A method for coarse and fine mode aerosol parametric inversion based on high spectral resolution lidar
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG UNIV
- Filing Date
- 2026-04-21
- Publication Date
- 2026-06-30
Smart Images
Figure CN122085303B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of atmospheric aerosol remote sensing lidar technology, and in particular to a method for coarse and fine mode aerosol parameter inversion based on high spectral resolution lidar. Background Technology
[0002] Aerosols are a core component of the Earth's atmospheric system. Their coarse and fine modes, classified by particle size, exhibit significant differences in atmospheric radiative forcing, cloud and precipitation formation, regional air quality evolution, and impacts on human health. Accurately acquiring the optical parameters of coarse and fine aerosol modes is a key foundation for quantifying aerosol-climate interactions and conducting atmospheric environmental monitoring and assessment, and has become an important research direction in the field of atmospheric remote sensing.
[0003] Currently, the mainstream methods for detecting coarse and fine modal aerosol parameters are mainly passive remote sensing technologies such as solar photometers and multispectral remote sensing satellites. LiDAR, as an active remote sensing device, is also being gradually applied in this field. Passive remote sensing technology can achieve whole-layer observation of aerosol optical thickness, while lidar, with its active detection characteristics, has the ability to observe vertical profiles and can preliminarily distinguish the types of coarse and fine modal particles by the degree of aerosol depolarization.
[0004] However, passive remote sensing technology can only obtain the optical thickness of the entire atmospheric column integral, and cannot provide vertical stratification information. Furthermore, it is limited by lighting and cloud conditions, and observations are interrupted during dawn, dusk, cloudy days, and nighttime, making continuous monitoring difficult. Lidar-based inversion methods also have significant shortcomings. Traditional methods based on Mie scattering rely on lidar ratio assumptions, resulting in significant spatiotemporal uncertainties and errors in the inversion results. While Raman scattering-based methods avoid this assumption, the weak signal leads to extremely low signal-to-noise ratios during daytime detections, making simultaneous verification with passive remote sensing observations difficult, resulting in shortcomings in consistency assessment. In related technologies, patents CN108897903A and CN110705089B disclose fine-modal optical thickness inversion methods based on MODIS and POLDER sensors. While such methods provide a way to obtain fine modal parameters under passive remote sensing, how to overcome the limitations of existing inversion methods in terms of accuracy, continuity and vertical resolution based on lidar observation, and achieve all-weather, high-reliability quantitative separation of coarse and fine modal aerosol optical parameters, remains a key technical challenge that urgently needs to be overcome. Summary of the Invention
[0005] In view of this, this application provides a method for coarse and fine mode aerosol parameter inversion based on high spectral resolution lidar, for implementation.
[0006] Specifically, this application is implemented through the following technical solution:
[0007] The first aspect of this application provides a method for coarse and fine mode aerosol parameter inversion based on high spectral resolution lidar, the method comprising:
[0008] Acquire the main channel scattering signal, depolarization channel elastic scattering signal, and molecular Rayleigh scattering signal of aerosols;
[0009] The volume depolarization ratio is calculated based on the ratio of the main channel scattering signal to the depolarization channel elastic scattering signal. The total backscattering coefficient is calculated based on the volume depolarization ratio and the molecular Rayleigh scattering signal, and the ratio to the lidar is calculated. No lidar ratio assumptions are required, and there are no day or night time restrictions.
[0010] The aerosol depolarization ratio is calculated based on the volume depolarization ratio and the total backscattering coefficient.
[0011] Based on the clustering results of historical data, the characteristic depolarization ratio constants of coarse mode and fine mode are obtained respectively. Based on the characteristic depolarization ratio constants, an aerosol multi-component mixing model is constructed. The aerosol depolarization ratio is input into the multi-component mixing model to obtain the coarse mode backscattering coefficient and the fine mode backscattering coefficient.
[0012] The lidar ratios for coarse and fine modes are obtained based on the clustering results of historical data. The coarse mode extinction coefficient and the fine mode extinction coefficient are calculated based on the coarse mode backscattering coefficient and the lidar ratio of the fine mode backscattering coefficient.
[0013] The total backscattering coefficient is calculated based on the coarse-mode backscattering coefficient and the fine-mode backscattering coefficient. The vertical range of the aerosol layer is determined based on the total backscattering coefficient. The coarse-mode extinction coefficient and the fine-mode extinction coefficient are vertically integrated within the height of the aerosol layer to obtain the coarse-mode optical thickness and the fine-mode optical thickness. The inversion is performed based on the coarse-mode optical thickness and the fine-mode optical thickness.
[0014] To address the shortcomings of existing lidar methods for retrieving coarse and fine-mode aerosols, this invention provides a method for coarse and fine-mode aerosol retrieval based on high-spectral-resolution lidar (HSRL). Compared to the Klett-Fernald method based on Mie scattering, HSRL uses a spectral discriminator to separate aerosol Mie scattering signals and molecular Rayleigh scattering signals, eliminating the need for lidar ratio assumptions and enabling real-time acquisition of aerosol optical properties. Compared to Raman scattering lidar, HSRL separates Rayleigh scattering signals with a stronger cross-section, resulting in a higher daytime detection signal-to-noise ratio. This invention establishes a multi-component mixing model and uses the aerosol depolarization ratio retrieved by HSRL to separate components, achieving all-day, real-time, and accurate retrieval of optical parameters for coarse and fine-mode aerosols.
[0015] 1. This invention proposes a coarse and fine mode aerosol parameter inversion method based on the HSRL system. By using a polarization separation coarse and fine mode estimation scheme, it solves the problems of passive nighttime observation interruption and low signal-to-noise ratio during the day, and realizes 24-hour continuous and accurate inversion of optical parameters such as coarse mode optical thickness / fine mode optical thickness (cAOD / fAOD).
[0016] 2. This invention uses the HSRL system to perform stratified integration of coarse and fine mode extinction coefficients, which can not only obtain the cAOD / fAOD of the entire atmosphere, but also reflect the distribution differences of different altitude layers (such as the boundary layer and the upper-level suspended layer), providing support for the monitoring of aerosol transport and stratified pollution characteristics.
[0017] Specifically, by acquiring multi-channel signals and calculating the volume depolarization ratio, total backscattering coefficient, and lidar ratio stepwise, this method eliminates the pre-defined assumptions about the lidar ratio and removes the spatiotemporal uncertainties and systematic errors inherent in traditional methods. By eliminating atmospheric molecular interference to calculate the aerosol depolarization ratio and constructing a multi-component mixing model using historical data clustering of typical values, it achieves accurate quantitative separation of coarse and fine modal backscattering coefficients. This solves the problem that traditional lidar can only qualitatively distinguish modes. Through the product of the modal lidar ratio and the backscattering coefficient, it efficiently completes the separation of coarse and fine modal backscattering coefficients. By separating and calculating the modal extinction coefficients, and then precisely defining the vertical range of the aerosol layer and performing layered vertical integration of the extinction coefficients, the coarse and fine modal optical thicknesses of each layer and the entire layer are obtained, thus completing the inversion. This not only achieves precise vertical layering of aerosol parameters, but also overcomes the limitations of traditional passive remote sensing based on illumination and cloud cover, enabling continuous observation throughout the day. It also solves the problem of low signal-to-noise ratio during the daytime with Raman scattering lidar, providing high-precision parameter support for refined research on atmospheric aerosols and environmental monitoring, and promoting the development of a collaborative active and passive remote sensing monitoring system. Attached Figure Description
[0018] Figure 1 A flowchart of Embodiment 1 of the coarse and fine mode aerosol parametric inversion method based on high spectral resolution lidar provided in this application;
[0019] Figure 2 A schematic diagram illustrating the HSRL-based inversion results for an exemplary embodiment of this application; Figure 2 (a) in the figure is a schematic diagram of the coarse mode aerosol extinction inversion results based on HSRL shown in an exemplary embodiment of this application; Figure 2 (b) is a schematic diagram of the fine-modal aerosol extinction inversion results based on HSRL, as shown in an exemplary embodiment of this application;
[0020] Figure 3The comparison results of the coarse mode optical thickness (cAOD), fine mode optical thickness (fAOD), and total optical thickness obtained by HSRL inversion with the time series observations of the solar photometer (SP) are shown as exemplary embodiments of this application.
