A method for estimating aerosol mass concentration based on lidar and optical property meter

Abstract

The invention relates to an aerosol mass concentration estimation method based on a laser radar and an optical property meter. Based on collected data, the extinction coefficient, backscattering coefficient and aerosol backscattering coefficient of atmospheric molecules are calculated; based on the initial assumption of the ratio of the aerosol extinction coefficient to the backscattering coefficient, the aerosol extinction coefficient is calculated; the difference between the aerosol extinction coefficient observed by an aerosol extinction meter and the aerosol extinction coefficient calculated by a laser radar inversion is analyzed, an error convergence index is constructed, and a correction strategy for the laser radar aerosol extinction coefficient is formulated; based on the observation data of an aerosol particle size spectrometer and the corrected laser radar aerosol extinction coefficient, an addition mathematical model and a logarithmic mathematical model are combined to design a sliding window width dynamic adjustment strategy, and a relationship model between the laser radar aerosol extinction coefficient and the mass concentration that integrates the linear and nonlinear model characteristics is constructed, thereby realizing accurate estimation of the aerosol mass concentration.

Description

Technical Field

[0001] The invention relates to the field of meteorological radar detection, and in particular to an aerosol mass concentration estimation method based on laser radar and an optical property meter. Background Art

[0002] Aerosols reflect and absorb solar radiation in the atmosphere, affecting the Earth's radiation balance and energy flow, thereby affecting the climate system. By estimating the mass concentration of aerosols, we can study the impact of aerosols on cloud formation, atmospheric temperature, and precipitation patterns, and thus more accurately predict climate change. In addition, the mass concentration of aerosols (such as PM2.5 and PM10) is an important indicator of air quality. High concentrations of aerosols can have adverse effects on the environment and health, so by estimating the mass concentration of aerosols, we can monitor air pollution levels in real time and help formulate air quality reports, pollution warnings, and governance policies.

[0003] At present, there are several common methods for obtaining aerosol mass concentration. These methods can select different technical means to measure aerosol particle concentration according to needs: (1) mass method (filter membrane weighing method). This method has the advantage of high accuracy and is often used as a calibration and reference method. However, its disadvantages are that the sampling process is time-consuming, the equipment is complex, and real-time monitoring is not possible; (2) light scattering method. This method has the advantages of real-time monitoring and convenient equipment, but its disadvantages are that it is sensitive to particle size and shape. The accuracy may be affected by the refractive index of the particles and requires calibration; (3) optical particle counter. This method can simultaneously measure the number and particle size distribution of particles, but it needs to be converted into mass concentration, which results in large errors.

[0004] Based on the above analysis, it is difficult to obtain aerosol mass concentration information in real time and accurately by relying solely on the measurement of a single aerosol mass concentration or remote sensing inversion. In addition, when using the light scattering method to realize the inversion of aerosol mass concentration, the accuracy of the aerosol extinction coefficient obtained by the backscattering integration method of the laser radar is often not corrected, and the relationship model between the aerosol extinction coefficient and the aerosol mass concentration is still unclear. Therefore, it is very necessary to carry out the calibration of the laser radar aerosol extinction coefficient and the inversion of aerosol mass concentration by comprehensively utilizing observation data such as ground-based laser radar, microwave radiometer, aerosol extinction meter and aerosol particle size spectrometer observation data. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide an aerosol mass concentration estimation method based on laser radar and optical property meter, which solves the shortcomings of the prior art.

[0006] The object of the present invention is achieved by the following technical solution: an aerosol mass concentration estimation method based on laser radar and optical property meter, the estimation method comprising:

[0007] S1. Collect parameter data of ground-based lidar, ground-based microwave radiometer and optical property meter, and obtain high temporal and spatial resolution gas molecule number density N0(z), gas molecule refractive index n0(z) and atmospheric molecule extinction coefficient α based on the collected parameter data. m (z), and obtain the backscattering coefficient β of atmospheric molecules at different altitudes m (z);

[0008] S2, based on the collected parameter data and α m (z) and β m (z) Calculate the aerosol extinction coefficient α at different altitudes a_C (z), and calculate the spherical distance difference △R and time stamp distance difference △t between the ground-based lidar parameter data and the optical property instrument parameter data in three-dimensional space, and obtain the aerosol extinction coefficient α of the ground-based lidar observation after time and space matching according to the screening criteria a_L (z) and the extinction coefficient α observed by the aerosol extinction instrument a_S (z);

