Convective Potential Monitoring Method and System Based on Infrared Hyperspectral of Fengyun Geostationary Satellite

Abstract

The present invention discloses a method and system for monitoring convective potential based on infrared hyperspectral data of Fengyun geostationary satellites, including: Step 1, obtaining an estimated value of gridded atmospheric temperature and humidity profile data for a target area; Step 2, obtaining a corrected atmospheric stability parameter for the target area; Step 3, determining the potential convective occurrence area of the target area reflected by each atmospheric stability parameter; Step 4, comprehensively evaluating the convective potential of the target area. It can be seen that the present invention can obtain regional atmospheric temperature and humidity profile data and multiple atmospheric stability parameters nearly in real time according to satellite real-time observation data, and finally obtain the regional convective potential result according to the optimal combination, so as to realize the evaluation and early warning of the future convective area. By giving play to the spatio-temporal advantages of satellite observation, it is possible to evaluate future convective occurrences in 45 minutes or more, and achieve monitoring and early warning with a 0.25-degree grid in space, effectively improving the understanding, prediction and early warning of convective processes, and having broad application prospects.

Description

Technical Field

[0001] The present invention relates to a method and system for monitoring convective potential based on infrared hyperspectral of Fengyun geostationary satellite, belonging to the technical field of weather forecasting. Background Art

[0002] The monitoring and early warning of severe convective weather have always been difficult problems in weather forecasting. With the continuous progress of observation means, especially weather radars, it is possible to effectively achieve refined detection of the internal structure, effectively improving the weather monitoring and early warning capabilities. Convective potential forecasting is to analyze the possible occurrence areas, types and intensity of convection according to the weather situation before the occurrence of convection. The method based on diagnosing the convective occurrence conditions established on the basis of the three basic conditions for the occurrence of deep convection, that is, diagnosing the indicative significance of different physical quantities for different severe convective weather and classifying to achieve convective monitoring and early warning, is one of the important methods for potential forecasting. At present, the calculation of convective weather diagnostic physical quantities and the analysis of atmospheric stability mainly rely on various ground-based and upper-air observations, as well as the forecast products of numerical models. This mainly includes calculating observation environment parameters based on sounding data for convective weather forecasting. Based on the construction of the new generation of Doppler operational radar network in China, through the networked observations of multiple cloud radars, dual-polarization radars, phased-array radars, etc., the understanding of the formation mechanism of convective weather in South China, the Yangtze-Huaihe region, North China and other regions in China and its monitoring and early warning capabilities have been significantly improved.

[0003] The advantages of satellite remote sensing, especially the continuous regional observation of geostationary meteorological satellites, have enabled the rapid development of our convective weather monitoring and early warning capabilities. Therefore, based on the analysis of the brightness temperature of the infrared channel of the geostationary satellite spectral imager and its evolution, the effective identification of the initial stage of convection can be achieved, which can be 10-30 minutes earlier than the monitoring of the initial stage of convection based on radar echoes.

[0004] However, through the analysis of the above monitoring means based on radars and satellite spectral imagers, it can be found that there are still great deficiencies in the existing convective monitoring and early warning technologies. The main problems include: 1) The ground-based and sounding observations with limited spatio-temporal resolution limit the identification and early warning of remote areas, over the ocean and rapidly developing local convective systems; 2) The existing methods at present, including radar observations and the monitoring of the initial stage of convection based on satellite spectral imagers, are more suitable for relatively mature convective systems with a long life history. However, there are still large gaps in the discrimination and monitoring of the process from scratch, especially in the early stage of convective development, especially before the initial stage of convection. Summary of the Invention

[0005] In view of the current deficiencies in convective monitoring and assessment, the present invention gives full play to the advantages of the regional continuous detection, hyperspectral, and high-resolution capabilities of the geostationary satellite infrared hyperspectral detector. On the basis of obtaining atmospheric temperature and humidity parameters, an optimal regional convective potential index is constructed to achieve the monitoring and assessment of convective potential, providing technical support for meteorological disaster prevention and mitigation.

[0006] To achieve the above technical objectives, the present invention will adopt the following technical solutions:

[0007] A method for monitoring convective potential based on the infrared hyperspectrum of Fengyun geostationary satellite includes the following steps:

[0008] Step 1: Obtain the estimated values of the gridded atmospheric temperature and humidity profile data for the target area:

[0009] Based on the infrared hyperspectral observation data of the target area and the atmospheric temperature and humidity inversion model, conduct remote sensing of the atmospheric temperature and humidity profile to obtain the estimated values of the gridded atmospheric temperature and humidity profile data for the target area;

[0010] Step 2: Obtain the corrected atmospheric stability parameters for the target area:

[0011] According to the estimated values of the atmospheric temperature and humidity profile data for the target area obtained in Step 1, construct multiple atmospheric stability parameters and correct them according to the atmospheric characteristics of the target area to obtain the corresponding corrected atmospheric stability parameters, so as to meet the analysis requirements of the atmospheric stability parameters in the target area;

[0012] The multiple obtained corrected atmospheric stability parameters include Convective Available Potential Energy (CAPE), Convective Inhibition Energy (CIN), corrected maximum convective stability parameter (mBI), corrected Showalter Index (mSI), and corrected K-index (mK);

[0013] Step 3: Determine the potential convective occurrence areas in the target area reflected by each atmospheric stability parameter:

[0014] According to the multiple corrected atmospheric stability parameters obtained in Step 2, analyze the statistical characteristics of each atmospheric stability parameter to determine the potential convective occurrence areas in the target area reflected by each atmospheric stability parameter;

[0015] Step 4: Comprehensively evaluate the convective potential of the target area:

[0016] According to the potential convective occurrence areas in the target area determined in Step 3, comprehensively evaluate the areas where convection may occur in the future in the target area and output the analysis results of the convective potential assessment.

