A method and system for calculating the attenuation coefficient profile of water body particles

Abstract

The present invention provides a method and system for calculating the profile of the attenuation coefficient of water body particles. First, the corresponding relationship between the chlorophyll concentration on the surface of the target water body and the attenuation coefficient of water body particles is obtained. Then, a preset algorithm is used to calculate the chlorophyll concentration on the surface corresponding to each pixel in the water body area, and the attenuation coefficient of the surface water body particles of each pixel is obtained according to the corresponding relationship. Finally, the profile type of the attenuation coefficient of water body particles of each pixel is identified, and the profile data of the attenuation coefficient of water body particles of each pixel is calculated according to the preset method corresponding to the profile type to which it belongs. By using the method and system disclosed in the present invention, it is possible to evaluate the profile data of the current attenuation coefficient of water body particles of the target water body based on long-term remote sensing observation data and construct a three-dimensional field of profile data, and at the same time, it is also possible to trace back the profile data of historical periods.

Description

Technical Field

[0001] The present invention belongs to the field of remote sensing technology, and particularly relates to a method and system for calculating the profile of the attenuation coefficient of water body particles. Background Art

[0002] In recent years, the rapid development of observation technologies has greatly expanded the observation scope of ocean dynamic processes. Ocean color satellite remote sensing has become an important means for monitoring the ocean environment because it can provide global or regional large-scale and continuous observation data. However, traditional ocean color remote sensing technology mainly relies on water body optical parameters, and the information obtained is mainly limited to the upper 20% area of the euphotic layer, and it is unable to effectively analyze the distribution and dynamic changes of subsurface phytoplankton biomass.

[0003] In-situ ocean observations have shown that there is a subsurface maximum phytoplankton biomass layer in the upper ocean. This phenomenon has caused significant errors in the estimation of phytoplankton biomass and net primary productivity. Existing studies have shown that compared with the case of assuming uniform biomass distribution, combining dynamic water body optical profile information can improve the remote sensing observation accuracy of phytoplankton biomass, light attenuation coefficient, and primary productivity by about 5-70%, 36-300%, and 20-54% respectively. Therefore, improving the observation ability of ocean color satellite remote sensing for water body optical profiles is crucial for improving the accuracy of ocean environment monitoring and model simulation capabilities.

[0004] The attenuation coefficient of water body particles (cp) describes the attenuation ability of suspended particulate matter in water to light beams, and its profile type refers to the distribution pattern of the particle attenuation coefficient with depth in the water body. Waters in different regions and seasons will exhibit different structural types of attenuation coefficient profiles due to differences in particulate matter concentration, composition, and hydrodynamic environment.

[0005] Current ocean color satellite remote sensing still faces many challenges in analyzing the complete profile information of the ocean euphotic layer, including: (1) Existing statistical models have good applicability on a global scale, but lack accurate characterization of local dynamic processes on a regional scale; (2) Machine learning methods based on observational data have advantages in refined modeling, but their dependence on Argo float data limits their large-scale application; (3) As a more comprehensive optical parameter, the analysis of the dynamic change characteristics of the attenuation coefficient of water body particles (cp) still needs to be further studied.

[0006] Currently, the classification method of the profile type of the attenuation coefficient of water body particles is mainly based on static identification of sea surface chlorophyll concentration and seasonal changes, ignoring the variation characteristics of the mixing of type I and type II seawater. At the same time, the static classification method results in unnatural jump phenomena in the profile structure, limiting its applicability in complex ocean environments. Summary of the Invention

[0007] The present invention provides a method and system for calculating the water body particle attenuation coefficient profile to solve the problems in the prior art, such as ignoring the changing characteristics of seawater mixing and the unnatural jump phenomenon in the profile structure caused by the static classification method.

[0008] To solve the above technical problems, the embodiments of the present invention disclose the following technical solutions:

[0009] One aspect of the present invention provides a method for calculating the water body particle attenuation coefficient profile, including:

[0010] Extracting the water body area in the remote sensing image of the target water body;

[0011] Obtaining the corresponding relationship between the surface chlorophyll concentration of the target water body and the water body particle attenuation coefficient;

[0012] Calculating the surface chlorophyll concentration corresponding to each pixel in the water body area by using a preset algorithm, and obtaining the surface water body particle attenuation coefficient of each pixel according to the corresponding relationship;

[0013] Identifying the water body particle attenuation coefficient profile type of each pixel, where the water body particle attenuation coefficient profile type includes a uniformly mixed type, a Gaussian-like type, and an exponential decay type;

[0014] Calculating the water body particle attenuation coefficient profile data of each pixel according to the preset method corresponding to the profile type, wherein the water body particle attenuation coefficient profile data of the pixels belonging to the Gaussian-like type is calculated based on the Bayesian optimization method.

