A method for estimating the coverage and biomass of herbaceous vegetation in coastal salt marshes and application thereof

Abstract

The present application belongs to the technical field of coastal salt marsh monitoring, and particularly relates to a method for estimating the coverage and biomass of coastal salt marsh herbaceous vegetation and application. The present application takes multispectral satellite images as the main data basis, adopts the normalized vegetation index threshold method to identify the salt marsh herbaceous vegetation coverage area, adopts the computer classification method to identify the salt marsh vegetation type, calculates the coverage of each type of vegetation on each pixel independently to generate a vegetation coverage result file, takes the coverage and biomass of the field investigation as the actual sample data to establish a quantitative inversion model of the coverage and biomass, and quantitatively inverts the biomass of the salt marsh herbaceous vegetation. The present application uses the differences in the vegetation spectral index under different coverages to invert the vegetation coverage from the high-resolution multispectral remote sensing image, and further uses the correlation between the coverage and the biomass to further invert the vegetation biomass, and performs the calculation or estimation pixel by pixel, which is suitable for the large-area and time-space synchronous coverage and biomass estimation of the salt marsh herbaceous vegetation.

Description

Technical Field

[0001] This invention belongs to the field of coastal salt marsh monitoring technology, specifically relating to a method and application for estimating the herbaceous vegetation cover and biomass of coastal salt marshes. Background Technology

[0002] Coastal salt marsh vegetation is an important component of the coastal ecosystem, playing a variety of important ecological functions, including coastal protection, disaster prevention and mitigation, carbon sequestration and climate regulation, water purification and pollutant interception, and biodiversity maintenance.

[0003] Currently, marine ecological authorities conduct field surveys of salt marsh vegetation cover and biomass, obtaining information on the cover and biomass of a specific type of salt marsh vegetation within each quadrat by setting up a certain number of quadrats. Regarding cover, the average cover of all quadrats is used to characterize the cover index of this salt marsh vegetation type; regarding biomass, the average biomass per unit area is used to characterize the biomass index of this salt marsh vegetation type by taking the average biomass of all quadrats. The distribution area of ​​salt marsh vegetation interpreted by remote sensing is used to estimate the overall biomass of the monitoring area. For field quadrat sampling surveys of salt marsh vegetation cover and biomass, although the scientific, systematic, representative, consistent, and accurate nature of the cross sections and survey stations were considered in the design of the survey plan and station layout, the long survey period and the inability to reach some stations due to natural conditions made it difficult to meet the expected representativeness of salt marsh vegetation cover and biomass obtained by field quadrat sampling surveys. It is also difficult to fully reflect the spatiotemporally synchronous cover and biomass information of the survey area.

[0004] Salt marsh vegetation cover and biomass are important indicators for evaluating the quality of coastal ecosystems and carbon storage, and play a crucial role in assessing the health of coastal ecosystems. Therefore, developing a large-scale, spatiotemporally synchronized method for monitoring or estimating vegetation cover and biomass is essential for the operationalization of salt marsh vegetation monitoring.

[0005] Remote sensing technology, especially high-resolution multispectral remote sensing, has an inherent advantage in accurately acquiring large-area, spatiotemporally synchronous, and continuous information about survey areas. Vegetation indices established from the spectral information in high-resolution multispectral remote sensing data can directly extract vegetation regions and also directly reflect vegetation density in numerical form. Domestic and international scholars have conducted extensive research on vegetation cover and biomass inversion based on vegetation indices, confirming the effectiveness and practicality of vegetation indices in vegetation cover and biomass inversion. Their research results provide important theoretical and methodological support for this application. Summary of the Invention

[0006] To address the problems existing in the prior art, this invention provides a method and application for estimating the coverage and biomass of herbaceous vegetation in coastal salt marshes.

