A HSRL seawater backscattering inversion method based on vertical deviation model

Abstract

The invention discloses a HSRL seawater backscattering inversion method based on a vertical deviation model, comprising: obtaining a particle scattering signal and a molecular scattering signal simulated by a HSRL system; calculating the backscattering directly detected by the HSRL, and calculating the backscattering vertical deviation according to the backscattering true value input by the simulation; obtaining a vertical deviation model according to system parameters, environmental parameters and the backscattering vertical deviation; processing near-water surface data, calculating the backscattering near-water surface value not affected by multiple scattering, and estimating the scattering coefficient near-water surface value in combination with the particle backscattering ratio; calculating the vertical deviation at different depths based on the vertical deviation model; and inverting the backscattering profile according to the obtained vertical deviation profile. The invention overcomes the HSRL seawater backscattering inversion error caused by multiple scattering, and significantly improves the detection accuracy of the ocean HSRL.

Description

Technical Field

[0001] The invention belongs to the technical field of ocean laser radar, and in particular relates to a seawater backscattering inversion method of a high spectral resolution laser radar (HSRL) based on a vertical deviation model. Background Art

[0002] Seawater backscatter is defined as the integral of the volume scattering function in the scattering direction of 90-180 degrees. It is one of the core parameters of the application of ocean optics in marine ecology and biogeochemistry. Seawater backscatter can be used in many fields such as evaluating particulate organic carbon, phytoplankton biomass, zooplankton, building carbon-based net primary productivity remote sensing algorithms, and estimating particulate organic carbon output. Therefore, monitoring the acquisition and analysis of backscatter information plays an extremely important role in the development of marine science, the construction of marine ecological environment, and many other aspects.

[0003] Among the existing monitoring methods, in-situ detection technology has high accuracy but relatively low efficiency, while LiDAR technology has unique capabilities in marine remote sensing, providing continuous profile information of seawater backscattering characteristics both during the day and at night. Among them, HSRL can detect both particle and molecular signals at the same time, because there is no need to assume that LiDAR has broken through the limitation of equation ill-posedness, and significantly improves the detection accuracy of seawater backscattering compared to traditional particle LiDAR.

[0004] At present, HSRL mainly uses the direct method to invert the backscattering characteristics of seawater, that is, according to the laser radar equation, the particle signal is compared with the molecular signal, and then the seawater backscattering is obtained through the conversion coefficient of 180-degree backscattering to backscattering. In the current laser radar, most of them use the direct method. For example, the Chinese patent with publication number CN105486664A discloses a device and method for detecting marine phytoplankton biomass and particulate organic carbon. The patent HSRL combines the marine phytoplankton biomass and particulate organic carbon (POC) inversion algorithm, and uses the particle volume scattering equation and the effective attenuation coefficient when the scattering angle is π in the laser radar equation to synchronously invert the phytoplankton biomass and POC. The patent uses an HSRL system working in the frequency doubling band, combined with the marine phytoplankton biomass and POC inversion algorithm, to perform high-precision measurements of marine phytoplankton biomass and POC in the ocean over a large range with high depth resolution without sampling.

[0005] However, the transmission of laser in seawater faces the inevitable multiple scattering effect, that is, the signal photon undergoes more than one scattering process in the seawater, which introduces errors in the backscattering inversion. Studies have found that the maximum error of ocean HSRL backscattering in the vertical direction due to multiple scattering can exceed 25%.

[0006] Therefore, in order to solve the above-mentioned deficiencies in existing research, it is urgent to design a backscattering vertical deviation model and a HSRL seawater backscattering inversion method to correct the deviation. Summary of the invention

[0007] The present invention provides a HSRL seawater backscattering inversion method based on a vertical deviation model, which overcomes the HSRL seawater backscattering inversion error caused by multiple scattering and significantly improves the detection accuracy of HSRL seawater backscattering.

[0008] A HSRL seawater backscattering inversion method based on a vertical deviation model comprises the following steps:

[0009] (1) Simulate and obtain the seawater particle scattering signal and molecular scattering signal of the HSRL system;

[0010] (2) Using the particle scattering signals and molecular scattering signals obtained by simulation, the backscattering directly detected by HSRL is inverted and the vertical deviation of the backscattering is calculated based on the true value of the backscattering input by simulation.

