Method for evaluating lake and reservoir nutrition state based on water color index and red edge band chroma angle

Abstract

The present application provides a kind of lake and reservoir nutrition state evaluation method based on water color index and red edge band chroma angle, comprising the following steps: S1, lake and reservoir water extraction and image cutting obtains lake and reservoir water range GF-6WFV / Sentinel-2MSI remote sensing reflectivity image;S2, water color index FUI and red edge band chroma angle α' calculation;S3, joint water color index FUI and red edge band chroma angle α' single time phase lake and reservoir nutrition state evaluation;S4, based on the spatial evaluation result of different time phase lake and reservoir nutrition state, realize the dynamic evaluation of lake and reservoir nutrition state.The present application can reduce the influence of high CDOM or suspended solids concentration on nutrition state evaluation, reduce the proportion of medium-nutrition state lake and reservoir water body dominated by CDOM or suspended solids being misidentified as eutrophic, and improve the precision of lake and reservoir nutrition state remote sensing evaluation.

Description

Technical Field

[0001] This invention relates to the field of remote sensing technology, and in particular to a method for evaluating the trophic status of lakes and reservoirs based on water color index and red-edge band chromaticity angle. Background Technology

[0002] Lakes and reservoirs are important carriers of surface water resources, providing abundant water while also serving vital functions such as flood control, aquaculture, navigation, and ecosystem services. With increasing human impact, eutrophication in lakes and reservoirs has become a serious problem, attracting global attention and urgently needing resolution. The 2021 "China Ecological Environment Status Bulletin" released the trophic status assessment results of 209 important lakes and reservoirs in my country, revealing that eutrophic lakes and reservoirs accounted for 27.3%, indicating that eutrophication remains a severe problem in my country. Conducting dynamic assessments of lake and reservoir trophic status is a prerequisite for studying the evolution and driving factors of trophic status, and it is also a requirement for lake and reservoir water environment management. Achieving dynamic assessments of lake and reservoir trophic status has significant scientific and practical value.

[0003] Traditional methods for assessing the trophic status of lakes and reservoirs involve on-site water sample collection and laboratory testing to obtain water quality parameters such as chlorophyll a concentration, transparency, total nitrogen, total phosphorus, and permanganate index. The trophic status index (TSI) or TLI is then calculated using a single or weighted approach based on multiple water quality indicators. The trophic status is then classified into five levels: oligotrophic, mesotrophic, slightly eutrophic, moderately eutrophic, and severely eutrophic. While traditional methods offer advantages such as comprehensive evaluation indicators and high accuracy, they involve multiple processes including manual sample collection and laboratory analysis, resulting in high time and labor costs. Furthermore, they only assess the trophic status at the sampling point, limiting the spatial representativeness of the results. In contrast, dynamic assessment of lake and reservoir trophic status using satellite remote sensing offers advantages such as low cost and high spatiotemporal coverage, and has become an important supplementary method for trophic status assessment.

[0004] Currently, the color index (FUI) of lakes and reservoirs is calculated using the reflectance of blue, green, and red bands in satellite remote sensing imagery. Based on the intrinsic relationship between FUI and the trophic status index calculated using chlorophyll a (Chl-a) concentration as the main parameter, a trophic status assessment model for lakes and reservoirs based on FUI classification is constructed. This model can classify lakes and reservoirs into three trophic states: oligotrophic, mesotrophic, and eutrophic. Some scholars have used FUI-based trophic status assessment models and imagery from MODIS and Landsat to conduct dynamic assessments of lake and reservoir trophic status at regional or global scales. However, since FUI is influenced by the three water color elements—chlorophyll a, total suspended solids, and yellow matter (CDOM)—for lakes and reservoirs dominated by CDOM and suspended solids concentration, the FUI-based trophic status assessment model may overestimate the trophic status of some mesotrophic lakes and reservoirs due to the influence of CDOM or suspended solids concentration. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes a method for assessing the trophic status of lakes and reservoirs based on the water color index and red-edge band chromaticity angle. Targeting WFV sensor images from the Gaofen-6 satellite and MSI images from the Sentinel-2 satellite, which possess red-edge bands, this method introduces the red-edge band chromaticity angle index and combines it with the water color index to assess the trophic status of lakes and reservoirs. This assessment method can more accurately identify the trophic status of lakes and reservoirs dominated by CDOM or suspended matter, thereby improving the accuracy of trophic status assessment.

