METHOD FOR INVERTING AEROSOL COMPONENTS USING LIDAR RATIO AND DEPOLARIZATION RATIO

Abstract

The present disclosure discloses a method for inverting aerosol components using a LiDAR ratio and a depolarization ratio, including: S1. identifying sand dust, a spherical aerosol and a mixture of the sand dust and the spherical aerosol based on a depolarization ratio; S2. calculating a proportion of the sand dust in the mixture of the sand dust and the spherical aerosol; and S3. identifying soot and a water-soluble aerosol in the spherical aerosol based on a LiDAR ratio. In the present disclosure, only a wavelength with a polarization channel is needed, to identify the aerosol components, achieving high accuracy with low detection costs.

Claims

1. A method for inverting aerosol components using a LiDAR ratio and a depolarization ratio, comprising: S1. identifying sand dust, a spherical aerosol and a mixture of the sand dust and the spherical aerosol based on a depolarization ratio; S2. calculating a proportion of the sand dust in the mixture of the sand dust and the spherical aerosol; and S3. identifying soot and a water-soluble aerosol in the spherical aerosol based on a LiDAR ratio, wherein step S3 comprises: S31. establishing a lookup table 1 for the extinction coefficient σ with respect to an extinction coefficient σ.sub.ws of a water-soluble aerosol and an extinction coefficient σ.sub.st of soot, and establishing an additional lookup table 2 with respect to a case that the extinction coefficient is greater than 1; S32. combining the lookup table 1 and the lookup table 2 for the extinction coefficient with the LiDAR ratio to establish a lookup table 3 for a backscattering coefficient β: $\beta =\frac{{\sigma }_{ws}}{S_1}+\frac{{\sigma }_{st}}{S_2}$ S33. if 0<σ.sub.sphere≤1, traversing the lookup table 1, or if 1<σ.sub.sphere≤3, traversing the lookup table 2, wherein σ.sub.sphere is an extinction coefficient for the spherical aerosol, σ.sub.sphere is the backscattering coefficient; and retrieving an array of extinction coefficients from the lookup table 3 whose errors relative to observed values meet a standard: $.Math.\frac{{\sigma }_{\mathrm{sphere}}-\sigma }{{\sigma }_{\mathrm{sphere}}}.Math.<0.01$ $.Math.\frac{{\beta }_{\mathrm{sphere}}-\beta }{{\beta }_{\mathrm{sphere}}}.Math.<0.01$ and S34. if the lookup table 3 does not match the observed values, selecting as an optimal solution a solution ensuring a minimum deviation between an observed value and a theoretical value, that is, a solution meeting the following condition: $\min (\sqrt{{.Math.\frac{{\sigma }_{\mathrm{sphere}}-\sigma }{{\sigma }_{\mathrm{sphere}}}.Math.}^2+{.Math.\frac{{\beta }_{\mathrm{sphere}}-\beta }{{\beta }_{\mathrm{sphere}}}.Math.}^2)}.$

2. The method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to claim 1, wherein step S1 comprises: calculating an aerosol depolarization ratio (ADR), wherein when the ADR is greater than 0.31, the aerosol components are considered to be the sand dust; when 0.05≤ADR≤0.31, the aerosol components are considered to be the mixture of the sand dust and the spherical aerosol; or when the ADR is less than 0.05, the aerosol components are considered to be the spherical aerosol.

3. The method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to claim 2, wherein the calculating the ADR comprises: directing a LiDAR signal through a polarizer with a 45° polarization direction at a 532 nm signal using a NIES Miescattering LiDAR system, to separate a horizontal signal from a vertical signal.

