An objective forecast method for thunderstorm gale based on Doppler radar and mesoscale model

Abstract

The invention relates to the technical field of meteorological forecasting and discloses an objective forecasting method for thunderstorms and gale based on Doppler radar and mesoscale model. The method comprises the following steps: step 1, establishing a data collection module for collecting various meteorological data, including ground observation data, radar echo data and satellite remote sensing data; step 2, performing model simulation and construction of a virtual environment data collection module, and constructing the virtual environment data collection module on the basis of the model; step 3, establishing a data calculation module, and performing fusion calculation on the collected data and the data generated by the virtual environment data collection module in the data calculation module; step 4, establishing a verification module, and performing formula verification by substituting historical stored data in the data collection module to verify the validity of the model; step 5, establishing an evaluation module, and evaluating the accuracy of the forecast according to the calculation results; and step 6, establishing a management module, and adjusting the model parameters according to the information of the evaluation module.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological forecasting, and in particular to an objective forecasting method for thunderstorms and gale winds based on Doppler radar and a mesoscale model. Background Art

[0002] The existing objective forecasting method for thunderstorms and gale winds combines advanced remote sensing technology and numerical simulation technology. Although it has achieved remarkable results in improving forecast accuracy, it still has some defects, mainly reflected in the forecast timeliness and spatial resolution.

[0003] First, existing methods have limitations in forecast timeliness. Due to the sudden and localized nature of thunderstorms and gale-force winds, their generation and development processes are complex and dynamic, making them difficult to accurately predict. Consequently, existing forecasting methods often fail to provide sufficient lead time for relevant departments to implement countermeasures. This is primarily because current forecasting models still exhibit a certain lag when dealing with rapidly changing meteorological conditions, making it impossible to capture the generation and evolution of thunderstorms and gale-force winds in real time.

[0004] Secondly, the spatial resolution of existing methods needs to be improved. Although the spatial resolution of mesoscale models is already high, they still cannot capture all key small and medium-scale features, which will affect the detail and accuracy of the forecast. For example, in some areas with complex terrain, mesoscale models may not accurately simulate the impact of terrain on thunderstorm winds, resulting in deviations in forecast results. In addition, although Doppler radar can provide observation data with high temporal and spatial resolution, its coverage is limited and it is easily affected by factors such as weather conditions and terrain obstruction, resulting in missing or erroneous observation data.

[0005] In summary, in order to further improve the accuracy and timeliness of objective forecasts of thunderstorms and gale winds, it is necessary to continuously improve and perfect existing forecasting methods and technical means. Summary of the Invention

[0006] (1) Technical problems solved

[0007] In response to the shortcomings of the existing technology, the present invention provides an objective forecasting method for thunderstorms and gale winds based on Doppler radar and mesoscale models, which has the advantages of no limitations in forecast timeliness and high spatial resolution, and solves the problems of certain limitations in forecast timeliness and low spatial resolution in the existing technology.

[0008] (2) Technical solution

[0009] To achieve the above object, the present invention provides the following technical solution: a method for objectively forecasting thunderstorm gale based on Doppler radar and mesoscale model, comprising the following steps:

[0010] Step 1: Establish a data collection module to collect various meteorological data, including ground observation data, radar echo data and satellite remote sensing data;

[0011] Step 2: Conduct model simulation and construct a virtual environment data collection module. Construct a virtual environment data collection module based on the model.

[0012] Step 3: Establish a data calculation module, in which the collected data is integrated with the data generated by the virtual environment data collection module;

[0013] Step 4: Establish a verification module to verify the formula by substituting historical data stored in the data collection module to verify the validity of the model;

[0014] Step 5: Establish an evaluation module to evaluate the accuracy of the forecast based on the calculation results;

[0015] Step 6: Establish a management module and adjust the model parameters and forecasting strategies based on the information from the evaluation module.