[0007] The technical solution adopted by this invention to solve its technical problem is as follows: A method for estimating the cover and biomass of herbaceous vegetation in coastal salt marshes, comprising the following steps: S1. Based on multispectral satellite imagery with a spatial resolution better than 2 meters, the multispectral data includes at least two spectral bands: red and near-infrared. S2. The normalized vegetation index threshold method was used to identify the herbaceous vegetation cover area in the salt marsh. S3. Use computer classification methods to identify the types of salt marsh vegetation in areas covered by salt marsh vegetation; S4. Using the identified salt marsh herbaceous vegetation classification results as a mask file, calculate the coverage of each type of vegetation in each pixel independently on the NDVI calculation result file, and generate a vegetation coverage result file. S5. Using the coverage and biomass from the field survey as actual sample data, establish a quantitative inversion model for the coverage and biomass of herbaceous vegetation in salt marshes. S6. Quantitative inversion of biomass of salt marsh herbaceous vegetation: Based on the quantitative inversion model between biomass and coverage of different salt marsh herbaceous vegetation established in step S5, and taking into account the ratio of field survey plot size to remote sensing pixel size, the biomass of each type of vegetation in each pixel is calculated independently for different salt marsh herbaceous vegetation types, and a quantitative inversion result file of vegetation biomass is generated.

[0008] Preferably, in step S1, multispectral UAV orthophotos are used as the data basis for key areas or areas where satellite images of salt marsh herbaceous vegetation are difficult to identify.

[0009] Preferably, in step S2, the NDVI recognition threshold of the vegetation-covered area fluctuates depending on the actual vegetation conditions reflected in the image.

[0010] Preferably, in step S3, the computer classification method includes, but is not limited to, supervised classification, unsupervised classification, object-oriented classification, machine learning-based classification, and deep learning-based classification.

[0011] Preferably, in step S3, the identification results are supplemented with necessary on-site verification and manual correction to ensure the accuracy of the classification results.

[0012] Preferably, in step S4, the formula for calculating vegetation coverage is: ; In the formula, FVC is the vegetation coverage, NDVInon = NDVImin - 0.01, NDVImin is the minimum NDVI value identified as the vegetation type in the corresponding salt marsh herbaceous vegetation area, and NDVImax is theoretically the maximum NDVI value of the corresponding salt marsh vegetation type in the corresponding vegetation area. If the number of pixels with a value of NDVImax is small and cannot represent the NDVI situation of the densest area of ​​the vegetation type, then based on the pixel-level NDVI statistics, the NDVI value taken from NDVImax down to 0.5‰-5‰ of the total effective pixels in the area can be used as an approximate NDVImax value to participate in the vegetation coverage calculation.

[0013] Preferably, in step S6, taking the quantitative inversion model of cover and biomass as a linear model as an example, the formula for calculating biomass per unit pixel in remote sensing images is as follows: ; in, Biomass per unit pixel in a remotely sensed image. These are the constant parameters determined in the linear model regression analysis. Vegetation cover calculated per pixel in a remotely sensed image. per unit pixel area To investigate the area of ​​the sample plot.

[0014] Preferably, when step S3 determines that there is only a single type of salt marsh vegetation in the target area, the salt marsh vegetation area extracted by the NDVI threshold method in step S2 is the salt marsh vegetation coverage area of ​​that type.

[0015] The above-mentioned methods for estimating the cover and biomass of coastal salt marshes herbaceous vegetation were applied to the zoning evaluation of herbaceous vegetation cover and biomass in salt marshes.

[0016] Preferably, based on the salt marsh herbaceous vegetation cover result file calculated in step S4 and the salt marsh herbaceous vegetation biomass result file calculated in step S6, the region of interest or the region to be evaluated is delineated as needed, and the cover and biomass are statistically analyzed according to the delineated region. Thus, the relevant evaluation work of the salt marsh herbaceous vegetation in a specific region can be carried out based on the two indicators of cover and biomass.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This application employs remote sensing technology, particularly high-resolution satellite remote sensing technology, to invert vegetation cover from high-resolution multispectral remote sensing images by leveraging the differences in vegetation spectral indices under different cover conditions. Furthermore, it utilizes the correlation between cover and biomass to invert vegetation biomass. This allows for pixel-by-pixel calculation or estimation of cover and biomass in the entire salt marsh vegetation distribution area. Then, by delineating regions of interest or target regions, it enables flexible regional and hierarchical evaluation of salt marsh vegetation based on cover and biomass.

[0018] 2. This application presents an inversion method applicable to the large-scale spatiotemporal synchronous estimation of cover and biomass of herbaceous vegetation in salt marshes. It guides technical personnel in functional departments to conduct related cover and biomass estimation work in professional remote sensing data processing software or other self-developed information software based on this method. It enables synchronous and uniform estimation of cover and biomass of herbaceous vegetation in salt marshes on a large spatial scale, overcomes the limitations of traditional sample surveys, provides reliable methodological support for early warning monitoring and evaluation of typical salt marsh vegetation ecosystems, assessment of ecological restoration effects, and evaluation of ecosystem health, and lays the foundation for estimating carbon storage in blue carbon vegetation. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the target area determined according to work requirements in the embodiment.