[0021] Figure 4 The data correlation analysis results obtained by HSRL inversion are shown as an exemplary embodiment of this application; Figure 4 (a) in this example illustrates the correlation analysis results between cAOD obtained by HSRL inversion and SP synchronous observation data as shown in the exemplary embodiment of this application; Figure 4 (b) in this embodiment shows the correlation analysis results between the fAOD obtained by HSRL inversion and the SP synchronous observation data. Detailed Implementation
[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.
[0023] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.
[0024] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."
[0025] The following specific embodiments are given to illustrate the technical solution of this application in detail.
[0026] Figure 1 This is a flowchart of Example 1 of the coarse and fine mode aerosol parametric inversion method based on high spectral resolution lidar provided in this application. Please refer to... Figure 1 The method provided in this embodiment may include:
[0027] S101. Obtain the main channel scattering signal, depolarization channel elastic scattering signal, and molecular Rayleigh scattering signal of aerosol.
[0028] Specifically, a high-spectral-resolution lidar system is employed to simultaneously receive the main channel scattering signal, the depolarization channel elastic scattering signal, and the molecular Rayleigh scattering signal, while performing a full-process preprocessing operation on the acquired raw signals. The detection module of the high-spectral-resolution lidar system acquires the main channel elastic scattering signal and the depolarization channel elastic scattering signal of aerosols, respectively. Then, using the narrowband frequency discriminator on the system, the Mie scattering signal of aerosols and the Rayleigh scattering signal of atmospheric molecules are separated at the spectral level, extracting the Rayleigh scattering signal of the molecular channel. The three types of raw main channel scattering signals, depolarization channel elastic scattering signals, and molecular Rayleigh scattering signals are then preprocessed sequentially, including background noise subtraction, distance-squared correction, polarization gain ratio correction, and system overlap factor correction, eliminating various system interferences and environmental influences, providing accurate and effective basic signal data for subsequent parametric inversion.
[0029] S102. Calculate the volume depolarization ratio based on the ratio of the main channel scattering signal to the depolarization channel elastic scattering signal, and calculate the total backscattering coefficient and lidar ratio based on the volume depolarization ratio and the molecular Rayleigh scattering signal. No lidar ratio assumption is required, and there are no day or night time restrictions.
[0030] Specifically, the volume depolarization ratio is calculated based on the preprocessed main channel scattering signal and depolarization channel signal; combining the volume depolarization ratio with the molecular Rayleigh scattering signal, the total backscattering coefficient is inverted based on HSRL and the lidar ratio, and the inversion process does not require preset lidar ratio related assumptions.
[0031] Furthermore, the ratio of the preprocessed main channel scattering signal to the depolarized channel elastic scattering signal is calculated to obtain the aerosol depolarization ratio. Using the preprocessed molecular Rayleigh scattering signal, the molecular extinction coefficient, molecular backscattering coefficient, and molecular depolarization ratio at the corresponding atmospheric altitude are obtained. At the same time, the spectral transmittance of the spectral discriminator for molecules and aerosols, and the signal ratio of the main channel and molecular channel are determined. The obtained volume depolarization ratio, molecular related optical parameters, spectral transmittance, and channel signal ratio are substituted into the high-spectral-resolution lidar-specific inversion formula for calculation, directly solving for the total backscattering coefficient and lidar ratio. There is no lidar ratio assumption throughout the process, ensuring the objectivity of the inversion results.
[0032] Optionally, when calculating the total backscattering coefficient and lidar ratio based on the volume depolarization ratio and the molecular Rayleigh scattering signal, no lidar ratio assumption is required, and there are no day / night restrictions.
[0033] Furthermore, the steps for calculating the ratio of the total backscattering coefficient to the lidar signal based on the volume deflection ratio and the molecular Rayleigh scattering signal include:
[0034] (1) Determine the first and second spectral transmittances of atmospheric molecules and aerosols in the spectral discriminator;
[0035] Specifically, through performance calibration experiments of the spectral frequency discriminator, its transmission efficiency for Rayleigh scattering signals of atmospheric molecules was tested to obtain the first spectral transmittance under this scenario; simultaneously, the transmission efficiency of the spectral frequency discriminator for aerosol mi scattering signals was tested to obtain the second spectral transmittance under this scenario; through multiple calibration experiments, multiple sets of transmittance data were obtained, and after removing outliers, stable quantitative values of the first and second spectral transmittances were determined, while the proportional relationship between the two was clarified, providing accurate equipment calibration parameters for subsequent optical parameter inversion.
[0036] (2) Obtain quantitative parameters of atmospheric molecules, wherein the quantitative parameters include at least the extinction coefficient, backscattering coefficient, depolarization ratio and Rayleigh scattering signal;
[0037] Specifically, by combining calculations using standard atmospheric molecular models with lidar detection and acquisition, the extinction coefficient, backscattering coefficient, depolarization ratio, and Rayleigh scattering signal of atmospheric molecules are obtained, ensuring the spatiotemporal matching of each quantitative parameter.
[0038] Furthermore, based on the geographical, meteorological, and altitude characteristics of the detection area, the corresponding standard atmospheric molecular model is invoked. The atmospheric molecular extinction coefficient, atmospheric molecular backscattering coefficient, and atmospheric molecular depolarization ratio are calculated at each altitude within the detection altitude range. Simultaneously, the molecular channel of the lidar is used to separate and collect the raw Rayleigh scattering signal of atmospheric molecules through a narrowband frequency discriminator. The raw signal is preprocessed by background noise subtraction, distance square correction, and system overlap factor correction to obtain accurate quantitative data of atmospheric molecular Rayleigh scattering signal, ensuring that the acquired quantitative parameters of various atmospheric molecules correspond one-to-one with the detection time and space altitude.
[0039] (3) Calculate the total backscattering coefficient and the lidar ratio by combining the first spectral transmittance, the second spectral transmittance, the volume depolarization ratio and the quantitative parameters.
[0040] Specifically, the total backscattering coefficient and the lidar ratio can be calculated using the following formula:
[0041] (1)
[0042] (2)
[0043] in, This represents the total backscattering coefficient;
[0044] For lidar ratio;
[0045] The transmittance of the first spectrum;
[0046] The transmittance of the second spectrum;
[0047] K represents the ratio of main channel signal to molecular channel signal;
[0048] The backscattering coefficient of atmospheric molecules;
[0049] This refers to the de-polarization ratio of atmospheric molecules;
[0050] For the body to retreat and be biased;
[0051] The extinction coefficient of atmospheric molecules;
[0052] This is the Rayleigh scattering signal from atmospheric molecules.
[0053] By calibrating the spectral transmittance of atmospheric molecules and aerosols using a spectral discriminator, and combining this with multi-dimensional quantitative parameters of atmospheric molecules obtained from standard model calculations and detection data, and then integrating the volume depolarization ratio to perform joint calculations of the total backscattering coefficient and lidar ratio, we can eliminate the pre-defined assumptions about the lidar ratio. Relying on the accurate parameters obtained from actual measurements and calibration, we can construct an objective inversion system, effectively eliminating the spatiotemporal uncertainties and systematic errors caused by assumptions, and significantly improving the accuracy and reliability of aerosol core optical parameter inversion. At the same time, this lays a precise and solid foundation of parameters for subsequent aerosol mode separation, extinction coefficient, and optical thickness inversion.