[0016] Preferably, the data collection module includes a ground weather station unit, a meteorological satellite unit and a Doppler radar unit. The ground weather station unit collects ground meteorological data through sensors, data collectors and operation monitoring equipment. The meteorological satellite unit collects meteorological satellite forecast data through satellite payloads and ground station receiving stations. The Doppler radar unit collects Doppler radar measurement data through antennas, transmitters and signal processors.

[0017] Preferably, the virtual environment data collection module collects virtual environment prediction data through a model simulation environment.

[0018] Preferably, the data calculation module includes a forecast timeliness evaluation unit, a spatial resolution test unit and a precipitation forecast scoring unit. The forecast timeliness evaluation unit calculates the model forecast timeliness accuracy Az based on the virtual environment prediction data, the spatial resolution test unit calculates the spatial resolution overlap rate Sx based on the Doppler radar measurement data and the virtual environment prediction data, and the precipitation forecast scoring unit calculates the simulation model prediction accuracy Dh based on the ground meteorological data and the virtual environment prediction data.

[0019] Preferably, the ground meteorological station unit numbers the actual accumulated precipitation in a period of time according to the characteristics of the ground meteorological data, and the actual accumulated precipitation in a period of time is numbered as q1, q2, q3, ... q n The meteorological satellite unit numbers the predicted cumulative precipitation within a period of time according to the characteristics of the meteorological satellite prediction data. The predicted cumulative precipitation within a period of time is numbered as y1, y2, y3, ...y nThe Doppler radar unit numbers the number of antenna elements, wavelength, and effective aperture of the antenna array according to the characteristics of the Doppler radar measurement data. The number of antenna elements, wavelength, and effective aperture of the antenna array are numbered T, α, and D, respectively.

[0020] Preferably, the virtual environment data collection module numbers the number of times the model actually occurs and is predicted to occur, the number of times it actually does not occur but is predicted to occur, and the number of times it actually occurs but is predicted not to occur according to the virtual environment prediction data characteristics. The number of times it actually occurs and is predicted to occur, the number of times it actually does not occur but is predicted to occur, and the number of times it actually occurs but is predicted not to occur are numbered n1, n2, and n3, respectively.

[0021] Preferably, the forecast timeliness evaluation unit calculates the forecast timeliness accuracy Az of the model based on the virtual environment prediction data, and the calculation formula is:

[0022]

[0023] In the formula, Az represents the accuracy of the model forecast timeliness, n1, n2, and n3 represent the number of times that the event actually occurred and was predicted to occur, the number of times that the event actually did not occur but was predicted to occur, and the number of times that the event actually occurred but was not predicted to occur, respectively. It represents the proportion of actual thunderstorms and gales that were also predicted by the forecast, that is, the hit rate P. It represents the proportion of thunderstorms and gales that did not actually occur but were predicted to occur, that is, the false alarm rate X. It represents the ratio of correctly predicted thunderstorms and gale times to the total number of predicted thunderstorms and gale times, that is, the accuracy V.

[0024] Preferably, the spatial resolution test unit calculates the spatial resolution overlap rate Sx based on the Doppler radar measurement data and the virtual environment prediction data, and the calculation formula is:

[0025]

[0026] In the formula, Sx represents the spatial resolution overlap rate, T represents the number of antenna elements, α represents the wavelength, and D represents the effective aperture of the antenna array. represents the spatial resolution of the radar measurement data, K represents the spatial resolution of the virtual environment prediction data, which is a fixed parameter set by the model, and C represents the maximum spatial resolution, which is the maximum spatial resolution between the spatial resolution of the radar measurement data and the spatial resolution of the virtual environment prediction data. The spatial resolution of the radar measurement data and the spatial resolution of the virtual environment prediction data are compared, and the larger one is selected as the maximum spatial resolution.

[0027] Preferably, the precipitation forecast scoring unit calculates the cumulative precipitation forecast score Dh based on ground meteorological data and meteorological satellite forecast data, and the calculation formula is:

[0028]

[0029] In the formula, Dh represents the cumulative precipitation forecast score, m represents the number of observation points, q1, q2, q3, ...q n Indicates the actual cumulative precipitation over a period of time, y1, y2, y3, ...y n Indicates the predicted cumulative precipitation over a period of time, q i and y i They represent the actual cumulative precipitation and predicted cumulative precipitation at the i-th observation point over a period of time, respectively.