[0017] Preferably, in Step 3, to determine the potential convective occurrence areas reflected by each atmospheric stability parameter, it specifically includes the following steps:

[0018] Step 301: Calculate the mean value of each atmospheric stability parameter in the target area respectively and the standard deviation ;

[0019] Step 302: Construct the threshold criteria for each atmospheric stability parameter in the target area respectively;

[0020] Based on the mean value and the standard deviation of each atmospheric stability parameter in the target area calculated in Step 301, construct the threshold criteria for the corresponding atmospheric stability parameter;

[0021] There are two types of the above-mentioned threshold criteria, corresponding to the large-value threshold criterion and the small-value threshold criterion , where:

[0022] ;

[0023] ;

[0024] In the above formula: represents the large-value threshold criterion of the th atmospheric stability parameter; represents the mean value of the th atmospheric stability parameter; represents the standard deviation of the th atmospheric stability parameter; represents the small-value threshold criterion of the th atmospheric stability parameter; represents the type of the atmospheric stability parameter, and the value range is [1, 5]. Among them: the first type of atmospheric stability parameter is Convective Available Potential Energy (CAPE), the second type of atmospheric stability parameter is Convective Inhibition Energy (CIN), the third type of atmospheric stability parameter is Modified Bulk Richardson Number (mBI), the fourth type of atmospheric stability parameter is Modified Showalter Index (mSI), and the fifth type of atmospheric stability parameter is Modified K Index (mK);

[0025] Step 303: Mark the grid point types in the area:

[0026] Compare each atmospheric stability parameter corresponding to any grid point in the area with the threshold criteria of the corresponding atmospheric stability parameter constructed in Step 302, and mark the corresponding grid point type according to the comparison result: when the comparison result shows , mark the corresponding grid point as the large-value area of the th atmospheric stability parameter: when the comparison result shows , mark the corresponding grid point The small-value region marked as the th atmospheric stability parameter;

[0027] Step 304: Determine the potential convective occurrence areas reflected by different atmospheric stability parameters:

[0028] The large-value region of the convective available potential energy CAPE is the potential convective occurrence area reflected by the convective available potential energy CAPE, the small-value region of the convective inhibition energy CIN is the potential convective occurrence area reflected by the convective inhibition energy CIN, the large-value region of the modified K-index mK is the potential convective occurrence area reflected by the modified K-index mK; the small-value region of the modified maximum convective stability index mBI is the potential convective occurrence area reflected by the modified maximum convective stability index mBI, and the small-value region of the modified Showalter index mSI is the potential convective occurrence area reflected by the modified Showalter index mSI.

[0029] Preferably, in step four, comprehensively evaluate the convective potential of the target area, which specifically includes the following steps:

[0030] First, determine whether the number of consecutive grid points in the potential convective occurrence area reflected by the convective available potential energy CAPE meets the preset requirements. If the judgment result shows that the number of consecutive grid points in the potential convective occurrence area reflected by the convective available potential energy CAPE meets the preset requirements, then continue to determine whether these grid points are simultaneously the potential convective occurrence areas reflected by the modified maximum convective stability index mBI, the modified Showalter index mSI, and the modified K-index mK. If the judgment result shows that these grid points are simultaneously the potential convective occurrence areas reflected by the modified maximum convective stability index mBI, the modified Showalter index mSI, and the modified K-index mK, then the area corresponding to these grid points can be comprehensively judged as the area where future convection may occur;

[0031] On the contrary, if the judgment result shows that the number of consecutive grid points in the potential convective occurrence area reflected by the convective available potential energy CAPE does not meet the preset requirements, then continue to determine whether these grid points are simultaneously the potential convective occurrence areas reflected by the convective inhibition energy CIN, the modified maximum convective stability index mBI, the modified Showalter index mSI, and the modified K-index mK. If the judgment result shows that these grid points are simultaneously the potential convective occurrence areas reflected by the convective inhibition energy CIN, the modified maximum convective stability index mBI, the modified Showalter index mSI, and the modified K-index mK, then the area corresponding to these grid points can be comprehensively judged as the area where future convection may occur.

[0032] Preferably, in step two, the convective available potential energy CAPE and the convective inhibition energy CIN are calculated respectively by the following formulas:

[0033] ;

[0034] ;

[0035] In the formula: is the free convection height; is the equilibrium height; represents the virtual temperature of the rising air parcel; is the virtual temperature of the environmental air; is the bottom layer of the atmospheric height layer; is the acceleration of gravity, represents the height change;

[0036] When the target area is the eastern and coastal regions of China, the modified maximum convective stability parameter mBI, the modified Showalter index mSI, and the modified K index mK are calculated respectively by the following formulas:

[0037] ;

[0038] ;

[0039] ;

[0040] In the above formula, represents the pseudo-equivalent potential temperature of the atmospheric height layer between 650 - 500 hPa; represents the pseudo-equivalent potential temperature of the atmospheric height layer between 1013 - 850; represents the environmental temperature of the atmospheric height layer at 500 hPa, represents the air parcel temperature of the atmospheric height layer at 500 hPa; represents the temperature of the atmospheric height layer at 850 hPa; represents the dew point temperature of the atmospheric height layer at 850 hPa; represents the temperature of the atmospheric height layer at 500 hPa; represents the temperature of the atmospheric height layer at 700 hPa; represents the dew point temperature of the atmospheric height layer at 700 hPa;

[0041] When the target area is the plateau region, the modified maximum convective stability index mBI, the modified Showalter index mSI, and the modified K index mK are calculated respectively by the following formulas:

[0042] ;

[0043] ;

[0044] ;

[0045] In the above formula, represents the pseudo-equivalent potential temperature of the atmospheric height layer between 500 - 300 hPa; represents the ambient temperature at an atmospheric altitude layer of 250 hPa, represents the temperature of an air parcel at an atmospheric altitude layer of 250 hPa; represents the temperature at an atmospheric altitude layer of 500 hPa; represents the dew point temperature at an atmospheric altitude layer of 500 hPa; represents the temperature at an atmospheric altitude layer of 300 hPa; represents the temperature at an atmospheric altitude layer of 400 hPa; represents the dew point temperature at an atmospheric altitude layer of 400 hPa.

