Odds Are Good That Rainfall Predictions That Use Percentages Can Cloud the Issue

George (Dr. George) Fischbeck says it’s one of the questions he’s most frequently asked. His colleagues on television say they get pelted with it, too.

The meteorological mystery concerns what weathercasters mean when they throw percentages into their predictions--when they say, for example, that there’s “a 30% chance of rain” in Los Angeles.

And the answer?

“That’s a good story!” enthused Fischbeck, the KABC-TV (Channel 7) weather prognosticator. “But it’s a complicated question. The best thing for you to do is call the National Weather Service.”

And so for the answer we take you to the National Weather Service, which supplies the forecasts used by the media.

The question of percentage predictions couldn’t have come up in the 1960s, when the weather service still talked about the wet stuff only in terms of trusty adjectives and adverbs. Then, in 1970, the service inaugurated its Precipitation Probability Program.

Numerals, supposedly more precise, began to appear in place of words: 10%-20% (for “slight chance”), 30%-50% (for “chance”), 60%-70% (for “likely”) and 80%-100% (for the more categorical “rain”).

The conversion system is still somewhat cloudy, however.

While “chance” equals a 30%-50% probability, the weather service may occasionally forecast an “80% chance” of rain.

“It gets a bit tricky,” acknowledged Rod Becker, the agency’s public weather program leader. “But we put in ‘chance’ because it’s easier for the rip ‘n’ readers (television and radio forecasters) to say than ‘probability.’ ”

But even with the advent of the Precipitation Probability Program and model output statistics, some people don’t know whether to leave home with their umbrellas, admitted Art Lessard, the weather service’s chief meteorologist in the Southern California region.

“There’s a lot of confusion over the numbers,” Lessard said. “When we say there’s a 30% chance of rain, people don’t know whether we’re saying there’s a 30% chance anywhere in the area or that only 30% of the area is going to get wet. And some people just hear the word ‘rain’ and think no matter what number you give, it’s going to rain.”

So what’s the answer?

“Generally,” Lessard said, “if we say there’s a 30% chance in L.A. and the vicinity, we’re talking about the odds of it raining anywhere in that area. We’re saying there’s a 3-in-10 chance.”

But. . . .

“But in a larger area like the Mojave Desert, where you’re talking about thousands and thousands of miles, we would also take into account the percentage of area that might be affected,” Lessard said. “If we think that there’s a 50% chance that a storm is going to move in and cover just half the area, then we’d say there’s a 25% chance of rain.”

The Weather Service claims its numbers are accurate 85% of the time. In other words, in 30%-chance-of-rain forecasts, it rains 25%-35% of the time, or in 25%-35% of the area, depending on how you look at it.

Lessard admits that rain statistics can be misleading because of another factor, the method of measurement.

Rain is defined by the Weather Service as “the occurrence of a measured amount that is greater than or equal to 1/100th of an inch . . . of liquid precipitation . . . during a specific period of time (normally 12 hours).”

The agency’s official measuring device for Los Angeles is a gauge that sits on the roof of a downtown Department of Water and Power building.

“It could be raining hard in several areas (of the city), whereas the gauge at Civic Center is receiving much less rain,” Lessard said. “So our figure would be misleading (too small). But we can’t install gauges at every street corner. It would be too expensive, and besides, they’d probably get stolen.”