction of $5,000,000 to slow the transition of advanced modeling research into operations for improved warnings and forecasts. This affects the Next Generation Global Prediction System, Hurricane Forecast Improvement Program, NOAA Environmental Modeling System, and other model coupling, data assimilation, and collaborative research efforts.

Science and Technology Integration: Reduce Investment in the National Water Model: NOAA requests a reduction of $3,101,000 to slow the incorporation of upgrades into the National Water Model. In FY 2016, NOAA launched the first operational National Water Model, which significantly improved flood forecasting. NOAA will continue to provide valuable river forecast guidance to emergency managers and the public.

Overall, Trump’s budget would cut NOAA funding by 16 percent, with much of that being shouldered by the Office of Oceanic and Atmospheric Research, which funds climate studies, among other work. 

However, weather satellite funding would also be reduced, and the cuts mentioned above would hit forecasters’ ability to better predict the fickle movements of deadly hurricanes and floods. 

In an interview with the Washington Post, former NOAA chief operating officer David Titley decried the proposed cuts, saying: “This budget would ensure that NOAA-NWS becomes a second- or third-tier weather forecasting enterprise, frozen in the early 2000s.”

The GFS model is in need of significant improvements in how it brings in weather data and represents the atmosphere’s behavior. It’s currently experiencing a sharp dropoff in accuracy compared to the model run by the European Center for Medium-Range Weather Forecasting, or ECMWF, based in Reading, England. 

The agency is planning to roll out a new computer model, known as FV3, in 2019, but that, too, may start out behind the capabilities of international partners.

As the Post points out, some of these cuts may not be legally permissible considering Trump signed a bill into law in April that requires NOAA to come up with a plan to regain its leadership position in computer modeling and weather forecasting.

If anything resembling Trump’s budget request passes Congress, such a goal won’t be achievable anymore.