[0046] Preferably, the virtual temperature of the ascending air parcel and the virtual temperature of the ambient air are both calculated by the following formula:

[0047] ;

[0048] where: represents the virtual temperature of any pressure layer; is the actual temperature of the corresponding pressure layer, is the water vapor pressure of the corresponding pressure layer, is the atmospheric pressure of the corresponding pressure layer.

[0049] Preferably, the pseudo-equivalent potential temperature , the pseudo-equivalent potential temperature and the pseudo-equivalent potential temperature are all calculated by the following formula:

[0050] ;

[0051] where: represents the pseudo-equivalent potential temperature of the atmosphere at any pressure layer; represents the temperature of the atmosphere at the corresponding pressure layer, represents the latent heat of water vapor, represents the specific heat capacity of the atmosphere at any pressure layer, represents the specific humidity of the atmosphere at any pressure layer.

[0052] Preferably, in step one, the atmospheric temperature and humidity inversion model is a deep learning-based atmospheric temperature and humidity inversion model. The learning of the atmospheric temperature and humidity inversion model uses a training data set, the hyperparameter tuning of the atmospheric temperature and humidity inversion model uses a validation data set, and the generalization performance evaluation of the atmospheric temperature and humidity inversion model uses a test data set;

[0053] The training data set, the validation data set, and the test data set all include historical temperature and humidity profile observation data and historical infrared hyperspectral observation data that are spatio-temporally matched with the historical temperature and humidity profile observation data;

[0054] The historical infrared hyperspectral observation data from the training data set is processed multiple times by using an atmospheric temperature and humidity inversion model based on deep learning, and the output of the atmospheric temperature and humidity inversion model based on deep learning is compared with the historical temperature and humidity profile observation data in the input training data set to calculate a loss function to train the atmospheric temperature and humidity inversion model based on deep learning.

[0055] Preferably, the training data set, the validation data set, and the test data set are obtained through the following steps:

[0056] Step 101: Extract the historical temperature and humidity profile observation data and infrared hyperspectral observation data of the target area:

[0057] Extract the historical temperature and humidity profile observation data within a preset atmospheric height layer in a historical year from the atmospheric reanalysis data, and vertically divide the preset atmospheric height layer into several pressure layers;

[0058] Extract the infrared hyperspectral observation data in the infrared hyperspectral observation data corresponding to the time of the historical temperature and humidity profile observation data;

[0059] Step 102. Construct the grid data of the target area:

[0060] Match the historical temperature and humidity profile observation data and the full-channel spectral data extracted in Step 101 in space and time to form grid data;

[0061] Step 103. Calculate statistical parameters:

[0062] Calculate the average value and standard deviation of the historical temperature and humidity profile observation data in each grid data at each pressure layer one by one; calculate the average value and standard deviation of the infrared hyperspectral observation data in each grid data one by one;

[0063] Step 104. Standardize the grid data:

[0064] Based on the average value and standard deviation of each historical temperature and humidity profile observation data and the average value and standard deviation of each infrared hyperspectral observation data calculated in Step 103, standardize the historical temperature and humidity profile observation data and the infrared hyperspectral observation data in the corresponding grid data to obtain standardized grid data;

[0065] Step 105. Construct the training data set, the validation data set, and the test data set;

[0066] After sorting the grid data obtained in Step 104 in chronological order, divide it into three parts, one part is the training data set, one part is the validation data set, and the other part is the test data set.

[0067] Preferably, in step 104, the normalization process of the grid data specifically includes: subtracting the average value of each data in the grid data and then dividing by the corresponding standard deviation.

[0068] Another technical object of the present invention is to provide a convective potential monitoring system based on the infrared hyperspectral of Fengyun geostationary satellite, which includes being programmed to execute the above-mentioned convective potential monitoring method based on the infrared hyperspectral of Fengyun geostationary satellite.

[0069] Based on the above technical objects, compared with the prior art, the present invention has the following advantages:

[0070] The present invention provides a convective potential monitoring method and system based on the infrared hyperspectral of Fengyun geostationary satellite, which can obtain regional atmospheric temperature and humidity profile data and multiple atmospheric stability parameters in near real-time according to satellite real-time observation data, and finally obtain the regional convective potential result according to the optimal combination, so as to realize the evaluation and early warning of the future convective area. By giving full play to the spatio-temporal advantages of satellite observation, it is possible to evaluate the future convection occurrence within 45 minutes or more, and realize the monitoring and early warning with a 0.25-degree grid in space, effectively improving the understanding, prediction and early warning of the convective process, and having a wide range of application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a flowchart of the method for the convective potential monitoring method and system based on the infrared hyperspectral of Fengyun geostationary satellite described in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0072] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The description of at least one exemplary embodiment is actually only illustrative and in no way restrictive of the present invention and its application or use. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Unless otherwise specifically stated, the relative arrangements, expressions and numerical values of the components and steps set forth in these embodiments do not limit the scope of the present invention. Technologies, methods and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods and devices should be regarded as part of the specification. In all the examples shown and discussed here, any specific value should be interpreted as merely exemplary and not as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0073] The convective potential monitoring method based on the infrared hyperspectral of Fengyun geostationary satellite described in the present invention, as Figure 1As shown, it includes the following steps:

[0074] Step 1: Obtain the estimated gridded atmospheric temperature and humidity profile data of the target area:

[0075] Based on the infrared hyperspectral observation data of the target area and the atmospheric temperature and humidity inversion model, conduct remote sensing of the atmospheric temperature and humidity profile to obtain the estimated gridded atmospheric temperature and humidity profile data of the target area.