[0011] (3) Obtaining a vertical deviation model based on system parameters, environmental parameters, and backscatter vertical deviation;

[0012] (4) Process the simulated near-water surface data, use the backscattering calculation formula to calculate the backscattering near-water surface value that is not affected by multiple scattering, and combine it with the particle backscattering ratio Estimate the scattering coefficient near the water surface ; Among them, the particle backscatter ratio The definition is as follows:

[0013] ;

[0014] In the formula is the scattering ratio of particles in the backward direction, is the scattering ratio of particles in the forward direction;

[0015] (5) The obtained scattering coefficient near the water surface value Divide the depth intervals into multiple depth intervals according to the equal depth intervals. Scattering coefficient at depth Approximately the scattering coefficient at shallow depths , and calculate the depth interval based on the vertical deviation model vertical deviation within , and further obtain the vertical deviation profile;

[0016] (6) Invert the backscattering profile based on the obtained vertical deviation profile.

[0017] The specific process of step (1) is as follows:

[0018] Using the HSRL Monte Carlo simulation method, more than 50 sets of simulations with different system parameters and environmental parameters as simulation variables were designed to collect the particle scattering signals and molecular scattering signals of HSRL under different conditions.

[0019] In step (2), the formula for calculating backscatter is:

[0020] ;

[0021] In the formula, is the backscattered signal value directly detected by HSRL, is the particle scattering signal, is the molecular scattering signal, is the 180-degree volume scattering function of the molecule, is the molecular backscattering, To detect depth; is the conversion factor from 180 degree body scattering to backscattering.

[0022] The specific process of step (3) is as follows:

[0023] Filter height in system parameters H , Field of view FOV And the scattering coefficient in the environmental parameters As the key parameter affecting the backscatter deviation, the following vertical deviation function is established:

[0024] ;

[0025] Further establish the following fitting function

[0026] ;

[0027] in, , The fitting function consists of height H, field of view FOV, and scattering coefficient The coefficient of determination;

[0028] By establishing equations or lookup tables with system parameters and environmental parameters, a vertical deviation model is obtained.

[0029] In step (4), the near-water surface data is processed as follows:

[0030] Through the laser radar system response deconvolution technology and the sea surface-seawater decomposition method, the influence of the sea surface signal on the seawater signal is removed, the seawater component in the particle and molecular scattering signal is obtained, and the near-water surface signal that is not affected by multiple scattering is obtained.

[0031] In step (5), the depth interval Inside, and The depth interval between them is less than or equal to 1 m.

[0032] The specific process of step (6) is as follows:

[0033] According to the vertical deviation And the backscattered signal value directly detected by HSRL After correction, high-precision seawater backscattering is obtained by inversion. The formula is as follows:

[0034] ;

[0035] when When the detection limit depth is not exceeded, the calculation is performed ,get To estimate the scattering coefficient , and then approximate it as the depth interval The scattering coefficient at depth is ;when When the detection limit depth is exceeded, all depths will be inverted The output is the backscatter profile.

[0036] In step (6), after inverting the backscatter profile, the following steps are also included:

[0037] In the near-water area, the HSRL system parameters were experimentally determined;

[0038] Based on typical sea area experimental data, preprocess the signal and invert it to calculate the vertical deviation and seawater backscattering;

[0039] In-situ instruments are used to measure the scattering coefficient and backscattering coefficient, and the estimation accuracy of the HSRL vertical deviation and the inversion accuracy of seawater backscatter are evaluated.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention obtains the simulation experimental data required for establishing the vertical deviation model based on the previously studied HSRL Monte Carlo simulation method, then finds the main parameters affecting the deviation in the system parameters and environmental parameters, and analyzes the relevant mechanisms affecting the vertical deviation of seawater backscattering. Based on this, a vertical deviation model is established, and the measured backscattering signal is inverted. Compared with previous studies, the method of the present invention overcomes the HSRL seawater backscattering inversion error caused by multiple scattering, significantly improves the detection accuracy of the ocean HSRL, and after the calibration experiment, the physical and bio-optical properties of the water body can be detected with high precision and high efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 The present invention is a flow chart of a HSRL seawater backscattering inversion method based on a vertical deviation model.

