A rapid warning method for identifying sudden visibility drop

Abstract

The present invention discloses an early warning method for quickly identifying sudden drops in visibility, belonging to the field of meteorological early warning technology. The method comprises the following steps: S1, selecting low clouds, wind direction, wind speed, precipitation, the absolute value of the temperature-water difference, haze, dust, and special weather as the main meteorological factors affecting sudden drops in visibility; S2, determining the thresholds and coefficients of the effects of each meteorological element on sudden drops in visibility; S3, establishing a sudden drop in visibility early warning model; and S4, using observation equipment to obtain current meteorological element observation data, substituting the data into the early warning model to obtain a result, and taking measures based on the early warning result. By establishing an early warning model, the present invention can more quickly and accurately identify the trend of sudden drops in visibility, effectively improving the safety and efficiency of flight tests.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological monitoring, and in particular to an early warning method for rapidly identifying sudden drop in visibility. Background Art

[0002] According to relevant statistics, meteorological factors account for 20% of international civil aviation accident causes, second only to crew error. Low visibility accounts for 16% of all accidents, exceeding thunderstorms. Domestically, low visibility accounts for 29.2% of all accidents, demonstrating the significant impact of visibility on flight safety. Sudden drops in visibility, in particular, pose a significant risk to aircraft takeoff and landing.

[0003] Numerous scholars, both domestically and internationally, have studied the changing trends and characteristics of atmospheric visibility. Research from the UK, the US, and other countries has found that haze, formed by precipitation and water vapor, is the primary cause of low visibility. A close relationship exists between aerosol concentrations in the troposphere and atmospheric visibility, with visibility most closely linked to PM2.5 concentrations, sulfate, and nitrate concentrations. Furthermore, studies using chemical composition analysis to study the scattering extinction coefficients of pollutants have shown that the reduced visibility caused by haze days can be attributed to fine particulate matter.

[0004] my country has also achieved some results in studying changes in atmospheric visibility. The main conclusions are that atmospheric pollutants and humidity are the key factors affecting atmospheric visibility; visibility is negatively correlated with relative humidity and pollutant concentrations, positively correlated with wind speed, and sometimes positively and sometimes negatively correlated with temperature, with no significant relationship to air pressure.

[0005] However, most of the aforementioned research focuses on visibility forecasts and severe low visibility and foggy weather conditions, but has yet to address the rapid deterioration of local visibility. Sudden drops in visibility pose a significant threat to flight safety, severely impacting the safety and efficiency of scientific research flight testing. Therefore, developing an algorithmic model to rapidly identify sudden drops in visibility and applying appropriate response methods are key to ensuring the safe and efficient conduct of scientific research flight testing.

[0006] Existing methods for warning sudden drops in visibility typically rely on observing a clear downward trend in the daily visibility curve. Forecasters and observers, based on their personal experience, confirm the changes in key meteorological factors affecting visibility and then alert the flight commander. This method lacks appropriate thresholds and coefficients, and lacks triggering criteria and action measures for sudden drops in visibility. This results in short warning times, false alarms, missed alerts, and a low accuracy rate (averaging 73.6%), severely impacting the safety and efficiency of scientific research test flights.

[0007] After searching and searching the literature and patents related to the impact of meteorological factors on visibility, most of them use visibility level analysis, ridit analysis, single factor variance analysis, Pearson correlation analysis, linear regression and other methods to explore the relationship between meteorological factors (temperature, precipitation, relative humidity, wind speed, air pressure) and atmospheric pollutants (PM 2.5 The degree and correlation of atmospheric quality impacts of CO, O₃, and SO₂ concentrations are known. However, due to the numerous meteorological factors influencing visibility and the complex interactions between some of these factors, most visibility forecasting models struggle to accurately predict visibility trends. Especially for predicting sudden drops in local visibility, current visibility forecasts typically use a rolling forecast based on current conditions to recursively predict future trends. However, this rolling forecast model is prone to error accumulation, leading to erroneous forecast results. Summary of the Invention

[0008] The present invention aims to solve the problems in the prior art of short warning time for sudden visibility drop, no warning threshold and coefficient, no triggering standard and insufficient response measures, and provide an early warning method for quickly identifying sudden visibility drop. The method can quickly and accurately detect the trend of sudden visibility drop in the airport area.