[0009] S3. Construct the error convergence index ER and find the α corresponding to the highest value of ER. a (z0) and β a (z0) ratio C0, and then calculate the aerosol extinction coefficient α of different altitude layers of the corrected ground-based lidar a (z);

[0010] S4. Compare the parameter data of the optical characteristic instrument with α a (z) Perform direct fitting and logarithm fitting to construct linear and nonlinear models respectively, and obtain residual C1 and C2 sequences;

[0011] S5, according to the residual C1 and C2 sequences and α a (z) Establish the average gradient change △d of C1 sequence respectively i _The average gradient change △d of C1 and C2 sequences i _C2 and α a (z) relationship model, respectively extracting a (z) △d i _C1 and △d i _The minimum absolute value of C2△d i _C, combined to get △d i _C and α a (z) relational model;

[0012] S6. According to α a (z) and the new relationship model to calculate △d i _C and mark the type of mathematical model, and calculate the mass concentration m of aerosol based on the constructed linear model or nonlinear model a (z).

[0013] The parameter data collected from the ground-based laser radar, ground-based microwave radiometer and optical property meter include:

[0014] Collect ground-based lidar detection data and observation parameters, collect ground-based microwave radiometer observation data; collect observation data from optical property instruments placed at different heights of the vertical observation tower, including aerosol extinction meters and aerosol particle size spectrometers;

[0015] Ground-based lidar detection data include echo power at different altitude layers;

[0016] Ground-based lidar observation parameters include wavelength, pulse energy, observation elevation angle, observation azimuth, temporal and spatial resolution, noise level, observation time, observation altitude, system constants, and the longitude and latitude of the instrument.

[0017] Ground-based microwave radiometer observation data include profiles of atmospheric temperature, humidity, pressure, etc. at different altitudes with high temporal and spatial resolutions;

[0018] The aerosol extinction instrument observation data include information such as the aerosol extinction coefficient measured at the same wavelength as the ground-based lidar, the observation time, the observation altitude, and the latitude and longitude of the instrument;

[0019] The aerosol particle size spectrometer observation data includes the observed aerosol mass concentration data at different altitudes.

[0020] The parameter data collected by the method can obtain the gas molecule number density N0(z), the refractive index n0(z) of the gas molecules and the extinction coefficient α of the atmospheric molecules with high temporal and spatial resolution. m (z), and obtain the backscattering coefficient β of atmospheric molecules at different altitudes m (z) include:

[0021] The collected high-temporal-resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z are input into the ideal gas law equation to obtain the high-temporal-resolution gas molecule number density N0(z);

[0022] The high temporal and spatial resolution atmospheric temperature T(z) and pressure P(z) profiles collected at different altitudes z are input into the empirical formula of atmospheric refractive index to obtain the refractive index n0(z) of gas molecules;

[0023] The collected high-temporal and spatial resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z, the laser wavelength λ, and the calculated N0(z) and n0(z) are input into the Rayleigh scattering molecular extinction coefficient calculation formula to obtain the atmospheric molecular extinction coefficient α m (z);

[0024] The main component ratio of the gas is set to a, and the collected ground-based lidar observation parameters, the high temporal and spatial resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z, and the calculated N0(z) and n0(z) are input into the Rayleigh scattering model to obtain the atmospheric molecular backscattering coefficients β at different altitudes. m (z).

[0025] The S2 includes:

[0026] Initial setup of ground-based lidar to measure the extinction coefficient α of the most distant aerosol a (z0) and the backscatter coefficient β a The ratio of (z0) is C0, and the collected ground-based lidar system constant C and the echo power P(z) at different altitudes, as well as the calculated α m (z) and β m (z) are input into the backward integration solver of the ground-based lidar equation. After iterative calculation until the result converges, the aerosol extinction coefficient α at different altitudes is derived. a_C (z);

[0027] According to the collected ground-based lidar observation altitude, longitude and latitude, observation time, aerosol extinction coefficient, and the calculated aerosol extinction coefficient α at different altitudes a_C (z) Calculate the spherical distance difference △R and time stamp distance difference △t between the ground-based lidar observation data and the aerosol extinction instrument observation data in three-dimensional space respectively, and calculate the spherical distance difference less than d max And the timestamp distance difference is less than t max The screening criteria are matched in time and space to obtain the aerosol extinction coefficient α observed by the ground-based lidar after time and space matching. a_L (z) and the extinction coefficient α observed by the aerosol extinction instrument a_S (z).