[0030] Preferably, the evaluation module evaluates the forecast accuracy of the model simulation environment according to the model forecast time accuracy Az, evaluates the forecast accuracy of the Doppler radar according to the spatial resolution overlap rate Sx, and evaluates the forecast accuracy of the meteorological satellite according to the cumulative precipitation forecast score Dh;

[0031] The management module adjusts the physical process and algorithm of the model according to the information of the evaluation module.

[0032] Compared with the existing technology, the present invention provides an objective forecasting method for thunderstorms and gale winds based on Doppler radar and mesoscale models, which has the following beneficial effects:

[0033] 1. The present invention calculates the model's forecast timeliness accuracy Az. The formula comprehensively measures the model's forecasting capabilities at different time scales by taking into account the number of times an event actually occurred and was predicted to occur, the number of times an event actually did not occur but was predicted to occur, and the number of times an event actually occurred but was not predicted to occur. This is to assess the model's accuracy in capturing thunderstorm and gale events, thereby reducing false positives and omissions in forecasts.

[0034] 2. By calculating the spatial resolution overlap rate Sx, the present invention can effectively evaluate the accuracy of Doppler radar in capturing the detailed characteristics of thunderstorm and gale, improve the precision of the forecast, and enhance the model's forecasting ability for thunderstorm and gale events.

[0035] 3. The present invention monitors satellite forecasts in real time by calculating the cumulative precipitation forecast score Dh, and can monitor the accuracy of meteorological satellite forecast data in real time. This method can promptly detect the difference between the forecast and the actual situation. When the cumulative precipitation forecast score Dh is monitored to be lower than 95%, the difference is adjusted through a rapid feedback mechanism, thereby improving the timeliness of satellite forecasts. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a structural schematic diagram of the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0038] See also Figure 1 , an objective forecasting method for thunderstorm gale based on Doppler radar and mesoscale model, comprising the following steps:

[0039] Step 1: Establish a data collection module to collect various meteorological data, including ground observation data, radar echo data and satellite remote sensing data;

[0040] Step 2: Conduct model simulation and construct a virtual environment data collection module (using a high-resolution mesoscale numerical model to simulate thunderstorm winds, using a three-dimensional meteorological field as the initial field to simulate the spatiotemporal variation characteristics of the thunderstorm cloud's macrodynamics, microphysical processes, and electrical structure, and then combining it with a long short-term memory (LSTM) network). Based on the model, a virtual environment data collection module is constructed to better capture the complex evolution of thunderstorm winds.

[0041] Step 3: Establish a data calculation module, in which the collected data is integrated with the data generated by the virtual environment data collection module;

[0042] Step 4: Establish a verification module to verify the formula by substituting historical data stored in the data collection module to verify the validity of the model;

[0043] Step 5: Establish an evaluation module to evaluate the accuracy of the forecast based on the calculation results;

[0044] Step 6: Establish a management module to adjust model parameters and forecast strategies based on the information from the evaluation module. At the same time, formulate targeted response measures based on actual weather conditions and terrain characteristics to improve the practicality and effectiveness of the forecast.

[0045] The data collection module includes a ground meteorological station unit, a meteorological satellite unit and a Doppler radar unit. The ground meteorological station unit collects ground meteorological data through sensors, data collectors and operation monitoring equipment. The meteorological satellite unit collects meteorological satellite forecast data through satellite payloads and ground station receiving stations. The Doppler radar unit collects Doppler radar measurement data through antennas, transmitters and signal processors.

[0046] The virtual environment data collection module collects virtual environment prediction data through the model simulation environment.

[0047] The data calculation module includes a forecast timeliness evaluation unit, a spatial resolution test unit and a precipitation forecast scoring unit. The forecast timeliness evaluation unit calculates the model forecast timeliness accuracy Az based on the virtual environment prediction data. The spatial resolution test unit calculates the spatial resolution overlap rate Sx based on the Doppler radar measurement data and the virtual environment prediction data. The precipitation forecast scoring unit calculates the simulation model prediction accuracy Dh based on the ground meteorological data and the virtual environment prediction data.