[0015] Optionally, the obtaining the corresponding relationship between the surface chlorophyll concentration of the target water body and the water body particle attenuation coefficient includes:

[0016] Collecting the historical data of the surface chlorophyll concentration and the historical data of the water body particle attenuation coefficient at multiple observation stations in the target water body;

[0017] Obtaining the following corresponding relationship according to the historical data of the above two:

[0018] log(cp660)=B1 × log(Chla) 2 +B2×log(Chla) -INT

[0019] where cp660 is the surface water body particle attenuation coefficient at a wavelength of 660 nm below the water surface; Chla is the surface chlorophyll concentration; B1, B2, and INT are the regression coefficients obtained by comparing the historical data of the above two through regression analysis.

[0020] Optionally, the calculating the surface chlorophyll concentration corresponding to each pixel in the water body area by using a preset algorithm includes:

[0021] Preprocess the remote sensing image of the target water body;

[0022] Obtain the remote sensing reflectance of each pixel within the water body area;

[0023] Calculate the surface chlorophyll concentration corresponding to the remote sensing reflectance of each pixel using a preset chlorophyll concentration inversion algorithm.

[0024] Optionally, identifying the type of water body particle attenuation coefficient profile for each of the pixels includes:

[0025] Calculate the euphotic layer depth at the pixel based on the surface chlorophyll concentration;

[0026] Obtain the water depth and mixed layer depth at the pixel;

[0027] Calculate the LaRaD ratio of the pixel, LaRaD = water depth / mixed layer depth;

[0028] Judge whether the euphotic layer depth at the pixel is greater than 50,

[0029] When the euphotic layer depth at the pixel is greater than 50, judge whether the LaRaD ratio is less than 7.5,

[0030] If so, determine that the type of water body particle attenuation coefficient profile of the pixel is the well - mixed type;

[0031] If not, determine that the type of water body particle attenuation coefficient profile of the pixel is the Gaussian - like type;

[0032] When the euphotic layer depth at the pixel is not greater than 50, judge whether the LaRaD ratio is less than 2,

[0033] If so, determine that the type of water body particle attenuation coefficient profile of the pixel is the well - mixed type;

[0034] If not, determine that the type of water body particle attenuation coefficient profile of the pixel is the exponential attenuation type.

[0035] Optionally, calculate the euphotic layer depth at the pixel using the following formula:

[0036] Zeu = 10^(1.35 - 0.4026×log10(Chla)+0.0375×log10(Chl)^2)

[0037] Where Zeu is the euphotic layer depth; Chla is the surface chlorophyll concentration of the pixel.

[0038] Optionally, calculating the water body particle attenuation coefficient profile data of each pixel according to the preset method corresponding to the profile type includes:

[0039] When the water body particle attenuation coefficient profile type of the pixel is the exponential attenuation type, the following formula is used to calculate the profile data of the water body particle attenuation coefficient:

[0040] cp660 z = cp660 surf , (z < MLD);

[0041] cp660 z = cp660 surf × exp(-k exp / (z - MLD)), (z ≥ MLD);

[0042] Among them, k exp is a preset constant parameter; MLD is the mixed layer depth at the pixel; z is the currently calculated profile depth; cp660 surf is the surface particle attenuation coefficient corresponding to the pixel; cp660 z is the water body particle attenuation coefficient of the pixel at the profile depth of z.

[0043] Optionally, calculating the water body particle attenuation coefficient profile data of each pixel according to the preset method corresponding to the profile type includes:

[0044] When the water body particle attenuation coefficient profile type of the pixel is the Gaussian-like type, the following formula is used to calculate the profile data of the water body particle attenuation coefficient:

[0045] cp660 z = cp660 surf × (A gau × exp(-(z - U) 2 / D 2 ) - B gau × z + 1);

[0046] Among them, z is the currently calculated profile depth; cp660 z is the water body particle attenuation coefficient of the pixel at the profile depth of z; cp660 surf is the surface particle attenuation coefficient corresponding to the pixel; A gau , B gau , U, D are constant parameters determined by the Bayesian optimization method respectively.