4. The method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to claim 1, wherein step S2 comprises: S21. setting depolarization ratios of the sand dust and the spherical aerosol as δ.sub.1 and δ.sub.2 respectively, and defining the depolarization ratios of the sand dust and the spherical aerosol as δ.sub.i=P.sub.i⊥/P.sub.i∥, wherein P.sub.i⊥ and P.sub.i∥ are vertical and horizontal polarization components of an aerosol backscattered signal respectively; S22. defining δ.sub.i′=P.sub.i⊥/(P.sub.i⊥+P.sub.i∥), then δ.sub.i′=δ.sub.i/(δ.sub.i+1); and S23. assuming that x represents an optical proportion of the sand dust in an aerosol mixture, wherein a polarization component of the backscattered signal is expressed as:
P.sub.⊥=[xδ.sub.1′+(1−x)δ.sub.2′]P
P.sub.∥=[x(1−δ.sub.1′)+(1−x)(1−δ.sub.2′)]P wherein P=P.sub.⊥+P.sub.∥, and therefore, the aerosol depolarization ratio δ is expressed as: $\delta =\frac{P_{\perp }}{P_{.Math.}}=\frac{x{\delta }_1^{\prime }+\left(1-x\right){\delta }_2^{\prime }}{x\left(1-{\delta }_1^{\prime }\right)+\left(1-x\right)\left(1-{\delta }_2^{\prime }\right)}$ and therefore, the optical proportion of the sand dust x can be calculated using an equation: $x=\frac{\left(\delta -{\delta }_2\right)\left(1+{\delta }_1\right)}{\left(1+\delta \right)\left({\delta }_1-{\delta }_2\right)}$

5. The method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to claim 4, wherein the aerosol depolarization ratio δ is obtained through the following steps: assuming $R=\frac{{\beta }_1+{\beta }_2}{{\beta }_2},$ wherein β.sub.1 represents a backscattering coefficient of an aerosol particle, and β.sub.2 represents a backscattering coefficient of an atmospheric molecule; and $\delta =\frac{\left(1+{\delta }_m\right){\delta }_{v}R-\left(1+{\delta }_v\right){\delta }_m}{\left(1+{\delta }_m\right)R-\left(1+{\delta }_v\right)}$ substituting R into the equation: $\delta =\frac{\left(1+{\delta }_m\right){\delta }_{v}R-\left(1+{\delta }_v\right){\delta }_m}{\left(1+{\delta }_m\right)R-\left(1+{\delta }_v\right)}$ wherein δ.sub.v represents a signal depolarization ratio, and δ.sub.m represents a molecule depolarization ratio.

6. The method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to claim 5, wherein δ.sub.1 is obtained through the following steps: measuring the Miescattering caused by the aerosol and the Rayleigh scattering caused by the atmospheric molecule separately, wherein a LiDAR equation is expressed as
P(z)=CP.sub.0z.sup.−2[β.sub.1(z)+β.sub.2(z)]exp[−2∫.sub.0.sup.zσ.sub.1(z)dz]exp[−2∫.sub.0.sup.zσ.sub.2(z)dz] wherein C represents a radar correction constant; P.sub.0 represents radar transmission power; and β represents the backscattering coefficient, σ represents an extinction coefficient, and subscripts of 1 and 2 represent the aerosol particle and the atmospheric molecule respectively; assuming that a relationship between the extinction coefficient σ and the backscattering coefficient β as follows:
S=σ/β and solving the foregoing LiDAR equation gives: ${\beta }_1=-{\beta }_2+\frac{X\left(z\right)\exp \left[-2\left(S_1-S_2\right){\int }_{z_c}^z{\beta }_2\left(z\right)dz\right]}{\frac{X\left(z_c\right)}{{\beta }_1\left(z_c\right)+{\beta }_2\left(z_c\right)}-2S_1{\int }_{z_c}^{z}P\left(z\right)z^2\exp \left[-2\left(S_1-S_2\right){\int }_{z_c}^z{\beta }_2\left(z^{\prime }\right){\mathrm{dz}}^{\prime }\right]\mathrm{dz}}$ wherein X(z)=P(z)z.sup.2; and β.sub.1(z.sub.c) and β.sub.2(z.sub.c) are boundary values at a far end z.sub.c.