[0020] Figure 2 This is a schematic diagram of the preprocessed multispectral remote sensing image of the target area in the embodiment.

[0021] Figure 3 This is a schematic diagram of the NDVI calculation results for the target area in the embodiment.

[0022] Figure 4 This is a schematic diagram of the screening results of the target area salt marsh vegetation coverage area in the embodiment.

[0023] Figure 5 This is a schematic diagram of the target area salt marsh vegetation type identification results in the example.

[0024] Figure 6 This is a schematic diagram of the NDVI calculation results for the salt marsh-Suaeda salsa region in the example.

[0025] Figure 7 This is a schematic diagram of the NDVI calculation results of the salt marsh-reed area in the example.

[0026] Figure 8 This is a schematic diagram showing the calculation results of the salt marsh-Suaeda salsa coverage in the target area in the example.

[0027] Figure 9 shows the percentage of salt marsh-Suaeda salsa coverage in the target area in the embodiment.

[0028] Figure 10This is a schematic diagram showing the calculation results of salt marsh-Suaeda biomass in the target area in the example.

[0029] Figure 11 This is a schematic diagram illustrating the delineation of the evaluation area within the target region in the embodiment.

[0030] Figure 12 This is a schematic diagram of the salt marsh-Suaeda salsa coverage results in the evaluation area of ​​the example.

[0031] Figure 13 This is a schematic diagram of the salt marsh-Suaeda biomass results in the evaluation area of ​​the example.

[0032] Figure 14 This is a flowchart of the estimation method used in this application. Detailed Implementation

[0033] To facilitate understanding of the present invention, it will be described in more detail below with reference to the accompanying drawings and specific embodiments. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in this specification. Rather, these embodiments are provided to provide a more thorough and complete understanding of the disclosure of the present invention.

[0034] A method for estimating the cover and biomass of herbaceous vegetation in coastal salt marshes is proposed, and further applied to the zonal evaluation of herbaceous vegetation cover and biomass in salt marshes, combined with... Figure 14 Understanding, Solution Introduction: 1. Selection of remote sensing data Using multispectral satellite imagery with a spatial resolution better than 2 meters as the main data basis, multispectral UAV orthophotos are used as the data basis for key areas or areas where herbaceous vegetation in salt marshes is difficult to identify from satellite imagery. Necessary radiometric correction, geometric correction, and image cropping, mosaicking, and fusion work are carried out to ensure that the quality of the remote sensing data used is normalized and that the survey area is fully covered. The multispectral data should include at least two spectral bands: red and near-infrared.

[0035] 2. Determination of vegetation areas in salt marshes Normalized Difference Vegetation Index (NDVI) is used to identify areas of herbaceous vegetation cover in salt marshes. Generally, areas with NDVI > 0 are selected as vegetation-covered areas. However, the NDVI threshold for identifying vegetation-covered areas can be adjusted appropriately based on the actual vegetation conditions reflected in the imagery. The formula for calculating the Normalized Difference Vegetation Index is as follows: ; In the formula, Reflectivity in the near-infrared band; This refers to the reflectivity in the red light band.

[0036] 3. Determination of vegetation types in salt marshes When the target area contains multiple types of salt marsh vegetation, computer classification methods (including but not limited to supervised classification, unsupervised classification, object-oriented classification, machine learning-based classification, and deep learning-based classification) are used to identify the salt marsh vegetation types within the salt marsh vegetation coverage area. The identification results are then supplemented with necessary on-site verification and manual correction to ensure accuracy. When the target area contains only a single type of salt marsh vegetation, the salt marsh vegetation area extracted in step 2 using the NDVI thresholding method is the salt marsh vegetation coverage area for that type.

[0037] 4. Calculation of herbaceous vegetation cover in salt marshes Using the identified salt marsh herbaceous vegetation classification results as a mask file, the vegetation cover of each type of vegetation in each pixel is independently calculated on the NDVI calculation result file according to the following formula, and a vegetation cover result file is generated.

