CN108572402A - Forecasting method of convective weather - Google Patents

Abstract

The invention discloses a convective weather prediction method. The prediction method first obtains a plurality of representative physical quantities and the threshold values of each said representative physical quantity according to years of convective weather phenomena and physical data; Grouping: dividing the grouped representative physical quantities into five levels according to the threshold segmental linearization. The weather phenomenon is related to a plurality of said representative physical quantities, and the said representative physical quantities related to each weather phenomenon are respectively determined; each physical quantity of the actual data is compared with the said threshold value of each said representative physical quantity, if a certain weather phenomenon is satisfied If the threshold values of all the physical quantities representing the phenomenon are met, the weather phenomenon is predicted to occur; otherwise, the weather phenomenon will not occur. This prediction method can be applied to different numerical weather prediction models, and it can objectively predict convective weather, and the prediction effect is better.

Description

Forecasting method of convective weather

technical field

The present invention relates to the technical field of weather forecasting, in particular to a forecasting method for convective weather.

Background technique

Convective weather has the characteristics of strong locality, strong suddenness, small scale, and short life history. According to the impact and destructiveness, although different countries pay attention to different types of strong convective weather, the forecast of severe convective weather is an international problem.

Thunder and lightning, short-term heavy rainfall, convective gale, hail, etc. are all caused by convective weather, and it is very difficult to forecast. For strong convective weather such as short-term heavy precipitation, convective wind and hail in China, the forecast of severe convective weather in the vicinity of 0-2 hours can be solved with the help of extrapolation technology of remote sensing data such as weather early warning radar and meteorological satellites, 2-6 Severe convective weather forecast in the short-term period of 1 hour can be carried out with the help of the rapid assimilation system and the fusion technology of remote sensing and numerical forecasting, but the severe convective forecast in the short-term period of 6 hours to several days mainly depends on the numerical forecast model.

For a long time, the short-term weather forecast of the National Meteorological Center has mainly relied on the artificial production of meteorologists, mainly for the analysis of environmental conditions. The forecast results rely on the subjective judgment of meteorologists, which has certain limitations.

Therefore, how to invent a prediction method, which can not only be applied to different numerical weather prediction models, but also can objectively predict convective weather, is a technical problem urgently needed to be solved by those skilled in the art.

Contents of the invention

The object of the present invention is to provide a method for predicting convective weather, which can be applied to different numerical weather prediction models, can objectively predict convective weather, and has better forecasting effect.

In order to achieve the above technical purpose, the present invention provides a method for forecasting convective weather, comprising the following steps:

S1, according to many years of convective weather conditions and physical data to obtain a plurality of representative physical quantities and the threshold value of each said representative physical quantity;

S2, grouping according to the representative meaning of the representative physical quantity; dividing the grouped said representative physical quantity into five levels according to the threshold segmental linearization;

S3, the convective weather phenomenon is related to a plurality of said representative physical quantities, and respectively determine the said representative physical quantities related to each convective weather phenomenon;

S4. Comparing the physical quantities of the actual data with the thresholds of the representative physical quantities respectively, if the thresholds of all the representative physical quantities of a certain convective weather phenomenon are satisfied, the weather phenomenon is predicted to occur; otherwise, no This weather phenomenon occurs.

Optionally, in step S2, when linearizing the representative physical quantity in pieces, according to the threshold, it can be divided into five levels: weak, weak to medium, medium, medium to strong, and strong, and use 1, 2, 3 respectively , 4, 5 represent the corresponding grade value.

Optionally, in step S4, the level values of the physical quantities satisfying a certain convective weather phenomenon are summed up, and the greater the obtained value, the greater the probability of occurrence of the corresponding convective weather phenomenon.

Optionally, the characteristic physical quantities related to short-term heavy rainfall include: precipitable water in the entire layer of the atmosphere, relative humidity, uplift index of the most unstable layer, K index, and bottom divergence.

Optionally, the characteristic physical quantities related to hail include: the precipitable water volume of the whole layer of the atmosphere, the most unstable layer uplift index, the low layer divergence, 0-6km vertical wind shear, 0 degree layer height, and convective effective potential energy.

Optionally, the characteristic physical quantities related to lightning include: the most unstable layer uplift index, convective effective potential energy, low layer divergence, and the relative humidity of the whole layer of the atmosphere.

The present invention provides a convective weather forecasting method. The forecasting method first obtains a plurality of representative physical quantities and the threshold values of each said representative physical quantity according to years of convective weather live conditions and physical data; then, according to the representative meaning of said representative physical quantities performing grouping; dividing the grouped representative physical quantities into five levels according to the threshold segmental linearization. The weather phenomenon is related to a plurality of the representative physical quantities, and the representative physical quantities related to each weather phenomenon are respectively determined; after that, each physical quantity of the actual data is compared with the threshold value of each said representative physical quantity, and if a certain If the threshold values of all the physical quantities representing a weather phenomenon are predicted, the weather phenomenon will appear; otherwise, the weather phenomenon will not appear.