[0009] In order to achieve the above-mentioned object of the invention, the technical solution of the present invention is as follows:

[0010] A method for quickly identifying a sudden drop in visibility, comprising the following steps:

[0011] Step S1: Determine the meteorological factors that affect the sudden drop in visibility

[0012] Combined with the historical meteorological data statistics of the target monitoring area, low clouds, wind direction, wind speed, precipitation, absolute value of temperature-water difference, haze, floating dust and special weather are selected as the main meteorological factors affecting the sudden drop in visibility;

[0013] Step S2: Determine the threshold and coefficient of meteorological elements

[0014] Historical meteorological data were used to identify meteorological elements that caused visibility to drop from 6 km or higher to the minimum required level within one hour. Classification and statistical analysis were then performed. The change in a single element and the 1 km visibility drop were used as thresholds. Ridit analysis, one-way analysis of variance, and the weighted contribution of each element to the visibility drop were used to determine the coefficient of influence of each meteorological element on the sudden drop in visibility.

[0015] Step S3, establishing a sudden visibility drop warning model: determining a sudden visibility drop warning model based on the threshold and coefficient determined in step S2;

[0016] Step S4: Use observation equipment to obtain current meteorological element observation data, substitute it into the early warning model to obtain the result, and take action based on the early warning result.

[0017] Furthermore, the established visibility drop warning model is as follows:

[0018] ∑=0.11*A1+0.15*A2+0.31*A3+0.21*A4+0.13*A5+0.09*A6;

[0019] Among them, A1 to A6 represent different meteorological elements. A1 represents low clouds, A2 represents wind direction and speed, A3 represents precipitation, A4 represents temperature difference, A5 represents haze and dust, and A6 represents special weather.

[0020] When each meteorological element appears and reaches the threshold, the value is 1; if not, the value is 0.

[0021] Furthermore, the conditions for each meteorological element to appear and reach the threshold are as follows:

[0022] .

[0023] Furthermore, it also includes step S5, model optimization: shortening the time variable of establishing the model, processing special weather that has a particularly large impact on visibility separately, and correcting the coefficient of special weather to 1 in the model, and adding its original coefficient to the meteorological element A5 representing haze and dust, and averaging after the coefficients of each meteorological element appear or reach the threshold.

[0024] Furthermore, the expression of the early warning model is:

[0025] ;

[0026] Among them, A1 indicates low clouds; A2 indicates wind direction and speed; A3 indicates precipitation; A4 indicates temperature difference; A5 indicates haze and dust; A6 indicates special weather conditions;

[0027] When each meteorological element appears or reaches the threshold, the value is 1, and if not, the value is 0.

[0028] Furthermore, the conditions for each meteorological element to appear and reach the threshold are as follows:

[0029] .

[0030] Furthermore, all observation data are automatically collected by observation equipment, refreshed every minute, and recorded in observation reports.

[0031] Furthermore, in step S4, special weather conditions are determined by manual observation; visibility is collected using a visibility observation instrument; total cloud cover and cloud height are measured using an all-sky ground-based cloud meter; wind direction and wind speed are measured using an ultrasonic anemometer; precipitation is obtained using a precipitation observation instrument; and temperature, humidity, water vapor pressure, haze, and dust are measured using a six-element automatic weather station.

[0032] In summary, the present invention has the following advantages:

[0033] 1. This invention utilizes meteorological data that can be collected in real time at the airport, replacing the rolling forecast method to recursively predict future change trends, which is prone to error accumulation. This method can effectively improve the accuracy of sudden visibility drop estimation. The prediction method based on this model can more quickly and accurately identify sudden visibility drop trends, effectively improving flight test safety and efficiency.

[0034] 2. The present invention summarizes the meteorological factors that affect sudden visibility drop. The time variable is 1 hour. By comparing the changes in meteorological factors before and after the hour, and taking the change value of a single factor and the visibility reduction of 1000 meters as the threshold, the Ridit analysis method and the one-way variance analysis method are combined with the statistical analysis of the weight of the impact of a single factor on visibility reduction to determine the coefficient of each meteorological factor in the model, thereby establishing a sudden visibility drop warning model.

[0035] 3. Based on the established visibility drop warning model, the present invention uses observation equipment to automatically collect current meteorological element observation data in real time. Computer-assisted calculations are used to ensure that meteorological operators can promptly detect visibility drop trends and make relevant recommendations and measures based on the model calculation results to ensure flight test safety and efficiency.

