Dr. Gary Lackmann is a professor of atmospheric sciences at N.C. State. Here he explains how meteorologists forecast summer weather. Questions and answers have been edited.
Q. Do last winter’s brutal temperatures mean we are in for a hot summer?
A. No. Using data from this area, we find only a weak connection between winter temperatures and those of the following summer. Although the connection is very weak, there is a slight tendency for cold winters to be followed by cooler-than-average summers, and for warm winters to be followed by warmer-than-average summers.
Note this past winter was in the warmer half of winters, ranking as the 30th-warmest out of the last 70 for Raleigh. In Raleigh, there were only two days with a low below 10 degrees; in Charlotte, only three. In contrast, both December and February featured three days with highs above 70, so the extremes went both ways.
Q. What kind of tools do forecasters use to predict the weather today, next week, next month?
A. For today’s forecast, tools include very detailed computer-model forecasts and different kinds of weather observations, including satellite and radar data, soundings measured by weather balloons, and surface weather station data.
For short-term forecasts, a key tool is “numerical weather prediction.” These tools are essentially computer programs that represent the atmosphere through the laws of physics, like the law of conservation of energy. The computer can integrate these equations forward in time to produce “future” weather maps.
For some forecast aspects, such as rainfall, a useful approach is a group of computer forecasts known as an “ensemble.” Ensembles can be made of different models or the same model with a slightly different starting point. Chaos theory tells us that if we change the starting point slightly, the “butterfly effect” will eventually yield a completely different forecast. Knowing how fast the different solutions move apart is important, because it can tell us how much confidence to have in the prediction.
For monthlong or longer extended-range forecasts, the weather cannot be predicted the same way. Because of the chaotic nature of the atmosphere, tiny differences in the initial conditions amplify, and the best forecasters can do is predict the temperature or precipitation will be “above average” or “below average.”
Q. How does climate change factor into long-range weather forecasting?
A. For long-range seasonal forecasts, a prediction of “above average” temperatures will be correct more often than a prediction of “below average” if the average used is for a fairly long time period, say 30 years. Because the weather is naturally variable, these variations often drown out the climate-change signal and make it difficult to detect for a given year or a given season. For that reason, climate change is best diagnosed using temperature data averaged over the entire globe and over many years.