[0076] In the present invention, the adopted atmospheric temperature and humidity inversion model is a deep learning-based atmospheric temperature and humidity inversion model (U-Net model). The U-Net model is a convolutional neural network widely used in the field of image processing. Through the architecture of convolution, pooling, and decoding, this model extracts multi-scale features layer by layer from the input data. First, the convolutional layer is used to extract spatial features, then the pooling layer is used to reduce the dimension and capture global information, and finally the decoder is used to restore the original resolution. During the training process, the standardized infrared hyperspectral observations of the geostationary meteorological satellite are used as the input, and through continuous adjustment, the prediction of the atmospheric temperature and humidity profile is optimized. The training objective of the model is to minimize the error between the output and the actual observed values to ensure that the model can accurately invert the atmospheric temperature and humidity conditions.

[0077] The training of the atmospheric temperature and humidity inversion model uses the training dataset, the hyperparameter tuning of the atmospheric temperature and humidity inversion model uses the validation dataset, and the generalization performance evaluation of the atmospheric temperature and humidity inversion model uses the test dataset. The training dataset, validation dataset, and test dataset all include historical temperature and humidity profile observation data and historical infrared hyperspectral observation data that are spatio-temporally matched with the historical temperature and humidity profile observation data.

[0078] By using the deep learning-based atmospheric temperature and humidity inversion model to process the historical infrared hyperspectral observation data from the training dataset multiple times, and comparing the output of the deep learning-based atmospheric temperature and humidity inversion model with the historical temperature and humidity profile observation data in the input training dataset, the deep learning-based atmospheric temperature and humidity inversion model is trained by calculating the loss function.

[0079] At the end of each training cycle, use the validation set data to evaluate the atmospheric temperature and humidity inversion model, calculate the mean absolute error (MAE) on each channel, and record these metrics to ensure that the atmospheric temperature and humidity inversion model does not overfit. Finally, select the model weights with the best performance on the validation set as the final model. Use the test dataset for testing to verify the inversion results of the atmospheric temperature and humidity profile.

[0080] In the present invention, the training dataset, validation dataset, and test dataset are obtained through the following steps:

[0081] Step 101: Extract the historical temperature and humidity profile observation data and infrared hyperspectral observation data of the target area:

[0082] Extract historical temperature and humidity profile observation data within a preset atmospheric height layer from atmospheric reanalysis data; in the present invention, when extracting atmospheric reanalysis data, a total of 98 pressure layers from 1013 hPa to 0.05 hPa are selected, and the atmospheric temperature (T) and atmospheric humidity (relative humidity RH, absolute humidity Q) in the fifth-generation atmospheric reanalysis data (ERA5) of the European Centre for Medium-Range Weather Forecasts from 2019 to 2022 are extracted, with a spatial resolution of 0.25 degrees × 0.25 degrees. This data is used to train a deep learning atmospheric temperature and humidity inversion model to help the deep learning atmospheric temperature and humidity inversion model learn the laws of atmospheric temperature and humidity changes;

[0083] Extract infrared hyperspectral observation data from the infrared hyperspectral observation data corresponding to the time of the historical temperature and humidity profile observation data; specifically, obtain the infrared hyperspectral observation data of geostationary satellites from 2019 to 2022, with a spatial resolution of 0.25 degrees × 0.25 degrees. The infrared hyperspectral observation data provides multi-channel brightness temperature information and can reflect the temperature and humidity structure of the atmosphere.

[0084] Step 102. Construct grid data for the target area:

[0085] Perform spatio-temporal matching on the historical temperature and humidity profile observation data and the historical infrared hyperspectral observation data extracted in Step 101 to form grid data;

[0086] Step 103. Calculate statistical parameters:

[0087] Calculate the mean and standard deviation of the historical temperature and humidity profile observation data in each grid data at each atmospheric height layer one by one; calculate the mean and standard deviation of the infrared hyperspectral observation data in each grid data one by one;

[0088] Step 104. Standardize the grid data:

[0089] Based on the mean and standard deviation of each historical temperature and humidity profile observation data and the mean and standard deviation of each infrared hyperspectral observation data calculated in Step 103, standardize the historical temperature and humidity profile observation data and the infrared hyperspectral observation data in the corresponding grid data to obtain standardized grid data; specifically, standardize the atmospheric temperature, atmospheric humidity, and infrared hyperspectral observation data corresponding to each pressure layer in ERA5, that is, subtract the mean from each sample and then divide by the corresponding standard deviation. This is aimed at eliminating the scale differences between different data, ensuring that the data is distributed within a consistent range, and improving the stability of model training.

[0090] Step 105. Construct a training dataset, a validation dataset, and a test dataset;

[0091] After sorting the grid data obtained in step 104 in chronological order, it is divided into three parts, one part is the training data set, one part is the validation data set, and the other part is the test data set. By means of time segmentation, it is ensured that the atmospheric temperature and humidity inversion model can correctly capture the spatio-temporal characteristics of the atmospheric temperature and humidity profiles and avoid data leakage in time.

[0092] The final model is applied to the near-real-time infrared hyperspectral observations of geostationary meteorological satellites, and near-real-time regional atmospheric temperature and humidity profile data are output. The spatial horizontal resolution of the output data is 0.25 degrees, and the vertical direction is divided into 98 pressure layers, from 1013 hPa to 0.05 hPa.