[0038] The formula for calculating vegetation cover is: ; In the formula, FVC is the vegetation coverage, NDVInon = NDVImin - 0.01 (NDVImin is the minimum NDVI value identified as the vegetation type within the corresponding salt marsh herbaceous vegetation area), and NDVImax is theoretically the maximum NDVI value of the corresponding salt marsh vegetation type within the area. If the number of pixels with a value of NDVImax is statistically small and cannot represent the NDVI situation of the densest area of ​​the vegetation type, then based on the pixel-level NDVI statistics, the NDVI value taken from NDVImax down to 0.5‰-5‰ of the total effective pixels in the area can be used as an approximate NDVImax value to participate in the vegetation coverage calculation.

[0039] 5. Determination of the quantitative inversion model for herbaceous vegetation cover and biomass in salt marshes Using the vegetation cover and biomass from field surveys as actual sample data, a quantitative inversion model of vegetation cover and biomass is established. Taking the linear model as an example, the formula for calculating vegetation biomass is: ; In the formula, For vegetation biomass, is the value of vegetation coverage, and k is a parameter.

[0040] 6. Quantitative inversion of biomass of herbaceous vegetation in salt marshes Based on the calculated cover results of salt marsh herbaceous vegetation, following the quantitative inversion models established in step 5 for different salt marsh herbaceous vegetation biomass and cover, and considering the ratio of field survey plot size to remote sensing pixel size, the biomass of each type of vegetation in each pixel is calculated independently according to the following formula, generating a quantitative inversion result file for vegetation biomass. Taking the quantitative inversion model of cover and biomass as a linear model as an example, the calculation formula is as follows: ; Biomass per unit pixel in a remotely sensed image. These are the constant parameters determined in step 5 of the regression analysis. Vegetation cover calculated per pixel in a remotely sensed image. per unit pixel area To determine the area of ​​the sample plots, the units for the unit pixel area and the unit of the sample plot area should be consistent during the calculation process. When using satellite remote sensing imagery as the data source, it is recommended to use the unit in square meters; when using sub-decimeter-level UAV remote sensing imagery as the data source, it is recommended to use the unit in square centimeters.

[0041] 7. Evaluation of herbaceous vegetation cover and biomass zoning in salt marshes Based on the salt marsh herbaceous vegetation cover results file calculated in step 4 and the salt marsh herbaceous vegetation biomass results file calculated in step 6, the region of interest or the region to be evaluated can be delineated as needed. Coverage and biomass can be statistically analyzed according to the delineated region. Based on the two indicators of cover and biomass, relevant evaluation work on the salt marsh herbaceous vegetation in a specific region can be carried out.

[0042] This application, based on actual data obtained from the operational work of marine ecological functional departments, and according to the distribution information and characteristics of salt marsh herbaceous vegetation, and building upon field surveys of salt marsh vegetation cover and biomass, and considering the actual needs of large-area cover and biological monitoring surveys, has developed a method for estimating salt marsh herbaceous vegetation cover and biomass based on multispectral remote sensing technology. This method can guide technical personnel in functional departments to conduct operational work related to cover and biomass estimation, achieving synchronous and uniform estimation of salt marsh herbaceous vegetation cover and biomass at large spatial scales. It can also guide the development of relevant salt marsh vegetation cover calculation and biomass inversion functional modules in domestically produced information software, thereby automating and informatizing the estimation of salt marsh herbaceous vegetation cover and biomass indicators.

[0043] The specific implementation of this method is described here, taking the joint use of remote sensing image processing software ENVI and geographic information processing software ArcGIS to estimate the herbaceous vegetation cover and biomass in salt marshes as an example.

[0044] 1. Determine the target area The target area is determined according to the needs of the survey and monitoring work. The target area is determined based on the purpose, task requirements and relevant standards of the work. Taking coastal zone work as an example, the upper and lower boundaries of the salt marsh vegetation monitoring and survey area are the seaward side of the shoreline to the 0-meter isobath, and the left and right boundaries are the administrative or geographical boundaries determined according to specific needs.

[0045] This embodiment takes the coastal salt marsh vegetation survey and monitoring work in a certain area as an example to determine the target area for remote sensing monitoring and survey, such as... Figure 1 As shown, the blue box indicates the target area.

[0046] 2. Selection and preprocessing of remote sensing data Within the target area, multispectral satellite imagery with imaging dates between July and September and a spatial resolution better than 2 meters is selected as the main data basis. For key areas or areas where salt marsh herbaceous vegetation is difficult to identify through satellite imagery, UAV multispectral orthophotos are selected as the data basis to ensure that the remote sensing data fully covers the survey area and that there is no cloud cover in the survey area. The selected multispectral remote sensing data should include at least two spectral bands: red light and near-infrared light.