[0028] The S3 includes:

[0029] Statistics α a_L (z) and α a_S (z) similarity index R and error index MAE and RMSE, construct the error convergence index ER = R-|MAE|-RMSE;

[0030] The initial proportional value C0 is gradually increased at the same interval, and the ER index under different C0 conditions is calculated respectively. The C0 corresponding to the highest value of the ER index is found, and the C0 at this time is substituted into the backward integral solver of the lidar equation to calculate the corrected aerosol extinction coefficient α of different altitude layers of the ground-based lidar. a (z).

[0031] The S4 includes:

[0032] The aerosol mass concentration ma(z) at different altitudes observed by the aerosol size spectrometer is compared with the aerosol extinction coefficient α at different altitudes obtained by the calibrated ground-based lidar. a(z) , matching time and space according to the method of time-space matching;

[0033] Based on the additive mathematical model, m a (z) and α a (z) is fitted to construct m a (z) and α a (z) A specific linear model is used to obtain the residual C1 sequence;

[0034] Based on the logarithmic mathematical model, m a (z) and α a (z) Take the logarithm and perform fitting processing to construct m a (z) and α a (z) The specific nonlinear model is used to obtain the residual C2 sequence.

[0035] The S5 includes:

[0036] According to the residual C1 sequence C1 and the corresponding α in the obtained linear model a(z) , calculate the global variance δ_g of the residual C1 sequence, set the initial width of the sliding window to W0 and the moving step of the sliding window to 1, calculate the local variance δ_i of the residual C1 sequence in the window, and reset the window width W according to the dynamic adjustment strategy of the window width i ;

[0037] Get the window width W i The residual C1 sequence in the window is linearly fitted to obtain the average gradient change △d of the C1 sequence in the window. i _C1, thus establishing △d i _C1 and α a (z) relational model;

[0038] The same method is used to process the residual C2 and the corresponding α in the nonlinear model a(z) , establish △d i _C2 and α a (z) relational model;

[0039] Compare △d i _C1 and △d i The absolute value of _C2 is extracted at different α a (z) △d i _C1 and △d i _The minimum absolute value of C2△d i _C, thus forming a new △d i _C and α a (z), and mark each segment △d i _C corresponds to a linear model or a nonlinear model.

[0040] The S6 includes:

[0041] The corrected aerosol extinction coefficient α at different altitudes is a (z) are input into the combined △d i _C and α a (z), calculate △d i _C and mark the category of mathematical model;

[0042] According to the marked mathematical model category, take out the corresponding established linear model or nonlinear model and set α a (z) and △d i _C are input into the model to calculate the mass concentration of aerosol m a (z).

[0043] The linear model includes: a (z)=A1×α a (z)+C1; the nonlinear model includes: ln[m a (z)]=A2×ln[α a (z)]+C2.

[0044] The window width dynamic adjustment strategy is W i = W0×(δ_g / δ_i).

[0045] The present invention has the following advantages: a method for estimating aerosol mass concentration based on laser radar and optical property meter, based on the temperature and temperature isoprofiles observed by microwave radiometer, realizes the calculation of gas molecule number density and refractive index, thereby combining with the Rayleigh scattering molecular extinction coefficient calculation model to realize accurate estimation of atmospheric molecular extinction coefficient; based on laser radar observation parameters and microwave radiometer observation data, the Rayleigh scattering model is used to calculate the atmospheric molecular backscattering coefficient; based on the calculated atmospheric molecular backscattering coefficient and the estimated atmospheric molecular extinction coefficient, combined with the laser radar echo signal power, the atmospheric molecular backscattering coefficient is estimated. , the laser radar backward integrator was used to estimate the aerosol extinction coefficient; based on the difference between the aerosol extinction coefficient observed by the aerosol extinction meter and the aerosol extinction coefficient estimated by the laser radar, an error convergence index was constructed, and the correction of the laser radar aerosol extinction coefficient was achieved through multiple iterations; based on the observation data of the aerosol particle size spectrometer and the corrected aerosol extinction coefficient, a dynamic adjustment strategy for the sliding window width was designed, and a relationship model between the laser radar aerosol extinction coefficient and mass concentration that integrated the characteristics of the linear and nonlinear models was constructed, thus achieving accurate estimation of the aerosol mass concentration. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the process of the present invention;