[0054] S103. Calculate the aerosol depolarization ratio based on the volume depolarization ratio and the total backscattering coefficient.
[0055] Specifically, using volume depolarization ratio and total backscattering coefficient as input parameters, and combining atmospheric molecule backscattering coefficient and molecular depolarization ratio, the aerosol depolarization ratio, which reflects only the characteristics of aerosols themselves, is extracted from the volume depolarization ratio, which is affected by both molecules and aerosols.
[0056] Furthermore, the molecular backscattering coefficient and molecular depolarization ratio calculated by the standard atmospheric molecular model at the corresponding atmospheric altitude are retrieved. The volume depolarization ratio and the total backscattering coefficient are substituted into the aerosol depolarization ratio calculation formula. By eliminating the influence of atmospheric molecules in the volume depolarization ratio through formula calculation, the aerosol depolarization ratio that only characterizes the polarization scattering characteristics of aerosol particles is obtained. The aerosol depolarization ratio is the key basis for subsequent realization of coarse and fine mode aerosol separation.
[0057] Furthermore, the specific implementation steps include:
[0058] (1) Determine the atmospheric molecule backscattering coefficient and atmospheric molecule depolarization ratio;
[0059] Specifically, by combining the detection area and altitude of the lidar with the real-time meteorological background, a suitable standard atmospheric molecular model is retrieved. The molecular backscattering coefficient of the atmosphere at different vertical heights is accurately calculated through the model. At the same time, the fixed atmospheric molecular depolarization ratio value under the atmospheric environment is obtained. This ensures that the two parameters are completely matched with the subsequent aerosol parameters in terms of time, space and altitude, providing an accurate molecular parameter basis for the calculation of the aerosol depolarization ratio.
[0060] (2) Calculate the aerosol depolarization ratio based on the atmospheric molecule backscattering coefficient, the atmospheric molecule depolarization ratio, the volume depolarization ratio and the total backscattering coefficient.
[0061] Specifically, the aerosol deflection ratio can be calculated using the following formula:
[0062] (3)
[0063] in, This refers to the aerosol depolarization ratio;
[0064] This represents the total backscattering coefficient;
[0065] The backscattering coefficient of atmospheric molecules;
[0066] This refers to the de-polarization ratio of atmospheric molecules;
[0067] The body is biased towards the retreat.
[0068] By determining the atmospheric molecule backscattering coefficient and atmospheric molecule depolarization ratio using standard methods, and then combining the aerosol depolarization ratio and total backscattering coefficient to perform a comprehensive calculation of the aerosol depolarization ratio, interference components caused by atmospheric molecule scattering in the volume depolarization ratio can be effectively eliminated. This allows for the accurate extraction of the aerosol depolarization ratio, which only reflects the polarization scattering characteristics of the aerosol itself, ensuring authenticity and accuracy. This provides reliable parameter support for the subsequent construction of aerosol multi-component mixing models and the quantitative separation of coarse and fine modal aerosols, thus laying the foundation for the accurate inversion of various optical parameters of aerosols.
[0069] S104. Based on the clustering results of historical data, obtain the characteristic depolarization ratio constants of coarse mode and fine mode respectively. Construct an aerosol multi-component mixing model based on the characteristic depolarization ratio constants. Input the aerosol depolarization ratio into the multi-component mixing model to obtain the coarse mode backscattering coefficient and the fine mode backscattering coefficient.
[0070] Specifically, statistical cluster analysis was conducted based on historical data of aerosol depolarization ratios from long-term observations to determine the characteristic depolarization ratio constants of coarse and fine-mode aerosols. A binary component mixing model was constructed based on typical values, and the aerosol depolarization ratios calculated in real time were input into the model. The coarse backscattering coefficient and the fine-mode backscattering coefficient were obtained by solving the simultaneous equations.
[0071] Furthermore, statistical clustering was performed on massive historical aerosol depolarization ratio observation data. Based on the differences in polarization scattering characteristics of aerosol particles, representative coarse-mode and fine-mode aerosol characteristic depolarization ratio constants were selected. Based on the binary component mixing principle, an aerosol multi-component mixing model was constructed, and the constraint relationship that the total backscattering coefficient is the sum of the coarse-mode and fine-mode backscattering coefficients was determined. The real-time aerosol depolarization ratio was input into the model, and a system of simultaneous equations was constructed in combination with the determined coarse-mode and fine-mode characteristic depolarization ratio constants. The system of equations was solved through algebraic operations to obtain the coarse-mode and fine-mode backscattering coefficients, respectively, thus achieving quantitative separation of the total backscattering coefficient by mode.
[0072] Furthermore, the steps for obtaining the feature debiasing ratio constants for coarse and fine modes based on historical data clustering results include:
[0073] (1) Collect historical observation data on aerosol depolarization ratio;
[0074] Specifically, the study systematically collected measured aerosol depolarization ratio (DBR) data under different time and weather conditions within the study area, constructing a multi-dimensional, large-sample historical DBR dataset to provide a sufficient and representative data foundation for subsequent cluster analysis. Utilizing aerosol detection equipment such as high-spectral-resolution lidar, the study continuously collected measured DBR data at different altitudes, time periods, and meteorological conditions within the study area, while simultaneously recording auxiliary information such as the detection time and space, meteorological background, and aerosol type. The collected raw measured data underwent preprocessing to remove outliers and missing values caused by equipment malfunctions or environmental interference, resulting in a comprehensive, high-quality, and sufficiently large historical DBR dataset, ensuring that the data accurately reflects the actual distribution characteristics of the DBR within the region.
[0075] (2) Perform statistical cluster analysis on the historical observation data to divide the data into the debiasing ratio data clusters of coarse-mode aerosols and the debiasing ratio data clusters of fine-mode aerosols;
[0076] Specifically, a density- or distance-based statistical clustering algorithm is selected. The pre-processed historical data of aerosol depolarization ratio observations are used as the algorithm input. The aerosol depolarization ratio value is used as the core clustering feature. The algorithm calculates the similarity and difference between the data and divides the overall dataset into two independent data clusters according to the numerical clustering characteristics. Combining the physical characteristics of coarse and fine aerosols, coarse aerosols, due to their large particle size and significant polarization scattering characteristics, correspond to the dataset cluster with a higher depolarization ratio value, while fine aerosols, due to their small particle size and weak polarization scattering characteristics, correspond to the dataset cluster with a lower depolarization ratio value. This completes the accurate division and correspondence of coarse and fine aerosol depolarization ratio data clusters.
[0077] (3) Calculate the statistical characteristic values of the debiasing ratio data clusters of coarse-mode aerosols and fine-mode aerosols respectively, and use them as the characteristic debiasing ratio constants of coarse-mode aerosols and fine-mode aerosols.
[0078] Specifically, for the coarse-mode aerosol debiasing ratio data clusters and fine-mode aerosol debiasing ratio data clusters, representative statistical characteristic values such as arithmetic mean and median are calculated respectively. Statistical values that can stably reflect the overall level of the data cluster and are less affected by extreme values are selected first. The calculated statistical characteristic values of the coarse-mode data cluster are calibrated as the characteristic debiasing ratio constants of the coarse-mode aerosol, and the statistical characteristic values of the fine-mode data cluster are calibrated as the characteristic debiasing ratio constants of the fine-mode aerosol. The two types of constants are stored in the model parameter library as fixed basic parameters for subsequent construction of aerosol multi-component mixing models and the realization of modal quantitative separation, and will no longer change with real-time detection data.