[0006] The specific technical solution is as follows:

[0007] The method for assessing the trophic status of lakes and reservoirs based on water color index and red-edge band chromaticity angle includes the following steps:

[0008] S1. Extracting and cropping images of lake and reservoir water bodies to obtain GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images of the lake and reservoir water areas.

[0009] Preset lake and reservoir buffer vectors, use the buffer vectors to crop the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images covering the lake and reservoir, automatically extract the water body vector range, and obtain the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images of the lake and reservoir water area through a mask.

[0010] S2, calculation of water color index FUI and red-edge band chromaticity angle α′

[0011] The water color index FUI and the chromaticity angle α′ of the red edge band were calculated using the blue (B1 / B2), green (B2 / B3), and red (B3 / B4) bands and the red (B3 / B4), red edge 1 (B5 / B5), and red edge 2 (B6 / B6) bands of the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image of the lake and reservoir area, respectively.

[0012] S3, combined water color index FUI and red-edge band chromaticity angle α′ single-phase lake / reservoir trophic status assessment

[0013] An improved decision tree evaluation model for the trophic status of lakes and reservoirs was constructed by combining the water color index FUI and the red-edge band chromaticity angle α′. The spatial distribution of the trophic status of lakes and reservoirs was obtained by inputting the water color index FUI and the red-edge band chromaticity angle α′ based on GF-6 WFV imagery.

[0014] S4. Based on the spatial evaluation results of lake and reservoir trophic status in different time phases, realize dynamic evaluation of lake and reservoir trophic status.

[0015] The specific steps are as follows:

[0016] S1. Extracting and cropping images of lake and reservoir water bodies to obtain GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images of the lake and reservoir water areas.

[0017] A preset lake / reservoir buffer vector is used. The GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image covering the lake / reservoir is then cropped using this buffer vector. The water body vector range is automatically extracted, and a mask is used to obtain the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image of the lake / reservoir area. The specific steps are as follows:

[0018] Based on prior knowledge of the geometric characteristics of lakes and reservoirs, buffer vectors are drawn. If an approximate vector range of the lake or reservoir exists, a buffer zone can be directly generated, with the total area of ​​the buffer zone being 2-3 times the area of ​​the lake or reservoir. The buffer vectors are used to clip GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images that have undergone geometrical, radiometric, and atmospheric corrections. An automatic lake / reservoir water body extraction method based on object-oriented and bimodal thresholding is used to automatically extract the vector range of the lake / reservoir water body. The GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images are then masked using the lake / reservoir water body vector range to generate a GF-6 WFV image of the lake / reservoir water area.

[0019] S2, calculation of water color index FUI and red-edge band chromaticity angle α′

[0020] The water color index FUI and the chromaticity angle α′ of the red edge band were calculated using the blue (B1 / B2), green (B2 / B3), and red (B3 / B4) bands and the red (B3 / B4), red edge 1 (B5 / B5), and red edge 2 (B6 / B6) bands of the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image of the lake and reservoir area, respectively.

[0021] Based on the chromaticity angle and color calculation rules in the CIE-XYZ color system proposed by the International Commission on Illumination (CIE), the water color index FUI and the chromaticity angle α′ of the red edge band were calculated using the blue (B1 / B2), green (B2 / B3), and red (B3 / B4) bands and the red (B3 / B4), red edge 1 (B5 / B5), and red edge 2 (B6 / B6) bands of the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image of the lake and reservoir area.

[0022] S3, combined water color index FUI and red-edge band chromaticity angle α′ single-phase lake / reservoir trophic status assessment

[0023] An improved decision tree evaluation model for the trophic status of lakes and reservoirs was constructed by combining the water color index FUI and the red edge band chromaticity angle α′. The spatial distribution of the trophic status of lakes and reservoirs was obtained by inputting the water color index FUI and the red edge band chromaticity angle α′ based on GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images.