7. The method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to claim 1, wherein in step S33, the extinction coefficient σ.sub.sphere of the spherical aerosol is equal to (1−x)σ.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0035] FIG. 1 is a flowchart of a method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to the present disclosure;

[0036] FIG. 2 is a diagram of testing a heavily polluted weather in a method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to the present disclosure; and

[0037] FIG. 3 is a diagram of testing a clear weather in a method for inverting aerosol components using a LiDAR ratio and a depolarization ratio according to the present disclosure.

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] The specific embodiments of the present disclosure are further described in detail with reference to the accompanying drawings. It should be noted here that the description of these embodiments is intended to facilitate understanding of the present disclosure, but does not constitute a limitation to the present disclosure. Further, the technical features involved in the various embodiments of the present disclosure described below may be combined with each other as long as they do not conflict with each other.

[0039] As shown in FIG. 1, a method for inverting aerosol components using a LiDAR ratio and a depolarization ratio is provided, including steps of: S1. identifying sand dust, a spherical aerosol and a mixture of the sand dust and the spherical aerosol based on a depolarization ratio; S2. calculating a proportion of the sand dust in the mixture of the sand dust and the spherical aerosol; and S3. identifying soot and a water-soluble aerosol in the spherical aerosol based on a LiDAR ratio.

[0040] In a Fernald inversion method, Miescattering caused by the aerosol and Rayleigh scattering caused by an atmospheric molecule are measured separately (Fernald, 1984), and therefore, a LiDAR equation may be expressed as:

P(z)=CP.sub.0z.sup.−2[β.sub.1(z)+β.sub.2(z)]exp[−2∫.sub.0.sup.zσ.sub.1(z)dz]exp[−2∫.sub.0.sup.zσ.sub.2(z)dz]   (8)

where C represents a radar correction constant; P.sub.0 represents radar transmission power; and β represents the backscattering coefficient, σ represents an extinction coefficient, and subscripts of 1 and 2 represent the aerosol particle and the atmospheric molecule respectively.

[0041] To solve the equation, a relationship between the extinction coefficient and the backscattering coefficient needs to be assumed, that is, S=σ/β. In this case, the equation (8) can be solved:

[00015]$\begin{array}{cc}{\beta }_1=-{\beta }_2+\frac{X\left(z\right)\exp \left[-2\left(S_1-S_2\right){\int }_{z_c}^z{\beta }_2\left(z\right)\mathrm{dz}\right]}{\begin{array}{c}\frac{X\left(z_c\right)}{{\beta }_1\left(z_c\right)+{\beta }_2\left(z_c\right)}-\\ 2S_1{\int }_{z_c}^{z}P\left(z\right)z^2\exp \left[-2\left(S_1-S_2\right){\int }_{z_c}^z{\beta }_2\left(z^{\prime }\right){\mathrm{dz}}^{\prime }\right]\mathrm{dz}\end{array}}& \left(9\right)\end{array}$

where X(z)=P(z)z.sup.2; and β.sub.1(z.sub.c) and β.sub.2(z.sub.c) are boundary values at a far end z.sub.c separately. Herein,

[00016]$S_2=\frac{8\pi }{3}$

is a fixed value for the atmospheric molecule. In addition, β.sub.2 is known for an atmospheric model and meteorological observation. Herein, it is assumed that the LiDAR ratio S.sub.1 of the aerosol is an empirical value of 50 sr. When optical data of the aerosol provided by AERONET is accurate, the selected LiDAR ratio of the aerosol can also be adjusted based on a calculation result. Therefore, the extinction coefficient of the aerosol is calculated using the equation σ.sub.1=50β.sub.1.