[0093] Step 2: Obtain the corrected atmospheric stability parameters of the target area:

[0094] Based on the estimated values of the atmospheric temperature and humidity profile data of the target area obtained in step 1, multiple atmospheric stability parameters are constructed and corrected according to the atmospheric characteristics of the target area to obtain the corresponding corrected atmospheric stability parameters, so as to meet the needs of analyzing the atmospheric stability parameters of the target area; the multiple corrected atmospheric stability parameters obtained include Convective Available Potential Energy (CAPE), Convective Inhibition Energy (CIN), corrected maximum convective stability parameter (mBI), corrected Showalter Index (mSI), and corrected K Index (mK).

[0095] The Convective Available Potential Energy (CAPE) and Convective Inhibition Energy (CIN) are calculated respectively by the following formulas:

[0096] ;

[0097] ;

[0098] In the formulas: is the level of free convection; is the equilibrium level; represents the virtual temperature of the rising air parcel; is the virtual temperature of the environmental air; is the bottom layer of the atmospheric height layer; is the acceleration due to gravity, represents the change in height;

[0099] In the calculation formulas of the Convective Available Potential Energy (CAPE) and Convective Inhibition Energy (CIN), the virtual temperature of the rising air parcel and the virtual temperature of the environmental air are both calculated by the following formula:

[0100] ;

[0101] Where: represents the virtual temperature of any pressure layer; is the actual temperature of the corresponding pressure layer, is the water vapor pressure of the corresponding pressure layer, is the atmospheric pressure of the corresponding pressure layer.

[0102] When the target area is the eastern and coastal regions of China, the corrected maximum convective stability parameter mBI, the corrected Showalter index mSI, and the corrected K index mK are calculated respectively by the following formulas:

[0103] ;

[0104] ;

[0105] ;

[0106] In the above formulas, represents the pseudo-equivalent potential temperature when the atmospheric height layer is between 650 - 500 hPa; represents the pseudo-equivalent potential temperature when the atmospheric height layer is between 1013 - 850; represents the environmental temperature when the atmospheric height layer is 500 hPa, represents the parcel temperature when the atmospheric height layer is 500 hPa; represents the temperature when the atmospheric height layer is 850 hPa; represents the dew point temperature when the atmospheric height layer is 850 hPa; represents the temperature when the atmospheric height layer is 500 hPa; represents the temperature when the atmospheric height layer is 700 hPa; represents the dew point temperature when the atmospheric height layer is 700 hPa;

[0107] In the calculation formula of the corrected K index mK, represents the low-level and high-level stratification stability, is the low-level humidity, is the mid-level humidity. The larger the corrected K index, the more unstable the atmosphere is.

[0108] When the target area is the plateau region, the corrected maximum convective stability index mBI, the corrected Showalter index mSI, and the corrected K index mK are calculated respectively by the following formulas:

[0109] ;

[0110] ;

[0111] ;

[0112] In the above formulas, represents the pseudo-equivalent potential temperature when the atmospheric height layer is between 500 - 300 hPa; represents the environmental temperature at an atmospheric altitude layer of 250 hPa, represents the temperature of an air parcel at an atmospheric altitude layer of 250 hPa; represents the temperature at an atmospheric altitude layer of 500 hPa; represents the dew point temperature at an atmospheric altitude layer of 500 hPa; represents the temperature at an atmospheric altitude layer of 300 hPa; represents the temperature at an atmospheric altitude layer of 400 hPa; represents the dew point temperature at an atmospheric altitude layer of 400 hPa.

[0113] In the calculation formulas of the corrected maximum convective stability parameter mBI, the corrected Showalter index mSI, and the corrected K index mK, the pseudo-equivalent potential temperature , the pseudo-equivalent potential temperature , and the pseudo-equivalent potential temperature are all calculated by the following formula:

[0114] ;

[0115] where: represents the pseudo-equivalent potential temperature of the atmosphere at any pressure layer; represents the temperature of the atmosphere at the corresponding pressure layer, represents the latent heat of water vapor, represents the specific heat capacity of the atmosphere at any pressure layer, represents the specific humidity of the atmosphere at any pressure layer.

[0116] The stable change of the air parcel follows that the unsaturated moist air parcel first changes according to the dry adiabatic process. That is, in this process, the water vapor pressure in the air parcel decreases as the environmental pressure decreases, so the dew point also decreases. When the temperature and dew point of the air parcel are equal at a certain height, saturation will occur, and the height where it occurs is the lifting condensation height. After that, it changes according to the wet adiabatic process. Thus, the temperature of the air parcel can be jointly calculated from the environmental pressure (determined by the given pressure layer in the atmospheric altitude layer), the environmental temperature at the bottom layer of the atmospheric altitude layer (i.e., the above-mentioned layer), and the dew point temperature.

[0117] Step 3. Determine the potential convective occurrence area of the target area reflected by each atmospheric stability parameter:

[0118] According to the multiple corrected atmospheric stability parameters obtained in Step 2, analyze the statistical characteristics of each atmospheric stability parameter, and determine the potential convective occurrence area of the target area reflected by each atmospheric stability parameter; specifically, it includes the following steps:

[0119] Step 301. Calculate the mean value and the standard deviation ; The specific calculation method is as follows:

[0120] ;

[0121] ;

[0122] Among them is the mean value of the total grid point index of the corresponding atmospheric stability at the corresponding moment, is the corresponding standard deviation, and N is the total amount of grid point data. The standard deviation reflects the discrete situation of the atmospheric stability parameters in the area at this moment.

[0123] Step 302: Construct the threshold criteria for each atmospheric stability parameter in the target area respectively;

[0124] Based on the mean values and standard deviations of each atmospheric stability parameter in the target area calculated in Step 301, according to the historical observation convective sample data, construct the threshold criteria for each atmospheric stability parameter;

[0125] ;

[0126] Among them and are the above-mentioned mean value and standard deviation, and x and y are given coefficients. For each atmospheric stability parameter, compare and test the consistency between the indication significance of each atmospheric stability parameter for severe convective weather and the actual observation results under different threshold schemes, and finally determine the values of coefficients x and y to obtain the comprehensive threshold result.