[0048] The ground meteorological station unit numbers the actual cumulative precipitation in a period of time according to the characteristics of the ground meteorological data. The actual cumulative precipitation in a period of time is numbered as q1, q2, q3, ...q n The meteorological satellite unit numbers the predicted cumulative precipitation over a period of time according to the characteristics of the meteorological satellite prediction data. The predicted cumulative precipitation over a period of time is numbered as y1, y2, y3, ...y n The Doppler radar unit numbers the number of antenna elements, wavelength, and effective aperture of the antenna array according to the characteristics of the Doppler radar measurement data. The number of antenna elements, wavelength, and effective aperture of the antenna array are numbered as T, α, and D respectively.

[0049] The virtual environment data collection module numbers the number of times the model actually occurred and was predicted to occur, the number of times the model actually did not occur but was predicted to occur, and the number of times the model actually occurred but was not predicted to occur according to the characteristics of the virtual environment prediction data. The number of times the model actually occurred and was predicted to occur, the number of times the model actually did not occur but was predicted to occur, and the number of times the model actually occurred but was not predicted to occur are numbered n1, n2, and n3 respectively.

[0050] The forecast timeliness evaluation unit calculates the model forecast timeliness accuracy Az based on the virtual environment prediction data. The calculation formula is:

[0051]

[0052] In the formula, Az represents the accuracy of the model forecast timeliness, n1, n2, and n3 represent the number of times that the event actually occurred and was predicted to occur, the number of times that the event actually did not occur but was predicted to occur, and the number of times that the event actually occurred but was not predicted to occur, respectively. It represents the proportion of actual thunderstorms and gales that were also predicted by the forecast, that is, the hit rate P. It represents the proportion of thunderstorms and gales that did not actually occur but were predicted to occur, that is, the false alarm rate X. It represents the ratio of correctly predicted thunderstorms and gale times to the total number of predicted thunderstorms and gale times, that is, the accuracy V.

[0053] The advantages are: by calculating the model's forecast timeliness accuracy Az, the formula comprehensively measures the model's forecasting capabilities at different time scales through the number of times an event actually occurred and was predicted to occur, the number of times an event actually did not occur but was predicted to occur, and the number of times an event actually occurred but was not predicted to occur, so as to evaluate the model's accuracy in capturing thunderstorm and gale events, thereby reducing forecast false alarms and missed alarms.

[0054] The spatial resolution test unit calculates the spatial resolution overlap rate Sx based on the Doppler radar measurement data and the virtual environment prediction data. The calculation formula is:

[0055]

[0056] In the formula, Sx represents the spatial resolution overlap rate, T represents the number of antenna elements, α represents the wavelength, and D represents the effective aperture of the antenna array. represents the spatial resolution of the radar measurement data, K represents the spatial resolution of the virtual environment prediction data, which is a fixed parameter set by the model, and C represents the maximum spatial resolution, which is the maximum spatial resolution between the spatial resolution of the radar measurement data and the spatial resolution of the virtual environment prediction data. The spatial resolution of the radar measurement data and the spatial resolution of the virtual environment prediction data are compared, and the larger one is selected as the maximum spatial resolution.

[0057] The advantages are: by calculating the spatial resolution overlap rate Sx, the accuracy of Doppler radar in capturing the detailed characteristics of thunderstorms and strong winds can be effectively evaluated, the precision of the forecast can be improved, and the model's forecasting ability for thunderstorms and strong winds can be enhanced.

[0058] The precipitation forecast scoring unit calculates the cumulative precipitation forecast score Dh based on ground meteorological data and meteorological satellite forecast data. The calculation formula is:

[0059]

[0060] In the formula, Dh represents the cumulative precipitation forecast score, m represents the number of observation points, q1, q2, q3, ...q n Indicates the actual cumulative precipitation over a period of time, y1, y2, y3, ...y n Indicates the predicted cumulative precipitation over a period of time, q i and y i They represent the actual cumulative precipitation and predicted cumulative precipitation at the i-th observation point over a period of time, respectively.