[0047] Necessary radiometric and geometric corrections, as well as image cropping, mosaicking, and fusion, were performed on the selected remote sensing data to ensure the normalization of the data quality. A schematic diagram of the preprocessed multispectral remote sensing image of the target area is shown below. Figure 2 As shown.

[0048] In this embodiment, the following is a schematic diagram of the subsequent work of the specific implementation method of this application, based on the area in the above target area schematic diagram, selecting a multispectral satellite remote sensing image of a certain year.

[0049] 3. Extracting vegetation cover areas from salt marshes (1) Perform NDVI index calculation on the preprocessed multispectral remote sensing image and obtain the NDVI calculation results, such as Figure 3 As shown.

[0050] (2) Based on the NDVI calculation results of the survey area, and in conjunction with the multispectral remote sensing image, extract the NDVI results of more than 20 vegetation boundary points. After arithmetic averaging, take the NDVI arithmetic mean as the threshold for extracting the vegetation coverage area of ​​the survey area. Perform condition filtering on the NDVI result file (NDVI>= boundary point NDVI arithmetic mean). Convert the filtered result file (binarized raster) into a vector file and use the vector file as the ROI. Perform image cropping on the multispectral image. The cropping result is the salt marsh vegetation coverage area of ​​the survey area.

[0051] Taking the NDVI calculation results in this embodiment as an example, after arithmetic averaging, NDVI=0.08 is taken as the threshold for extracting vegetation cover in the survey area. The final target area of ​​salt marsh vegetation cover is as follows: Figure 4 As shown.

[0052] 4. Identify vegetation types in salt marshes Within the salt marsh vegetation coverage area, monitoring and classification are conducted based on remote sensing interpretation markers established during operational surveys in the region to identify the salt marsh vegetation types and coverage status within the area. Figure 5 This is a schematic diagram showing the results of identifying two types of salt marsh herbaceous vegetation, salt marsh-reed and salt marsh-Suaeda, in the target area in this embodiment.

[0053] After converting the classification results into vectors, spatial calculations can be performed on the corresponding salt marsh vegetation to obtain the coverage area of ​​the salt marsh vegetation. Taking the salt marsh-Suaeda salsa in the example, the calculated coverage area of ​​the salt marsh-Suaeda salsa is 5210.46 hectares.

[0054] 5. Calculate the herbaceous vegetation cover of the salt marsh. (1) Using the identified salt marsh herbaceous vegetation classification results as a mask file, the NDVI calculation result file is cropped category by category to obtain the NDVI calculation results under each salt marsh herbaceous vegetation category. In the example, the NDVI calculation results of the salt marsh-Suaeda salsa area identified in the target area are as follows: Figure 6 As shown. Figure 7 The NDVI calculation results for the salt marsh-reed area are shown.

[0055] (2) According to the salt marsh herbaceous vegetation category, the NDVI results of the corresponding category are statistically analyzed. Taking the NDVI of the target area salt marsh-Suaeda salsa calculated by the multispectral remote sensing data used in this embodiment as an example, the NDVI calculation results are shown in Table 1.

[0056] Table 1. Statistical table of NDVI calculation results for the target area salt marsh-Suaeda salsa region. Based on statistical results and the vegetation cover calculation formula: Following the principle that "NDVInon = NDVImin - 0.01 (NDVImin is the minimum NDVI value identified as the vegetation type within the corresponding salt marsh herbaceous vegetation area), and NDVImax is theoretically the maximum NDVI value for the corresponding salt marsh vegetation type within the area, if statistically the number of pixels with a value of NDVImax is small and cannot represent the NDVI situation of the densest area of ​​that vegetation type, then based on pixel-level NDVI statistics, NDVI values ​​from NDVImax down to 1‰-5‰ of the total effective pixels in the area can be used as approximate NDVImax values ​​for vegetation coverage calculation," NDVI = 0.07 was determined as the value of NDVInon, and NDVI = 0.33 as the value of NDVImax. The salt marsh-Suaeda salsa coverage calculation for the target area was then performed, and the final calculation formula was determined as follows: .

