How to report the weather

Weather forecast is to predict the growth, decline, migration and intensity changes of various weather systems in a certain period in the future. It is the basis of meteorological element forecast. The methods of situation forecasting can be divided into two categories: one is numerical forecasting method, that is, directly integrating atmospheric equations or its simplified equations, and forecasting the future pressure field, temperature field and wind field according to the obtained results; The other is the weather chart method. The latter has the following practices:

Empirical extrapolation method

Also known as trend method, it infers the future position and intensity of various weather systems according to their past moving paths and intensity change trends on the weather map. This method is effective when the movement and intensity of the weather system do not suddenly change, or when the weather system does not regenerate or die. But when it suddenly changes or the weather system is alive or dead, the forecast is often not realistic.

Similar to situational method.

Also known as pattern method, it is to find some similar weather conditions from a large number of historical weather maps and summarize them into certain patterns. If the current weather situation is similar to that before a certain model, it can be predicted with reference to the later evolution of the model. Because similarity is always relative and can't be exactly the same, this method often leads to errors.

Statistical data method

Also known as correlation method, it is to use historical data to count the occurrence, development and movement of various weather systems in different seasons in history, get their average moving speed, find forecast indicators (such as cyclone generation and typhoon turning), and make predictions. This method cannot be applied to examples that have never appeared in history or examples that are fast and slow.

Physical analysis method

Firstly, the physical factors of the birth, decline, migration and strong change of the weather system are analyzed, and then the weather forecast is made. This method usually works well. However, when the simplification and assumption of the motion equation reflecting these physical factors are not realistic, it will often lead to prediction errors, even far from the actual situation.

The above four methods have their own advantages and disadvantages, and they need to complement each other, learn from each other's strong points and make comprehensive consideration to achieve better results.

The elements of weather forecast, forecasting and editing, refer to this section, that is, forecasting the changes of temperature, wind, clouds, precipitation and weather phenomena in a certain period in the future. There are several ways to predict elements:

Empirical prediction method

Based on the situation forecast of the weather map, the future weather distribution is predicted according to the future position and intensity of the weather system. For example, when the low pressure moves and strengthens, it can be predicted that there will be rainy weather or heavy precipitation in the future. The accuracy of this method depends largely on the experience of forecasters, and the forecast effect is unstable because the weather system and weather phenomena are not in one-to-one correspondence.

Weather sign: statistical forecasting method of rain turning fine

Analyze the historical data of weather, seek the correlation between the change of atmospheric state and the previous meteorological factors, screen the forecast factors by using regression equation and probability theory, and establish the forecast equation. Substituting the recent meteorological elements into the equation, the required forecast value can be obtained. The effect of this method mainly depends on the correct choice of factors.

Dynamic statistical prediction method

The future meteorological parameters calculated by numerical forecasting method are used as forecasting factors, and a set of forecasting formulas are obtained by regression equation to forecast the factors. With the improvement of numerical model, the accuracy of this method may be improved steadily.

According to the time limit of weather forecast, this catalogue can be divided into:

Short-term forecast: according to the radar and satellite detection data, monitor the local severe storm system on the spot and predict its future movement of 1-6 hours.

Short-term forecast: forecast the weather in the next 24-48 hours.

Medium-term forecast: the forecast for the next 3- 15 days, mainly including what kind of weather process will be affected, whether there will be disastrous weather and the main weather change trend.

Long-term forecast: usually refers to the forecast from 1 month to 1 year. The statistical method is mainly based on the deviation between the monthly average of meteorological elements and the multi-year average. The method of long-term forecasting by numerical forecasting method is being tested and some progress has been made. The forecast time limit of 1-5 years is called ultra-long-term forecast, and the forecast time limit of 5 years and above 10 years is called climate outlook.

According to the size of the forecast range, the weather forecast can be divided into:

1, large-scale forecast. Generally speaking, it refers to global forecast, hemispheric forecast, continental or national forecast. Mainly produced by World Meteorological Center, Regional Meteorological Center and National Meteorological Center.

2. Medium-term forecast. Often refers to the provincial (district), state and regional forecasts, which are made by provincial, municipal or state meteorological stations and regional meteorological stations.

3. Small-scale forecasting. Such as the forecast of a county, city, reservoir, airport, port, etc. These forecasts are made by the local weather station.