Why Is the Weather Hard to Predict?

Every morning, millions of people check the weather forecast before they leave home. Should I take an umbrella? Will it be cold this evening? Today's forecasts are much more accurate than they were fifty years ago, but they are still sometimes wrong. Why is it so difficult to predict the weather?

To make a forecast, scientists first collect huge amounts of data about the air, such as temperature, wind, and humidity. This information comes from satellites, weather stations, and even balloons. Powerful computers then use the data to calculate how the air will move over the next few hours and days.

The problem is that the air is always changing, and a very small difference at the start can lead to a big difference later. If there is even a small error in the temperature measured in one area, the computer's prediction for several days later may be completely different from what really happens. This is why forecasts for tomorrow are usually reliable, while predictions for next week are much less certain.

To deal with this problem, forecasters do not run their computer programs just once. Instead, they run them many times, each time starting with slightly different data. If most of the results are similar, the forecast is likely to be right. If the results are very different from each other, the forecasters know that the weather is harder to predict, and they are more careful about what they say.

Because forecasts can never be perfect, they often use numbers like "a 70 percent chance of rain." In Japan, this means that when this forecast is given 100 times, at least one millimeter of rain falls on about 70 of those occasions. It does not mean that it will rain for 70 percent of the day. Understanding what a forecast really says can help you make better decisions, such as whether to carry that umbrella.