[0047] Figure 2 The present invention is a flowchart for calculating the atmospheric molecule extinction coefficient;

[0048] Figure 3 A flowchart for calculating the atmospheric molecule backscattering coefficient of the present invention;

[0049] Figure 4 The present invention is a flowchart for calculating the aerosol extinction coefficient of the laser radar;

[0050] Figure 5 This is a flow chart for realizing the correction of the laser radar aerosol extinction coefficient of the present invention;

[0051] Figure 6 The figure is a flow chart for realizing the aerosol mass concentration estimation of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided below in conjunction with the drawings is not intended to limit the scope of protection of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application. The present invention is further described below in conjunction with the drawings.

[0053] like Figure 1 and Figure 6 As shown, the present invention specifically relates to an aerosol mass concentration estimation method based on a laser radar and an optical property meter, which specifically includes the following contents:

[0054] Step 1: Collect ground-based lidar detection data and observation parameters; collect ground-based microwave radiometer observation data; collect observation data from optical property instruments placed at different heights of the vertical observation tower, including aerosol extinction meters and aerosol particle size spectrometers;

[0055] Among them, the ground-based lidar detection data mainly refers to the echo power at different altitude layers;

[0056] The ground-based lidar observation parameters mainly refer to the wavelength of 532 nm, pulse energy, observation elevation angle, observation azimuth, temporal and spatial resolution, noise level, observation time, observation altitude, system constants and the longitude and latitude of the instrument.

[0057] The ground-based microwave radiometer observation data mainly refers to the atmospheric temperature, humidity, pressure and other profiles with high temporal and spatial resolution at different altitudes;

[0058] The aerosol extinction instrument observation data mainly refers to the aerosol extinction coefficient measured at a wavelength of 532 nm, the observation time, the observation altitude, and the longitude and latitude of the instrument;

[0059] Aerosol particle size spectrometer observation data mainly refers to aerosol mass concentration data;

[0060] Step 2: If Figure 2As shown, the high-temporal-resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z collected in step 1 are input into the ideal gas law equation to obtain the high-temporal-resolution gas molecule number density N0(z); the high-temporal-resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z collected in step 1 are input into the empirical formula of atmospheric refractive index to obtain the refractive index n0(z) of the gas molecules. The high-temporal-resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z collected in step 1, the laser wavelength λ, and the calculated N0(z) and n0(z) are input together into the Rayleigh scattering molecular extinction coefficient calculation formula to obtain the atmospheric molecular extinction coefficient α m (z).

[0061] Among them, the ideal gas equation is: N0(z)=P(z) / (K B ×T(z)), where K B is the Boltzmann constant.

[0062] The empirical formula for atmospheric refractive index is: n0(z)=1+77.6×P(z) / (T(z)×10 6 ).

[0063] The formula for calculating the extinction coefficient of Rayleigh scattering molecules is:

[0064]

[0065] Step 3: If Figure 3 As shown in Figure 1, assuming that the ratio of the main components of the gas is 3.71:1, the ground-based lidar observation parameters collected in step 1, the high temporal and spatial resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z, and N0(z) and n0(z) calculated in step 2 are input into the Rayleigh scattering model to obtain the atmospheric molecular backscattering coefficients β at different altitudes. m (z).

[0066] Step 4: Figure 4 As shown, it is preliminarily assumed that the ground-based lidar measures the extinction coefficient α of the farthest aerosol a (z0) and the backscatter coefficient β a The ratio of (z0) is C0=20~40 sr (sr is steradian), and the ground-based lidar system constant C and the echo power P(z) at different altitudes collected in step 1, as well as α calculated in steps 2 and 3 are m (z) and β m (z) are input into the backward integration solver of the ground-based lidar equation. After iterative calculation until the result converges, the aerosol extinction coefficient α at different altitudes is derived. a_C (z).