[0079] By systematically collecting historical data on aerosol depolarization ratio observations, conducting statistical cluster analysis, and dividing the data into depolarization ratio clusters corresponding to coarse and fine modes, and calculating the statistical characteristic values of each data cluster as the characteristic depolarization ratio constants of the corresponding modes, the inherent characteristic patterns of aerosol depolarization ratios of coarse and fine modes can be mined based on large-sample historical observation data. The obtained characteristic depolarization ratio constants have both regional adaptability and numerical stability, and can be used as fixed basic parameters for subsequent construction of aerosol multi-component mixing models. This provides objective and reliable parameter support for achieving accurate quantitative separation of backscattering coefficients of coarse and fine modes, effectively avoiding mode separation errors caused by subjective parameter selection, and also improving the consistency and accuracy of subsequent aerosol mode parameter inversion.
[0080] Furthermore, the steps for inputting the aerosol depolarization ratio into the multi-component mixing model to obtain the coarse-mode backscattering coefficient and the fine-mode backscattering coefficient include:
[0081] (1) The multi-component mixing model uses the characteristic depolarization ratio constant as the basic constant to retrieve the constraint relationship of the total backscattering coefficient;
[0082] Specifically, the multi-component hybrid model first solidifies the depolarization ratio constants of the coarse and fine modes as the basic constants for computation. At the same time, it retrieves the inherent constraint relationship of the total aerosol backscattering coefficient, which is formed by the superposition of the backscattering coefficients of the coarse and fine modes, to build a basic framework for subsequent simultaneous calculations.
[0083] Furthermore, the coarse and fine mode characteristic depolarization ratio constants obtained from the cluster analysis of historical aerosol depolarization ratio data are preloaded and used as fixed basic computational constants, no longer changing with real-time detection data; at the same time, the model retrieves the physical constraint relationship of the total aerosol backscattering coefficient, which is determined based on the physical characteristics of independent scattering of aerosol coarse and fine modes, providing core equation constraints for subsequent numerical solutions.
[0084] Furthermore, the constraint relationship can be expressed by the following formula:
[0085] (4)
[0086] in, This represents the total backscattering coefficient;
[0087] The coarse-mode backscattering coefficient;
[0088] This represents the fine-mode backscattering coefficient.
[0089] (2) Construct a mixing equation for the depolarization ratio based on the linear mixing principle of aerosol depolarization ratio;
[0090] Specifically, based on the physical principle that the aerosol depolarization ratio is linearly superimposed by the coarse and fine mode depolarization ratios according to their respective backscattering contributions, a depolarization ratio hybrid equation is constructed with the aerosol depolarization ratio, the coarse and fine mode characteristic depolarization ratio constants, and the backscattering ratios of the two as variables, thus establishing the correlation between the depolarization ratio and modal backscattering.
[0091] Furthermore, based on the linear mixing physics principle of aerosol depolarization ratio, the overall aerosol depolarization ratio detected in real time is obtained by multiplying the coarse-mode aerosol characteristic depolarization ratio constant and the fine-mode aerosol characteristic depolarization ratio constant by the proportion of their respective backscattering coefficients to the total backscattering coefficients, and then summing them. According to this principle, with the real-time aerosol depolarization ratio as the dependent variable, the coarse and fine-mode characteristic depolarization ratio constants as known constants, and the proportion of coarse and fine-mode backscattering coefficients as unknown variables, a corresponding depolarization ratio mixing equation is constructed to realize the mathematical correlation between the overall aerosol depolarization ratio, the coarse and fine-mode depolarization ratios, and the modal backscattering contribution.
[0092] Furthermore, the depolarization ratio mixing equation can be expressed by the following formula:
[0093] (5)
[0094] (6)
[0095] in, This represents the total backscattering coefficient;
[0096] The coarse-mode backscattering coefficient;
[0097] The fine-mode backscattering coefficient;
[0098] This refers to the aerosol depolarization ratio;
[0099] This is the debias ratio constant for coarse mode characteristics;
[0100] This is the debias ratio constant for fine-mode characteristics.
[0101] (3) Using the aerosol depolarization ratio as the independent variable, the constraint relationship and the depolarization ratio mixed equation are combined to calculate the coarse mode backscattering coefficient and the fine mode backscattering coefficient.
[0102] Specifically, the real-time detected aerosol depolarization ratio is substituted into the depolarization ratio mixing equation, and the total backscattering coefficient constraint relationship is combined with the depolarization ratio mixing equation. The unknown variables in the two equations are unified as coarse and fine mode backscattering coefficients. The equation system is solved through algebraic operations to obtain the quantitative values of the backscattering coefficients of the two modes.
[0103] Furthermore, the aerosol depolarization ratio obtained from real-time inversion is used as the independent variable and substituted into the depolarization ratio mixing equation constructed in step (2) to make the depolarization ratio mixing equation an equation containing only two unknowns: coarse and fine mode backscattering coefficients. Then, this equation is combined with the total backscattering coefficient constraint relationship equation obtained in step (1) to form a set of equations. By solving this set of equations, the backscattering coefficient of one type of mode is calculated first, and then substituted into the constraint relationship equation to obtain the backscattering coefficient of another type of mode. Finally, the quantitative values of the coarse mode backscattering coefficient and the fine mode backscattering coefficient are accurately solved, and the coarse and fine modes of the total backscattering coefficient are quantitatively separated.
[0104] By using the coarse and fine modal depolarization ratio constant as the fundamental constant of the multi-component mixing model and invoking the constraint relationship of the total backscattering coefficient, and then constructing the depolarization ratio mixing equation based on the linear mixing principle of aerosol depolarization ratio, and solving the above constraint relationship and mixing equation with the real-time aerosol depolarization ratio as the independent variable, it is possible to achieve accurate quantitative separation of the backscattering coefficients of coarse and fine modes. Relying on the combination of physical principles and mathematical operations, the mode separation results are made to fit the actual scattering characteristics of aerosols, effectively avoiding the limitations of qualitative differentiation, and providing accurate modal fundamental parameters for the subsequent inversion of coarse and fine mode extinction coefficients and optical thickness. At the same time, this process relies on the real-time detection of aerosol depolarization ratio to complete the calculation, ensuring the real-time performance and accuracy of the modal backscattering coefficient inversion.
[0105] S105. Based on the clustering results of historical data, obtain the characteristic lidar ratio constants of coarse mode and fine mode respectively, and calculate the coarse mode extinction coefficient and fine mode extinction coefficient based on the coarse mode backscattering coefficient, the fine mode backscattering coefficient and the characteristic lidar ratio constant.
[0106] Specifically, statistical cluster analysis is conducted based on historical lidar ratio data from long-term observations to determine the characteristic lidar ratio constants corresponding to coarse-mode and fine-mode aerosols; and the coarse-mode extinction coefficient and fine-mode extinction coefficient are calculated using the correlation between lidar ratio and backscattering coefficient.
[0107] Furthermore, statistical cluster analysis was performed on the historical data of aerosol lidar ratio obtained from long-term observation. Based on the differences in optical properties between coarse and fine-mode aerosols, the characteristic lidar ratio constants corresponding to coarse-mode and fine-mode aerosols were determined. Taking the product of lidar ratio and backscattering coefficient as the core correlation, the coarse-mode backscattering coefficient was multiplied with the coarse-mode lidar ratio to obtain the coarse-mode extinction coefficient. At the same time, the fine-mode backscattering coefficient was multiplied with the fine-mode lidar ratio to obtain the fine-mode extinction coefficient.