[0127] There are two threshold criteria, corresponding to the large-value threshold criterion and the small-value threshold criterion , where:

[0128] ;

[0129] ;

[0130] In the above formula: represents the large-value threshold criterion of the th atmospheric stability parameter; represents the mean value of the th atmospheric stability parameter; represents the standard deviation of the th atmospheric stability parameter; represents the small-value threshold criterion of the th atmospheric stability parameter; Indicates the types of atmospheric stability parameters, with values in the range of [1, 5]. Among them: the first type of atmospheric stability parameter is Convective Available Potential Energy (CAPE), the second type is Convective Inhibition Energy (CIN), the third type is Modified Bulk Richardson Index (mBI), the fourth type is Modified Showalter Index (mSI), and the fifth type is Modified K - index (mK).

[0131] Step 303: Marking the grid - point types within the region

[0132] For any grid - point within the region Compare each corresponding atmospheric stability parameter with the threshold criteria of the corresponding atmospheric stability parameters constructed in Step 302, and mark the corresponding grid - point types according to the comparison results: When the comparison result shows then mark the corresponding grid - point as the large - value region of the th type of atmospheric stability parameter; when the comparison result shows then mark the corresponding grid - point as the small - value region of the th type of atmospheric stability parameter;

[0133] Step 304: Determine the potential convective occurrence areas reflected by different atmospheric stability parameters

[0134] The large - value region of Convective Available Potential Energy (CAPE) is the potential convective occurrence area reflected by CAPE, the small - value region of Convective Inhibition Energy (CIN) is the potential convective occurrence area reflected by CIN, the large - value region of Modified K - index (mK) is the potential convective occurrence area reflected by mK; the small - value region of Modified Bulk Richardson Index (mBI) is the potential convective occurrence area reflected by mBI, and the small - value region of Modified Showalter Index (mSI) is the potential convective occurrence area reflected by mSI.

[0135] Step Four: Comprehensively evaluate the convective potential of the target region

[0136] Based on the potential convective occurrence areas reflected by each atmospheric stability parameter determined in Step Three, comprehensively evaluate the areas where convection may occur in the future in the target region, and output the analysis results of the convective potential assessment.

[0137] Specifically, first determine whether the number of consecutive grid points in the potential convective occurrence area reflected by the convective available potential energy (CAPE) meets the preset requirements (for example, the number of consecutive grid points can be 8). If the judgment result shows that the number of consecutive grid points in the potential convective occurrence area reflected by the CAPE meets the preset requirements, then continue to determine whether these grid points are simultaneously the potential convective occurrence areas reflected by the modified maximum convective stability index (mBI), the modified Showalter index (mSI), and the modified K index (mK). If the judgment result shows that these grid points are simultaneously the potential convective occurrence areas reflected by the mBI, mSI, and mK, the area corresponding to these grid points can be comprehensively judged as the area where future convection may occur; mark all potential convective occurrence areas as 1, otherwise mark them as 0.

[0138] Conversely, if the judgment result shows that the number of consecutive grid points in the potential convective occurrence area reflected by the CAPE does not meet the preset requirements, then continue to determine whether these grid points are simultaneously the potential convective occurrence areas reflected by the convective inhibition energy (CIN), the mBI, the mSI, and the mK. If the judgment result shows that these grid points are simultaneously the potential convective occurrence areas reflected by the CIN, mBI, mSI, and mK, the area corresponding to these grid points can be comprehensively judged as the area where future convection may occur.

[0139] Based on the above results, give the convective potential assessment and analysis area, and perform data marking and image display.

[0140] It can be seen that the convective potential monitoring and assessment method described in the present invention can effectively utilize the advantages of geostationary satellite infrared hyperspectral detection, obtain the convective potential results, make up for the lack of convective monitoring ability in the areas lacking marine and some ground-based observation data in China, and provide technical support for improving the defense ability of regional severe weather.

[0141] Application Example

[0142] In the present invention, taking the infrared hyperspectral observation data of Fengyun geostationary satellite as the main input, convert the radiation parameters therein into brightness temperature parameters, and the formula used is the Planck function , where represents the brightness temperature, represents the radiation intensity, is the wave number corresponding to the band, is the speed of light, with a value of 3*10 8 m / s, is the Planck constant, with a value of 6.626*10 -34 J*S, is the Boltzmann constant, with a value of 1.3806*10 -23 J / K. At the same time, the spatio-temporal information of the target area is obtained for sub-region implementation.

[0143] In the present invention, according to the pre-established atmospheric temperature and humidity retrieval model based on deep learning algorithm and applicable to geostationary satellite infrared hyperspectral, the radiation parameters in the infrared hyperspectral observation data of Fengyun geostationary satellite are directly input into the atmospheric temperature and humidity retrieval model to obtain the gridded estimated values of atmospheric temperature and humidity profiles in the region, with a spatial horizontal resolution of 0.25 degrees and altitude levels from 1013 hPa to 0.05 hPa.

[0144] Calculate all atmospheric stability parameters.

[0145] Calculate the mean and standard deviation of all atmospheric stability parameters, screen out the large-value and small-value regions of all stability parameters, further make an optimal combination, and finally mark the regions where future convection may occur to obtain the convective potential index. Among them: the regions where future convection may occur are marked as "1".

[0146] Based on the method and system of the present invention, it is possible to better predict in advance the regions where future convection may occur, which is of great significance for improving the monitoring and early warning of severe convective weather.