[0067] Step 5: Based on the ground-based lidar observation altitude, longitude and latitude, observation time, aerosol extinction coefficient, and aerosol extinction coefficient α of different altitude layers calculated in step 4, the aerosol extinction coefficient α of different altitude layers calculated in step 4 is calculated. a_C (z) Calculate the spherical distance difference △R and timestamp distance difference △t between the ground-based lidar observation data and the aerosol extinction instrument observation data in three-dimensional space. According to the screening criteria of spherical distance difference less than 30 meters and timestamp distance difference less than 1 minute, time and space matching is performed to obtain the aerosol extinction coefficient α of the ground-based lidar observation after time and space matching. a_L (z) and the extinction coefficient α observed by the aerosol extinction instrument a_S (z);

[0068] Step 6: Figure 5 As shown, the α of statistical step 5 is a_L (z) and α a_S The similarity index R and error indexes MAE and RMSE of (z) are used to construct the error convergence index ER=R-|MAE|-RMSE. The ratio value C0 initially set in step 4 is gradually increased at the same interval, and the ER index under different C0 conditions is calculated respectively. Find the C0 corresponding to the highest value of the ER index, and substitute the C0 at this time into the backward integral solver of the ground-based lidar equation to calculate the corrected aerosol extinction coefficient α of different altitude layers of the ground-based lidar a (z).

[0069] Step 7: Compare the aerosol mass concentration ma(z) at different altitudes observed by the aerosol particle size spectrometer in step 1 with the aerosol extinction coefficient α of different altitudes of the ground-based lidar obtained after correction in step 6. a(z) , and perform time and space matching according to the time and space matching method in step 5. Based on the additive mathematical model, m a (z) and α a (z) is fitted to construct m a (z) and α a (z) specific linear model, we can get the residual C1 sequence; based on the logarithmic mathematical model, m a (z) and α a (z) Take the logarithm and perform fitting processing to construct m a (z) and α a (z) The specific nonlinear model can obtain the residual C2 sequence;

[0070] Among them, the linear model refers to: m a (z)=A1×α a (z)+C1, A1 are the fitting coefficients;

[0071] Nonlinear model refers to: ln[m a (z)]=A2×ln[α a (z)]+C2, A2 is the fitting coefficient;

[0072] Step 8: Take out the residual C1 sequence and the corresponding α in the linear model obtained in step 7 a(z) , calculate the global variance δ_g of the residual C1 sequence. Set the initial width of the sliding window to W0 and the moving step of the sliding window to 1, calculate the local variance δ_i of the residual C1 sequence in the window, and reset the window width W according to the dynamic adjustment strategy of the window width i . Get the window width W i The residual C1 sequence in the window is processed by linear fitting to obtain the average gradient change △d of the C1 sequence in the window. i _C1, thus establishing △d i _C1 and α a The same method is used to process the residual C2 and the corresponding α in the nonlinear model. a(z) , we can establish △d i _C2 and α a (z) relationship model. Compare △d i _C1 and △d i The absolute value of _C2 is extracted at different α a (z) △d i _C1 and △d i _The smaller value of the absolute value of C2△d i _C, thus forming a new △d i _C and α a (z), and mark each segment △d i _C corresponds to a linear model or a nonlinear model.

[0073] Among them, the window width dynamic adjustment strategy is W i = W0×(δ_g / δ_i).

[0074] Step 9: The corrected aerosol extinction coefficient α of different altitude layers obtained in step 6 a (z) are input into the △d formed in step 8 i _C and α a (z), calculate △d i _C and mark the type of mathematical model (linear model or nonlinear model). According to the marked type of mathematical model, take out the corresponding established linear model or nonlinear model from step 7 and replace α a (z) and △d i_C is input into the model to calculate the mass concentration of aerosol m a (z).