[0047] Optionally, the following method is used to determine the initial values of the constant parameters U and D:

[0048] Based on the historical data of the surface chlorophyll concentration, the subsurface maximum depth, and the subsurface maximum width, obtain the correlation between the surface chlorophyll concentration Chla and the subsurface maximum depth U:

[0049] log(U) = 0.446 × log(Chla) + 0.413;

[0050] And, obtain the correlation between the surface chlorophyll concentration Chla and the width D of the subsurface maximum:

[0051] log(D) = 0.406 × log(Chla) - 0.132;

[0052] Use the above correlations to calculate the initial values of the constant parameters U and D corresponding to the pixel.

[0053] Optionally, the method further includes:

[0054] Adopt the following method to determine the observation constraint conditions:

[0055] cp660 Mean-Zeu = M1 × exp (M2 × cp660 surf );

[0056] cp660 Mean-Zmax = N1 × exp (N2 × cp660 surf );

[0057] cp660 Mean-2*Zmax =P1 ×cp660 surf + P2;

[0058] Wherein, cp660 Mean-Zeu is the average value of the water body particle attenuation coefficient in the euphotic layer; cp660 Mean-Zmax is the average value of the water body particle attenuation coefficient from the water surface to the maximum depth of the subsurface layer; cp660 Mean-2*Zmax is the average value of the water body particle attenuation coefficient from the water surface to twice the maximum depth of the subsurface layer; M1, M2, N1, N2, P1, and P2 are respectively preset constant parameters.

[0059] Optionally, adopt the following method to determine the final values of the constant parameters A gau , B gau , U, D for each pixel:

[0060] Calculate the water body particle attenuation coefficient cp660 between the maximum depth of the subsurface layer and the preset maximum water depth according to the following formula Deep-Zmax :

[0061] cp660 Deep-Zmax =cp660 surf × exp(-0.025 × (z - z max ))

[0062] Among them, z max is the maximum depth of the subsurface layer, obtained by multiplying U by Zeu;

[0063] Construct the following objective function:

[0064] Dif = Dif Zeu 2 + Dif Zmax 2 + Dif 2-Zmax 2 + Dif U 2 + (Dif D / 10) 2 + Dif model 2 ;

[0065] Among them, Dif Zeu , Dif Zmax and Dif 2-Zmax respectively represent the differences between the calculation results of cp660 Mean-Zeu , cp660 Mean-Zmax and cp660 Mean-2*Zmax and the true observed data; Dif U and Dif D are respectively the differences between the calculation results of U and D and the true observed data; Dif model is the difference between the calculation result of cp660 Deep-Zmax and the true observed data;

[0066] Use the Bayesian optimization method to optimize the objective function to obtain the final values of A gau , B gau , U, and D.

[0067] Another aspect of the present invention discloses a water body particle attenuation coefficient profile calculation system, and the system executes the water body particle attenuation coefficient profile calculation method described in the foregoing aspect.

[0068] A method and system for calculating the profile of the attenuation coefficient of water body particles disclosed by the present invention. First, obtain the corresponding relationship between the chlorophyll concentration on the surface of the target water body and the attenuation coefficient of water body particles. Then, use a preset algorithm to calculate the chlorophyll concentration on the surface corresponding to each pixel in the water body area, and obtain the attenuation coefficient of the surface water body particles of each pixel according to the corresponding relationship; finally, identify the profile type of the attenuation coefficient of water body particles of each pixel, and calculate the profile data of the attenuation coefficient of water body particles of each pixel according to the preset method corresponding to the profile type. By using the method and system disclosed by the present invention, the profile data of the current attenuation coefficient of water body particles of the target water body can be evaluated based on long-term remote sensing observation data, and a three-dimensional field of profile data can be constructed. At the same time, the profile data of historical periods can also be traced back.

[0069] The invention content part is provided to introduce the selection of concepts in a simplified form, which will be further described in the specific implementation manner below. The invention content part is not intended to identify the important features or essential features of the present disclosure, nor is it intended to limit the scope of the present disclosure. Brief Description of the Drawings

[0070] By describing the exemplary embodiments of the present disclosure in more detail in conjunction with the drawings, the above and other objects, features, and advantages of the present disclosure will become more obvious. Among them, in the exemplary embodiments of the present disclosure, the same reference numerals generally represent the same components.

