What is real-time forecasting?

The development trend of meteorology, especially in big cities, is a kind of "real-time forecast", that is, very short-term forecast can last 1 ~ 3 hours. Real-time prediction is the product of data development in 1990s. Small-scale meteorological system, fast computer network, accurate engineering pictures, automatic observation, new tools like Doppler radar, and regional analysis of global weather forecast model called meteorological scale model make real-time forecast logical. Thanks to these systems, forecasters can now predict the movement of small-scale weather features, such as storms, heavy snow or wind direction. These will not be realized before 10 or 15.

Some meteorological observation and early warning belong to another kind of real-time forecast. One of them greatly reduced the casualties caused by bad weather. When the atmospheric conditions are bad, it can be observed. In the United States, violent storms and tornadoes lasting for several hours and hurricanes lasting for two days can only be observed on a large scale and models after being signed by the National Special Meteorological Center. Only when bad weather appears and approaches you will an alarm be issued.

In 1950s, the local meteorological department issued a tornado and rainstorm warning 30 minutes after 1 hour. Continuous prediction depends on the completion of a large number of control equation simulation data obtained by computer. These equations can describe the atmospheric conditions better than observations. Many weather forecasting units released a large number of forecasts about temperature and rainfall 3 ~ 10 days ago. These predictions mainly rely on a large number of long-term simulation results, which are slightly better than climatology. However, they are also very important for weather-sensitive fields such as agriculture and transportation. Even a hint of the weather will save a lot of wealth.

Chaos science tells us that even with the best observation instruments and computers, it is impossible to know all kinds of weather conditions two weeks in advance, but this does not prevent researchers from using new forecasting tools. Forecasters can accurately predict the weather conditions in 3 ~ 10 days in advance through all-round three-dimensional performance.

According to this method, first, each model runs with subtle and global changes, which leads to the separation of models at any time. According to the degree of separation, people can judge whether the provided meteorological forecast is reliable, and the greater the degree of separation, the more inaccurate the forecast is. Forecasters use the "graph" of the pressure field to observe whether the lines drawn by different models are consistent or intertwined like dough.

Seasonal characteristics can be predicted, such as whether winter is dry, warm or cold and humid. Weather forecast mainly depends on ocean changes, which is characterized by slow progress but great influence. This connection between meteorological conditions in a certain place and those in a distant place is called teleconnection.

In all long-term forecasts, global climate models are tracking greenhouse gases added to the atmosphere. This climate model predicts that the global temperature will rise in the next century, and the regional and regional impact of this trend is difficult to predict. This is a challenge for the weatherman in the new century, but instability is also part of it. From the philosopher's point of view, this situation can be completely faced. As napier Shaw, a British weatherman in the19th century, once wrote, "The weatherman's heart will know more about the bitterness. The layman is not for fun, but for interference. "