[0057] In ENVI software, the above coverage formula is expressed as: (b1 lt 0.08)*0+(b1 gt 0.33)*1+(b1 ge 0.08 and b1 le 0.33)*((b1-0.08) / (0.33-0.08)).

[0058] The calculation results of the salt marsh-Suaeda salsa coverage in the target area were obtained, such as... Figure 8 As shown.

[0059] The calculation results of salt marsh-Suaeda salsa coverage in the target area of ​​this embodiment were statistically analyzed pixel by pixel, and the statistical results are shown in Table 2.

[0060] Table 2. Statistical results of calculation of salt marsh-Suaeda salsa coverage in the target area. Based on the coverage statistics, flexible coverage interval statistical evaluation can be carried out on the target area. For example, the coverage interval ratio can be evaluated based on certain evaluation criteria. In this embodiment, the coverage interval ratio statistical evaluation of salt marsh-Suaeda salsa in the target area is shown in Table 3 and Figure 9, and is displayed using a bar chart as shown in Figure 9(a) or a pie chart as shown in Figure 9(b).

[0061] Table 3. Statistical table of the proportion of salt marsh-Suaeda salsa coverage in the target area. 6. Determine the quantitative inversion model for herbaceous vegetation cover and biomass in salt marshes. Using the canopy cover and biomass from the field survey as actual sample data, a quantitative inversion model of canopy cover and biomass is established. Taking the multispectral remote sensing image of the target area selected in this embodiment as an example, for salt marsh-Suaeda salsa, the correlation model between canopy cover and biomass of salt marsh-Suaeda salsa determined based on the field survey data is as follows: ; In the formula, For salt marsh-Suaeda biomass, The value represents the salt marsh-Suaeda salsa coverage, and 0.385 is the correlation coefficient between the salt marsh-Suaeda salsa coverage and biomass, in kilograms per square meter.

[0062] 7. Quantitative inversion of herbaceous vegetation biomass in salt marshes (1) Calculation of single pixel area of ​​remote sensing image Before applying the established correlation model between salt marsh herbaceous vegetation cover and biomass, it is necessary to first convert the ratio of field survey quadrat size to remote sensing image cell size before conducting quantitative inversion of salt marsh herbaceous vegetation biomass ...

[0063] Due to the imaging method of remote sensing images, there is a difference in area between nadir pixels and edge pixels, and remote sensing pixels are not strictly square. Therefore, when calculating the area of ​​a single pixel in a remote sensing image, it is not recommended to directly use the square of the remote sensing image resolution as the area of ​​a single pixel. To minimize the introduction of external errors in the calculation of salt marsh vegetation biomass, it is recommended to calculate the total number of pixels and the total pixel area for a specific salt marsh vegetation type before calculating the area of ​​a single pixel. The formula for calculating the area of ​​a single pixel is: ; Taking the multispectral remote sensing image of the target area selected in this embodiment as an example, for the salt marsh-Suaeda vegetation type, the single pixel area under the salt marsh-Suaeda type is calculated to be 4.747 square meters.

[0064] (2) Quantitative inversion of biomass After obtaining the unit pixel area under each salt marsh herbaceous vegetation type, the biomass of each type of vegetation in each pixel is independently calculated on the remote sensing image processing software according to the following formula, and a quantitative inversion result file of vegetation biomass is generated.

[0065] ; Biomass per unit pixel in a remotely sensed image. This represents the correlation coefficient between vegetation cover and biomass in a specific salt marsh. Vegetation cover calculated per pixel in a remotely sensed image. per unit pixel area To determine the area of ​​the sample plots, the units for the unit pixel area and the unit of the sample plot area should be consistent during the calculation process. When using satellite remote sensing imagery as the data source, it is recommended to use the unit in square meters; when using sub-decimeter-level UAV remote sensing imagery as the data source, it is recommended to use the unit in square centimeters.

[0066] Taking the multispectral remote sensing image of the target area in this embodiment and the salt marsh-Suaeda salsa identified based on the image as examples, the correlation model applied to the quantitative inversion of salt marsh-Suaeda salsa biomass in the example multispectral remote sensing image is as follows: ; This represents the biomass per pixel in the salt marsh-Suaeda salsa coverage area in this embodiment. This refers to the vegetation cover calculated per pixel in the salt marsh-Suaeda salsa area from remote sensing imagery.

[0067] According to the above formula, the biomass calculation results of the salt marsh-Suaeda salsa covered area in this embodiment were obtained in the remote sensing image processing software as follows: Figure 10 As shown.