[0036] 4. The early warning model of the present invention fully considers the problem of the mutual influence of various meteorological factors. The model is further improved, the time variable is shortened to 10 minutes, and the coefficients of the occurrence or reaching of the threshold of each meteorological factor are added and averaged. The improved early warning model can issue an early warning about 30 minutes in advance with an accuracy rate of 95.6%. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a warning flow chart of Example 1 of the present invention;

[0038] Figure 2 This is a warning flow chart of Example 2 of the present invention. DETAILED DESCRIPTION

[0039] The present invention will be further described in detail below with reference to the examples, but the embodiments of the present invention are not limited thereto.

[0040] Example 1

[0041] This embodiment 1 provides a method for quickly identifying a sudden drop in visibility. The method includes the following steps:

[0042] Step 1: Determine the meteorological factors that affect the sudden drop in visibility

[0043] Traditionally, the determination of meteorological factors affecting visibility is primarily based on environmental factors, but the meteorological factors affecting visibility in flight meteorological support differ slightly. Based on the meteorological factors collected in real time at this airport, we applied the statistical results of the single factor method and combined them with historical meteorological data to preliminarily select low cloud, wind direction and speed, precipitation, temperature and humidity, dust, haze, and special weather conditions as the main meteorological factors affecting sudden visibility drops.

[0044] When compiling historical meteorological data, it was found that there was no obvious change in the above-mentioned single meteorological factors one hour before and after some sudden drops in visibility occurred. Careful comparison of all the data of these cases revealed that the absolute value of the difference between temperature and water vapor pressure (absolute humidity) between 2.0 and 3.0 is likely to cause a sudden drop in visibility.

[0045] Based on the above conclusions, low clouds, wind direction, wind speed, precipitation, absolute value of temperature-water difference, haze, floating dust and special weather were finally selected as the main meteorological factors affecting sudden drop in visibility.

[0046] Step 2: Determine the thresholds and coefficients of meteorological elements

[0047] Historical meteorological data was used to screen for meteorological elements whose visibility dropped from ≥6 km to below the airport's minimum visibility requirement within one hour (the minimum visibility for airports or users with higher visibility requirements is generally 3000 meters, while general users can adjust the screening level based on their minimum visibility requirements). A categorized statistical analysis (a total of 216 samples in this example) was conducted to examine the relationship between the range of change of each meteorological element within one hour and visibility reduction. Using the change value of a single element and a visibility reduction of 1000 meters as the threshold, a Ridit analysis, one-way analysis of variance, and statistical analysis of the weight of each element's impact on visibility reduction were used to determine the coefficient of influence of each meteorological element on the sudden drop in visibility, as shown in Table 1.

[0048] Table 1 Coefficients and thresholds of meteorological factors affecting sudden visibility drop

[0049]

[0050] Low clouds: This observation found that the low clouds have a height of <500 meters and a cloud cover of ≥50%. Visibility will be reduced, and the time it takes to drop by 1000 meters is about 30 minutes on average, which accounts for about 10% of the impact on the sudden drop in visibility.

[0051] Wind direction and speed: The wind direction in this area is 90°~180°, with a wind speed of 3m / s. Visibility will be reduced, and the average time for a reduction of 1000 meters is about 20 minutes; the impact on the sudden drop in visibility is about 15%.

[0052] Precipitation (Snow): When continuous rain (snow) occurs within or within a 5,000-meter radius of the site, while precipitation can remove and flush out air pollutants and reduce atmospheric pollution, it also reduces temperatures, increases water vapor in the air, and increases humidity. This, combined with an increase in low cloud cover, lower cloud height, and dimming light, can lead to a rapid decrease in visibility. On average, it takes about 20 minutes for visibility to drop by 1,000 meters. Considering precipitation alone, the contribution of precipitation to a sudden drop in visibility is approximately 30%. The greater the intensity of the precipitation, the faster the visibility decreases.

