An evaluation method for shoreline change-tidal flat erosion and deposition coordination

Abstract

This invention discloses a method for assessing the coordinated erosion and deposition of shoreline changes and tidal flats. First, a study area is selected, and multi-temporal high-resolution remote sensing images and corresponding tidal level data at the imaging times are acquired and validated. Second, coastline information is extracted and corrected using measured and revised shorelines. Then, a digital shoreline analysis system is used to quantitatively assess the characteristics of shoreline changes. Subsequently, based on each remote sensing image, water body information extraction, filtering and denoising, image binarization, morphological optimization, and edge vector extraction are performed sequentially to obtain time-series remote sensing waterline data. Finally, relying on waterline elevation control points, a digital elevation model is constructed using a tidal flat topography inversion method. After validation, the dynamics of erosion and deposition in the tidal flats are analyzed, and the system reveals the laws of coastal evolution. This method achieves a coordinated assessment of shoreline and tidal flat changes, possesses strong practicality and scalability, and improves the accuracy and efficiency of monitoring coastal evolution processes.

Description

Technical Field

[0001] This invention relates to the field of tidal flat landform evolution assessment technology, and in particular to an assessment method for shoreline change-tidal flat erosion and deposition synergy. Background Technology

[0002] As a zone where land and sea interact, the coastal zone possesses unique ecological vulnerability and sensitivity to environmental change, while simultaneously harboring abundant natural resources and biodiversity. The coastal zone exhibits a dual nature, serving both ecological services and economic development value. Human activity is evident, making it a region crucial for human survival and development, and a site of highly exploited economic resources. The interaction between land and sea forms tidal flats. As a vital component of the coastal zone, tidal flats are defined as the coastal area between the multi-year average high tide line and the low tide line, serving as sedimentary beaches receiving sediment from both land and sea. As an important part of coastal wetlands, the ecological function of tidal flats is not only to provide habitats for various organisms but also to effectively mitigate the damage caused by extreme storm surges and to provide flexible indicators of sea-level rise. Through continuous siltation and seaward extension, tidal flats possess the potential to become a dynamically renewable spatial resource reserve, suitable for mariculture, land reclamation, salt production, and other activities, thus providing crucial support for human economic activities.

[0003] In recent years, with global climate change and intensive human development, coastal zones are facing unprecedented challenges, and the ecological functions of tidal flats are being seriously threatened. The risk of coastal erosion disasters affecting coastal areas of my country is increasing. Therefore, conducting research on coastal erosion and understanding the dynamic processes of coastal changes is crucial for protecting human communities and the natural environment, and promoting marine economic development.

[0004] However, traditional survey methods face numerous challenges in monitoring shoreline changes and obtaining elevation information. The rapid evolution of tidal flat topography, the complexity of the hydrodynamic environment, and the consumption of human and material resources make the limitations of conventional methods increasingly apparent. In addition, some historical topographic data are too old to meet the current needs for in-depth research on tidal flat evolution. Summary of the Invention

[0005] To address the above technical problems, this invention provides a method for assessing the synergistic effect of shoreline change and tidal flat erosion and deposition, comprising the following steps:

[0006] S1. Select the study area, acquire Landsat series satellite remote sensing images and TPXO tide level data of the corresponding imaging time, and verify the tide level simulated by TPXO by combining the actual measured station tide level data.

[0007] S2. Extract coastline information and correct it in conjunction with the reconstructed coastline. Statistically analyze the net coastline movement distance and endpoint change rate of the coastline within the study area. Use the digital coastline analysis system based on the ArcGIS platform to perform quantitative analysis of the coastline in the study area. The net coastline movement distance is expressed as the distance between coastlines at different times, which is used to reflect the overall horizontal movement of the coastline. The endpoint change rate is expressed as the rate of change of the coastline distance over time, which is used to measure the amount of change in the coastline per unit time.

[0008] S3. Based on the collected remote sensing images, remote sensing water boundary data at different times are obtained sequentially through water body information extraction, filtering, image binarization, morphological noise reduction, and edge extraction vector methods.