[0068] By statistically analyzing the biomass calculation results of the salt marsh herbaceous vegetation coverage area, the total biomass of a certain type of salt marsh herbaceous vegetation in the target area can be obtained. Taking the salt marsh-Suaeda biomass calculation results of this embodiment as an example, the statistics are shown in Table 4.

[0069] Table 4. Statistical table of calculation results of salt marsh-Suaeda biomass in the target area. Based on the statistical results, the total number of pixels and total biomass of a certain salt marsh vegetation type within the target area can be obtained. Alternatively, the average biomass of the target area can be obtained based on the total biomass and the total coverage area of ​​this type of salt marsh vegetation. The formula for calculating the average biomass is as follows: ; The recommended unit for average biomass is "kg / m²" or "g / m²".

[0070] Taking the salt marsh-Suaeda salsa in this embodiment as an example, the total biomass of the salt marsh-Suaeda salsa in the target area is calculated to be 5609.92 tons, and the average biomass is 0.108 kg / m².

[0071] 8. Evaluation of herbaceous vegetation cover and biomass in salt marshes Based on the calculated salt marsh herbaceous vegetation cover and biomass results, one or more areas to be evaluated can be delineated according to actual needs. Cover and biomass zoning statistics can then be carried out according to the delineated areas to conduct zoning evaluation of salt marsh herbaceous vegetation cover and biomass in the relevant areas.

[0072] Taking the salt marsh-Suaeda salsa in the target area of ​​this embodiment as an example, the evaluation area (red polygonal area) is delineated, such as... Figure 11 As shown.

[0073] The salt marsh-Suaeda cover and biomass results calculated in the previous work were cropped using the evaluation area vector to obtain the salt marsh-Suaeda cover and biomass results within the evaluation area, such as... Figure 12-13 As shown.

[0074] The coverage and biomass of salt marsh and Suaeda salsa within the evaluation area are statistically analyzed to obtain corresponding statistical files. Based on these files, necessary analyses are performed to conduct the evaluation of the coverage and biomass of salt marsh and Suaeda salsa within the evaluation area. This includes statistics on the percentage of coverage intervals within the evaluation area, the average coverage of the evaluation area, the average biomass of the evaluation area, and the total biomass of the evaluation area. This part of the work is similar to the statistical analysis of the coverage and biomass of salt marsh vegetation in the target area and will not be elaborated further.

[0075] This application leverages the differences in vegetation spectral indices under different vegetation cap conditions to invert vegetation cap from high-resolution multispectral remote sensing images. Furthermore, it utilizes the correlation between cap and biomass to invert vegetation biomass. Based on multispectral high-resolution remote sensing data, a method for large-area spatiotemporal synchronous estimation of cap and biomass in salt marsh herbaceous vegetation is developed. This method guides technical personnel in relevant departments to conduct cap and biomass estimation work using professional remote sensing data processing software or other self-developed information software. It achieves synchronous and uniform estimation of cap and biomass in salt marsh herbaceous vegetation on a large spatial scale, overcoming the limitations of traditional quadrat sampling surveys. This provides reliable methodological support for early warning monitoring and evaluation of typical salt marsh vegetation ecosystems, assessment of ecological restoration effects, and ecosystem health evaluation, laying the foundation for estimating carbon storage in blue carbon vegetation.

[0076] Compared with traditional field survey techniques, which offer the advantage of obtaining accurate vegetation cover and biomass information, obtaining sufficiently representative quadrat data over large areas requires significant manpower and resources, and is also limited by the difficulty of field sampling, resulting in substantial time and labor costs. This application utilizes remote sensing technology, particularly high-resolution satellite remote sensing, which offers significant advantages in large-scale, integrated, refined, and visualized vegetation monitoring. This technology allows for pixel-by-pixel calculation or estimation of vegetation cover and biomass across the entire salt marsh vegetation distribution area. Then, by delineating regions of interest or target areas, flexible zoning and grading assessments of salt marsh vegetation based on cover and biomass can be conducted. The technical solution provided in this application is an effective means of monitoring large-area salt marsh vegetation cover and biomass.