[0053] Absolute value of the temperature-water difference: The relationship between temperature, humidity, and visibility generally refers to the relationship between temperature and relative humidity. When relative humidity is ≥90%, rising temperatures cause water vapor to rise, favoring the formation of low clouds. Lower temperatures promote the formation of fog, which reduces visibility. The absolute value of the difference between temperature and vapor pressure (absolute humidity) is referred to as the temperature-water difference. Although the units of temperature and vapor pressure are different, the diurnal curve of the temperature-water difference closely resembles that of visibility (between 07:00 and 17:00), and it also slightly precedes changes in visibility. When the temperature-water difference decreases between 2.0 and 3.0 degrees per hour, the probability of a sudden drop in visibility is >83%, and a 1000-meter drop takes an average of just over 20 minutes, contributing approximately 21% to the sudden drop in visibility.

[0054] Haze and dust: Haze and dust are fine particles suspended in the air and are indicators of air pollution (PM2.5). When exposed to sunlight, they easily form photochemical smog (since there is no photochemical smog weather symbol, it is only recorded as haze). Haze and dust affect visibility by reducing it by 1,000 meters for an average of about 30 minutes, contributing to approximately 13% of sudden visibility drops.

[0055] Special weather conditions: Sandstorms, volcanic ash, and thick smoke can cause a rapid drop in visibility. Although rare, these conditions can have a significant impact on visibility, with visibility dropping by 1,000 meters lasting an average of 10 minutes. These conditions contribute to approximately 9% of sudden drops in visibility.

[0056] Step 3: Based on the above conclusions, simulate the early warning model for sudden visibility drop

[0057] A1 to A6 represent different meteorological elements. Let A1 = low cloud, A2 = wind direction and speed, A3 = precipitation, A4 = temperature difference, A5 = haze and dust, and A6 = special weather. The warning model is as follows:

[0058] ∑=0.11*A1+0.15*A2+0.31*A3+0.21*A4+0.13*A5+0.09*A6

[0059] When each meteorological element appears or reaches the threshold, it is 1; if not, it is 0.

[0060] Step 4: Use the above warning model to predict visibility

[0061] To ensure the accuracy and continuity of the meteorological monitoring data required by the present invention, such as cloud cover, cloud base height, wind direction, wind speed, temperature, humidity, precipitation, haze, and dust, all observation data are automatically collected using observation equipment, refreshed every minute, and recorded in observation reports. Visibility can be collected using forward scattering or transmission-type visibility meters. Total cloud cover, cloud shape, and cloud height can be measured using a ground-based all-sky ceilometer. Cloud height can also be measured using a ceilometer. Visibility and runway visual range can be measured using a transmission-type visibility meter. Wind direction and wind speed can be measured using an ultrasonic anemometer. Precipitation phenomena and precipitation amounts can be measured using a precipitation phenomenon meter. Temperature, humidity, water vapor pressure, haze, and dust can be measured using a six-element automatic weather station.

[0062] Substitute the above measurement data into the early warning model and take action based on the model calculation results:

[0063] When ∑≥0.3, immediately notify the main shift and field forecasters that visibility is decreasing and remind the commander; pay attention to changes in visibility in a timely manner and keep relevant records.

[0064] When ∑≥0.4, immediately notify the main shift and field forecaster that visibility is decreasing and alert the commander; pay attention to changes in visibility in a timely manner, recommend not to take off again, notify aircraft flying in the airspace to prepare to return, and keep relevant records.

[0065] When ∑≥0.5, immediately notify the main shift and field forecasters that visibility is decreasing rapidly and alert the commander; pay close attention to changes in visibility, recommend grounding, notify aircraft flying in the airspace to return immediately, and keep relevant records.

[0066] When ∑≥0.6, immediately notify the main shift and field forecasters that visibility is decreasing rapidly and alert the commander; pay close attention to changes in visibility, recommend grounding, notify aircraft flying in the airspace to return immediately, contact the alternate airport in a timely manner, make preparations for the alternate landing, and keep relevant records.

[0067] Due to the changes in meteorological factors, the decrease in visibility is non-uniform and generally shows a trend of gradual acceleration. Therefore, when ∑≥0.3, the commander should be reminded and corresponding suggestions should be made. The larger the ∑, the faster the visibility will decrease. Observers should strengthen observation and, based on the ∑ value and the changes in meteorological factors, give timely and repeated reminders, and even make preparations for diversion to ensure flight safety.

[0068] Example 2

[0069] The present invention has been verified in application and has issued warnings 10-30 minutes in advance many times with an accuracy rate of 90.8%. There have been no missed reports, and visibility has not dropped below 3 kilometers on several occasions. The actual effect of this method is accurate and effective, but there is also the problem of unsatisfactory warning time.