[0009] S4. Based on the waterline elevation control points, a digital elevation model is constructed and verified using the tidal flat topographic inversion method. Using remote sensing images acquired by Landsat series satellites at low tide times with similar tide levels over the years, exposed tidal flats in the study area are identified. Similar tide levels mean that the average error of the tide level values ​​at multiple feature points is less than 0.15m. Contour lines are generated by combining the waterline data obtained in step S3. The contour lines are imported into the ArcGIS platform to construct a digital elevation model of the tidal flats. Finally, the tidal flat erosion and deposition trends and the dynamic changes of the coastline are analyzed.

[0010] The technical solution further defined in this invention is:

[0011] Furthermore, in step S1, the Landsat series satellite remote sensing images include Landsat 5 TM, Landsat 7ETM+, Landsat 8 OLI, and Landsat 9 OLI satellites, and the remote sensing images are from the USGS official data website; the tide data uses the tide data simulated by the TPXO 9.0 model as the data source, and the simulation results of the TPXO model are verified by selecting measured data from multiple tide stations.

[0012] As described above, in the assessment method of shoreline change and tidal flat erosion and deposition, the calculation process of the digital shoreline analysis system in step S2 is as follows: First, a baseline is drawn, which is parallel to the shoreline; second, the shoreline is set, and all shoreline data involved in the calculation are in the same feature class layer; then, a cross section is generated, and the digital shoreline analysis system is used to generate several cross sections perpendicular to the baseline on the seaward side; finally, the net shoreline movement distance and the endpoint change rate (EPR) are calculated by the distance between the intersection points of each shoreline and the cross section line.

[0013] As described above, the assessment method for the synergy between shoreline change and tidal flat erosion and deposition calculates the net shoreline movement distance (NSM) using the following formula:

[0014] ;

[0015] in, This refers to the most recent coastline. This indicates a later phase of the coastline;

[0016] The endpoint rate of change (EPR) is calculated using the following formula:

[0017] ;

[0018] in, This represents the rate of change of the endpoint of the shoreline along a certain cross-section between adjacent years. This represents the distance from the j-th phase coastline to the baseline along this cross-section line. This represents the distance from the i-th phase coastline to the baseline along the same cross-section line. This represents the difference between the j-th and i-th years of the coastline along this cross-section.

[0019] As described above, in the assessment method for the synergistic effect of shoreline change and tidal flat erosion and deposition, step S3 involves using the enhanced water body index method to extract water body information and enhance the water body information. The formula is as follows:

[0020] ;

[0021] Green represents the green light band, which is the second band in TM and ETM+ satellite imagery and the third band in OLI imagery; NIR represents the near-infrared band, which is the fourth band in TM and ETM+ imagery and the fifth band in OLI; SWIR1 represents the shortwave infrared band 1, which is the fifth band in TM and ETM+ imagery and the sixth band in OLI.

[0022] As described above, in the evaluation method for the synergistic effect of shoreline change and tidal flat erosion and deposition, step S3 involves using convolutional filtering. Assuming the original image is f(i,j), an M×N window is opened in the upper left corner of the original image. Then, an M×N order filter function K(m,n) is selected. This function is then used to weight the information within the window. Finally, the processed result is superimposed onto the original image to form the filtered image Q(i,j), represented as:

[0023] ;

[0024] Where Q(i,j) represents the result of filtering the image at position (i,j), f(i,j) represents the initial preimage at position (i,j), K(m,n) represents the filtering kernel matrix function, and m and n represent the parameters of the matrix function in the x and y directions, respectively.