[0077] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A method for estimating the cover and biomass of herbaceous vegetation in coastal salt marshes, characterized in that, Includes the following steps: S1. Based on multispectral satellite imagery with a spatial resolution better than 2 meters, the multispectral data includes at least two spectral bands: red and near-infrared. S2. The normalized vegetation index threshold method was used to identify the herbaceous vegetation cover area in the salt marsh. S3. Use computer classification methods to identify the types of salt marsh vegetation in areas covered by salt marsh vegetation; S4. Using the identified salt marsh herbaceous vegetation classification results as a mask file, calculate the coverage of each type of vegetation in each pixel independently on the NDVI calculation result file, and generate a vegetation coverage result file. S5. Using the coverage and biomass from the field survey as actual sample data, establish a quantitative inversion model for the coverage and biomass of herbaceous vegetation in salt marshes. S6. Quantitative inversion of biomass of salt marsh herbaceous vegetation: Based on the quantitative inversion model between biomass and coverage of different salt marsh herbaceous vegetation established in step S5, and taking into account the ratio of field survey plot size to remote sensing pixel size, the biomass of each type of vegetation in each pixel is calculated independently for different salt marsh herbaceous vegetation types, and a quantitative inversion result file of vegetation biomass is generated.

2. The method for estimating the cover and biomass of coastal salt marsh herbaceous vegetation according to claim 1, characterized in that, In step S1, multispectral UAV orthophotos are used as the data basis for key areas or areas where herbaceous vegetation in salt marshes is difficult to identify from satellite images.

3. The method for estimating the cover and biomass of coastal salt marsh herbaceous vegetation according to claim 1, characterized in that, In step S2, the NDVI recognition threshold of the vegetation-covered area fluctuates depending on the actual vegetation conditions reflected in the image.

4. The method for estimating the cover and biomass of coastal salt marsh herbaceous vegetation according to claim 1, characterized in that, In step S3, computer classification methods include, but are not limited to, supervised classification, unsupervised classification, object-oriented classification, machine learning-based classification, and deep learning-based classification.

5. The method for estimating the cover and biomass of coastal salt marsh herbaceous vegetation according to claim 1, characterized in that, In step S3, the identification results are supplemented with necessary on-site verification and manual correction to ensure the accuracy of the classification results.

6. The method for estimating the cover and biomass of coastal salt marsh herbaceous vegetation according to claim 1, characterized in that, In step S4, the formula for calculating vegetation coverage is: ； In the formula, FVC is the vegetation coverage, NDVInon = NDVImin - 0.01, NDVImin is the minimum NDVI value identified as the vegetation type in the corresponding salt marsh herbaceous vegetation area, and NDVImax is theoretically the maximum NDVI value of the corresponding salt marsh vegetation type in the corresponding vegetation area. If the number of pixels with a value of NDVImax is small and cannot represent the NDVI situation of the densest area of ​​the vegetation type, then based on the pixel-level NDVI statistics, the NDVI value taken from NDVImax down to 0.5‰-5‰ of the total effective pixels in the area can be used as an approximate NDVImax value to participate in the vegetation coverage calculation.

7. The method for estimating the cover and biomass of coastal salt marsh herbaceous vegetation according to claim 1, characterized in that, In step S6, taking the quantitative inversion model of cover and biomass as a linear model as an example, the formula for calculating biomass per unit pixel on remote sensing images is as follows: ； in, Biomass per unit pixel in a remotely sensed image. These are the constant parameters determined in the linear model regression analysis. Vegetation cover calculated per pixel in a remotely sensed image. per unit pixel area To investigate the area of ​​the sample plot.

8. The method for estimating the cover and biomass of coastal salt marsh herbaceous vegetation according to claim 1, characterized in that, When step S3 determines that there is only a single type of salt marsh vegetation in the target area, the salt marsh vegetation area extracted by the NDVI threshold method in step S2 is the salt marsh vegetation coverage area of ​​that type.

9. The application of the method for estimating the cover and biomass of coastal salt marsh herbaceous vegetation as described in any one of claims 1-8, characterized in that, It is applied to the evaluation of herbaceous vegetation cover and biomass zoning in salt marshes.

10. The application according to claim 9, characterized in that, Based on the salt marsh herbaceous vegetation cover results file calculated in step S4 and the salt marsh herbaceous vegetation biomass results file calculated in step S6, the region of interest or the region to be evaluated can be delineated as needed. Coverage and biomass zoning statistics can be performed according to the delineated regions, so that relevant evaluation work can be carried out on the salt marsh herbaceous vegetation in a specific region based on the two indicators of cover and biomass.

Images