[0075] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used for various other combinations, modifications and improvements, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for estimating aerosol mass concentration based on laser radar and optical property meter, characterized in that: The estimation method includes: S1. Collect parameter data of ground-based lidar, ground-based microwave radiometer and optical property meter, and obtain high temporal and spatial resolution gas molecule number density N0(z), gas molecule refractive index n0(z) and atmospheric molecule extinction coefficient a based on the collected parameter data. m (z), and obtain the backscattering coefficient b of atmospheric molecules at different altitudes m (z); S2, based on the collected parameter data and a m (z) and b m (z) Calculate the aerosol extinction coefficient a at different altitudes a_C (z), and calculate the spherical distance difference △R and time stamp distance difference △t between the ground-based lidar parameter data and the optical property instrument parameter data in three-dimensional space, and obtain the aerosol extinction coefficient a of the ground-based lidar observation after time and space matching according to the screening criteria a_L (z) and the extinction coefficient a observed by the aerosol extinction instrument a_S (z); S3. Construct the error convergence index ER and find the a corresponding to the maximum value of ER. a (z0) and b a (z0) ratio C0, and then calculate the aerosol extinction coefficient a of different altitude layers of the corrected ground-based lidar a (z); S4. Compare the parameter data of the optical characteristic instrument with a a (z) Perform direct fitting and logarithm fitting to construct linear and nonlinear models respectively, and obtain residual C1 and C2 sequences; S5, according to the residual C1 and C2 sequences and a a (z) Establish the average gradient change △d of C1 sequence respectively i _The average gradient change △d of C1 and C2 sequences i _C2 and a a (z) relational model, respectively extracting a (z) △d i _C1 and △d i _The minimum absolute value of C2△d i _C, combined to get △d i _C and a a (z) relational model; S6. According to a a (z) and the new relationship model to calculate △d i _C and mark the type of mathematical model, and calculate the mass concentration m of aerosol based on the constructed linear model or nonlinear model a (z); The S2 includes: Initial setup of ground-based lidar to measure the extinction coefficient a of the most distant aerosol a (z0) and the backscattering coefficient b a The ratio of (z0) is C0, and the collected ground-based lidar system constant C and the echo power P(z) at different altitudes, as well as the calculated a m (z) and b m (z) are input into the backward integration solver of the ground-based lidar equation, and the aerosol extinction coefficient a at different altitudes is derived through iterative calculation until the result converges. a_C (z); According to the collected ground-based lidar observation altitude, longitude and latitude, observation time, aerosol extinction coefficient, and the calculated aerosol extinction coefficient a at different altitudes, a_C (z) Calculate the spherical distance difference △R and time stamp distance difference △t between the ground-based lidar observation data and the aerosol extinction instrument observation data in three-dimensional space respectively, and calculate the spherical distance difference less than d max And the timestamp distance difference is less than t max The screening criteria are matched in time and space to obtain the aerosol extinction coefficient a observed by the ground-based lidar after time and space matching. a_L (z) and the extinction coefficient a observed by the aerosol extinction instrument a_S (z); The S3 includes: Statisticsa a_L (z) and a a_S (z) similarity index R and error index MAE and RMSE, construct the error convergence index ER = R-|MAE|-RMSE; The initial proportional value C0 is gradually increased at the same interval, and the ER index under different C0 conditions is calculated respectively. The C0 corresponding to the highest value of the ER index is found, and the C0 at this time is substituted into the backward integral solver of the ground-based lidar equation to calculate the corrected aerosol extinction coefficient a of different altitude layers of the ground-based lidar. a (z).

2. The aerosol mass concentration estimation method based on laser radar and optical property meter according to claim 1 is characterized in that: The parameter data collected from the ground-based laser radar, ground-based microwave radiometer and optical property meter include: Collect ground-based lidar detection data and observation parameters, collect ground-based microwave radiometer observation data; collect observation data from optical property instruments placed at different heights of the vertical observation tower, including aerosol extinction meters and aerosol particle size spectrometers; Ground-based lidar detection data include echo power at different altitude layers; Ground-based lidar observation parameters include wavelength, pulse energy, observation elevation angle, observation azimuth, temporal and spatial resolution, noise level, observation time, observation altitude, system constants, and data on the latitude and longitude of the instrument; Ground-based microwave radiometer observation data include profiles of atmospheric temperature, humidity, and pressure at different altitudes with high temporal and spatial resolutions; The aerosol extinction instrument observation data include the aerosol extinction coefficient measured at the same wavelength as the ground-based lidar, the observation time, the observation altitude and the latitude and longitude of the instrument; The aerosol particle size spectrometer observation data includes the observed aerosol mass concentration data at different altitudes.