[0025] As described above, in the evaluation method for the synergistic effect of shoreline change and tidal flat erosion and deposition, step S3 involves obtaining the optimal threshold using the Otsu method and binarizing the image. Let the number of image pixels be N, the grayscale range be [0, K], the image grayscale level be L, and the number of pixels with grayscale level i be... Then the probability of grayscale value i appearing is:

[0026] ;

[0027] in, This means that a pixel is randomly selected from the entire image, and the gray value of this pixel is the probability of i. express Summation;

[0028] The pixels in the image are divided into class A and class B according to a grayscale threshold t. Class A consists of pixels with grayscale values ​​between [0, t], and class B consists of pixels with grayscale values ​​between [t+1, K]. The probabilities of class A and class B are as follows:

[0029] ;

[0030] The mean gray values ​​for class A and class B are as follows:

[0031] ;

[0032] The average gray level of the entire image is:

[0033] ;

[0034] Define the between-class variance as:

[0035] ;

[0036] in, and Let A and B represent the probabilities of A and B, respectively. and Let A and B represent the mean gray values ​​respectively; let t be in the range [0,K], and increment by 1 step, when The t value corresponding to the maximum value is the optimal threshold.

[0037] As described above, in the evaluation method of shoreline change-tidal flat erosion and deposition synergy, in step S3, the opening and closing operations in morphological filtering are used for noise reduction. The opening operation based on morphological filtering first applies erosion to peel off the bright areas, and then uses dilation operation to restore the main outline; the closing operation first dilates and then erodes, which is used to remove isolated noise points in the water area.

[0038] As described above, in the evaluation method of shoreline change-tidal flat erosion and deposition, in step S3, the Laplacian operator is used to obtain continuous waterline information for the denoised binarized image. The binarized image is then processed by Laplacian filtering in the software ENVI to detect edge information in the image, and instantaneous waterline data of a certain scene image is obtained, thus completing the extraction of the waterline.

[0039] As described above, in the assessment method of shoreline change-tidal flat erosion and deposition, in step S4, the instantaneous tide level result at the time of remote sensing image imaging is simulated using TPXO and assigned to the waterline obtained in step S3 to obtain contour data. The contour data is then imported into ArcGIS to construct an irregular triangular grid. The irregular triangular grid data is then converted into raster data using ArcGIS's function of automatically generating irregular triangular grids from features, thus constructing a digital elevation model of the tidal flat. When identifying exposed tidal flats, a true color band combination of Landsat series satellite imagery, namely red, green, and blue bands, is used to ultimately highlight the land features of the study area.

[0040] The beneficial effects of this invention are:

[0041] (1) In this invention, by integrating image processing algorithms such as Enhanced Water Index (EWI), Otsu's method for adaptive thresholding, and morphological denoising, manual intervention is significantly reduced and noise interference is effectively suppressed, thereby ensuring the accuracy and consistency of waterline extraction and laying a reliable data foundation for subsequent terrain inversion.

[0042] (2) In this invention, the time-series Landsat imagery and TPXO tidal model data are innovatively used to invert the tidal flat topography through the waterline elevation control point, which effectively overcomes the limitations of high cost and long cycle of on-site measurement, and provides an economical and efficient technical means for large-scale and long-term tidal flat geomorphological monitoring.

[0043] (3) In this invention, based on the mature ArcGIS platform and ENVI and other tools, the process is clear, the steps are standardized, and it is easy to promote and apply to different types of coastal zones, providing strong technical support for coastal zone resource management, erosion prevention and control and ecological protection decision-making. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0045] Figure 2Figure 1 is a schematic diagram of the TPXO tide level data verification results in the embodiments of the present invention, wherein (a) is a schematic diagram of the TPXO tide level data verification results at the mouth of the Guanhe River, (b) is a schematic diagram of the TPXO tide level data verification results at the mouth of the Sheyang River, and (c) is a schematic diagram of the TPXO tide level data verification results at Dafeng Port.

[0046] Figure 3 This is a schematic diagram of the average EPR distribution in the study area in an embodiment of the present invention. Detailed Implementation

[0047] This embodiment provides an assessment method for the synergistic effect of shoreline change and tidal flat erosion and deposition, such as... Figure 1 As shown, it includes the following steps:

[0048] S1. Select the study area, acquire Landsat series satellite remote sensing images and corresponding TPXO (tide level prediction model) tide level data at the imaging time, and verify the tide level simulated by TPXO by combining the actual measured station tide level data.