[0043] Figure 2 The present invention calculates the vertical deviation under different scattering coefficient environments through the vertical deviation model and obtains the curve of the actual vertical deviation changing with depth.

[0044] Figure 3 For the present invention Figure 2 Medium scattering coefficient b =0.3, the vertical deviation is used to correct the variation curve of HSRL measurement value and depth. DETAILED DESCRIPTION

[0045] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be pointed out that the embodiments described below are intended to facilitate the understanding of the present invention and do not have any limiting effect on the present invention.

[0046] like Figure 1 As shown, this embodiment provides a HSRL seawater backscattering inversion method based on a vertical deviation model, which mainly includes three parts: a mechanism model, an inversion method and a detection experiment.

[0047] Part I, a method for establishing a seawater backscatter vertical deviation model, includes the following steps:

[0048] (1) Based on the HSRL particle and molecular signal radiation transmission model in the ocean, i.e., the HSRL Monte Carlo simulation method developed in the early stage, more than 50 sets of simulations were designed with different system parameters (working height from the sea surface, field of view, etc.) and environmental parameters (scattering, absorption, etc.) as simulation variables to collect the HSRL particle and molecular signal intensities under different conditions.

[0049] The detection wavelength of the laser radar is 532nm. H The telescope field of view angle is FOV Take appropriate range according to different heights. Scattering coefficient , backscatter coefficient , absorption coefficient The main considerations are various situations under actual conditions such as open sea, near sea, and turbid water. In this example, the scattering coefficient is 0.1m -1 , 0.2m -1 , 0.3m -1 ; Sea level when on board H Take 10m, 15m, 20m; telescope field of view FOV Take 100mrad, 200mrad, and 300mrad.

[0050] (2) HSRL particle signal obtained based on simulation and molecular signals Calculate the backscattering detected directly by HSRL and express it as:

[0051] ;

[0052] in, and are the 180-degree backscattering and backscattering of the molecule, respectively, and z is the detection depth. , use the following formula to calculate the vertical deviation:

[0053] ;

[0054] Finally, the influence mechanism of system parameters and environmental parameters on vertical deviation is analyzed by controlling variables and combining multiple scattering theory. Then, the value range and value accuracy of each parameter are further optimized according to the influence of different system parameters and environmental parameters on vertical deviation.

[0055] (3) Construct a vertical deviation model. Based on the results obtained in step (2), the main parameters that affect the backscatter deviation are screened out. According to previous studies, the main parameters are estimated to be height H, field of view FOV, and scattering coefficient b. Therefore, the following vertical deviation function γ can be established:

[0056] ;

[0057] Then, according to the function characteristics, the appropriate function type is selected to fit the function, mainly considering exponential function or polynomial function, and the following fitting function can be established:

[0058] ;

[0059] in , The fitting function consists of height H, field of view FOV, and scattering coefficient By establishing mathematical relationships such as equations or lookup tables with system and environmental parameters, the basic model of vertical deviation can be obtained.

[0060] Part II, a method for establishing a seawater backscatter vertical deviation model, includes the following steps:

[0061] Step S1, perform near-water surface data processing, analyze and reduce the impact of system response and sea surface signal on particle and molecular signals. Through the previously developed laser radar system response deconvolution technology, the interference of different particle and molecular channel system responses can be reduced first. Secondly, the sea surface-seawater decomposition method is used to remove the sea surface particle scattering and molecular scattering components in the received backscattered signal, so as to obtain a relatively pure near-water surface signal affected by multiple scattering. Through the above formula (1), the backscattered near-water surface value that is not affected by multiple scattering can be obtained, and combined with the particle backscattering ratio Estimate the scattering coefficient near the water surface .

[0062] Radiation coefficient near water surface value .

