But Longer-Range Predictions Can Be Like Checking Caterpillars : Farmers Glued to Weather Reports

Like many drought-stricken farmers, Terry Grainer is spending a lot of time this summer watching and listening to weather reports.

Last month, after forecasters predicted scattered rainfall, he sold part of his grain crop before market prices took the plunge that inevitably follows reports of rain. Two weeks ago, despite his reservations about the accuracy of rain forecasts, the Iowa farmer decided to use hundreds of dollars worth of insecticide to protect a soybean crop that would have died anyway unless it rained soon.

“The farmers are listening all the time this summer,” said Russ Solheim, manager of a farm in Bremer County, Iowa. “They keep their tractor radios tuned into the weather, they listen on their barn radios and they watch TV weather at night.”

May Avert Total Loss

Although they are always of considerable interest, weather forecasts have taken on a new degree of importance this summer for farmers in the nation’s parched midsection as they make decisions that could mean the difference between salvaging a meager crop and total loss.

Mostly, they are monitoring highly accurate short-range forecasts that predict weather up to five or six days. But, as the drought has worn on, farmers have relied increasingly on longer-range forecasts, whose accuracy even forecasters seldom rate above 40%. Despite growing computerization, weather experts say, longer-range forecasting often relies as much on the instincts of experienced forecasters as on concrete scientific methods.

The main reason that long-range 90-day forecasts and medium-range one-month predictions tend to be of little use to farmers is that they monitor only basic weather factors and are usually expressed in general terms that give farmers little useful information. A monthly forecast, for example, simply may predict lower-than-normal rainfall and higher-than-normal temperatures for the Midwest.

“I guess, even if it’s wrong, it’s the best thing going. So, naturally, we’re going to pay some attention (to longer-range forecasts),” said J. Howard Mueller, a farmer in Washington County, Iowa. “But sometimes we think it’s as good as counting the rings on a woolly caterpillar.”

Scarcity of Indicators

Shaky estimates result from incomplete understanding of the vagaries of weather movement and from a scarcity of reliable indicators in the atmosphere, according to Donald Gilman, director of the predictions branch of the National Weather Service, which is responsible for most of the nation’s longer-range forecasting.

“It’s like reconstructing a crime with not very much evidence,” said Gilman, who is in his fourth decade of forecasting. “There’s a lot of room for subjectivity and a lot of room for error.”

Long- and medium-range forecasts are based on temperature and air pressure measurements taken by satellites and weather balloons orbiting high above the Earth’s surface.

Weather service forecasters track abnormalities--areas where the weather is significantly different from seasonal norms--over several months and predict how the aberrations will continue to move.

Instinct a Factor

Once they have deduced the likely paths of the abnormalities, forecasters use information from periods of similar weather in previous years, statistical models, judgments based on years of experience and more than a little instinct to predict the weather that will result on the Earth’s surface from the abnormalities moving overhead.

Nevertheless, the weather service has been generally accurate in its seasonal predictions for the drought areas this spring and summer.

Experts disagree over whether increased funding for weather research would enable scientists to provide farmers with more useful information or would have allowed them to predict the drought more accurately.

But Gilman says that Congress must approve funds for a new computer system if the United States is to catch up with European forecasting agencies, which have taken over the lead from the United States in forecasting technology.

Agency ‘Computer-Limited’

“It’s clear right now that we’re rather computer-limited,” Gilman said. “It’s not because of the state of the art in computing but because of the failure to have the top of the line.”

Even if the new computer system is funded, farmers and others who would like to rely more heavily on long-range forecasting should not expect quick results, he said.