This prediction method is mainly used in the prediction of convective weather phenomena in short-term periods, and can objectively predict convective weather phenomena. The prediction method first automatically recognizes environmental conditions, and compares them with various representative physical quantities to determine whether the environmental conditions meet convective weather conditions. The degree of the weather phenomenon is described, so as to obtain the forecast structure of the convective weather phenomenon. This prediction method can be applied to different numerical weather prediction models, and it can objectively predict convective weather, and the prediction effect is better.

Description of drawings

The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the description, together with the following specific embodiments, are used to explain the present invention, but do not constitute a limitation to the present invention.

FIG. 1 is a flow chart of the method for predicting convective weather provided by the present invention.

Detailed ways

Specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

Please refer to FIG. 1 , which is a flow chart of the convective weather forecasting method provided by the present invention.

In a specific embodiment, the present invention provides a method for predicting convective weather, comprising the following steps:

Step S1, according to years of convective weather conditions and physical data to obtain a plurality of representative physical quantities and the threshold value of each said representative physical quantity;

Step S2, grouping according to the representative meaning of the representative physical quantity; dividing the grouped said representative physical quantity into five levels according to the threshold segmental linearization;

Step S3, the convective weather phenomenon is related to a plurality of said representative physical quantities, and respectively determine the said representative physical quantities related to each convective weather phenomenon;

Step S4, comparing the physical quantities of the actual data with the thresholds of the representative physical quantities, if the thresholds of all the representative physical quantities of a certain convective weather phenomenon are met, then the weather phenomenon is predicted to occur; otherwise, no This weather phenomenon occurs.

This prediction method is mainly used in the forecast of convective weather phenomena in a short period of time. It can objectively forecast convective weather phenomena. First, it automatically recognizes the environmental conditions. By comparing with various representative physical quantities, the environmental conditions meet certain convective weather phenomena. To describe the degree of the convective weather phenomenon, so as to obtain the forecast results of the convective weather phenomenon.

The prediction process of this prediction method does not involve the physical process of convective weather phenomena, but the environmental conditions of convective weather phenomena. it is good

Take precipitation as an example. From a synoptic point of view, precipitation is the product of condensation from uplifted moist air. Sufficient water vapor content, upward motion of the air, and rapid uplift are necessary conditions for high rainfall rates. For short-term heavy precipitation, the ambient atmosphere also needs to meet some conditions in terms of water vapor, heat and dynamics, and these conditions can be characterized by physical quantities.

In order to obtain physical quantities with representative significance, multiple representative physical quantities can be analyzed in two different ways, one is based on the strength of short-term heavy precipitation environmental indications based on box plots, and the other is based on test scores The physical quantity indicates the order of meaning. The former is based on the classification of physical quantities, while the latter is purely mathematical statistics without considering the physical meaning of physical quantities. But in comparison, the physical meaning of the former is clearer. The prediction method is based on the identification of the physical quantity.

This prediction method is based on experience and statistics. Compared with pure experience methods, it has higher credibility. At the same time, due to the use of statistical results, the technology has higher reliability, strong objectivity, and coverage. The scope is wide, the production time of products is short, and the universality is strong. In addition, compared with the forecast method based on experience, this technology has better portability.

In a further specific implementation, in step S2, when linearizing the representative physical quantity in pieces, according to the threshold, it is divided into five levels: weak, weak to medium, medium, medium to strong, and strong, and respectively use 1, 2, 3, 4, 5 represent the corresponding grade values.

In this step, each representative physical quantity is first grouped, and then each group of representative physical quantities is graded. Through piecewise linearization processing, according to the specific threshold, the actual environmental conditions represented by each physical quantity are displayed as strong, medium, and weak, and are not randomized. Changes with the seasons.

After classifying according to the representative meaning of the physical quantity, the indicative meaning is analyzed. In the process of physical quantity classification, it has been defaulted that the same type of physical quantity plays an equally important role in characterizing the environmental characteristics of convective weather. Therefore, for the selected physical quantity, By default, equal weight is given to each group of physical quantities.

In a further embodiment, in step S4, the grade values of the physical quantities satisfying a certain convective weather phenomenon are summed up, and the greater the obtained value, the greater the probability of occurrence of the corresponding convective weather phenomenon.

The grade values of each physical quantity are added, and the obtained value can be converted into the probability of occurrence of the weather phenomenon, and the forecast result of a certain weather phenomenon is displayed in a probabilistic manner.

In each of the specific implementations above, the characteristic physical quantities related to short-term heavy rainfall include: the precipitable water volume of the entire layer of the atmosphere, relative humidity, the most unstable layer uplift index, K index, and bottom layer divergence.