[0070] After research and summary, it was found that the main reason for the unsatisfactory warning time is that the warning model in Example 1 is mainly designed for the sudden drop in visibility caused by changes in a single meteorological element. In reality, various meteorological elements will interact with each other. For example, precipitation can reduce the impact of haze and dust, but it will increase water vapor and form low clouds. It will also affect the absolute value of the difference between temperature and water vapor pressure (absolute humidity). The model is based on the change of a single meteorological element with a time variable of 1 hour and cannot accurately fit the change in visibility.

[0071] To address the above issues, the time variable was changed to 10 minutes, and the coefficients of each meteorological element appearing or reaching the threshold were added and then averaged. Due to the short interval, the mutual influence of meteorological elements can be more accurately reflected. To enhance versatility, special weather that has a significant impact on visibility is treated separately, with a coefficient of 1. When special weather is discovered, a direct warning is issued. Due to its similar physical properties to haze and dust, its original coefficient is added to haze and dust. The optimized and adjusted coefficients and thresholds of meteorological elements that affect a sudden drop in visibility are shown in Table 2.

[0072] Table 2 Coefficients and thresholds of meteorological factors affecting sudden visibility drop

[0073]

[0074] The optimized and improved model is as follows: A1 = low cloud, A2 = wind direction and speed, A3 = precipitation, A4 = temperature-water difference, A5 = haze and dust, A6 = special weather, and n is the number of meteorological elements that reach the threshold; the improved model is as follows:

[0075] ;

[0076] When each meteorological element appears or reaches a threshold of 1, and if it does not appear it is 0, the warning method for sudden visibility drop remains unchanged.

[0077] The improved model has been verified through application, with the warning time being about 30 minutes earlier and an accuracy rate of 95.6%. There was no missed report, and only one time did the visibility drop below 3 kilometers. This method is accurate, effective, and highly versatile, and is especially suitable for warning of sudden visibility drops in the Chengdu area.

[0078] Example 3

[0079] Here's an example of how the method of the present invention can be used to quickly detect sudden visibility drops, issue timely warnings, and avoid flight accidents: On August 1, 2019, an aircraft (with an out-of-town pilot) was conducting a test flight at our airport. At 1:00 PM, the weather conditions at our airport indicated six low clouds at 500 meters. Visibility was greater than 6 kilometers, meeting flight conditions, and the aircraft took off. At 1:12 PM, based on the results of the sudden visibility drop warning model, the observer determined that changes in meteorological factors had triggered the sudden visibility drop warning threshold. Furthermore, convective clouds were developing vigorously at our airport, with the cloud base darkening and beginning to form on the landing path. The lead forecaster, combined with radar echoes indicating precipitation echoes moving toward our airport, immediately notified the out-of-town forecaster that the weather was deteriorating and recommended that the aircraft return immediately.

[0080] The lead forecaster convened a meeting with the station's forecasters to discuss the weather and to assess the alternate airfield conditions, allowing for emergency contact. Trainee forecasters also participated throughout the meeting, familiarizing themselves with the procedures for handling sudden weather changes. Field forecasters remained at their command posts, reporting on weather developments. The station prepared for any emergency situations, awaiting the aircraft's return. At 1:22 PM, visibility at the control tower reached 6,113 meters. This decreased to 2,661 meters at 1:24 PM, and the aircraft began its descent. At 1:27 PM, the aircraft landed safely, with visibility dropping to 703 meters. Precipitation began at the airport, soaking the entire road surface within about 10 seconds. The rain was moderate to heavy. Building 608, located 1 kilometer from the control tower, also experienced a simultaneous decrease in visibility. At 1:26 PM, visibility reached 7,015 meters, dropping to 560 meters at 1:30 PM. A minute later, the aircraft would have been unable to land at the airport. The rain continued for an hour and a half, ending at 3:00 PM.

[0081] This was a typical case of a sudden drop in local visibility during flight. Due to its brief and sudden onset, it significantly impacted flight safety. Thanks to the meteorological station's early warning, the commander's correct command, the test pilot's precise operation, and proper handling by ground support, this dangerous situation caused by sudden weather changes was avoided, ensuring flight safety.

