The Washington PostDemocracy Dies in Darkness

What does ‘probability of precipitation’ actually mean? (POLL)

What does it take for you to pack the umbrella? (Dennis Dimick via Flickr)

There are lot of things about weather forecasts that are not well understood by the people they’re intended for. What is a watch, and how is that different than a warning? Why is a “northwest” wind from the northwest? Why does the National Weather Service use over 100 colors for their weather hazards map? (Good question!)

In particular, probability-based forecasts have been known to lead to confusion. While they are excellent for verifying whether or not a forecast was accurate, and although they are the obvious way to quantify weather risk, they are often not well-understood by the general public when all you really want to know is what the weather is going to be.

The most widely used of these probability-based forecasts is probability of precipitation, or POP. Basically, it’s the “chance of rain” forecast.

Before we dig into how POP is actually calculated, here’s a poll. The first question was taken from a public survey conducted by the National Center for Atmospheric Research in 2007.

What the POP?

This is a non-scientific user poll. Results are not statistically valid and cannot be assumed to reflect the views of Washington Post users as a group or the general population.

Technically, by the National Weather Service definition, POP is a combination of spatial coverage of rain — what percentage of the area will see any amount of rain — and the confidence level that at least somewhere in the area will see measurable precipitation. Here’s the math:

POP = Area × Confidence

For example, if you that 50 percent of the area will see rain tomorrow, and your confidence is 100 percent, then the POP (or chance of rain) is 50 percent. Often, it’s a little more complicated; if you’re 80 percent confident that 40 percent of the area will see rain, then the POP is about 30 percent.

What the math eventually boils down to is this, for example: if you collect all the days with a 30 percent chance of rain in the forecast, 30 percent of those days will have had measurable rain in the forecast area. In other words, “it will rain on 60 percent of days like tomorrow” was technically the correct answer in the poll above.

Make sense? Maybe. At the end of the day it doesn’t really matter if you understand the technical definition of POP. What matters is that the forecast gives you the best information to prepare for your day.

To this end, a weather consulting group is investigating the way POP is used and understood by conducting a survey. “Even when a forecaster has a clear picture in mind of when and how precipitation may be distributed over an area, it may be difficult to express that in a single number,” the group writes on “It is possible that forecasters are using statistical probability; areal coverage; intensity; duration; quantity; impact; and overall forecast confidence in arriving at a single percentage.”

They will be presenting the results of their survey at a conference of the American Meteorological Society on June 10.

Do you have a fantastic idea of how to improve chance of rain forecasts? Tell us about it in the comments!