Claims

1. A method for monitoring convective potential based on the infrared hyperspectral spectrum of Fengyun geostationary satellites, characterized in that: The steps include: Step 1: Obtain the estimated values ​​of the gridded atmospheric temperature and humidity profile data of the target area: According to the infrared hyperspectral observation data of the target area, based on the atmospheric temperature and humidity inversion model, remote sensing of atmospheric temperature and humidity profiles is carried out to obtain the estimated values ​​of gridded atmospheric temperature and humidity profile data of the target area; Step 2: Obtain the corrected atmospheric stability parameters of the target area: According to the estimated values ​​of atmospheric temperature and humidity profile data of the target area obtained in step 1, multiple atmospheric stability parameters are constructed, and they are corrected according to the atmospheric characteristics of the target area to obtain corresponding corrected atmospheric stability parameters, thereby meeting the needs of atmospheric stability parameter analysis in the target area; The obtained multiple corrected atmospheric stability parameters include convective effective potential energy CAPE, convective inhibition energy CIN, corrected maximum convective stability parameter mBI, corrected Sha's index mSI and corrected K index mK; Step 3: Determine the potential convection occurrence area in the target area reflected by each atmospheric stability parameter: According to the multiple corrected atmospheric stability parameters obtained in step 2, the statistical characteristics of each atmospheric stability parameter are analyzed to determine the potential convection occurrence area of ​​the target area reflected by each atmospheric stability parameter; Step 4: Comprehensively evaluate the convection potential of the target area: Based on the potential convection occurrence areas reflected by the atmospheric stability parameters determined in step 3, comprehensively evaluate the areas where future convection may occur in the target area, and output the convection potential assessment analysis results; In step 3, the potential convection occurrence area reflected by each atmospheric stability parameter is determined, which specifically includes the following steps: Step 301: Calculate the mean values ​​of the atmospheric stability parameters in the target area respectively. and standard deviation ; Step 302, respectively constructing threshold standards for each atmospheric stability parameter in the target area; Based on the average value of each atmospheric stability parameter in the target area calculated in step 301 and standard deviation , construct threshold standards for corresponding atmospheric stability parameters; There are two threshold standards, corresponding to the large value threshold standard And the small value threshold standard ,in: ； ； In the above formula: Indicates The maximum threshold standard of atmospheric stability parameters; Indicates The mean value of the atmospheric stability parameters; Indicates Standard deviation of the atmospheric stability parameters; Indicates Small threshold standards for atmospheric stability parameters; Indicates the type of atmospheric stability parameter, with a value of [1,5], where: the first atmospheric stability parameter is the convective effective potential energy CAPE, the second atmospheric stability parameter is the convective inhibition energy CIN, the third atmospheric stability parameter is the modified maximum convective stability parameter mBI, the fourth atmospheric stability parameter is the modified Sha's index mSI, and the fifth atmospheric stability parameter is the modified K index mK; Step 303: Marking the grid point type in the region: Any grid point in the region Corresponding atmospheric stability parameters are compared with the threshold standard of the corresponding atmospheric stability parameter constructed in step 302, and the corresponding grid point type is marked according to the comparison result: When Marked as The large value area of ​​atmospheric stability parameters: When the comparison results show When Marked as The area with small values ​​of various atmospheric stability parameters; Step 304: determine potential convection occurrence areas reflected by different atmospheric stability parameters: The large value area of ​​convective effective potential energy CAPE is the potential convection occurrence area reflected by convective effective potential energy CAPE, the small value area of ​​convective inhibition energy CIN is the potential convection occurrence area reflected by convective inhibition energy CIN, the large value area of ​​modified K index mK is the potential convection occurrence area reflected by modified K index mK; the small value area of ​​modified maximum convection stability index mBI is the potential convection occurrence area reflected by modified maximum convection stability index mBI, the small value area of ​​modified Sha's index mSI is the potential convection occurrence area reflected by modified Sha's index mSI; In step 4, the convection potential of the target area is comprehensively evaluated, which includes the following steps: First, determine whether the number of consecutive grid points in the potential convection occurrence area reflected by the convection effective potential energy CAPE meets the preset requirements. If the judgment result shows that the number of consecutive grid points in the potential convection occurrence area reflected by the convection effective potential energy CAPE meets the preset requirements, continue to determine whether these grid points are potential convection occurrence areas reflected by the modified maximum convection stability index mBI, the modified Sha's index mSI and the modified K index mK at the same time. If the judgment result shows that these grid points are potential convection occurrence areas reflected by the modified maximum convection stability index mBI, the modified Sha's index mSI and the modified K index mK at the same time, the areas corresponding to these grid points can be comprehensively determined as areas where future convection may occur. On the contrary, if the judgment result shows that the number of continuous grid points in the potential convection occurrence area reflected by the convection effective potential energy CAPE does not meet the preset requirements, it is then continued to be judged whether these grid points are potential convection occurrence areas reflected by the convection inhibition energy CIN, the modified maximum convection stability index mBI, the modified Sachs index mSI and the modified K index mK at the same time. If the judgment result shows that these grid points are potential convection occurrence areas reflected by the convection inhibition energy CIN, the modified maximum convection stability index mBI, the modified Sachs index mSI and the modified K index mK at the same time, the areas corresponding to these grid points can be comprehensively judged as areas where future convection may occur.