[0108] Furthermore, the specific implementation steps include:
[0109] (1) Retrieve the coarse mode characteristic lidar ratio constant and the fine mode characteristic lidar ratio constant obtained by clustering based on historical lidar ratio data;
[0110] Specifically, coarse and fine modal characteristic lidar ratio constants obtained from long-term lidar ratio observation data statistical clustering analysis are retrieved from the preset parameter library. These parameters are fixed reference values adapted to the aerosol characteristics of the detection area and are directly used as the basic constants for calculating the modal extinction coefficient.
[0111] Furthermore, based on the long-term aerosol lidar ratio observations conducted in the detection area in the early stage, statistical cluster analysis was performed on massive historical data. According to the differences in optical properties of coarse and fine mode aerosols, regionally representative and numerically stable coarse mode characteristic lidar ratio constants and fine mode characteristic lidar ratio constants were selected and stored in the model parameter library. When calculating the extinction coefficient, these two sets of typical values were directly retrieved from the parameter library without recalculation, ensuring computational efficiency and parameter adaptability.
[0112] (2) Calculate the coarse mode extinction coefficient based on the product of the coarse mode backscattering coefficient and the coarse mode characteristic lidar constant;
[0113] Specifically, the coarse mode extinction coefficient can be calculated using the following formula:
[0114] (7)
[0115] in, The coarse mode extinction coefficient;
[0116] The ratio constant is used for coarse-mode characteristic lidar;
[0117] This represents the coarse-mode backscattering coefficient.
[0118] (3) Calculate the fine mode extinction coefficient based on the product of the fine mode backscattering coefficient and the fine mode characteristic lidar constant.
[0119] Specifically, the fine-mode extinction coefficient can be calculated using the following formula:
[0120] (8)
[0121] The extinction coefficient is the fine mode extinction coefficient.
[0122] The ratio of fine-mode characteristic lidar to constant;
[0123] This represents the fine-mode backscattering coefficient.
[0124] By retrieving the coarse and fine modal characteristic lidar ratio constants obtained through clustering of historical lidar ratio data, and then multiplying the coarse and fine modal backscattering coefficients with the corresponding characteristic lidar ratio constants, the aerosol extinction coefficients of the corresponding modes are obtained. The typical values determined by historical observation data ensure the regional adaptability and stability of the parameters. The quantitative conversion is completed by following the inherent physical relationship between lidar ratio, backscattering coefficient, and extinction coefficient. The calculation logic is simple and accurate, realizing the efficient and accurate separation and quantitative acquisition of coarse and fine modal extinction coefficients. It effectively avoids additional errors in parameter conversion and provides accurate basic parameters for modal extinction coefficients for subsequent vertical integration of aerosol layers and optical thickness inversion, while ensuring the efficiency and reliability of extinction coefficient inversion.
[0125] S106. Calculate the total backscattering coefficient based on the coarse-mode backscattering coefficient and the fine-mode backscattering coefficient. Determine the vertical range of the aerosol layer based on the total backscattering coefficient. Perform vertical integration on the coarse-mode extinction coefficient and the fine-mode extinction coefficient within the height of the aerosol layer to obtain the coarse-mode optical thickness and the fine-mode optical thickness. Perform inversion based on the coarse-mode optical thickness and the fine-mode optical thickness.
[0126] Specifically, the total backscattering coefficient of the aerosol is reconstructed by summing the backscattering coefficients of the coarse and fine modes; the vertical range of the aerosol layer is identified based on the vertical profile of the total backscattering coefficient using threshold or gradient analysis methods; the extinction coefficients of the coarse and fine modes are vertically integrated within the identified aerosol layer height to obtain the corresponding modal optical thickness; the modal optical thickness is compared and verified with synchronous data from a solar photometer to complete the final coarse and fine mode aerosol parametric inversion.
[0127] Furthermore, the coarse-mode backscattering coefficient and the fine-mode backscattering coefficient are summed to obtain the total aerosol backscattering coefficient, thus acquiring complete atmospheric vertical profile data. The total backscattering coefficient is then compared with the corresponding atmospheric molecular backscattering coefficients at each altitude. Threshold analysis or gradient analysis is used to identify the altitude at which the total backscattering coefficient significantly increases as the aerosol layer bottom height, and the altitude at which it falls back to the atmospheric background level as the aerosol layer top height, thereby defining the vertical height range of each independent aerosol layer. Within the identified aerosol layer height ranges, the coarse-mode extinction coefficient is calculated separately. Vertical integration of the fine-mode extinction coefficients yields the coarse-mode optical thickness and fine-mode optical thickness of each layer and the entire atmosphere. These coarse-mode and fine-mode optical thicknesses are then spatiotemporally matched with synchronous observation data from a solar photometer. The accuracy of the distance and time matching between the comparison stations is carefully controlled, and invalid values such as cloud interference and time-period anomalies in the solar photometer data are eliminated. The consistency between the two sets of data is analyzed by calculating statistical indicators such as correlation coefficients and root mean square error, thus verifying the inversion accuracy. The verified coarse and fine-mode optical thicknesses are then output, achieving a complete inversion of coarse and fine-mode aerosol parameters.
[0128] Furthermore, the steps for determining the vertical range of the aerosol layer based on the total backscattering coefficient include:
[0129] (1) The total backscattering coefficient and the atmospheric molecule backscattering coefficient are compared at each height to identify the bottom and top height of the aerosol;
[0130] Specifically, based on the detection height gradient of the lidar, the total backscattering coefficient at the same height is compared with the backscattering coefficient of atmospheric molecules one by one. Based on the numerical change characteristics of the scattering coefficient, the bottom height where the aerosol layer begins to appear and the top height where it disappears are determined, thus achieving the preliminary identification of the aerosol layer height boundary.
[0131] Furthermore, quantitative data on the total backscattering coefficient and atmospheric molecular backscattering coefficient at each altitude within the lidar detection range are obtained. The two sets of data are compared and analyzed at the same altitude in ascending order of vertical height. When the total backscattering coefficient shows a significant increase and remains higher than the atmospheric molecular backscattering coefficient at the corresponding altitude, this altitude is determined as the bottom height of the aerosol layer, indicating that aerosol particles begin to accumulate at this altitude. When the total backscattering coefficient falls back to the background level that is equal to or close to the atmospheric molecular backscattering coefficient at the corresponding altitude, this altitude is determined as the top height of the aerosol layer, indicating that aerosol particles above this altitude have basically dissipated. This completes the accurate identification of the bottom and top heights of a single or multiple aerosol layers.
[0132] (2) Determine the vertical range based on the bottom height and the top height.
[0133] Specifically, the bottom height of the identified aerosol layer is used as the lower boundary of the vertical range, and the top height is used as the upper boundary. A unique and clear vertical height range is defined for each independent aerosol layer. At the same time, the independent vertical range of each multi-layer aerosol is clearly defined to avoid confusion in height range.
[0134] Furthermore, the bottom height and top height of each identified aerosol layer are paired, with the bottom height value as the lower limit and the top height value as the upper limit, to define the exclusive vertical height range of the aerosol layer. If multiple independent aerosol layers are detected in the atmosphere, the bottom height and top height of each layer are paired and defined separately to clarify the non-overlapping vertical height range of each layer, forming complete vertical range distribution information of aerosol layers. This provides a clear and accurate height definition basis for subsequent vertical integration of extinction coefficients only within the aerosol layer range.
[0135] Furthermore, by analyzing the vertical profile of the total backscattering coefficient obtained through gradient analysis or thresholding, all aerosol layers present in the atmosphere are identified. Each individual aerosol layer i is determined by its base height z. b,i and top z t,i Let i = 1, 2, ..., N, where N is the total number of identified aerosol layers. Let E be the total set of aerosol layer heights detected by the lidar:
[0136] (9)
[0137] On the height set E, piecewise integration should be performed on the coarse and fine modal extinction coefficients to calculate the modal optical thickness of each aerosol layer and the whole.