[0049] Landsat satellite remote sensing imagery includes Landsat 5 TM, Landsat 7 ETM+, Landsat 8 OLI, and Landsat 9 OLI satellites. The imagery is sourced from the USGS official data website (https: / / earthexplorer.usgs.gov). Tide level data was generated using TPXO 9.0 (Tidal Model Exchange-Ocean 9.0) model simulations as the data source, and measured data from multiple tide level stations were selected to validate the TPXO model simulation results.

[0050] S2. Extract coastline information and correct it by combining it with the latest revised coastline. Statistically calculate the net shoreline movement (NSM) and end point rate (EPR) of the coastline within the study area. Use the Digital Shoreline Analysis System (DSAS) to conduct quantitative analysis of the coastline in the study area. DSAS is a professional shoreline change analysis tool developed by the U.S. Geological Survey based on the ArcGIS platform. It aims to assess the spatiotemporal evolution characteristics of coastlines through quantitative methods, mainly calculating the NSM and EPR of the coastline within the study area. It is widely used in studies on coastal erosion, siltation, and the impact of human activities.

[0051] The DSAS calculation process begins with drawing the baseline, with the principle being that the baseline should be as parallel to the coastline as possible. Next, the coastline is set up, and all coastline data involved in the calculation must be within the same feature class layer. Then, cross-sections are generated, using DSAS to create several sections perpendicular to the baseline on the seaward side. Finally, the NSM and EPR are calculated based on the distance between the intersection points of each coastline and the cross-section lines.

[0052] NSM represents the distance between coastlines at different times, reflecting the overall horizontal movement of the coastline. It is a comprehensive indicator, and its formula is:

[0053] ;

[0054] in, This refers to the most recent coastline. It indicates a later phase of the coastline.

[0055] EPR is a specific representation of the rate of change of coastline distance over time. It is an important indicator for measuring the amount of change in coastline per unit time, reflecting the spatial differences in coastline changes. Its formula is:

[0056] ;

[0057] in, This represents the rate of change of the endpoint of the shoreline along a certain cross-section between adjacent years. This represents the distance from the j-th phase coastline to the baseline along this cross-section line. This represents the distance from the i-th phase coastline to the baseline along the same cross-section line. This represents the difference between the j-th and i-th years of the coastline along this cross-section.

[0058] S3. The waterline serves as the instantaneous boundary between land and water. Based on the collected remote sensing images, remote sensing waterline data at different times are obtained through methods such as water body information extraction, filtering, image binarization, morphological noise reduction, and edge extraction vectors.

[0059] Water body information extraction refers to the enhancement of water body information using the Enhanced Water Index (EWI) method. EWI not only has good sensitivity to water bodies but also exhibits considerable inhibition of non-water body areas. Its formula is as follows:

[0060] ;

[0061] Green represents the green light band, which is the second band in TM and ETM+ satellite imagery and the third band in OLI imagery; NIR represents the near-infrared band, which is the fourth band in TM and ETM+ imagery and the fifth band in OLI; SWIR1 represents the shortwave infrared band 1, which is the fifth band in TM and ETM+ imagery and the sixth band in OLI.

[0062] Next, filtering is performed. Based on the EWI grayscale remote sensing image, convolutional filtering is used to sharpen and smooth the edges of ground features. If the final appearance of ground features is not clear enough, high-pass filtering is used to remove the low-frequency components of the image while maintaining the integrity of the high-frequency components containing object boundary information; if the ground features appear coarse, low-pass filtering is used to weaken the high-frequency components of the image, thereby achieving the effect of smoothing the image.

[0063] Assuming the original image is f(i,j), an M×N window is opened in the upper left corner of the original image. Then, an M×N order filter function K(m,n) is selected. This function is then used to weight the information within the window. Finally, the processed result is superimposed onto the original image to form the filtered image Q(i,j), represented as:

[0064] ;

[0065] Where Q(i,j) represents the result of filtering the image at position (i,j), f(i,j) represents the initial preimage at position (i,j), K(m,n) represents the filtering kernel matrix function, and m and n represent the parameters of the matrix function in the x and y directions, respectively.