[0071] Figure 1 It is a schematic flowchart of a method for calculating the profile of the attenuation coefficient of water body particles provided by an embodiment of the present invention;

[0072] Figure 2 It is to implement one provided by an embodiment of the present invention Figure 1 The schematic flowchart of step S300 in;

[0073] Figure 3 It is to implement one provided by an embodiment of the present invention Figure 1 The schematic flowchart of step S400 in. Detailed Description of the Embodiments

[0074] The embodiments of the present disclosure will be described in more detail below with reference to the drawings. Although the embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0075] As used herein, the term "including" and its variations denote open inclusion, i.e., "including but not limited to". Unless otherwise specified, the term "or" means "and / or". The term "based on" means "at least partially based on". The terms "an example embodiment" and "an embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one additional embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included hereinafter.

[0076] Figure 1 The flowchart of a method for calculating the water body particle attenuation coefficient profile provided for the disclosed embodiment of the present invention is as Figure 1 shown, and the method includes the following steps:

[0077] Step S100: Extract the water body area from the remote sensing image of the target water body.

[0078] The water body area is extracted from the remote sensing image of the target water body by using an existing method. For example, the water body area is calculated through the Normalized Difference Water Index (NDWI) or the Modified Normalized Difference Water Index (MNDWI), and a threshold is set to segment the water body; or, machine algorithms such as Support Vector Machine (SVM), Random Forest (RF), and Convolutional Neural Network (CNN) are used to classify pixels to determine the water body area.

[0079] Step S200: Obtain the corresponding relationship between the surface chlorophyll concentration of the target water body and the water body particle attenuation coefficient.

[0080] In an embodiment disclosed in the present invention, step S200 is implemented in the following manner:

[0081] (1) Collect historical data of the surface chlorophyll concentration and historical data of the water body particle attenuation coefficient at the observation stations in the target water body.

[0082] Multiple observation stations are set in the target water body. For each observation station, surface chlorophyll concentration data and water body particle attenuation coefficient data at multiple moments within a historical period are collected, so as to obtain multiple sets of historical data that match each other.

[0083] (2) Obtain the following corresponding relationship based on the historical data of the above two:

[0084] log(cp660)=B1 × log(Chla) 2 +B2×log(Chla) -INT

[0085] Among them, cp660 is the attenuation coefficient of surface water particles at a wavelength of 660 nm (red light band) below the water surface. Since the absorption of the 660-nm wavelength in water is strong, but it can still be scattered by particles. At the same time, this band is also the absorption band of phytoplankton chlorophyll. Therefore, in the disclosed embodiments of the present invention, cp660 is used as the required attenuation coefficient of surface water particles.

[0086] Chla is the surface chlorophyll concentration; B1, B2, and INT are the regression coefficients obtained by comparing the historical data of the above two through regression analysis. In a specific embodiment disclosed in the present invention, B1 = 0.59039, B2 = 1.40173, and INT = 0.6381.

[0087] Step S300: Calculate the surface chlorophyll concentration corresponding to each pixel in the water body area using a preset algorithm, and obtain the attenuation coefficient of surface water particles for each pixel according to the corresponding relationship.

[0088] In an embodiment disclosed in the present invention, a preset algorithm is used to calculate the surface chlorophyll concentration corresponding to each pixel in the water body area, as Figure 2 shown, including:

[0089] Step S301: Preprocess the remote sensing image of the target water body:

[0090] For example, the following methods can be used to preprocess the remote sensing image.

[0091] (1) Radiometric correction: Convert the digital number value (DN) of the image to a physical unit (such as radiance, reflectance, etc.).

[0092] (2) Atmospheric correction: For example, Rayleigh scattering correction to remove the influence of atmospheric molecular scattering; or aerosol correction to reduce the interference of suspended particles on the water reflectance.

[0093] (3) Geometric correction: Project the image onto a unified coordinate system (WGS84, UTM) to correct the geometric distortion caused by terrain and sensor tilt.

[0094] (4) Noise removal: For example, cloud removal processing, using a cloud mask to remove the cloud-covered area in the image; or water body masking processing to exclude areas with serious influence on the water body edge and sediment.

[0095] Step S302: Obtain the remote sensing reflectance of each pixel in the water body area.

[0096] The remote sensing reflectance of each pixel in the water body area is obtained by using existing methods, which will not be elaborated here.

[0097] Step S303: Calculate the surface chlorophyll concentration corresponding to the remote sensing reflectance of each pixel using a preset chlorophyll concentration inversion algorithm.

[0098] Input the remote sensing reflectance of each pixel into a preset chlorophyll concentration inversion algorithm. For example, use a single-band empirical model, ratio index method, machine learning model, etc. to calculate the corresponding surface chlorophyll concentration value.

