How Do You Really Know When It’s Going to Rain?

Guest blogger Sara Prince is filling in for a couple of days while Martha Woodroof explores the upper Midwest.

Before our week-long bout of drought in the Valley, I was driving in the car with my sister and she pointed to a tree growing adjacent to the parking lot of Harrisonburg’s south-side Wal-Mart and said, “Look, the tree has upside-down leaves.”

“Huh?” was my well nuanced reply.

She then proceeded to remind me that our parents had once told us that when the wind is blowing in such a way to make the leaves look as if they are growing upside down, it’s going to rain.

To find out if that tradition had any truth behind it for those of us in Virginia, I contacted Steve Keighton at the National Weather Service.

“It may be a sign of an easterly wind,” he said, which indicates a better chance of rain in the region. But Keighton went on to say, “It depends on the kind of tree and how strong the wind is blowing, so I certainly wouldn't trust that one.”

Good point. It never did rain that day, despite the upside-down leaves and the thunderclouds up above – another sign of potential rainfall.

But it got me thinking. Are there any rain prediction beliefs that are actually reliable?

The Wikipedia site, How to Predict the Weather Without a Forecast, has a number of different ways to predict rainfall that are based on folklore passed down from generation to generation. One of the traditions is to look for a rainbow in the west. Because storms generally move east to west, and rainbows indicate moisture in the air, it’s fairly likely that a storm is imminent. Keighton agrees with this in general, but says that it indicates the likelihood of a storm that is only minutes away.

Since weather prediction folklore seems to only hold true under some circumstances and not others, are we better off relying on technology for predicting natural phenomena (like weather), as we are for many other aspects of our day to day life?

Keighton admits that meteorology is an inexact science – and it all depends on a variety of circumstances. As he explains it, there are three basic strategies for predicting rain – Doppler radar, satellite imagery, and upstream rain gauges. That gets us our weather for the upcoming hour or two. Beyond that, says Keighton, the NWS uses numerical prediction computer models.

And accuracy? Well, accuracy comes…and accuracy goes. It often takes off for vacation during the summer.

“With summer time thunderstorms, which can be quite scattered or isolated (versus widespread coverage), the models do a good job of predicting that they will form in general areas, but to show that any one location will get hit (say Harrisonburg versus Staunton) is beyond their capability,” said Keighton.

However, in the last ten years the computer models have improved their ability to determine the impact that terrain has on precipitation.

In the science of predicting rain further into the future, the NWS also looks at global patterns, such as El Nino and La Nina. But by this point, accuracy is determined by federal departments. Local NWS forecasts, according to Keighton, are only about 60% accurate up to seven days out, while the Climate Prediction Center is more focused (and presumably, more accurate) on seasonal predictions for rainfall.

So, back to my original question: how do you know when it’s gonna rain? I wish I could give you an answer. But at least I can arm you with a pocketful of folklore (How to Predict the Weather Without a Forecast) and technology (thank you, NWS) that, when doubled-up, will get you the most reliable forecast you (and your garden) can get.