[0138] (10)
[0139] (11)
[0140] in, Let the bottom height of the i-th aerosol layer be denoted as .
[0141] Let i be the top height of the i-th aerosol layer;
[0142] and Let E represent the coarse and fine modal optical thicknesses, respectively, integrated over the height set E. This integration form can output not only the total optical thickness of the entire layer but also the modal optical thicknesses of each individual aerosol layer, i.e.:
[0143] (12)
[0144] (13)
[0145] in, The coarse mode optical thickness of the i-th aerosol layer;
[0146] The fine modal optical thickness of the i-th aerosol layer;
[0147] The extinction coefficient is the fine mode extinction coefficient.
[0148] The coarse mode extinction coefficient;
[0149] Let the bottom height of the i-th aerosol layer be denoted as .
[0150] Let be the top height of the i-th aerosol layer.
[0151] By conducting a height-by-height comparative analysis of the total backscattering coefficient and the atmospheric molecular backscattering coefficient, the bottom height where the aerosol layer appears and the top height where it disappears are accurately identified. Then, based on the identified bottom and top height values, the specific vertical range of each independent aerosol layer is clearly defined. This method relies on the difference in backscattering characteristics between aerosols and molecules to achieve precise positioning of aerosol layers. It can clearly define the height boundaries of different aerosol layers, providing an accurate and clear height range basis for subsequent vertical integration of extinction coefficient and calculation of modal aerosol optical thickness within a specified height range, ensuring the relevance of subsequent integration calculations and the accuracy of the results.
[0152] Furthermore, after inverting based on the coarse-mode optical thickness and the fine-mode optical thickness, the method provided in this embodiment also includes:
[0153] (1) Obtain coarse mode optical thickness data and fine mode optical thickness data retrieved from the solar photometer;
[0154] Specifically, synchronous observation data from solar photometer stations matching the lidar detection area are retrieved, and coarse-mode and fine-mode aerosol optical thickness data obtained from the solar photometer stations using a standard inversion algorithm are extracted. The acquired solar photometer data are preprocessed to remove outliers and invalid values caused by factors such as cloud cover, equipment calibration deviations, and environmental interference. Coarse and fine-mode optical thickness reference data that are of acceptable quality and match the lidar detection period are selected to provide a reliable comparison benchmark for subsequent accuracy verification.
[0155] (2) Spatiotemporal matching of coarse mode optical thickness, fine mode optical thickness, coarse mode optical thickness data and fine mode optical thickness data, consistency analysis, and verification of inversion accuracy based on statistical indicators.
[0156] Specifically, based on the detection time and area of the lidar, the coarse and fine modal optical thicknesses obtained from lidar inversion are spatiotemporally matched with the coarse and fine modal optical thickness data from the photometer. The temporal matching accuracy and the adaptability to the spatial distance between the stations are controlled to ensure that the two sets of comparative data are aerosol parameter observation results within the same spatiotemporal range. Consistency analysis is then conducted between the spatiotemporally matched lidar inversion data and the photometer reference data. Quantitative statistical indicators such as correlation coefficient, root mean square error, and mean absolute error are calculated to quantify the degree of fit between the two sets of data. The lidar inversion accuracy is determined based on the numerical characteristics of the statistical indicators. Higher correlation coefficients and lower error indicators indicate better consistency between the inversion results and the standard reference data, resulting in higher inversion accuracy. This verifies the accuracy of the lidar coarse and fine modal aerosol optical thickness inversion results.
[0157] By acquiring measured data of coarse and fine modal optical thicknesses retrieved from a solar photometer, the coarse and fine modal optical thicknesses retrieved from lidar are precisely matched with these measured data in the spatiotemporal dimensions. Consistency analysis is then performed on the two sets of matched data, and relevant statistical indicators are used to verify the lidar retrieval accuracy. This approach leverages precise measured data from a solar photometer to effectively verify the lidar retrieval results, eliminating comparison errors caused by spatiotemporal misalignment. Statistical indicators intuitively quantify the accuracy and reliability of the retrieval results, ensuring the accuracy of the coarse and fine modal aerosol optical thickness parameters retrieved from lidar. Furthermore, verification allows for further optimization of the retrieval method, improving the overall accuracy of subsequent aerosol modal parameter retrieval and making the retrieval results more closely match the actual distribution characteristics of atmospheric aerosols, thus providing more reliable parameter support for refined atmospheric aerosol research.
[0158] Furthermore, this application will illustrate the solution through a complete embodiment:
[0159] (i) Using the HSRL system, three key signals are received synchronously: the main channel elastic scattering signal, the depolarization channel elastic scattering signal, and the molecular Rayleigh scattering signal separated by the frequency discriminator;
[0160] (II) Based on the main channel and polarization channel signals acquired in step (I), the aerosol depolarization ratio (δ) is inverted. v By combining the molecular Rayleigh scattering signal of the molecular channel, the total backscattering coefficient (β) can be directly inverted. a ) and lidar ratio (S a );
[0161] (iii) Using the body deterioration ratio (δ) obtained in step (ii) v ) and total backscattering coefficient (β) a ), calculate the aerosol depolarization ratio (δ) a );
[0162] (iv) Aerosol de-bias ratio (δ) retrieved from step (iii) a Historical data cluster analysis was used to determine the characteristic debiasing ratio constant (δ) of coarse-mode aerosols. c Characteristic debiasing constant (δ) of fine-mode aerosols f A multi-component aerosol mixing model is constructed, assuming that the total backscattering coefficient is the sum of the backscattering coefficients of the coarse and fine modes (β). a = β c + β f ), combined with the aerosol depolarization ratio (δ) calculated in step (iii) a Real-time data, solving simultaneous equations to obtain the coarse mode backscattering coefficient (β) c ) and fine mode backscattering coefficient (β) f This enables the separation of the two modalities;
[0163] (v) The lidar ratio (S) retrieved based on step (ii) a Historical data clustering analysis was used to determine the characteristic lidar ratio constant (S) of coarse and fine modal aerosols. c S f Using the correlation between lidar ratio and backscattering coefficient (α = S×β), the coarse mode extinction coefficient (α) is calculated. c ) and fine mode extinction coefficient (α) f );
[0164] (vi) The total backscattering coefficient (β) obtained from step (ii) a The vertical height range of several aerosol layers is determined by threshold identification; within this height range, the coarse mode extinction coefficient (α) obtained in step (v) is... c ) and fine mode extinction coefficient (α) f By performing vertical integration, the coarse mode optical thickness (cAOD) and the fine mode optical thickness (fAOD) are obtained respectively.
[0165] (vii) The coarse mode optical thickness (cAOD) and fine mode optical thickness (fAOD) results output in step (vi) are synchronously observed with the observation data of the solar photometer (SP), and the inversion accuracy is verified based on the consistency of the spatiotemporal matching data.
[0166] Furthermore, in step (i), the molecular channel should use a narrowband frequency discriminator to spectrally separate the Mie scattering signal of the aerosol from the Rayleigh scattering signal of the molecule, and extract the molecular Rayleigh scattering signal. In addition, it should also include conventional signal preprocessing, such as background noise subtraction, distance square correction, polarization gain ratio correction and system overlap factor correction, to provide a basis for subsequent quantitative inversion.
[0167] In step (ii), based on the above formula (1), the total backscattering coefficient (β) is obtained by HSRL inversion. a ), and based on the above formula (2) combined with the HSRL inverted lidar ratio (S a ); where δ v T is obtained by comparing the debiasing channel signal with the main channel signal obtained in step (i). m T a α represents the spectral transmittance of molecules and aerosols in the spectral discriminator, K is the ratio of the main channel signal to the molecular channel signal obtained in step (i), and α is the spectral transmittance of molecules and aerosols in the spectral discriminator. m , β m and δ m These are the atmospheric molecular extinction coefficient, backscattering coefficient, and depolarization ratio; The signal is the Rayleigh scattering signal of atmospheric molecules separated by the frequency discriminator.