[0024] S4. Based on the spatial evaluation results of lake and reservoir trophic status in different time phases, realize the spatiotemporal dynamic evaluation of lake and reservoir trophic status.

[0025] The trophic status of lakes and reservoirs at different time phases was obtained using GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images. The proportion of pixels with different trophic statuses was used as a comprehensive evaluation index to conduct a dynamic evaluation of the trophic status of lakes and reservoirs.

[0026] The advantages of this invention are:

[0027] By introducing the red edge bands of GF-6 WFV / Sentinel-2 MSI remote sensing reflectance imagery, and using the red (B3), red edge 1 (B5), and red edge 2 (B6) bands of GF-6 WFV remote sensing reflectance imagery and the red (B3), red edge 1 (B4), and red edge 2 (B5) bands of Sentinel-2 MSI remote sensing reflectance imagery, the chromaticity angle α′ of the red edge bands is calculated. The chromaticity angle α′ of the red edge bands is used to assist in determining the trophic status of lakes and reservoirs when FUI ≥ 11. This can reduce the impact of CDOM or suspended matter on trophic status assessment, reduce the proportion of mesotrophic lakes and reservoirs dominated by CDOM or suspended matter that are incorrectly identified as eutrophic, and improve the accuracy of remote sensing assessment of trophic status of lakes and reservoirs. Attached Figure Description

[0028] Figure 1 This is a flowchart of the invention;

[0029] Figure 2 It is the two-dimensional chromaticity map in the embodiment;

[0030] Figure 3 This is a schematic diagram of the division of chromaticity coordinates, chromaticity angles, and water color index in the CIE-XYZ chromaticity diagram in the embodiment;

[0031] Figure 4a This is one of the simulated spectra of GF-6 and Sentinel-2 in the embodiments;

[0032] Figure 4b This is the second of the simulated spectra of GF-6 and Sentinel-2 in the embodiments;

[0033] Figure 4c This is the third of the simulated spectra of GF-6 and Sentinel-2 in the embodiments;

[0034] Figure 5a This is the relationship model between GF-6 WFV and FUI, α′ and TLI in the embodiment;

[0035] Figure 5b This is the S2 MSI and FUI, α′ and TLI relationship model in the embodiment;

[0036] Figure 6 This is the lake and reservoir trophic status evaluation model that combines the water color index and the red-edge band chromaticity angle in the embodiment.

[0037] Figure 7 This is a schematic diagram of the spatial distribution of trophic states in some lakes obtained based on GF-6 WFV in the embodiment. Detailed Implementation

[0038] To make the technical problems, solutions, and advantages of this invention clearer, a detailed description will be provided below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the protection scope of this invention.

[0039] like Figure 1 As shown in the embodiment of the present invention, the method for evaluating the trophic status of lakes and reservoirs based on water color index and red-edge band chromaticity angle includes:

[0040] S1. Extracting and cropping images of lake and reservoir water bodies to obtain GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images of the lake and reservoir water areas.

[0041] A preset lake / reservoir buffer vector is used. The GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image covering the lake / reservoir is then cropped using this buffer vector. The water body vector range is automatically extracted, and a mask is used to obtain the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image of the lake / reservoir area. The specific steps are as follows:

[0042] Based on prior knowledge of the geometric characteristics of lakes and reservoirs, buffer vectors are drawn. If an approximate vector range of the lake or reservoir exists, a buffer zone can be directly generated, with the total area of ​​the buffer zone being 2-3 times the area of ​​the lake or reservoir. The buffer vectors are used to clip GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images that have undergone geometrical, radiometric, and atmospheric corrections. An automatic lake / reservoir water body extraction method based on object-oriented and bimodal thresholding is used to automatically extract the vector range of the lake / reservoir water body. The GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images are then masked using the lake / reservoir water body vector range to generate a GF-6 WFV image of the lake / reservoir water area.

[0043] S2, calculation of water color index FUI and red-edge band chromaticity angle α′

[0044] The water color index FUI and the chromaticity angle α′ of the red edge band were calculated using the blue (B1 / B2), green (B2 / B3), and red (B3 / B4) bands and the red (B3 / B4), red edge 1 (B5 / B5), and red edge 2 (B6 / B6) bands of the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image of the lake and reservoir area, respectively.