[0066] After grayscale images are processed by high-pass and low-pass filtering, they are then binarized. The binarization process mainly utilizes the Otsu method to obtain the optimal threshold. The Otsu method analyzes the statistical distribution characteristics of the pixel grayscale image and uses variance as a measure of separability to automatically determine the optimal threshold. When the inter-class variance between the target and background regions reaches its maximum value, the binarization result obtained at this point is considered to have the best separation.

[0067] The image is binarized; let the number of pixels in the image be N (the total number of pixels in the image), the grayscale range be [0, K] (referring to the range of possible pixel brightness values ​​in the image; in a common 8-bit grayscale image, the grayscale value ranges from 0 to 255, where 0 represents the darkest and 255 represents the brightest), the image grayscale level be L, and the number of pixels with grayscale level i is... Then the probability of grayscale value i appearing is (that is, the proportion of pixels with grayscale value i to the total number N):

[0068] ;

[0069] in, This means that a pixel is randomly selected from the entire image, and the gray value of this pixel is exactly the probability of i. and It is a value, and it is usually written in uppercase when summing.

[0070] The pixels in an image are divided into class A and class B according to a grayscale threshold t. Class A consists of pixels with grayscale values ​​between [0, t], and class B consists of pixels with grayscale values ​​between [t+1, K]. The probabilities of class A and class B are respectively (the proportion of pixels belonging to class A and class B in the entire image):

[0071] ;

[0072] The mean gray values ​​for classes A and B are respectively (representing the average gray value of all pixels in classes A and B, i.e., "how bright the background area is overall" and "how bright the target area is overall"):

[0073] ;

[0074] The average grayscale value of the entire image is (the average grayscale value of all pixels in the entire image, equivalent to the overall brightness level of the image):

[0075] ;

[0076] Define the inter-class variance as (used to measure the degree of distinction between class A and class B):

[0077] ;

[0078] in, and Let A and B represent the probabilities of A and B, respectively. and Let A and B represent the mean gray values ​​respectively; let t be in the range [0,K], and increment by 1 step, when The t value corresponding to the maximum value is the optimal threshold.

[0079] Next is morphological denoising. The binarized image contains some noise during segmentation, making it difficult to obtain continuous vector edges. To eliminate this noise, opening and closing operations from morphological filtering are used for denoising.

[0080] Opening operations prioritize applying erosion to remove small, bright areas (the erosion process primarily targets small, weakly connected bright noise points, making them unrecoverable during subsequent dilation, thus suppressing noise), and then use dilation operations to restore the main outline. This approach is highly effective in suppressing minor interference. It can highlight key features in the image, especially in land areas, while also separating narrow sections between land and water, smoothing land-water boundaries and outlines, and preserving the morphology of larger tidal flats as much as possible.

[0081] The closing operation, which operates in reverse to the opening operation, involves dilation followed by erosion. It is effective at removing isolated noise points in water areas, filling in voids or small defects within tidal flats, making the target area more coherent and complete, and effectively smoothing boundaries. Through these operations, noise generated after segmentation can be effectively eliminated, resulting in a more coherent and clear image.

[0082] Finally, edge vector extraction is performed. For the denoised binarized image, the Laplacian operator is used to obtain continuous waterline information. The binarized image is then processed by Laplacian filtering in the software ENVI (The Environment for Visualizing Images) to detect edge information in the image, obtaining instantaneous waterline data for a scene, and finally completing the accurate extraction of the waterline.

[0083] In water body information extraction, the enhanced water body index simultaneously utilizes green band, near-infrared band, and short-wave infrared band for normalization processing to extract water body information. In the filtering process, the Otsu method analyzes the statistical distribution characteristics of the pixel grayscale image, using variance as a measure of separability to automatically determine the optimal threshold. When the inter-class variance between the target and background regions reaches its maximum value, the binarized result obtained at this point is considered to have the best separation degree.