3. The aerosol mass concentration estimation method based on laser radar and optical property meter according to claim 2 is characterized in that: The method obtains the gas molecule number density N0(z), the refractive index n0(z) of the gas molecules and the atmospheric molecule extinction coefficient a with high temporal and spatial resolution according to the collected parameter data. m (z), and obtain the backscattering coefficient b of atmospheric molecules at different altitudes m (z) include: The collected high-temporal-resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z are input into the ideal gas law equation to obtain the high-temporal-resolution gas molecule number density N0(z); The high temporal and spatial resolution atmospheric temperature T(z) and pressure P(z) profiles collected at different altitudes z are input into the empirical formula of atmospheric refractive index to obtain the refractive index n0(z) of gas molecules; The collected high-temporal and spatial resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z, the laser wavelength l, and the calculated N0(z) and n0(z) are input into the Rayleigh scattering molecular extinction coefficient calculation formula to obtain the atmospheric molecular extinction coefficient a m (z); The main component ratio of the gas is set to a, and the collected ground-based lidar observation parameters, the high temporal and spatial resolution atmospheric temperature T(z) and pressure P(z) profiles at different altitudes z, and the calculated N0(z) and n0(z) are input into the Rayleigh scattering model to obtain the atmospheric molecular backscattering coefficients b at different altitudes. m (z).

4. The aerosol mass concentration estimation method based on laser radar and optical property meter according to claim 1 is characterized in that: The S4 includes: The aerosol mass concentration ma(z) at different altitudes observed by the aerosol size spectrometer is compared with the aerosol extinction coefficient a at different altitudes obtained by the calibrated ground-based lidar. a(z) , matching time and space according to the method of time-space matching; Based on the additive mathematical model, m a (z) and a a (z) is fitted to construct m a (z) and a a (z) A specific linear model is used to obtain the residual C1 sequence; Based on the logarithmic mathematical model, m a (z) and a a (z) Take the logarithm and perform fitting processing to construct m a (z) and a a (z) The specific nonlinear model is used to obtain the residual C2 sequence.

5. The aerosol mass concentration estimation method based on laser radar and optical property meter according to claim 4 is characterized in that: The S5 includes: According to the residual C1 sequence C1 and the corresponding a in the obtained linear model a(z) , calculate the global variance δ_g of the residual C1 sequence, set the initial width of the sliding window to W0 and the moving step of the sliding window to 1, calculate the local variance δ_i of the residual C1 sequence in the window, and reset the window width W according to the dynamic adjustment strategy of the window width i ; Get the window width W i The residual C1 sequence in the window is linearly fitted to obtain the average gradient change △d of the C1 sequence in the window. i _C1, thus establishing △d i _C1 and a a (z) relational model; The same method is used to process the residual C2 and the corresponding a in the nonlinear model a(z) , establish △d i _C2 and a a (z) relational model; Compare △d i _C1 and △d i _C2 absolute value, respectively extract the a (z) △d i _C1 and △d i _The minimum absolute value of C2△d i _C, thus forming a new △d i _C and a a (z), and mark each segment △d i _C corresponds to a linear model or a nonlinear model.

6. The aerosol mass concentration estimation method based on laser radar and optical property meter according to claim 5 is characterized in that: The S6 includes: The corrected aerosol extinction coefficient a of different altitude layers is obtained a (z) are input into the combined △d i _C and a a (z), calculate △d i _C and mark the category of mathematical model; According to the marked mathematical model category, take out the corresponding established linear model or nonlinear model and convert a a (z) and △d i _C are input into the model to calculate the mass concentration of aerosol m a (z).

7. The aerosol mass concentration estimation method based on laser radar and optical property meter according to claim 4 is characterized in that: The linear model includes: a (z)=A1´a a (z)+C1; the nonlinear model includes: ln[m a (z)]=A2´ln[a a (z)]+C2, where A1 and A2 are fitting coefficients.

8. The aerosol mass concentration estimation method based on laser radar and optical property meter according to claim 5, characterized in that: The window width dynamic adjustment strategy is W i = W0´(δ_g / δ_i).

Images