[0045] Based on the chromaticity angle and color calculation rules in the CIE-XYZ color system proposed by the International Commission on Illumination (CIE), the water color index FUI and the chromaticity angle α′ of the red edge band were calculated using the blue (B1 / B2), green (B2 / B3), and red (B3 / B4) bands and the red (B3 / B4), red edge 1 (B5 / B5), and red edge 2 (B6 / B6) bands of the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance image of the lake and reservoir area.

[0046] (1) Calculation of water color index

[0047] Water color is the visual perception characteristic of visible light reflected from a body of water when observed by the human eye. Water color is related to the physical properties of the water body and is also influenced by factors such as lighting conditions, viewing angle, and the observer's visual perception. To achieve quantitative representation of color, the International Commission on Illumination (CIE) proposed the CIE-XYZ color system, using the three primary colors [X], [Y], and [Z] to replace the three primary colors [R], [G], and [B] in the CIE-RGB color system. The conversion relationship between CIE-RGB and CIE-XYZ is as follows (CIE, 1932):

[0048]

[0049] The formulas for calculating the three primary color values ​​of the CIE-XYZ system based on the continuous visible light spectrum of 380-700nm are as follows (CIE, 1932):

[0050]

[0051]

[0052] Φ(λ)=S(λ)ρ(λ) (4) In the formula: K is the adjustment coefficient; S(λ) is the relative spectral energy distribution of the illumination source; ρ(λ) is the remote sensing reflectance; Φ(λ) is the reflectance spectrum; and These are the color matching functions corresponding to [X], [Y], and [Z], respectively.

[0053] To better demonstrate the CIE-XYZ color system, CIE created a two-dimensional chromaticity diagram, such as... Figure 2 As shown, the two-dimensional coordinates x and y on the chromaticity diagram, X, Y, and Z, are calculated using the following formula:

[0054]

[0055] On a chromaticity diagram, any coordinate (x, y) represents a color. The chromaticity coordinates (0.3333, 0.3333) represent an equal-energy white light point, indicating an equal mixture of the three primary colors. The angle formed by rotating clockwise from this equal-energy white light point along the opposite direction of the y-axis is called the chromaticity angle. The formula for calculating the chromaticity angle is as follows:

[0056] α=ARCTAN2(x-0.3333,y-0.3333) (6)

[0057] In the formula: α is the chromaticity angle, α∈[0°, 360°]; ARCTAN2 is the bivariate arctangent function.

[0058] The Water Color Index (FUI) refers to the Forel-UIe colorimetric table, a water color grading standard proposed by Francois Alphonse Forel and Willi Ule in the 1890s. This table divides water bodies from blue-green to reddish-brown into 21 color levels, corresponding to FUI values ​​from 1 to 21. The earliest method for measuring FUI involved comparing the water color at half the transparency depth with the FUI colorimetric table, selecting the FUI value that best matched the water color as the water color index. Wernand et al. (2010, 2013) prepared solutions with FUI values ​​from 1 to 21 in the laboratory and determined the CIE chromaticity coordinates and chromaticity angles corresponding to different FUI values ​​by testing the transmission spectra of different solutions. Figure 3 As shown, a FUI lookup table based on chromaticity angle was established (Table 1), which lays the foundation for monitoring water color index using remote sensing technology.

[0059] Table 1

[0060]

[0061] Based on the above method for calculating the water color index, the calculation process for the water color index based on GF-6 WFV / Sentinel-2 MSI is as follows:

[0062] 1) Calculate the tristimulus values ​​X, Y, and Z based on the blue, green, and red bands of GF-6 WFV / Sentinel-2 MSI (Equation (1));

[0063] 2) Calculate the image pixel chromaticity coordinates (x, y) based on the X, Y, and Z values ​​(Equation (5));

[0064] 3) Calculate the chromaticity angle α corresponding to the image pixel in the two-dimensional chromaticity space (Equation (6));

[0065] 4) Determine the FUI value using the FUI lookup table and chromaticity angle, as shown in Table 1.

