Why is the weather forecast wrong?

Medium-and long-term forecasts can forecast the weather for months, years or even decades. Long-term weather forecast In weather forecast, long-term weather forecast is the most concerned issue for meteorologists, such as long-term drought and flood forecast or long-term extreme cold and heat forecast. On the weather map, the same weather process that appears repeatedly for a long time is a persistent abnormal weather phenomenon. What factors make similar weather processes appear again and again? It is generally believed that this problem is not mainly in the atmosphere, but because some factors outside the atmosphere are at work. This is a basic concept in long-term forecasting research in many countries at present. What are these external factors? At present, opinions are still very inconsistent. Some people think that the change of the sun is the main external factor of the long-term weather process, and many long-term forecasting methods are based on this view. However, some people think that the abnormal thermal characteristics of the underlying surface, such as seawater temperature, geothermal distribution, snow on the earth's surface and polar icing, are the main external factors that lead to the recurrence of similar weather processes in a month or a season. The anomalies of thermal characteristics of these underlying surfaces are caused by the atmospheric circulation in the previous and the same period, but in turn affect the long-term weather process. Both of the above viewpoints have some truth, but neither can clearly explain the physical causes of long-term weather processes. The relationship between them is complicated, and even some observation facts are unclear, let alone theoretical research. Therefore, the methods currently used for long-term weather forecasting are basically some forecasting experience and statistical methods. In recent years, some people have begun to carry out numerical experiments on long-term weather processes, trying to make long-term weather forecasts by dynamic methods, but there is still a long way to go. In recent 10 years, many people have noticed the role of the ocean in the long-term weather process. It is found that the temperature of the ocean surface (hundreds of meters thick) varies greatly from month to month or year to year. The change of SST is mainly caused by atmospheric circulation conditions, which can cause large-scale atmospheric circulation anomalies. For example, the storage of abnormal heat in the ocean is very deep (up to about 500 meters), which makes the abnormal sea surface temperature last for a long time. Therefore, the effect of SST, an external atmospheric factor, can lead to a wide range of abnormal weather. Many meteorological stations regard seawater temperature as a factor of long-term weather forecast, which is reasonable. In recent years, satellite forecasting has been used to collect data of land snow and polar icing, so people can study the relationship between their anomalies and large-scale weather anomalies, and thus obtain effective long-term weather forecasting factors. In the long-term weather forecast, the selection of forecast factors cannot be the same as the selection of indicators in the short-term weather forecast. Factors with large scale and slow change should be selected as predictors. For example, the change of seawater temperature is much slower than the air temperature, and it is estimated that it is only110, so it is very suitable as a long-term weather forecast factor. At present, the long-term forecast of many countries only predicts the average weather in the next month or quarter, that is, only predicts the deviation of temperature or precipitation from the average climate in that month or quarter. If we can find similar months or seasons from historical data, we can predict the average weather of every ten days of each month in this month or season, but it is not easy to predict the weather of every day in next month. Short-term, medium-term and long-term weather forecast/futures/weather/2001-08-2000034576.asp]