[0063] Step S2, estimate the vertical deviation. According to the developed shipborne ocean HSRL system, the system parameters such as working height and field of view are obtained. Then the scattering coefficient obtained in step S1 is Approximate by depth segmentation, The scattering coefficient is approximately the scattering coefficient in shallow water , and The depth interval between the two can be 1 m or less to avoid large errors in approximation. Based on the system parameters, processed environmental parameters and the vertical deviation basic model, the depth is obtained. Vertical error within the range , as follows:

[0064] ;

[0065] Step S3, invert backscatter. Using the And the backscattered signal value directly detected by HSRL After correction, the high-precision seawater backscattering is inverted as follows:

[0066] ;

[0067] Considering the influence of signal-to-noise ratio, when When the detection limit depth is not exceeded, the calculation is performed ,get To estimate the scattering coefficient , which is then approximated as the scattering coefficient at a deeper depth .when When the detection limit depth is exceeded, all depths will be inverted The output is the backscatter profile.

[0068] Step S4, inversion optimization. Considering the adaptability of this method in different water environments, it is necessary to use model simulation to evaluate the inversion method obtained in steps S1 to S3 and optimize it when necessary. Considering that the scattering coefficient of the water body at different depths varies little, the depth interval element is expanded. The approximation on and , in non-uniform water, there may be a scattering coefficient The situation of rapid changes makes the inversion result less accurate. Therefore, if the error is small, the original method can be used; if the error is large, the method should be optimized. The optimization method mainly considers translating the vertical deviation to the shallow water by one depth resolution and re-inverting the seawater backscatter.

[0069] The third part is a method for experimentally verifying the constructed vertical deviation model, including the following steps:

[0070] Step 1: Calibrate the HSRL system parameters by measuring HSRL particle and molecular signals.

[0071] In the near-surface area where multiple scattering is less affected, the backscattering of seawater at multiple calibration stations is measured using in-situ instruments. Combined with the measured HSRL particle and molecular signals, the response ratio between particle and molecular channels is calibrated. The working height H is calculated based on the distance between the laser emission and the sea surface signal and the incident angle. The laser radar field of view calibration method based on the overlap factor is used to obtain the field of view FOV value.

[0072] Step 2, experimental verification of HSRL detection of backscattering.

[0073] The experiment verifies the accuracy of HSRL detection backscatter inversion. Experiments were carried out in typical sea areas (areas with strong scattering) using the ship-borne HSRL. The original signal was obtained by HSRL, and foam removal, sea surface position correction, average denoising, background noise removal, calibration parameter correction and position matching were performed to obtain the preprocessed signal. Combined with the calibrated system parameters, the vertical deviation and seawater backscattering were calculated using the inversion method in the second part. The scattering coefficient was measured using the in-situ instrument Wetlabs acs, and the vertical deviation was simulated to evaluate the estimation accuracy of HSRL for vertical deviation; the backscattering coefficient measured using the in-situ instrument HOBILabsHS6P was used to evaluate the inversion accuracy of HSRL for seawater backscattering.

[0074] The vertical deviation obtained by inversion according to the embodiment is similar to the actual vertical deviation. Figure 2As shown, curves of different colors represent different scattering coefficients, the dotted line represents the vertical deviation under actual data, and the solid line represents the vertical deviation obtained by model inversion. The two are in good agreement, indicating that the inversion method is effective.

[0075] In this embodiment, the HSRL seawater backscattering inversion method based on the vertical deviation model of the present invention is used to realize the detection inversion of the vertical deviation of water body backscattering. The scattering coefficient after HSRL inversion of seawater backscattering is as follows: Figure 3 The blue line shows the difference with the in-situ instrument detection result ( Figure 3 The black solid line) shows a good contrast, which verifies the effectiveness of the present invention.