[0084] S4. Based on the waterline elevation control points, a digital elevation model (DEM) is constructed and validated using the tidal flat topographic inversion method. Remote sensing images acquired by Landsat series satellites at low tide times with similar tide levels (the average error of tide level values ​​at multiple feature points is less than 0.15m, i.e., the average error of TPXO in the central coastal waters of Jiangsu) are used to identify exposed tidal flats in the study area. Contour lines are generated by combining the multi-temporal waterline data obtained in step S3. The contour lines are then imported into the ArcGIS platform to construct the tidal flat digital elevation model, thereby analyzing the tidal flat erosion and deposition trends and revealing the dynamic changes in the coastline.

[0085] The method for establishing the tidal flat DEM is as follows: The instantaneous tidal level results at the time of remote sensing imagery are simulated using TPXO and assigned to the waterline of the multi-temporal sequence obtained earlier to obtain contour data. This contour data is then imported into ArcGIS to construct an irregular triangular grid (TIN). ArcGIS's automatic TIN generation function is then used to convert the TIN data into raster data, thus constructing the tidal flat DEM. When identifying exposed tidal flats, a true-color band combination of Landsat satellite imagery (red, green, and blue bands) is used to highlight the land cover features of the study area. True-color imagery displays the tidal flat area through natural colors, making it easier to identify different types of land cover (such as tidal flats and water bodies) and facilitating the monitoring of dynamic changes in the tidal flat area.

[0086] In this embodiment, the Yancheng sea area in central and northern Jiangsu Province is taken as an example. This area is a typical silty muddy coast. Affected by complex hydrodynamic conditions and changes in the sediment environment, the tidal flat landform has undergone significant evolution and is highly representative.

[0087] First, satellite remote sensing images and tidal data from the Landsat series at different imaging times were acquired. Landsat 9 images, selected at different imaging times, were used with a spatial resolution of 30 meters and encompassing seven bands from Band-1 to Band-7. Tidal data were derived from the TPXO large model, covering eight major tidal constituents (M2, S2, N2, K2, K1, O1, P1, Q1), two long-period tidal constituents (Mf and Mm), and three nonlinear tidal constituents (M4, MS4, MN4).

[0088] To comprehensively reflect different tidal conditions, the period from 00:00 on October 20, 2011 to 00:00 on November 4, 2011 was selected. Figure 2 (b) shows the Sheyang River estuary, from 0:00 on July 12, 2014 to 0:00 on July 22, 2014. Figure 2 (c) As shown in the figure, Dafeng Port) and from 0:00 on April 1, 2023 to 20:00 on April 22, 2023 ( Figure 2 Figure (a) shows the tidal data at the Guanhe Estuary during three time periods. The tidal data output by the TPXO model were compared and verified with the measured data. The verification results are as follows: Figure 2 As shown.

[0089] Next, coastline information for the nearshore waters of Yancheng from 1981 to 2023 was extracted, and the results were corrected by combining the coastline surveyed in 2019. The coastal section of the study area was divided into five parts: Guanhekou-Abandoned Yellow River Estuary, Abandoned Yellow River Estuary-Sheyang River Estuary, Sheyang River Estuary-Doulong Port, Doulong Port-Chuandong Port, and Chuandong Port-Laoba Port.

[0090] Then, the net shoreline movement distance (NSM) and the rate of change at the endpoint (EPR) were calculated, and the results are shown in Table 1.

[0091] Table 1. Statistics on the rate of shoreline change in different sections of the study area

[0092]

[0093] Based on the above calculations, from 1981 to 2023, the overall shoreline of central and northern Jiangsu Province showed a northward retreat and southward advance trend, with the Sheyang River estuary as the boundary. Furthermore, the reclamation intensity in the south was significantly higher than in the north. The boundary between erosion and deposition showed a southward trend year by year, gradually developing from the two sides of the abandoned Yellow River estuary towards Doulong Port: From 1981 to 2005, the overall shoreline showed a northward erosion and southward deposition trend, with the abandoned Yellow River estuary as the boundary; from 2005 to 2011, erosion intensified on both sides of the abandoned Yellow River estuary, with the maximum erosion distance exceeding 500m and an erosion rate of -89m / a. From 2011 to 2017, the shoreline in the section from the abandoned Yellow River estuary to the Sheyang River estuary continued to retreat inland, with a maximum retreat distance of nearly 400m and an erosion rate of -66.06m / a. From 2017 to 2023, the erosion status of the abandoned Yellow River Estuary-Sheyang River Estuary remained unchanged. The erosion range on the southern side of the Sheyang River Estuary further expanded to Doulong Port, with an average bank retreat of approximately 134 meters and an average annual erosion rate of -22.35 m / a, indicating severe erosion. The average EPR distribution in the study area is shown below. Figure 3 As shown.

