The weather has always been political. From Bible stories that describe divine forecasts of flood and storms, to the correlation between rainfall and political assassinations in ancient Rome, the ability to anticipate and respond to weather-related phenomena has long been understood to make or break leaders. For modern examples, look no further than the fallout after Hurricane Katrina, or the urgent fight currently taking place over climate change.

But only very recently has the uncertainty of weather prediction also started to be exploited for political gain.

On Sept. 1, President Trump infamously claimed that Alabama "would most likely be hit (much) harder than anticipated" by Hurricane Dorian, a statement that was then refuted by the National Oceanic and Atmospheric Administration (NOAA), then defended by the agency, and is now the subject of an internal investigation and a growing scandal involving Commerce Secretary Wilbur Ross. While the whole snafu snowballed into a kind of theater that was almost comical — involving doctored maps and rage tweets — the underlying implication of the whole affair is much more worrying: That the inherent unknowableness of a forecast can be a political tool, too, and one that can carry deadly consequences if not taken seriously. "This is the first time I've felt pressure from above to not say what truly is the forecast," one anonymous employee at the national meteorological agency said.

Predicting the weather is something America is surprisingly not great at. Compared to European weather computer models, NOAA in particular tends to be more prone to error — most devastatingly, in the case of its estimation that Hurricane Sandy would weaken out over the ocean, while in reality over 200 people died after the storm strengthened as it turned back into New York. The agency is also plagued by bureaucratic deficiencies ranging from a 40-year-overdue software update to concerns that interference caused by a 5G wireless network could set NOAA back "several decades" in its ability to model weather patterns (not to mention the president's attempts to severely slash its budget). As a result, there can be a lot of wiggle-room around NOAA's estimations, if you suddenly find yourself, say, needing to support a gaffe you made: Hurricane Dorian might not actually be on track to hit Alabama, but hey, it could!

You're not supposed to look at weather models this way, though; let's not forget that the cone-shaped projection of a hurricane is literally called "the path of uncertainty" for a reason. (Alabama was never "likely" to be "hit much harder than anticipated," no matter how you read one stray NOAA model). Even the European weather models, while having a slightly better track record, aren't correct every time. This lack of certainty can be obnoxious as a layperson: Just this past winter in New York City, the National Weather Service (NWS) — an arm of NOAA — warned of an approaching storm so big it could have dropped 10 inches of snow in Central Park overnight. Due to a slight change in temperature and direction, the model was a dud; there was so little slush on the ground when New Yorkers woke up that the agency's goof became a meme, and the affair an embarrassment for city officials.

But as infuriating as it can be to have a snowless snow day, it is of the utmost importance that Americans put their trust in NOAA regardless of its occasional misfires. Meteorologists are still a lot more accurate than you think they are! Plus, between the internet's plethora of amateur meteorologists and the easy dissemination of illegal viral weather hoaxes, there is already far too much conflicting information in circulation. Adding to the noise, scammy, popular weather apps exploit the uncertainty surrounding weather forecasts, not even bothering to use real scientists to put together their predictions. Even though private weather agencies can have the reputation of being more accurate than a government agency, it is NOAA and the NWS that actually have the infrastructure in place to do the high-quality data collection required to predict potentially devastating storms (or, you know, if you ought to take an umbrella with you to work). Terrifyingly, Trump's attack and NOAA's subsequent flip-flopping have now further tarnished its reputation as a reliable source.

Meanwhile, people are dying. For all that Hurricane Dorian might seem an example of an Orwellian turn by NOAA — the agency relegated to a puppet of the president's misinformation campaign to protect himself — the storm is really best interpreted as yet another warning shot by our warming world. Hurricane seasons are only going to get worse as climate change progresses, NOAA itself predicts; the years ahead will hold more Dorians and Marias and Sandys (the Bahamas, which bore the brunt of Dorian's intensity, will likely be struggling to recover for months to come). It is vital that we continue to pour our resources into limiting the gray area around predictions while also preparing for the worst; it could be the difference, quite literally, for thousands of lives.

Yet for as long as weather remains a final frontier in science, the potential for gray area exploitations will persist. The floodgates have been opened by Trump's cherry-picking of NOAA models to back-up his gaffe; it isn't such a stretch to imagine NOAA similarly being weaponized by politicians to mitigate blame for poor preparation (it wasn't supposed to hit us!) or hinge badly-needed infrastructure and zoning decisions on the basis of an if.

The fact of the matter is, the weather isn't political: It doesn't vote, or care who's president, or have an agenda about which states it decimates. It's up to us to ultimately react to the facts: That southeastern Alabama never got more than nine-mile-per-hour winds last week; that the Bahamas are badly in need of our aid; that Puerto Rico is still recovering from the aftereffects of Maria two years later; that hurricanes are going to get worse until our models and policies reflect the facts of climate change. That, we can do something about. That, at least, is black and white.

Want more essential commentary and analysis like this delivered straight to your inbox? Sign up for The Week's "Today's best articles" newsletter here.