[0082] The above content describes in detail the preferred embodiments of the present invention. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, a variety of simple variations can be made to the technical solution of the present invention, and these simple variations all fall within the scope of protection of the present invention. It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner unless there is any contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

Claims

1. A method for quickly identifying sudden visibility drop, characterized in that: The steps include: Step S1: Determine the meteorological factors that affect the sudden drop in visibility Combined with the historical meteorological data statistics of the target monitoring area, low clouds, wind direction, wind speed, precipitation, absolute value of temperature-water difference, haze, floating dust and special weather are selected as the main meteorological factors affecting the sudden drop in visibility; Step S2: Determine the threshold and coefficient of meteorological elements Historical meteorological data were used to screen meteorological elements that caused visibility to drop from 6 km or higher to the minimum required level within one hour for classification and statistical analysis. Ridit analysis, one-way analysis of variance, and the weighted contribution of each element to the sudden drop in visibility were used as the condition for the change in a single element reaching the threshold and a decrease in visibility of 1 km. The coefficient of influence of each meteorological element on the sudden drop in visibility was determined using the Ridit analysis method, one-way analysis of variance, and the weighted contribution of each element to the sudden drop in visibility. Step S3, establishing a sudden visibility drop warning model: determining a sudden visibility drop warning model based on the threshold and coefficient determined in step S2; Step S4: Use observation equipment to obtain current meteorological element observation data, substitute it into the early warning model to obtain the result, and take action based on the early warning result.

2. The early warning method for quickly identifying sudden visibility drop according to claim 1, characterized in that: In step S3, the visibility sudden drop warning model is established as follows: ∑=0.11*A1+0.15*A2+0.31*A3+0.21*A4+0.13*A5+0.09*A6; Among them, A1 to A6 represent different meteorological elements. A1 indicates low cloud, A2 indicates wind direction and speed, A3 indicates precipitation, A4 indicates temperature-water difference, A5 indicates haze and dust, and A6 indicates special weather conditions. For A3, A5, and A6, the value is 1 when the corresponding meteorological element appears, and the value is 0 when it does not appear. For A1, A2, and A4, the value is 1 when the corresponding meteorological element reaches the threshold, and the value is 0 when it does not reach the threshold. ∑ represents the result of the mathematical operation on the right side of the equation based on the corresponding values.

3. The early warning method for quickly identifying sudden visibility drop according to claim 2, characterized in that: The conditions for each meteorological element to appear and reach the threshold are as follows: 。 4. The early warning method for quickly identifying sudden visibility drop according to claim 2, characterized in that: It also includes step S5, model optimization: shortening the time variable for establishing the model, handling special weather that has a particularly large impact on visibility separately, and correcting the coefficient of special weather to 1 in the model, and adding its original coefficient to the meteorological element A5 representing haze and dust, and averaging the coefficients after each meteorological element appears or reaches a threshold.

5. The early warning method for quickly identifying sudden visibility drop according to claim 4, characterized in that: In step S5, the optimized early warning model expression is: ； Among them, A1 represents low cloud; A2 represents wind direction and speed; A3 represents precipitation; A4 represents temperature-water difference; A5 represents haze and dust; A6 represents special weather; n is the number of meteorological elements that reach the threshold; for A3, A5, and A6, the value is 1 when the corresponding meteorological element appears, and the value is 0 when it does not appear; for A1, A2, and A4, the value is 1 when the corresponding meteorological element reaches the threshold, and the value is 0 when it does not reach the threshold; ∑ represents the result of the mathematical operation on the right side of the equation based on the corresponding values.

6. The early warning method for quickly identifying sudden visibility drop according to claim 5, characterized in that: The conditions for each meteorological element to appear and reach the threshold are as follows: 。 7. The early warning method for quickly identifying sudden visibility drop according to claim 1, characterized in that: All observation data are automatically collected by observation equipment, refreshed every minute, and recorded in observation reports.

8. The early warning method for quickly identifying sudden visibility drop according to claim 7, characterized in that: In step S4, special weather conditions are determined by manual observation; visibility is collected using a visibility observation instrument; total cloud cover and cloud height are measured using an all-sky ground-based cloud meter; wind direction and wind speed are measured using an ultrasonic anemometer; precipitation is obtained using a precipitation observation instrument; and temperature, humidity, water vapor pressure, haze, and dust are measured using a six-element automatic weather station.

Images