[0042] The aerosol depolarization ratio (ADR) is different from the signal depolarization ratio (SDR), and the SDR also covers Rayleigh scattering of the atmospheric molecule. The ADR other than the SDR is required for an aerosol attribute characterized by LiDAR data at a low concentration. The aerosol depolarization ratio can be derived from the signal depolarization ratio and the backscattering coefficient, it is assumed that

[00017]$R=\frac{{\beta }_1+{\beta }_2}{{\beta }_2},$

and therefore,

[00018]$\delta =\frac{\left(1+{\delta }_m\right){\delta }_{v}R-\left(1+{\delta }_v\right){\delta }_m}{\left(1+{\delta }_m\right)R-\left(1+{\delta }_v\right)}$

[0043] Substituting R into the foregoing equation gives:

[00019]$\begin{array}{cc}\delta =\frac{\left(1+{\delta }_m\right){\delta }_{\nu }R-\left(1+{\delta }_{\nu }\right){\delta }_m}{\left(1+{\delta }_m\right)R-\left(1+{\delta }_{\nu }\right)}& \left(10\right)\end{array}$

where δ.sub.v represents the signal depolarization ratio; δ represents the aerosol depolarization ratio; and δ.sub.m represents the molecular depolarization ratio. Herein, 0.0044 is used for calculation. Herein, δ.sub.1 is obtained in the foregoing Fernald inversion method; and β.sub.2 of the atmospheric molecule can be calculated based on meteorological observation data. It should be noted that the ADR is vulnerable to noise in case of a small backscattering coefficient of the aerosol.

[0044] In the step of separating the sand dust from the spherical aerosol, it is considered that the sand dust is mixed with an external part of the spherical aerosol, and aerosol depolarization ratios are δ.sub.1 and β.sub.2 separately. Herein, the depolarization ratios of the two components are defined as δ.sub.i=P.sub.i⊥/P.sub.i∥, where P.sub.i⊥ and P.sub.i∥ are vertical and horizontal polarization components of an aerosol backscattered signal respectively. If it is defined that δ.sub.i′=P.sub.i⊥/(P.sub.i⊥+P.sub.i∥), δ.sub.i′=δ.sub.i/(δ.sub.i+1). It should be noted that these parameters are all functions of height. If it is assumed that x represents an optical proportion of the sand dust in an aerosol mixture, a polarization component of the backscattered signal can be expressed as:

P.sub.⊥=[xδ.sub.1′+(1−x)δ.sub.2′]P   (11)

P.sub.∥=[x(1 −δ.sub.1′)+(1−x)(1−δ.sub.2′)]P   (12)

where P=P.sub.⊥+P.sub.81, and therefore, the aerosol depolarization ratio δ can be expressed as:

[00020]$\begin{array}{cc}\delta =\frac{P\perp }{P_{.Math.}}=\frac{x{\delta }_1^{\prime }+\left(1-x\right){\delta }_2^{\prime }}{x\left(1-{\delta }_1^{\prime }\right)+\left(1-x\right)\left(1-{\delta }_2^{\prime }\right)}& \left(13\right)\end{array}$

and therefore, the optical proportion of the sand dust x can be calculated using an equation:

[00021]$\begin{array}{cc}x=\frac{\left(\delta -{\delta }_2\right)\left(1+{\delta }_1\right)}{\left(1+\delta \right)\left({\delta }_1-{\delta }_2\right)}& \left(14\right)\end{array}$

[0045] It should be noted that δ represents the aerosol depolarization ratio calculated in the equation (10). As shown in Table 1, the depolarization ratio δ.sub.1 of the sand dust and the depolarization ratio δ.sub.2 of the spherical aerosol are 0.31 and 0.05 separately. The extinction coefficient σ.sub.ds of the sand dust can be expressed as xσ, and the extinction coefficient σ.sub.sphere of the spherical aerosol is expressed as (1−x)σ, where σ is the extinction coefficient of the aerosol calculated in the foregoing Fernald method.

