Guidance, not gospel

Missteps in forecasting by those of us in meteorology are nothing new.

Weather predictions are, after all, a look into the future, so they always will be inherently flawed to some degree, despite our best efforts. To be certain, the large percentage of forecasts are amazingly accurate—a remarkable scientific achievement, considering the complexities of our atmosphere.

When forecasts go wrong, we rightly give explanations for the errors in our prognostications. But maybe it’s also time for some introspection.

The issues of forecast flaws are most often related to our over-confidence in weather models, which simply tell us the range of possibilities of what might happen with the weather based on the latest data. They serve as guides for the meteorologist to develop the forecast (hence called guidance) and are not the forecasts themselves.

That’s what I fear often gets lost in the sea of so much information on social media and elsewhere. The models are mathematical simulations churned through a computer. Data is ingested and digested, and the plumes of atmospheric possibilities then projected. They should never be interpreted by the public as the definitive forecast, yet they too often are because some meteorologists leave off the fine print.

When we publish model maps, sometimes days ahead of a storm, we need to be conscious of the risk of conveying the perception of precise predictability over great variability, and to make sure people appreciate that uncertainty. We surely need to keep showing the maps (the information is out there, whether we do or not), but we must also always explain the purpose of the models as guidance to our forecasts, and that they are never absolute truth.

Scientific literacy is to be embraced, not shunned. However, when the computer models in general are internalized by people as the forecast itself, which often happens, that’s when those of us as professionals in the field need to make sure there’s an understanding of model limitations and implications.

The proliferation of so much information does not have to be confusing. It’s vital that forecasts include the full range of possibilities, but it is also up to each one of us as meteorologists to ensure clarity of messaging amid the clutter.