[0094] Then, based on the remote sensing images collected by Landsat 9 OLI, the water quality index EWI was calculated. The remote sensing images were then converted to grayscale using the EWI, and the grayscale images were then processed by convolution filtering using the filtering tools in ENVI software.

[0095] Wherein, the high-pass filter matrix is The low-pass filter matrix is .

[0096] Then, based on the filtered grayscale image, the Otsu method is used to binarize the image until the inter-class variance between the target and background regions reaches its maximum value. Finally, opening and closing operations are used to eliminate noise generated during the segmentation process in the binarized image, resulting in a more coherent and clearer image.

[0097] Then, the binarized image is processed by Laplacian filtering in ENVI to detect edge information in the image, obtain instantaneous waterline data of a certain scene, and perform manual correction processing in a small area to finally complete the accurate extraction of the waterline.

[0098] Finally, the digital elevation model is constructed, which includes the following steps:

[0099] (1) Contour acquisition. Based on the instantaneous tide level results of the TPXO simulated remote sensing image at the imaging time, the values ​​were assigned to the waterline of the multi-temporal sequence obtained above to obtain contour data.

[0100] (2) Constructing the tidal flat DEM. Import the contour lines into ArcGIS to construct an irregular triangular grid (TIN). Using ArcGIS's automatic TIN generation function, convert the TIN data into raster data to construct the tidal flat DEM. The specific parameters are: output cell size 30m, maximum number of iterations 45, and roughness penalty coefficient 0.5.

[0101] (3) DEM accuracy verification. Topographic survey data collected in December 2009 for the Tiaozini area were compared with the inversion results from 2011. The errors were generally maintained within 0.3m, while the maximum errors at deeper tidal channels and high shoals were 0.9m and 0.7m, respectively. This error may be due to the limited coverage of the two areas by the acquired sequence of waterline data. Without considering the errors at tidal channels and high shoals, the root mean square error (RMSE) within the verification range was 0.279m.

[0102] (4) Dynamic changes of tidal flats. A DEM of tidal flats for 30 years (1993, 1999, 2005, 2011, 2017, and 2023) was constructed in the study area. The DEMs of adjacent years were subtracted from each other to obtain the changes in the planar distribution of tidal flat erosion and deposition in different time periods. The topographic changes and erosion and deposition trends in the study area were analyzed using 2017-2023 as an example.

[0103] The results show that from 2017 to 2023, the tidal flats from the Guanhe River estuary to the abandoned Yellow River estuary in the north remained in an erosion state, with only some sections of the estuary showing siltation. In the central section from the abandoned Yellow River estuary to the Yunliang River estuary, the changes were not significant over the six years due to the smaller amount of tidal flats. Due to the construction of the dike, there was significant siltation at the foot of the dike on both sides of the Sheyang River estuary. The lower part of the intertidal zone from the estuary to Zhugang Port showed a clear erosion trend, with the most severe erosion occurring on both sides of Xinyang Port, the south bank of Doulong Port, and the north bank of the Simaoyou River estuary, with the maximum erosion reaching 2.5m. The intertidal zone from Zhugang Port to Chuandong Port showed alternating erosion and siltation, with the tidal flats south of Chuandong Port generally showing siltation, with the maximum siltation exceeding 2m.