[0046] Because the water-soluble aerosol S.sub.1 (47 sr) and the soot S.sub.2 have significantly different LiDAR ratios (85 sr) as calculated in the Miescattering theory, that is, the two components have significantly different ratios of extinction coefficients to backscattering coefficients, which can be used for distinguishing between the two components. Specific operations are as follows:

[0047] (1) Establish a lookup table 1 for the extinction coefficient σ with respect to an extinction coefficient σ.sub.ws of the water-soluble aerosol and an extinction coefficient σ.sub.st of the soot. The extinction coefficient ranges from 0 km.sup.−1 to 1 km.sup.−1, and an interval step of the extinction coefficient of the component is 0.001 km.sup.−1. In addition, in consideration of an actual situation, the extinction coefficient does not strictly ranges from 0 km.sup.−1 to 1 km.sup.−1 under a pollution condition. Therefore, it is necessary to establish an additional lookup table 2 in a case that the extinction coefficient is greater than 1, an upper limit of the extinction coefficient is 3 km.sup.−1 (selected based on AERONET data), and the interval step is 0.005 km.sup.−1.

[0048] (2) Combine the lookup table for the extinction coefficient with the LiDAR ratio to establish a lookup table for a backscattering coefficient β:

[00022]$\begin{array}{cc}\beta =\frac{{\sigma }_{ws}}{S_1}+\frac{{\sigma }_{st}}{S_2}& \left(15\right)\end{array}$

[0049] (3) Based on the extinction coefficient or σ.sub.sphere and the backscattering coefficient β.sub.sphere of the spherical aerosol calculated in the step of separating the sand dust from the spherical aerosol, and if 0<σ.sub.sphere≤1, traverse the lookup table 1; or if 1<σ.sub.sphere≤3, traverse the lookup table 2, to retrieve an array of extinction coefficients from the lookup table whose errors relative to observed values meet a standard.

[00023]$\begin{array}{cc}.Math.\frac{{\sigma }_{\mathrm{sphere}}-\sigma }{{\sigma }_{\mathrm{sphere}}}.Math.<0.01& \left(16\right)\end{array}$$\begin{array}{cc}.Math.\frac{{\beta }_{\mathrm{sphere}}-\beta }{{\beta }_{\mathrm{sphere}}}.Math.<0.01& \left(17\right)\end{array}$

[0050] (4) If the lookup table does not match the observed values, select as an optimal solution a solution ensuring a minimum deviation between an observed value and a theoretical value, that is, a solution meeting the following condition:

[00024]$\begin{array}{cc}\min (\sqrt{{.Math.\frac{{\sigma }_{\mathrm{sphere}}-\sigma }{{\sigma }_{\mathrm{sphere}}}.Math.}^2+{.Math.\frac{{\beta }_{\mathrm{sphere}}-\beta }{{\beta }_{\mathrm{sphere}}}.Math.}^2)}& \left(18\right)\end{array}$

[0051] In the present disclosure, the solution may be achieved in a method for inverting aerosol components using a LiDAR ratio and a depolarization ratio with a requirement of only a wavelength with a polarization channel. The method was applied to actually observed data of the LiDAR in 2017, and the data in a heavily polluted weather (FIG. 2) and a clean weather (FIG. 3) was calculated. Inversion results of aerosol components from January 2 to Jan. 4, 2017 are shown in FIG. 2. It can be seen that under the heavily polluted weather, the sand dust is not close to the ground, and the water-soluble aerosol and the soot are dominant. Inversion results of aerosol components from January 8 to Jan. 10, 2017 are shown in FIG. 3. Under the clean weather, there is almost no sand dust, the soot does not have a high concentration, and water-soluble aerosols such as sulfate and nitrate are dominant.

[0052] The embodiments of the present disclosure are described in detail above with reference to the accompanying drawings, but the present disclosure is not limited to the described embodiments. For a person skilled in the art, changes, modifications, replacements, and variations made to these embodiments without departing from the principle and spirit of the present disclosure shall still fall within the protection scope of the present disclosure.