[0061] The advantages are: by calculating the cumulative precipitation forecast score Dh, satellite forecasts can be monitored in real time, and the accuracy of meteorological satellite forecast data can be monitored in real time. This method can promptly detect the difference between the forecast and the actual situation. When the cumulative precipitation forecast score Dh is monitored to be lower than 95%, the difference is adjusted through a rapid feedback mechanism, thereby improving the timeliness of satellite forecasts.

[0062] The evaluation module evaluates the forecast accuracy of the model's simulated environment based on the model's forecast timeliness accuracy Az (the evaluation module evaluates the timeliness of the model's forecast results at different time scales. This includes the accuracy of short-term, medium-term, and long-term forecasts to ensure that the model can provide timely and effective thunderstorm and gale warning information). The module also evaluates the Doppler radar's forecast accuracy based on the spatial resolution overlap rate Sx (using the high spatial resolution data of the Doppler radar to calculate the spatial overlap rate between the radar prediction and the actual observation. This indicator helps evaluate the radar's accuracy in capturing the detailed characteristics of thunderstorms and gale, thereby improving the precision of the forecast). The module also evaluates the meteorological satellite's forecast accuracy based on the cumulative precipitation forecast score Dh (the cumulative precipitation forecast score (APFS) is calculated based on ground-based meteorological data and meteorological satellite forecast data. This score is used to quantify the model's performance in predicting cumulative precipitation and ensure the accuracy and reliability of precipitation forecasts).

[0063] The management module adjusts the model's physical processes and algorithms based on the information from the evaluation module. At the same time, it formulates targeted response measures based on actual weather conditions and terrain characteristics (these measures include warning issuance, emergency response, and public education, aimed at improving the practicality and effectiveness of forecasts and reducing the risks and losses caused by thunderstorms and strong winds) to improve the practicality and effectiveness of forecasts, while enhancing the model's ability to forecast thunderstorms and strong winds, ensuring more accurate forecast results.

[0064] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An objective forecasting method for thunderstorms and gale based on Doppler radar and mesoscale model, characterized by: The following steps are involved: Step 1: Establish a data collection module to collect various meteorological data, including ground observation data, radar echo data and satellite remote sensing data; Step 2: Conduct model simulation and construct a virtual environment data collection module. Construct a virtual environment data collection module based on the model. Step 3: Establish a data calculation module, in which the collected data is integrated with the data generated by the virtual environment data collection module; Step 4: Establish a verification module to verify the formula by substituting historical data stored in the data collection module to verify the validity of the model; Step 5: Establish an evaluation module to evaluate the accuracy of the forecast based on the calculation results; Step 6: Establish a management module to adjust model parameters and forecast strategies based on the information from the evaluation module; The virtual environment data collection module collects virtual environment prediction data through a model simulation environment; The data calculation module includes a forecast timeliness evaluation unit, a spatial resolution test unit and a precipitation forecast scoring unit. The forecast timeliness evaluation unit calculates the model forecast timeliness accuracy Az based on the virtual environment prediction data. The spatial resolution test unit calculates the spatial resolution overlap rate Sx based on the Doppler radar measurement data and the virtual environment prediction data. The precipitation forecast scoring unit calculates the simulation model prediction accuracy Dh based on the ground meteorological data and the virtual environment prediction data. The virtual environment data collection module numbers the number of times the model actually occurs and is predicted to occur, the number of times it actually does not occur but is predicted to occur, and the number of times it actually occurs but is predicted not to occur according to the virtual environment prediction data characteristics. The number of times it actually occurs and is predicted to occur, the number of times it actually does not occur but is predicted to occur, and the number of times it actually occurs but is predicted not to occur are numbered n1, n2, and n3 respectively.