[0076] The embodiments described above provide a detailed description of the technical solutions and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, supplements and equivalent substitutions made within the scope of the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A HSRL seawater backscattering inversion method based on a vertical deviation model, characterized in that: The following steps are involved: (1) Simulate and obtain the seawater particle scattering signal and molecular scattering signal of the HSRL system; (2) Using the particle scattering signals and molecular scattering signals obtained by simulation, the backscattering directly detected by HSRL is inverted and the vertical deviation of the backscattering is calculated based on the true value of the backscattering input by simulation. (3) Based on the system parameters, environmental parameters and backscatter vertical deviation, the vertical deviation model is obtained. The specific process is as follows: Filter height in system parameters H , Field of view FOV And the scattering coefficient in the environmental parameters As the key parameter affecting the backscatter deviation, the following vertical deviation function is established: ； Further establish the following fitting function ； in, , The fitting function consists of height H, field of view FOV, and scattering coefficient The coefficient of determination; By establishing equations or lookup tables with system parameters and environmental parameters, a vertical deviation model is obtained; (4) Process the simulated near-water surface data, use the backscattering calculation formula to calculate the backscattering near-water surface value that is not affected by multiple scattering, and combine it with the particle backscattering ratio Estimate the scattering coefficient near the water surface ; Among them, the particle backscatter ratio The definition is as follows: ； In the formula, is the scattering ratio of particles in the backward direction, is the scattering ratio of particles in the forward direction; (5) The obtained scattering coefficient near the water surface value Divide the depth intervals into multiple depth intervals according to the equal depth intervals. Scattering coefficient at depth Approximately the scattering coefficient at shallow depths , and calculate the depth interval based on the vertical deviation model vertical deviation within , and further obtain the vertical deviation profile; (6) Invert the backscattering profile based on the obtained vertical deviation profile; the specific process is as follows: According to the vertical deviation And the backscattered signal value directly detected by HSRL After correction, high-precision seawater backscattering is obtained by inversion. The formula is as follows: ； when When the detection limit depth is not exceeded, the calculation is performed ,get To estimate the scattering coefficient , and then approximate it as the depth interval The scattering coefficient at depth is ;when When the detection limit depth is exceeded, all depths will be inverted The output is the backscatter profile.

2. The HSRL seawater backscattering inversion method based on the vertical deviation model according to claim 1 is characterized in that: The specific process of step (1) is as follows: Using the HSRL Monte Carlo simulation method, more than 50 sets of simulations with different system parameters and environmental parameters as simulation variables were designed to collect the particle scattering signals and molecular scattering signals of HSRL under different conditions.

3. The HSRL seawater backscattering inversion method based on the vertical deviation model according to claim 1 is characterized in that: In step (2), the formula for calculating backscatter is: ； In the formula, is the backscattered signal value directly detected by HSRL, is the particle scattering signal, is the molecular scattering signal, is the 180-degree volume scattering function of the molecule, is the molecular backscattering, To detect depth; is the conversion factor from 180 degree body scattering to backscattering.

4. The HSRL seawater backscattering inversion method based on the vertical deviation model according to claim 3 is characterized in that: In step (2), the backscatter vertical deviation is calculated based on the true backscatter value input by the simulation, and the formula is: ； In the formula, is the backscatter vertical deviation, The true value of backscattering input for the simulation.

5. The HSRL seawater backscattering inversion method based on the vertical deviation model according to claim 1 is characterized in that: In step (4), the near-water surface data is processed as follows: Through the laser radar system response deconvolution technology and the sea surface-seawater decomposition method, the influence of the sea surface signal on the seawater signal is removed, the seawater component in the particle and molecular scattering signal is obtained, and the near-water surface signal that is not affected by multiple scattering is obtained.

6. The HSRL seawater backscattering inversion method based on the vertical deviation model according to claim 1 is characterized in that: In step (5), the depth interval Inside, and The depth interval between them is less than or equal to 1 m.

7. The HSRL seawater backscattering inversion method based on the vertical deviation model according to claim 1 is characterized in that: In step (6), after inverting the backscatter profile, the following steps are also included: In the near-water area, the HSRL system parameters were experimentally determined; Based on typical sea area experimental data, preprocess the signal and invert it to calculate the vertical deviation and seawater backscattering; In-situ instruments are used to measure the scattering coefficient and backscattering coefficient, and the estimation accuracy of the HSRL vertical deviation and the inversion accuracy of seawater backscatter are evaluated.

Images