2. The method for monitoring convective potential based on the infrared hyperspectral spectrum of a Fengyun geostationary satellite according to claim 1 is characterized in that: In step 2, the convection effective potential energy CAPE and the convection inhibition energy CIN are calculated by the following formulas: ； ； Where: is the free convection height; To balance the height; Indicates the virtual temperature of the rising air mass; It is the virtual temperature of the ambient air; The bottom layer of the atmospheric altitude. is the acceleration due to gravity, Indicates a change in altitude; When the target area is the eastern and coastal areas of China, the modified maximum convective stability parameter mBI, the modified Sha's index mSI and the modified K index mK are calculated by the following formulas: ； ； ； In the above formula, Indicates the pseudo-equivalent potential temperature of the atmospheric layer between 650-500 hPa; Indicates the pseudo-equivalent potential temperature of the atmospheric layer between 1013-850°; Indicates the ambient temperature at an atmospheric altitude of 500 hPa. Indicates the temperature of the air parcel at an atmospheric altitude of 500 hPa; Indicates the temperature at the atmospheric altitude layer of 850 hPa; Indicates the dew point temperature at the atmospheric altitude layer of 850hPa; Indicates the temperature at the atmospheric altitude layer of 500 hPa; Indicates the temperature at the atmospheric altitude layer of 700 hPa; Indicates the dew point temperature at an atmospheric altitude of 700 hPa; When the target area is a plateau, the modified maximum convective stability index mBI, the modified Sha's index mSI and the modified K index mK are calculated by the following formulas: ； ； ； In the above formula, Indicates the pseudo-equivalent potential temperature of the atmospheric layer between 500-300 hPa; Indicates the ambient temperature at the atmospheric altitude layer of 250 hPa, Indicates the temperature of the air parcel at the atmospheric altitude layer of 250 hPa; Indicates the temperature at the atmospheric altitude layer of 500hPa; Indicates the dew point temperature at an atmospheric altitude of 500 hPa; Indicates the temperature at the atmospheric altitude layer of 300 hPa; Indicates the temperature at the atmospheric altitude layer of 400 hPa; Indicates the dew point temperature at an atmospheric altitude of 400 hPa.

3. The method for monitoring convective potential based on the infrared hyperspectral spectrum of a Fengyun geostationary satellite according to claim 2 is characterized in that: Virtual temperature of rising air mass , the virtual temperature of the ambient air All are calculated by the following formula: ； in: represents the virtual temperature of any pressure layer; is the actual temperature of the corresponding pressure layer, is the water vapor pressure of the corresponding pressure layer, is the atmospheric pressure of the corresponding pressure layer.

4. The method for monitoring convective potential based on the infrared hyperspectral spectrum of a Fengyun geostationary satellite according to claim 2 is characterized in that: Pseudo-equivalent temperature , hypothetical equivalent temperature and pseudo-equivalent potential temperature All are calculated by the following formula: ； in: represents the pseudo equivalent potential temperature of the atmosphere at any pressure layer; represents the temperature of the atmosphere in the corresponding pressure layer, is the latent heat of water vapor, represents the specific heat capacity of the atmosphere at any pressure layer, It represents the specific humidity of the atmosphere at any pressure layer.

5. The method for monitoring convective potential based on the FY geostationary satellite infrared hyperspectral spectrum according to claim 1 is characterized in that: In step 1, the atmospheric temperature and humidity inversion model is a deep learning atmospheric temperature and humidity inversion model, the atmospheric temperature and humidity inversion model is learned using a training data set, the hyperparameter tuning of the atmospheric temperature and humidity inversion model is performed using a validation data set, and the generalization performance evaluation of the atmospheric temperature and humidity inversion model is performed using a test data set; The training data set, validation data set, and test data set all include historical temperature and humidity profile observation data and historical infrared hyperspectral observation data that are temporally and spatially matched with the historical temperature and humidity profile observation data; The deep learning atmospheric temperature and humidity inversion model is used to process the historical infrared hyperspectral observation data from the training data set multiple times, and the output of the deep learning atmospheric temperature and humidity inversion model is compared with the historical temperature and humidity profile observation data in the input training data set to calculate the loss function to train the deep learning atmospheric temperature and humidity inversion model.

6. The method for monitoring convective potential based on the infrared hyperspectral spectrum of a Fengyun geostationary satellite according to claim 5 is characterized in that: The training data set is obtained through the following steps: Step 101: extract historical temperature and humidity profile observation data and infrared hyperspectral observation data of the target area: Extract historical temperature and humidity profile observation data within the preset atmospheric height layer in the historical year from the atmospheric reanalysis data, and divide the preset atmospheric height layer vertically into several pressure layers; Extract infrared hyperspectral observation data from infrared hyperspectral observation data corresponding to the time of historical temperature and humidity profile observation data; Step 102. Construct the grid data of the target area: Performing spatiotemporal matching on the historical temperature and humidity profile observation data and the full-channel spectral data extracted in step 101 to form grid data; Step 103. Calculate statistical parameters: Calculate the mean and standard deviation of the historical temperature and humidity profile observation data in each pressure layer one by one in each grid point data; calculate the mean and standard deviation of the infrared hyperspectral observation data in each grid point data one by one; Step 104. Standardization of grid data: Based on the average value and standard deviation of each historical temperature and humidity profile observation data and the average value and standard deviation of each infrared hyperspectral observation data calculated in step 103, the historical temperature and humidity profile observation data and the infrared hyperspectral observation data in the corresponding grid data are standardized to obtain standardized grid data; Step 105. Construct a training data set, a validation data set, and a test data set; After the grid point data obtained in step 104 are sorted in chronological order, they are divided into three parts, one of which is a training data set, one is a verification data set, and the other is a test data set.

7. The method for monitoring convective potential based on the infrared hyperspectral spectrum of a Fengyun geostationary satellite according to claim 6 is characterized in that: In step 104, the grid point data is standardized, specifically including: subtracting the average value from each data point in the grid point data, and then dividing by the corresponding standard deviation.

8. A convective potential monitoring system based on the infrared hyperspectral spectrum of Fengyun geostationary satellites, characterized in that: The convective potential monitoring system includes a system programmed to execute the convective potential monitoring method based on the infrared hyperspectrum of the Fengyun geostationary satellite as described in any one of claims 1-7.

Images