2. The objective forecasting method for thunderstorms and gale based on Doppler radar and mesoscale model according to claim 1, characterized in that: The data collection module includes a ground weather station unit, a weather satellite unit and a Doppler radar unit. The ground weather station unit collects ground weather data through sensors, data collectors and operation monitoring equipment. The weather satellite unit collects weather satellite forecast data through satellite payloads and ground station receiving stations. The Doppler radar unit collects Doppler radar measurement data through antennas, transmitters and signal processors.

3. The objective forecasting method for thunderstorms and gale based on Doppler radar and mesoscale model according to claim 2, characterized in that: The ground weather station unit numbers the actual accumulated precipitation in a period of time according to the characteristics of the ground weather data, and the actual accumulated precipitation in a period of time is numbered as q1, q2, q3, ...q n The meteorological satellite unit numbers the predicted cumulative precipitation within a period of time according to the characteristics of the meteorological satellite prediction data. The predicted cumulative precipitation within a period of time is numbered as y1, y2, y3, ...y n The Doppler radar unit numbers the number of antenna elements, wavelength, and effective aperture of the antenna array according to the characteristics of the Doppler radar measurement data. The number of antenna elements, wavelength, and effective aperture of the antenna array are numbered T, α, and D, respectively.

4. The objective forecasting method for thunderstorms and gale based on Doppler radar and mesoscale model according to claim 1, characterized in that: The forecast timeliness evaluation unit calculates the forecast timeliness accuracy Az of the model based on the virtual environment prediction data, and the calculation formula is: In the formula, Az represents the accuracy of the model forecast timeliness, n1, n2, and n3 represent the number of times that the event actually occurred and was predicted to occur, the number of times that the event actually did not occur but was predicted to occur, and the number of times that the event actually occurred but was not predicted to occur, respectively. It represents the proportion of actual thunderstorms and gales that were also predicted by the forecast, that is, the hit rate P. It represents the proportion of thunderstorms and gales that did not actually occur but were predicted to occur, that is, the false alarm rate X. It represents the ratio of correctly predicted thunderstorms and gale times to the total number of predicted thunderstorms and gale times, that is, the accuracy V.

5. The objective forecasting method for thunderstorms and gale based on Doppler radar and mesoscale model according to claim 1, characterized in that: The spatial resolution test unit calculates the spatial resolution overlap rate Sx based on the Doppler radar measurement data and the virtual environment prediction data, and the calculation formula is: In the formula, Sx represents the spatial resolution overlap rate, T represents the number of antenna elements, α represents the wavelength, and D represents the effective aperture of the antenna array. represents the spatial resolution of the radar measurement data, K represents the spatial resolution of the virtual environment prediction data, which is a fixed parameter set by the model, and C represents the maximum spatial resolution, which is the maximum spatial resolution between the spatial resolution of the radar measurement data and the spatial resolution of the virtual environment prediction data. The spatial resolution of the radar measurement data and the spatial resolution of the virtual environment prediction data are compared, and the larger one is selected as the maximum spatial resolution.

6. The objective forecasting method for thunderstorms and gale based on Doppler radar and mesoscale model according to claim 1, characterized in that: The precipitation forecast scoring unit calculates the cumulative precipitation forecast score Dh based on ground meteorological data and meteorological satellite forecast data. The calculation formula is: In the formula, Dh represents the cumulative precipitation forecast score, m represents the number of observation points, q1, q2, q3, ...q n Indicates the actual cumulative precipitation over a period of time, y1, y2, y3, ...y n Indicates the predicted cumulative precipitation over a period of time, q i and y i They represent the actual cumulative precipitation and predicted cumulative precipitation at the i-th observation point over a period of time, respectively.

7. The objective forecasting method for thunderstorms and gale based on Doppler radar and mesoscale model according to claim 1, characterized in that: The evaluation module evaluates the forecast accuracy of the model simulation environment according to the model forecast time accuracy Az, evaluates the forecast accuracy of the Doppler radar according to the spatial resolution overlap rate Sx, and evaluates the forecast accuracy of the meteorological satellite according to the cumulative precipitation forecast score Dh; The management module adjusts the physical process and algorithm of the model according to the information of the evaluation module.

Images