[0099] Step S400: Identify the type of water body particle attenuation coefficient profile for each pixel.

[0100] The types of water body particle attenuation coefficient profiles include uniformly mixed type, Gaussian-like type, and exponential decay type.

[0101] In an embodiment disclosed by the present invention, as Figure 3 shown, for each pixel, the type of water body particle attenuation coefficient profile is determined in the following manner:

[0102] Step S401: Calculate the euphotic layer depth at the pixel based on the surface chlorophyll concentration.

[0103] In an embodiment disclosed by the present invention, the following formula is used to calculate the euphotic layer depth at the pixel:

[0104] Zeu = 10 ^ (1.35 - 0.4026 × log10(Chla) + 0.0375 × log10(Chl) ^ 2)

[0105] The above formula is an empirical regression model established based on the historical data of the euphotic layer depth and surface chlorophyll concentration. Among them, Zeu is the euphotic layer depth; Chla is the surface chlorophyll concentration of the pixel.

[0106] Step S402: Obtain the water depth and mixed layer depth at the pixel.

[0107] Among them, the water depth is obtained from the ETOPO1 global geomorphology dataset. ETOPO1 (Earth Topography One Arc-Minute Global Relief Model) is a dataset recording the global terrain and ocean depth, covering all land and ocean areas of the world.

[0108] The mixed layer depth is obtained from HYCOM model data. HYCOM (Hybrid Coordinate Ocean Model) is a global ocean model that can provide high-resolution ocean physical data, such as temperature, salinity, ocean current, mixed layer depth, etc.

[0109] Based on the geographic coordinate information of the remote sensing image and the row and column numbers of the pixel in the image, determine the longitude and latitude corresponding to the pixel, and then obtain the water depth at this pixel through the ETOPO1 global geomorphology dataset, and obtain the mixed layer depth at this pixel through the HYCOM model data.

[0110] Step S403: Calculate the LaRaD ratio of the pixel.

[0111] LaRaD = water depth / mixed layer depth.

[0112] Step S404: Determine whether the euphotic layer depth at the pixel is greater than 50.

[0113] When the euphotic layer depth at the pixel is greater than 50, execute Step S405: Determine whether the LaRaD ratio is less than 7.5.

[0114] Generally, in spring and summer seasons, the depth of the euphotic layer of the water body is greater than 50. If the LaRaD ratio is less than 7.5, determine that the water body particle attenuation coefficient profile type of the pixel is the uniformly mixed type.

[0115] If the LaRaD ratio is not less than 7.5, determine that the water body particle attenuation coefficient profile type of the pixel is the Gaussian-like type.

[0116] When the euphotic layer depth at the pixel is not greater than 50, execute Step S406: Determine whether the LaRaD ratio is less than 2.

[0117] Generally, in autumn and winter seasons, the depth of the euphotic layer of the water body is not greater than 50. If the LaRaD ratio is less than 2, determine that the water body particle attenuation coefficient profile type of the pixel is the uniformly mixed type.

[0118] If the LaRaD ratio is not less than 2, determine that the water body particle attenuation coefficient profile type of the pixel is the exponential attenuation type.

[0119] Step S500: Calculate the water body particle attenuation coefficient profile data of each pixel according to the preset method corresponding to the profile type.

[0120] Among them, if the water body particle attenuation coefficient profile type corresponding to the pixel belongs to the uniformly mixed type, the water body particle attenuation coefficient profile data of this pixel is the surface water body particle attenuation coefficient, that is, within the water column represented by this pixel, for example, within a depth of 125 meters, the water body particle attenuation coefficient is the surface water body particle attenuation coefficient.

[0121] In an embodiment disclosed in the present invention, when the water body particle attenuation coefficient profile type of the pixel is the exponential attenuation type, the following formula is used to calculate the profile data of the water body particle attenuation coefficient:

[0122] cp660 z = cp660surf , (z < MLD);

[0123] cp660 z = cp660 surf × exp(-k exp / (z - MLD)), (z ≥ MLD);

[0124] where k exp is a preset constant parameter. In a specific embodiment disclosed in the present invention, k exp = 0.024; MLD is the mixed layer depth at the pixel; z is the current calculated profile depth, i.e., water depth; cp660 surf is the surface particle attenuation coefficient corresponding to the pixel; cp660 z is the water body particle attenuation coefficient of the pixel at the profile depth of z.