[0168] In step (iii), the aerosol depolarization ratio δ is calculated according to the above formula (3). a .
[0169] Step (iv) Through long-term observation of aerosol depolarization ratio δ a Data statistical clustering determines the debiasing ratio δ of coarse and fine modal aerosols. c δ f Based on the binary component mixing model, we have β a =β c +β f and through δ a β is obtained by calculating using the above formulas (5) and (6). c With β f .
[0170] Step (5) Through long-term observation of S a Data statistical clustering determines the coarse and fine modal aerosol radar ratio S c S f Based on the binary component mixing model, according to the above formulas (7) and (8), the coarse and fine mode backscattering coefficients β are... c With β f Converted to coarse and fine mode extinction coefficients.
[0171] In step (six), the total backscattering coefficient β obtained in step (two) should be determined through gradient analysis or the threshold method. a The vertical profiles identify all aerosol layers present in the atmosphere. Each individual aerosol layer i is determined by its base height z. b,i and top z t,i Define i = 1, 2, ..., N, where N is the total number of identified aerosol layers. Let E be the total height set of aerosol layers detected by the lidar as given by the above formula (9). On the height set E, for α c With α fPiecewise integration is performed, and the modal optical thickness of each aerosol layer and the whole is calculated based on the above formulas (10) and (11), where τ c and τ f Let f(AOD) and f(AOD) represent the coarse modal optical thickness (cAOD) and fine modal optical thickness (fAOD) respectively, integrated over the height set E. This integration form can output not only the total optical thickness of the entire layer, but also the modal optical thickness of each independent aerosol layer, i.e., formulas (12) and (13);
[0172] In step (vii), the cAOD and fAOD products of the solar photometer should be filtered to remove cloud interference and obvious abnormal data during the day-night transition period. The distance between the solar photometer and the HSRL comparison station should not exceed 50km. The time matching accuracy should be controlled within ±5 minutes to obtain synchronous observations with spatiotemporal matching. The consistency of the data is analyzed by calculating statistical indicators such as correlation coefficient (R²) and root mean square error (RMSE) to verify the accuracy of the HSRL inversion results.
[0173] Compared with the prior art, the present invention has the following beneficial effects:
[0174] 1. This invention proposes a coarse and fine mode aerosol parameter inversion method based on the HSRL system. By using a coarse and fine mode estimation scheme with polarization separation, it solves the problems of passive nighttime observation interruption and low signal-to-noise ratio during the day, and realizes 24-hour continuous and accurate inversion of optical parameters such as cAOD / fAOD.
[0175] 2. This invention uses the HSRL system to perform stratified integration of coarse and fine mode extinction coefficients, which can not only obtain the cAOD / fAOD of the entire atmosphere, but also reflect the distribution differences of different altitude layers (such as the boundary layer and the upper-level suspended layer), providing support for the monitoring of aerosol transport and stratified pollution characteristics.
[0176] Furthermore, Figure 2 A schematic diagram illustrating the HSRL-based inversion results for an exemplary embodiment of this application; Figure 2 (a) in the figure is a schematic diagram of the coarse mode aerosol extinction inversion results based on HSRL shown in an exemplary embodiment of this application; Figure 2 (b) is a schematic diagram of the fine-modal aerosol extinction inversion results based on HSRL, as shown in an exemplary embodiment of this application; Figure 3 The comparison results of the coarse mode optical thickness (cAOD), fine mode optical thickness (fAOD), and total optical thickness obtained by HSRL inversion with the time series observations of the solar photometer (SP) are shown as exemplary embodiments of this application. Figure 4 The data correlation analysis results obtained by HSRL inversion are shown as an exemplary embodiment of this application; Figure 4(a) in this example illustrates the correlation analysis results between cAOD obtained by HSRL inversion and SP synchronous observation data as shown in the exemplary embodiment of this application; Figure 4 (b) in this embodiment shows the correlation analysis results between the fAOD obtained by HSRL inversion and the SP synchronous observation data.
[0177] Furthermore, a method for coarse and fine mode aerosol parameter inversion based on high spectral resolution lidar includes the following steps:
[0178] S1: The HSRL system is used to simultaneously acquire signals from three channels, including the main channel (parallel channel) signal, the depolarization channel (vertical channel) signal, and the molecular channel signal. In this embodiment of the invention, the HSRL detection wavelength is 532 nm, the spatial resolution is 7.5 m, the temporal resolution is 1 min, and the data detection period is from March 30 to March 31, 2021.
[0179] S2: Based on the above formulas (1) and (2), the total backscattering coefficient β is obtained by HSRL inversion. a Compared to LiDAR, S a ;
[0180] S3: Body Regression Ratio δ v It is influenced by both aerosols and molecules. Therefore, it is necessary to extract the depolarization ratio δ, which only represents the characteristics of aerosols. a β obtained using S2 inversion a and known β m Calculated using the following formula:
[0181] ;
[0182] in, This refers to the aerosol depolarization ratio;
[0183] This represents the total backscattering coefficient;
[0184] The backscattering coefficient of atmospheric molecules;
[0185] This refers to the de-polarization ratio of atmospheric molecules;
[0186] For the body to retreat and be biased;
[0187] The obtained δ a These are key parameters for subsequent separation of coarse and fine modes.
[0188] S4: A multi-component mixing model is used to separate the coarse and fine modes in the total backscattering, i.e., β. a =β c +β fBased on long-term observation statistics or typical regional climate characteristics, the coarse and fine mode aerosol depolarization ratios are determined. In this embodiment, the statistical results of various coarse and fine modes that meet this condition are: coarse mode aerosol depolarization ratio (δ... c ): 0.306±0.015. Fine-modal aerosol depolarization ratio (δ) f ): 0.055±0.018, solve the following equations to find β. c β f :
[0189] ;
[0190] ;
[0191] S5: Based on long-term observation statistics or typical regional climate characteristics, determine the coarse and fine mode aerosol radar ratio. In this embodiment, the statistical results of each coarse and fine mode that meet this condition are: coarse mode radar ratio (S c The radar ratio is 41±6 sr; the fine mode radar ratio (S) is 41±6 sr. f ): 57±12sr; Based on the binary component mixing model, β c With β f Converted to extinction coefficient α c =S c β c With α f =S f β f ; Figure 2 (a) and Figure 2 (b) shows the α values measured by HSRL in the Beijing area on March 30-31, 2021. c With α f Time-elevation map.
[0192] S6: The total backscattering coefficient β obtained from the inversion of S2. a The vertical profile. In this embodiment, β is set. a >1.1β m To obtain an effective aerosol signal, all aerosol layers present in the atmosphere were identified. In this embodiment, approximately 1-5 aerosol layers were identified (the location with the most layers appeared around 08:00 during the day-night transition period). The α value of each layer was then analyzed. c With α f Numerical integration is performed to calculate the cAOD and fAOD of the corresponding height layer.
[0193] S7: Sum the total cAOD and fAOD columns obtained by this method and compare them with the cAOD and fAOD data acquired by the solar photometer (SP) during the corresponding time period (including real-time data under both sunlight and moonlight modes). In this embodiment, synchronous observation data from the Beijing_CAMS site and the HSRL were selected for comparison, such as... Figure 3 As shown, this overcomes the problems of SP being affected by cloud cover and lacking day / night cycles. In this observation case, as... Figure 4 As shown, the correlation coefficients for cAOD (R²=0.9826, RMSE=0.086) and fAOD (R²=0.8630, RMSE=0.043) demonstrate good consistency, validating the reliability of this method.