[0066] (2) Calculation of the chromaticity angle α′ of red / red edge 1 / red edge 2

[0067] Using the red / red-edge 1 / red-edge 2 bands to replace the blue / green / red bands respectively, calculate the X, Y, and Z values ​​of the three primary colors, and then calculate the chromaticity coordinates and the corresponding chromaticity angle α′.

[0068] S3, combined water color index FUI and red-edge band chromaticity angle α′ single-phase lake / reservoir trophic status assessment

[0069] An improved decision tree evaluation model for the trophic status of lakes and reservoirs was constructed by combining the water color index FUI and the red edge band chromaticity angle α′. The spatial distribution of the trophic status of lakes and reservoirs was obtained by inputting the water color index FUI and the red edge band chromaticity angle α′ calculated based on GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images.

[0070] The steps for constructing an improved lake / reservoir trophic status decision tree evaluation model that combines the water color index FUI and the red-edge band chromaticity angle α′ are as follows:

[0071] (1) Satellite spectral simulation based on water hyperspectral remote sensing reflectance and satellite spectral response function

[0072] Satellite spectra can be simulated using the hyperspectral reflectance of water bodies and satellite spectral response functions. The calculation formula is as follows:

[0073] R rs (Bi)=∫R rs ·f(Bi) / ∫f(Bi) (7) In the formula: R rs (Bi) represents the remote sensing reflectance of the GF-6 WFV / Sentinel-2MSIBi band; R rs is the measured hyperspectral remote sensing reflectance; f(Bi) is the spectral response value of the GF-6 WFV / Sentinel-2MSIBi band; ∫ is the integral calculation.

[0074] In this embodiment, the water body hyperspectral reflectance R is used. rs The data comes from the Hydrolight simulation dataset (IOCCG, 2006) released by the International Organization for Ocean Color Coordination (IOCCG). This dataset contains 500 simulated water spectra (400-800 nm) of different concentrations of phytoplankton pigments, colored soluble organic matter (CDOM), and non-pigmented suspended matter. These 500 high-spectral reflectance R... rs Data and simulations of the spectral response function of GF-6 WFV / Sentinel-2MSI multispectral spectral components, such as GF-6 WFV / Sentinel-2MSI. Figure 4a , Figure 4b and Figure 4c As shown.

[0075] (2) Calculation of water color index FUI and red-edge band chromaticity angle α′ based on satellite simulated spectrum

[0076] The water color index FUI and the red edge band α′ were calculated using the blue (B1 / B2), green (B2 / B3), and red (B3 / B4) bands and the red (B3 / B4), red edge 1 (B5 / B5), and red edge 2 (B6 / B6) bands of the simulated spectrum of GF-6 WFV / Sentinel-2MSI.

[0077] Because there are differences between the X, Y, and Z values ​​obtained by calculating the chromaticity angle using the blue-green-red three-band method (Equation (1)) and the hyperspectral integral calculated within the blue-green-red sampling range (Equation (2)), the chromaticity angles obtained by the two calculation methods are different. Therefore, it is necessary to construct a conversion model for calculating the chromaticity angle α(1) using Equation (1) and calculating the chromaticity angle α(2) using Equation (2). The idea for constructing the conversion model is as follows:

[0078] 1) Using the 500 water body spectra published by IOCCG, the X, Y, and Z values ​​were calculated using formula (2), and then 500 sets of chromaticity angles α (2) were calculated.

[0079] 2) Using the GF-6 WFV / Sentinel-2MSI simulated spectra of 500 water body spectra published by IOCCG, the X, Y, and Z values ​​were calculated using formula (1), and then 500 sets of chromaticity angles α (1) were calculated.