[0125] After calculating different z in the entire profile of the pixel, the water body particle attenuation coefficient profile data of the pixel is generated. Through this profile data, the water body particle attenuation coefficients at each depth within the water column represented by the pixel can be obtained.

[0126] For example, cp660 is calculated every 1 meter z , until z reaches 125 meters. The 125 calculated cp660 z discrete data is interpolated into the entire profile by interpolation (other existing methods can also be used) to make the profile have a continuous and smooth water body particle attenuation coefficient.

[0127] In the embodiment disclosed in the present invention, the water body particle attenuation coefficient profile data of the Gaussian - like pixel is calculated based on the Bayesian optimization method.

[0128] When the water body particle attenuation coefficient profile type of the pixel is Gaussian - like, the following formula is used to calculate the profile data of the water body particle attenuation coefficient:

[0129] cp660 z = cp660 surf × (A gau × exp(-(z - U) 2 / D 2 ) - B gau × z + 1);

[0130] where z is the current calculated profile depth; cp660 z is the water body particle attenuation coefficient of the pixel at the profile depth of z; cp660 surf is the surface particle attenuation coefficient corresponding to the pixel; A gau , B gau, U, and D are constant parameters determined by the Bayesian optimization method respectively.

[0131] In an embodiment disclosed by the present invention, the initial values of the constant parameters U and D are determined in the following manner:

[0132] (1) Obtain historical data of surface chlorophyll concentration, subsurface maximum depth, and subsurface maximum width. Among them, the surface chlorophyll concentration (Chla) refers to the chlorophyll concentration in the outermost layer of water (generally 0–5 m); the subsurface maximum depth (U) refers to the depth at which the chlorophyll concentration reaches the subsurface maximum in the water body, usually near the thermocline, representing the main growth layer of phytoplankton; the subsurface maximum width (D) refers to the distribution range of the chlorophyll concentration around the subsurface maximum, which is the vertical range where the subsurface chlorophyll concentration gradually decreases from the maximum to the background level, describing the thickness of the high-value area of the chlorophyll concentration.

[0133] Obtain the correlation between the surface chlorophyll concentration Chla and the subsurface maximum depth U based on the above historical data:

[0134] log(U) = 0.446 × log(Chla) + 0.413;

[0135] And obtain the correlation between the surface chlorophyll concentration Chla and the subsurface maximum width D:

[0136] log(D) = 0.406 × log(Chla) - 0.132;

[0137] (2) Calculate the initial values of the constant parameters U and D corresponding to the pixel using the above correlations.

[0138] In an embodiment disclosed by the present invention, the following method is required to determine the observation constraint conditions:

[0139] cp660 Mean-Zeu = M1 × exp (M2 × cp660 surf );

[0140] cp660 Mean-Zmax = N1 × exp (N2 × cp660 surf );

[0141] cp660 Mean-2*Zmax =P1 ×cpδ660 surf + P2;

[0142] Among them, cp660 Mean-Zeu is the average value of the water body particle attenuation coefficient in the euphotic layer; cp660 Mean-Zmax It should be noted that there seems to be a typo in "cp660 =P1 ×cp660 + P2" in the original text, which is likely to be "cp660 =P1 ×cpδ660 + P2" in the translation. If this is not a typo in the original, please check and correct the relevant content for a more accurate translation.is the average value of the particulate attenuation coefficient of the water body within the maximum depth from the water surface to the subsurface layer; cp660 Mean-2*Zmax is the average value of the particulate attenuation coefficient of the water body within the maximum depth from the water surface to twice the maximum depth of the subsurface layer; M1, M2, N1, N2, P1, and P2 are constant parameters calculated based on the corresponding historical data; in a specific embodiment disclosed in the present invention, M1 = 0.0189, M2 = 21.989, N1 = 0.0216, N2 = 19.782, P1 = 0.9501, and P2 = 0.0027.

[0143] In an embodiment disclosed in the present invention, the following method is used to determine the final values of the constant parameters Agau, Bgau, U, and D corresponding to each pixel:

[0144] Calculate the particulate attenuation coefficient cp660Deep-Zmax of the water body between the maximum depth of the subsurface layer and the preset maximum depth of the water body according to the following formula:

[0145] cp660 Deep-Zmax = cp660 surf × exp(-0.025 × (z - z max ))

[0146] where z max is the maximum depth of the subsurface layer, obtained by multiplying U by Zeu.