In each of the above specific implementations, the characteristic physical quantities related to hail include: the amount of precipitable water in the entire layer of the atmosphere, the most unstable layer uplift index, the low-level divergence, 0-6km vertical wind shear, 0-degree layer height, convection effective potential energy.

In each of the specific implementations above, the characteristic physical quantities related to lightning include: the most unstable layer uplift index, convective effective potential energy, low layer divergence, and the relative humidity of the entire layer of the atmosphere.

Taking short-term heavy precipitation as an example, the boxplot-based short-term heavy precipitation environment identification method uses 9 years of real-time and physical quantity data to analyze a number of quantities that can be used to characterize environmental atmospheric water vapor, thermal and dynamic conditions, including Precipitable water in the whole layer of the atmosphere, relative humidity, specific humidity, optimal uplift index, K index, total index, maximum convective effective potential energy, low-level divergence, vertical wind shear, temperature difference between upper and lower layers, and the results show that water vapor Among the physical quantities, the precipitable water in the whole layer of the atmosphere is the most significant. Among the physical quantities that characterize the thermal characteristics of the environment, the uplift index is comparable to the K index, but they are better than other physical quantities. Among the physical quantities that characterize the dynamic conditions, the 850-hPa However, the vertical wind shear condition is not good for short-term heavy precipitation, and the maximum convective effective potential energy is not good for short-term heavy precipitation. At the same time, it is also found that some basic conditions need to be met when short-term heavy precipitation occurs, such as 850-hPa temperature, etc. Relative humidity also has certain significance.

For these physical quantities, the amount of precipitable water in the entire atmosphere theoretically determines the maximum possible instantaneous surface precipitation. Both the optimal uplift index and the K index can be used to characterize the environmental instability conditions of short-term heavy precipitation in central and eastern my country. When a storm is formed, stronger unstable conditions often imply stronger upward movement, and stronger upward movement is a necessary condition for high-intensity short-term heavy precipitation. Relative humidity characterizes the degree of saturation of the atmosphere, and more than three-quarters of precipitation occurs in an environment with a relative humidity greater than 80%. Saturated air from the ground to high altitude is more conducive to the formation of precipitation, especially short-term heavy precipitation. Although multiple processes at various scales can cause the uplift of air masses, the large-scale uplift conditions of storms that produce large-scale high-intensity short-term heavy precipitation can all be characterized by 850-hPa divergence.

The forecast results cover the understanding of the physical mechanism of the environmental conditions that produce short-term heavy precipitation, rather than pure mathematical statistics.

It can be understood that, the above embodiments are only exemplary embodiments adopted for illustrating the principle of the present invention, but the present invention is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also regarded as the protection scope of the present invention.

Claims (6)

1. a kind of prediction technique of convection weather, which is characterized in that include the following steps：

S1, according to convection weather for many years is live and physical data obtains multiple characterization physical quantitys and each characterization physical quantity Threshold values；

S2 is grouped according to the symbolical meanings of the characterization physical quantity；By the characterization physical quantity after grouping according to Threshold values piecewise linearity is divided into Pyatyi；

S3, convection weather phenomenon is related with multiple characterization physical quantitys, determines institute related with each convection weather phenomenon respectively State characterization physical quantity；

Each physical quantity of real data is compared with the threshold values of each characterization physical quantity, if meet by S4 respectively The threshold values of all characterization physical quantitys of certain convection weather phenomenon：It is that then prediction the weather phenomenon will occur；It is no, then it predicts not It will appear the weather phenomenon.

2. the prediction technique of convection weather as described in claim 1, which is characterized in that in step S2, to the characterization physics When measuring piece-wise linearization, according to the threshold values be divided into it is weak, weak to it is medium, medium, in wait until strong, strong Pyatyi, and respectively with 1,2, 3,4, the 5 corresponding grade point of characterization.

3. the prediction technique of convection weather as claimed in claim 2, which is characterized in that in step S4, certain convection current will be met The grade point of each physical quantity of weather phenomenon is added, and obtained value is bigger, and the probability for corresponding convection weather phenomenon occur is bigger.

4. the prediction technique of convection weather as described in claims 1 to 3, which is characterized in that institute related with heavy showers in short-term Stating characterization physical quantity includes：Atmospheric precipitable water, relative humidity, the most instable significant level lifting index, K indexes, low layer divergence.

5. the prediction technique of convection weather as described in claims 1 to 3, which is characterized in that the characterization related with hail Physical quantity includes：Atmospheric precipitable water, the most instable significant level lifting index, low layer divergence, 0-6km vertical wind shears, 0 degree of layer Highly, convective available potential energy.

6. the prediction technique of convection weather as described in claims 1 to 3, which is characterized in that the characterization related with thunder and lightning Physical quantity includes：The most instable significant level is lifted index, convective available potential energy, low layer divergence, whole atmosphere relative humidity.