[0194] The coarse and fine mode aerosol parameter inversion method based on high spectral resolution lidar provided in this embodiment solves the problems of passive nighttime observation interruption and low signal-to-noise ratio during the day by using a polarization separation coarse and fine mode estimation scheme, and achieves 24-hour continuous and accurate inversion of optical parameters such as cAOD / fAOD. Based on the HSRL system, the coarse and fine mode extinction coefficients are integrated in layers, which can not only obtain the cAOD / fAOD of the entire atmosphere, but also reflect the distribution differences of different altitude layers (such as the boundary layer and the upper-level suspended layer), providing support for the monitoring of aerosol transport and stratified pollution characteristics. Furthermore, by synchronously acquiring multi-channel scattering signals and completing precise preprocessing, the total backscattering coefficient and lidar ratio are directly inverted based on measured and calibrated parameters. The entire process does not require lidar ratio assumptions, effectively eliminating the spatiotemporal uncertainties and systematic errors caused by assumptions in traditional methods. By eliminating atmospheric molecular interference, the aerosol depolarization ratio is accurately extracted. A multi-component mixing model is constructed by combining typical values obtained from historical data clustering, achieving precise quantitative separation of coarse and fine modal backscattering coefficients. Then, based on the physical correlation between lidar ratio and backscattering coefficient, the modal separation calculation of extinction coefficient is completed. At the same time, the vertical range of the aerosol layer is accurately defined by height-by-height comparison and layered vertical integration is carried out. Accuracy verification is completed by combining synchronous data from a solar photometer, achieving all-day, continuous, and accurate inversion of coarse and fine modal optical thickness. This technology overcomes the limitations of traditional passive remote sensing, such as lighting and cloud cover constraints, observation interruptions, and lack of vertical stratification information. It also solves the problems of low signal-to-noise ratio during the day and difficulty in synchronous verification with passive remote sensing by Raman scattering lidar. Furthermore, it achieves a breakthrough in the qualitative differentiation and precise quantification of aerosol modal parameters, enabling the simultaneous acquisition of coarse and fine modal aerosol optical parameters of the entire layer and each altitude layer. This provides high-precision and high-reliability technical support for aerosol transport, stratified pollution characteristic monitoring, and quantitative research on aerosol-climate interactions.
[0195] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A method for coarse and fine mode aerosol parameter inversion based on high spectral resolution lidar, characterized in that, The method includes: Acquire the main channel scattering signal, depolarization channel elastic scattering signal, and molecular Rayleigh scattering signal of aerosols; The volume depolarization ratio is calculated based on the ratio of the main channel scattering signal to the depolarization channel elastic scattering signal. The total backscattering coefficient is calculated based on the volume depolarization ratio and the molecular Rayleigh scattering signal, and the ratio to the lidar is calculated. No lidar ratio assumptions are required, and there are no day or night time restrictions. The aerosol depolarization ratio is calculated based on the volume depolarization ratio and the total backscattering coefficient. Based on the clustering results of historical data, the characteristic depolarization ratio constants of coarse mode and fine mode are obtained respectively. Based on the characteristic depolarization ratio constants, an aerosol multi-component mixing model is constructed. The aerosol depolarization ratio is input into the multi-component mixing model to obtain the coarse mode backscattering coefficient and the fine mode backscattering coefficient. Based on the clustering results of historical data, the characteristic lidar ratio constants of coarse mode and fine mode are obtained respectively. The coarse mode extinction coefficient and the fine mode extinction coefficient are calculated based on the coarse mode backscattering coefficient and the fine mode backscattering coefficient and the characteristic lidar ratio constant. The total backscattering coefficient is calculated based on the coarse-mode backscattering coefficient and the fine-mode backscattering coefficient. The vertical range of the aerosol layer is determined based on the total backscattering coefficient. The coarse-mode extinction coefficient and the fine-mode extinction coefficient are vertically integrated within the height of the aerosol layer to obtain the coarse-mode optical thickness and the fine-mode optical thickness. The inversion is performed based on the coarse-mode optical thickness and the fine-mode optical thickness.
2. The method according to claim 1, characterized in that, The step of calculating the total backscattering coefficient and its ratio to the lidar signal based on the volume depolarization ratio and the molecular Rayleigh scattering signal includes: Determine the first and second spectral transmittances of atmospheric molecules and aerosols in the spectral discriminator. Quantitative parameters of atmospheric molecules are obtained, including at least the extinction coefficient, backscattering coefficient, depolarization ratio, and Rayleigh scattering signal; The total backscattering coefficient and the lidar ratio are calculated by combining the first spectral transmittance, the second spectral transmittance, the volume depolarization ratio, and the quantitative parameters.
3. The method according to claim 1, characterized in that, The calculation of the aerosol depolarization ratio based on the volume depolarization ratio and the total backscattering coefficient includes: Determine the backscattering coefficient and depolarization ratio of atmospheric molecules; The aerosol depolarization ratio is calculated based on the atmospheric molecule backscattering coefficient, the atmospheric molecule depolarization ratio, the volume depolarization ratio, and the total backscattering coefficient.
4. The method according to claim 1, characterized in that, The step of inputting the aerosol depolarization ratio into the multi-component mixing model to obtain the coarse-mode backscattering coefficient and the fine-mode backscattering coefficient includes: The multi-component mixing model uses the characteristic depolarization ratio constant as the basic constant to retrieve the constraint relationship of the total backscattering coefficient; A debias ratio mixing equation is constructed based on the linear mixing principle of aerosol debias ratio; By using the aerosol depolarization ratio as the independent variable and simultaneously solving the constraint relationship and the depolarization ratio mixed equation, the coarse mode backscattering coefficient and the fine mode backscattering coefficient are calculated.
5. The method according to claim 1, characterized in that, The calculation of the coarse-mode extinction coefficient and the fine-mode extinction coefficient based on the coarse-mode backscattering coefficient and the characteristic lidar ratio constant includes: Retrieve the coarse-mode characteristic lidar ratio constant and the fine-mode characteristic lidar ratio constant obtained by clustering based on historical lidar ratio data; The coarse mode extinction coefficient is calculated based on the product of the coarse mode backscattering coefficient and the coarse mode characteristic lidar ratio constant. The fine-mode extinction coefficient is calculated based on the product of the fine-mode backscattering coefficient and the fine-mode characteristic lidar constant.
6. The method according to claim 1, characterized in that, Determining the vertical range of the aerosol layer based on the total backscattering coefficient includes: By comparing the total backscattering coefficient and the atmospheric molecule backscattering coefficient at each height, the bottom and top heights of the aerosols are identified. The vertical range is determined based on the base height and the top height.
7. The method according to claim 1, characterized in that, The step of obtaining the feature debiasing ratio constants for coarse and fine modes based on historical data clustering results includes: Collect historical observation data on aerosol depolarization ratio; Statistical cluster analysis was performed on the historical observation data to divide it into a debiasing ratio data cluster for coarse-mode aerosols and a debiasing ratio data cluster for fine-mode aerosols. Statistical characteristic values of the debiasing ratio data clusters for coarse-mode aerosols and fine-mode aerosols were calculated separately, and used as characteristic debiasing ratio constants for coarse-mode aerosols and fine-mode aerosols, respectively.
8. The method according to claim 1, characterized in that, After inversion based on the coarse-mode optical thickness and the fine-mode optical thickness, the method further includes: Obtain coarse-mode optical thickness data and fine-mode optical thickness data retrieved from a solar photometer; Spatiotemporal matching of coarse-mode optical thickness, fine-mode optical thickness, coarse-mode optical thickness data, and fine-mode optical thickness data was performed, and consistency analysis was conducted. The inversion accuracy was verified based on statistical indicators.