[0080] 3) Using chromaticity angle α(1) as the independent variable and chromaticity angle α(2) as the dependent variable, the correction equation is obtained by fitting. The chromaticity angle correction equation for GF-6WFV / Sentinel-2MSI is as follows:

[0081] ①GF-6 WFV Chromaticity Angle Correction Equation

[0082] α(2)=-0.0000156962α(1) 3 +0.0026374749α(1) 2 +1.2794577705α(1)-9.1015173159

[0083] ②Sentinel-2A MSI chromaticity angle correction equation

[0084] α(2)=-0.000027496348317925α(1) 3 +0.00817300651908366α(1) 2+0.569736190054841α(1)+11.326697659071

[0085] ③Sentinel-2B MSI chromaticity angle correction equation

[0086] α(2)=-0.0000305380099492325α(1) 3 +0.00934485019101149α(1) 2 +0.439946045414801α(1)+14.6142941650078

[0087] (3) Construction of an improved lake and reservoir trophic status decision tree evaluation model that combines the water color index FUI and the red-edge band chromaticity angle α′.

[0088] In this embodiment of the invention, 500 water body spectra and phytoplankton pigment chlorophyll a concentrations published by IOCCG are used as modeling data. The modeling steps are as follows:

[0089] 1) Calculate the water color index FUI and red-edge band chromaticity angle α′ using GF-WFV / Sentinel-2MSI simulated spectra.

[0090] 2) The total trophic status index (TLI) was calculated using the concentration of chlorophyll a, a phytoplankton pigment, and other methods. Figure 5a and Figure 5b The intrinsic relationships between FUI, α′, and TLI shown are used to construct an improved decision tree evaluation model for lake and reservoir trophic status.

[0091] TLI(Chl-a)=10*(2.5+1.086·Ln(C(Chl-a))) (8)

[0092] In the formula: TLI(Chl-a) is the nutrient status index based on chlorophyll a concentration; C(Chl-a) is the chlorophyll a concentration; Ln is the natural logarithm. According to TLI(Chl-a), nutrient status can be divided into oligotrophic (TLI<30), mesotrophic (30≤TLI≤50), and eutrophic (TLI>50).

[0093] Figure 6 An improved decision tree evaluation model for the trophic status of lakes and reservoirs was constructed by combining the water color index FUI and the red-edge band chromaticity angle α′. In the figure, Thre is the red-edge band chromaticity angle threshold, and the Thre values ​​for GF-6 WFV and Sentinel-2 MSI images are 69 and 75, respectively.

[0094] Compared to the lake and reservoir trophic status assessment method based on water color index (Model 1), the remote sensing monitoring and assessment method combining water color index and red-edge band chromaticity angle (Model 2) significantly improves the accuracy of monitoring and assessment of eutrophic water bodies. The accuracy of eutrophication monitoring and assessment based on simulated GF-6 WFV data for Model 1 and Model 2 is 84.1% (122 / 145) and 95.8% (113 / 118), respectively; the accuracy based on simulated Sentinel-2 MSI data for Model 1 and Model 2 is 85.0% (118 / 139) and 99.0% (107 / 108), respectively.

[0095] The red-edge band chromaticity angle α′ can characterize the reflection peak of chlorophyll a near 700 nm. For water bodies with a FUI ≥ 11, if the high FUI is due to high chlorophyll a concentration, the red-edge band chromaticity angle α′ will be larger; conversely, if the high FUI is due to high CDOM or high suspended solids concentration, the red-edge band chromaticity angle α′ will be smaller. This model can effectively distinguish between water bodies with high chlorophyll a concentration and those with high CDOM or high turbidity using the red-edge band chromaticity angle α′, significantly improving the accuracy of trophic status assessment.

[0096] For GF-6 WFV and Sentinel-2 MSI images, the color index FUI and red-edge band chromaticity angle α′ of the lake / reservoir are calculated using step S2. These values ​​are then input into the improved lake / reservoir trophic status decision tree evaluation model to obtain the spatial distribution of the lake / reservoir trophic status. Figure 7 This is a schematic diagram of the spatial distribution of trophic states in some lakes obtained based on GF-6 WFV.

[0097] S4. Based on the spatial evaluation results of lake and reservoir trophic status in different time phases, realize the spatiotemporal dynamic evaluation of lake and reservoir trophic status.

[0098] By using GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images of lakes and reservoirs at different time phases, the trophic status of lakes and reservoirs at different time phases can be obtained. The proportion of pixels with different trophic statuses can be used as a comprehensive evaluation index to carry out dynamic evaluation of the trophic status of lakes and reservoirs.