[0147] Construct the following objective function:

[0148] Dif = Dif Zeu 2 + Dif Zmax 2 + Dif 2-Zmax 2 + Dif U 2 +(Dif D / 10) 2 + Dif model 2 ;

[0149] where Dif Zeu , Dif Zmax , and Dif 2-Zmax respectively represent the differences between the calculation results of cp660 Mean-Zeu , cp660 Mean-Zmax , and cp660 Mean-2*Zmax and the true observed data; Dif U and Dif D are respectively the differences between the calculation results of U and D and the true observed data; Dif modelIt is cp660 Deep-Zmax The difference between the calculation result of Deep-Zmax and the true observation data.

[0150] Before participating in Bayesian optimization, U and D need to be normalized using the euphotic layer depth to standardize U and D. For example, U = U / Zeu, D = D / Zeu.

[0151] The Bayesian optimization method is used to optimize the objective function to obtain A corresponding to the case where the objective function is closest to zero gau , B gau , U, D, as the final parameter values.

[0152] In the above manner, the formula cp660 for calculating the water body particle attenuation coefficient corresponding to each pixel is obtained z = cp660 surf × (A gau × exp(-(z - U) 2 / D 2 ) - B gau ×z + 1), and calculate the water body particle attenuation coefficients at various depths within the water column represented by the pixel, and generate continuous and smooth profile data.

[0153] After obtaining the water body particle attenuation coefficient profile data of each pixel in the water body area by using the foregoing method, a three-dimensional scene of the water body particle attenuation coefficient of the entire water body area is constructed.

[0154] Another embodiment of the present invention discloses a water body particle attenuation coefficient profile calculation system, which executes the water body particle attenuation coefficient profile calculation method disclosed in the foregoing embodiment.

[0155] The above has described the embodiments of the present disclosure. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technologies in the market, or to enable other ordinary skill in the art in the technical field to understand the disclosed embodiments.

Claims

1. A method for calculating the attenuation coefficient profile of water body particles, characterized in that, Including: Extracting the water body area from the remote sensing image of the target water body; Obtaining the corresponding relationship between the surface chlorophyll concentration and the water body particle attenuation coefficient of the target water body; Calculating the surface chlorophyll concentration corresponding to each pixel in the water body area by using a preset algorithm, and obtaining the surface water body particle attenuation coefficient of each pixel according to the corresponding relationship; Identifying the water body particle attenuation coefficient profile type of each pixel, where the water body particle attenuation coefficient profile type includes a uniformly mixed type, a Gaussian-like type, and an exponential decay type; Calculating the water body particle attenuation coefficient profile data of each pixel according to the preset method corresponding to the profile type, where the water body particle attenuation coefficient profile data of the pixels belonging to the Gaussian-like type is calculated based on the Bayesian optimization method; including: When the water body particle attenuation coefficient profile type of the pixel is the exponential decay type, using the following formula to calculate the profile data of the water body particle attenuation coefficient: cp660 z = cp660 surf ,(z < MLD) cp660 z = cp660 surf × exp(-k exp / (z - MLD)), (z ≥ MLD) where k exp is a preset constant parameter; MLD is the mixed layer depth at the pixel; z is the currently calculated profile depth; cp660 surf is the surface particle attenuation coefficient corresponding to the pixel; cp660 z is the water body particle attenuation coefficient of the pixel at the profile depth of z; When the water body particle attenuation coefficient profile type of the pixel is the Gaussian-like type, using the following formula to calculate the profile data of the water body particle attenuation coefficient: cp660 z = cp660 surf ×(A gau × exp(-(z - U) 2 / D 2 ) - B gau × z + 1) where z is the current calculated profile depth; cp660 z is the attenuation coefficient of water particles of the pixel at the profile depth of z; cp660 surf is the surface particle attenuation coefficient corresponding to the pixel; A gau , B gau , U, and D are constant parameters determined by the Bayesian optimization method respectively.

2. The method according to claim 1, characterized in that, The obtaining the corresponding relationship between the surface chlorophyll concentration and the water body particle attenuation coefficient of the target water body includes: Collecting the historical data of the surface chlorophyll concentration and the historical data of the water body particle attenuation coefficient at multiple observation stations in the target water body; Obtaining the following corresponding relationship according to the historical data of the above two: log(cp660) = B1 × log(Chla) 2 + B2 × log(Chla) - INT Where, cp660 is the surface water body particle attenuation coefficient at a wavelength of 660 nm below the water surface; Chla is the surface chlorophyll concentration; B1, B2, and INT are the regression coefficients obtained by comparing the historical data of the above two through regression analysis.