Claims

1. A method for evaluating the trophic status of lakes and reservoirs based on water color index and red-edge band chromaticity angle, characterized in that, Includes the following steps: S1. Extraction and cropping of lake and reservoir water bodies to obtain GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images of the lake and reservoir water areas; Preset lake and reservoir buffer vectors, use the buffer vectors to crop the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images covering the lake and reservoir, automatically extract the water body vector range, and obtain the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images of the lake and reservoir water area through a mask; S2, Water Color Index (FUI), and Red Edge Band Chromaticity Angle calculate; The specific method for S2 is as follows: The water color index FUI and the chromaticity angle of the red edge band were calculated using the blue, green, and red bands and the red, red edge 1, and red edge 2 bands of the GF-6 WFV / Sentinel-2 MSI remote sensing reflectance imagery of the lake / reservoir area. ; The calculation process for the water color index is as follows: 1) Calculate the tristimulus values ​​X, Y, and Z based on the blue, green, and red bands of GF-6 WFV or Sentinel-2 MSI: 2) Calculate the image pixel chromaticity coordinates (x, y) based on the X, Y, and Z values; 3) Calculate the chromaticity angle α corresponding to the image pixel in the two-dimensional chromaticity space; 4) Determine the FUI value using the FUI lookup table and chromaticity angle; Calculate the chromaticity angle of the red band using the red, red-edge 1, and red-edge 2 bands. : Using the red, red-edge 1, and red-edge 2 bands to represent the blue, green, and red bands respectively, calculate the X, Y, and Z values ​​of the three primary colors, and then calculate the chromaticity coordinates and the corresponding chromaticity angles of the red-edge bands. ; S3, Combined Water Color Index (FUI) and Red Edge Band Chromaticity Angle Single-phase trophic status assessment of lakes and reservoirs; If FUI ≤ 7, then it is considered malnourished; If 8 ≤ FUI ≤ 10, then it is considered mesonutrient; If FUI ≥ 11, and the chromaticity angle of the red-edge band is... >Thre, then it is rich in nutrients; conversely, if If ≤Thre, then it is considered mesotrophic; Thre is the chromaticity angle of the red-edge band. Threshold values ​​for GF-6 WFV and Sentinel-2 MSI images were 69 and 75, respectively. S4. Based on the spatial evaluation results of lake and reservoir trophic status in different time phases, realize dynamic evaluation of lake and reservoir trophic status.

2. The method for evaluating the trophic status of lakes and reservoirs based on water color index and red-edge band chromaticity angle according to claim 1, characterized in that, The specific method for S1 is as follows: Based on prior knowledge of the geometric characteristics of lakes and reservoirs, buffer vectors are drawn. If an approximate vector range of the lake or reservoir exists, a buffer zone can be directly generated, with the total area of ​​the buffer zone being 2-3 times the area of ​​the lake or reservoir. The buffer vectors are used to clip GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images that have undergone geometric correction, radiometric calibration, and atmospheric correction. An automatic lake / reservoir water body extraction method based on object-oriented and bimodal thresholding is used to automatically extract the vector range of the lake / reservoir water body. The GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images are then masked using the lake / reservoir water body vector range to generate a GF-6 WFV image of the lake / reservoir water area.

3. The method for evaluating the trophic status of lakes and reservoirs based on water color index and red-edge band chromaticity angle according to claim 1, characterized in that, The specific method for S3 is as follows: An improved decision tree evaluation model for the trophic status of lakes and reservoirs was constructed by combining the water color index (FUI) and the red-edge band chromaticity angle (α). The inputs were the water color index (FUI) and the red-edge band chromaticity angle (α) of the lakes and reservoirs based on GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images. To obtain the spatial distribution of trophic status in lakes and reservoirs.

4. The method for evaluating the trophic status of lakes and reservoirs based on water color index and red-edge band chromaticity angle according to claim 1, characterized in that, The specific method for S4 is as follows: The trophic status of lakes and reservoirs at different time phases was obtained using GF-6 WFV / Sentinel-2 MSI remote sensing reflectance images. The proportion of pixels with different trophic statuses was used as a comprehensive evaluation index to conduct a dynamic evaluation of the trophic status of lakes and reservoirs.

Images