3. The method according to claim 2, wherein The calculating the surface chlorophyll concentration corresponding to each pixel in the water body area by using a preset algorithm includes: Preprocessing the remote sensing image of the target water body; Obtaining the remote sensing reflectance of each pixel in the water body area; Calculating the surface chlorophyll concentration corresponding to the remote sensing reflectance of each pixel by using a preset chlorophyll concentration inversion algorithm.

4. The method according to claim 1, wherein The identifying the water body particle attenuation coefficient profile type of each pixel includes: Calculating the euphotic layer depth at the pixel based on the surface chlorophyll concentration; Obtaining the water depth and the mixed layer depth at the pixel; Calculating the LaRaD ratio of the pixel, LaRaD = water depth / mixed layer depth; Judging whether the euphotic layer depth at the pixel is greater than 50, When the euphotic layer depth at the pixel is greater than 50, judging whether the LaRaD ratio is less than 7.5, If so, determining that the water body particle attenuation coefficient profile type of the pixel is the uniformly mixed type; If not, determining that the water body particle attenuation coefficient profile type of the pixel is the Gaussian-like type; When the euphotic layer depth at the pixel is not greater than 50, judging whether the LaRaD ratio is less than 2, If so, determining that the water body particle attenuation coefficient profile type of the pixel is the uniformly mixed type; If not, determining that the water body particle attenuation coefficient profile type of the pixel is the exponential decay type.

5. The method according to claim 4, characterized in that, Using the following formula to calculate the euphotic layer depth at the pixel: Zeu = 10 ^ (1.35 - 0.4026 × log10(Chla) + 0.0375 × log10(Chl) ^ 2) where Zeu is the euphotic layer depth; Chla is the surface chlorophyll concentration of the pixel.

6. The method according to claim 1, characterized in that, The initial values of the constant parameters U and D are determined in the following manner: Based on the historical data of the surface chlorophyll concentration, the subsurface maximum depth, and the subsurface maximum width, obtain the correlation between the surface chlorophyll concentration Chla and the subsurface maximum depth U: log(U) = 0.446 × log(Chla) + 0.413 And, obtain the correlation between the surface chlorophyll concentration Chla and the subsurface maximum width D: log(D) = 0.406 × log(Chla) - 0.132 Use the above correlations to calculate the initial values of the constant parameters U and D corresponding to the pixel.

7. The method according to claim 6, wherein The method further includes: Determine the observation constraint conditions by the following method: cp660 Mean-Zeu = M1 × exp(M2 × cp660 surf ) cp660 Mean-Zmax = N1 × exp(N2 × cp660 surf ) cp660 Mean-2*Zmax = P1 × cp660 surf + P2 Among them, cp660 Mean-Zeu is the average value of the water body particle attenuation coefficient in the euphotic layer; cp660 Mean-Zmax is the average value of the particulate attenuation coefficient of the water body within the maximum depth from the water surface to the subsurface layer; cp660 Mean-2*Zmax is the average value of the particulate attenuation coefficient of the water body within the depth from the water surface to twice the maximum value of the subsurface layer; M1, M2, N1, N2, P1, and P2 are respectively preset constant parameters.

8. The method according to claim 7, characterized in that Determine the constant parameters A, B, U, and D corresponding to each pixel in the following manner: gau , B gau , U, D final values: Calculate the particulate attenuation coefficient cp660 of the water body between the maximum depth of the subsurface layer and the maximum depth of the preset water body according to the following formula Deep-Zmax :[[]]END]] cp660 Deep-Zmax = cp660 surf × exp(-0.025 × (z - z max )) where z max is the maximum depth of the subsurface layer, obtained by multiplying U by Zeu; Construct the following objective function: Dif = Dif Zeu 2 + Dif Zmax 2 + Dif 2-Zmax 2 + Dif U 2 +(Dif D / 10) 2 + Dif model 2 Among them, Dif Zeu , Dif Zmax and Dif 2-Zmax respectively represent the differences between the calculation results of cp660 M ean-Zeu, cp660 M ean-Zmax and cp660 Mean-2*Zmax and the true observed data; Dif U and Dif D are respectively the differences between the calculation results of U and D and the true observed data; Dif model is the difference between the calculation result of cp660 Deep-Zmax and the true observed data; Optimize the objective function using the Bayesian optimization method to obtain A gau , B gau , and the final values of U and D.

9. A system for calculating the attenuation coefficient profile of water body particles, characterized in that, The system executes the method for calculating the water body particle attenuation coefficient profile according to any one of claims 1-8.

Images