[0104] This embodiment aims to improve the accuracy and efficiency of monitoring coastal evolution. First, a study area is selected, and multi-temporal high-resolution remote sensing images and corresponding tidal level data at the imaging times are acquired and validated. Second, coastline information is extracted and corrected using measured and revised coastlines. Then, a digital shoreline analysis system is used to quantitatively assess the characteristics of shoreline changes. Subsequently, based on each remote sensing image, water body information extraction, filtering and denoising, image binarization, morphological optimization, and edge vector extraction are performed sequentially to obtain time-series remote sensing waterline data. Finally, relying on waterline elevation control points, a digital elevation model is constructed using a tidal flat topography inversion method. After validation, the dynamics of tidal flat erosion and deposition are analyzed, and the system reveals the laws of coastal evolution. This method achieves a coordinated assessment of shoreline and tidal flat changes, possessing strong practicality and scalability. It can provide a scientific basis for coastal resource management, ecological protection, and sustainable development, and is of significant value in promoting the coordinated development of ecological security and the economy in coastal areas.

[0105] In addition to the embodiments described above, the present invention may have other implementations. All technical solutions formed by equivalent substitution or equivalent transformation fall within the protection scope claimed by the present invention.

Claims

1. A method for evaluating the synergistic effect of shoreline change and tidal flat erosion and deposition, characterized in that: Includes the following steps: S1. Select the study area, acquire Landsat series satellite remote sensing images and TPXO tide level data of the corresponding imaging time, and verify the tide level simulated by TPXO by combining the actual measured station tide level data. S2. Extract coastline information and correct it in conjunction with the reconstructed coastline. Statistically analyze the net coastline movement distance and endpoint change rate of the coastline within the study area. Use the digital coastline analysis system based on the ArcGIS platform to perform quantitative analysis of the coastline in the study area. The net coastline movement distance is expressed as the distance between coastlines at different times, which is used to reflect the overall horizontal movement of the coastline. The endpoint change rate is expressed as the rate of change of the coastline distance over time, which is used to measure the amount of change in the coastline per unit time. S3. Based on the collected remote sensing images, remote sensing water boundary data at different times are obtained sequentially through water body information extraction, filtering, image binarization, morphological noise reduction, and edge extraction vector methods. S4. Based on the waterline elevation control points, a digital elevation model is constructed and verified using the tidal flat topographic inversion method. Using remote sensing images acquired by Landsat series satellites at low tide times with similar tide levels over the years, exposed tidal flats in the study area are identified. Similar tide levels mean that the average error of the tide level values ​​at multiple feature points is less than 0.15m. Contour lines are generated by combining the waterline data obtained in step S3. The contour lines are imported into the ArcGIS platform to construct a digital elevation model of the tidal flats. Finally, the tidal flat erosion and deposition trends and the dynamic changes of the coastline are analyzed.

2. The assessment method for shoreline change-tidal flat erosion and deposition synergy according to claim 1, characterized in that: In step S1, the Landsat series satellite remote sensing images include Landsat 5 TM, Landsat 7 ETM+, Landsat 8 OLI, and Landsat 9 OLI satellites, and the remote sensing images are from the USGS official data website; the tide data uses the tide data simulated by the TPXO 9.0 model as the data source, and the simulation results of the TPXO model are verified by selecting measured data from multiple tide stations.

3. The assessment method for shoreline change-tidal flat erosion and deposition synergy according to claim 1, characterized in that: In step S2, the calculation process of the digital shoreline analysis system is as follows: First, a baseline is drawn, which is parallel to the shoreline; second, the shoreline is set, and all shoreline data involved in the calculation are in the same feature class layer; then, a cross section is generated, and the digital shoreline analysis system generates several cross sections perpendicular to the baseline on the seaward side; finally, the net shoreline movement distance and the endpoint change rate (EPR) are calculated by the distance between the intersection points of each shoreline and the cross section line.

4. The assessment method for shoreline change-tidal flat erosion and deposition synergy according to claim 3, characterized in that: The net shoreline movement distance NSM is calculated using the following formula: ； in, This refers to the most recent coastline. This indicates a later phase of the coastline; The endpoint rate of change (EPR) is calculated using the following formula: ； in, This represents the rate of change of the endpoint of the shoreline along a certain cross-section between adjacent years. This represents the distance from the j-th phase coastline to the baseline along this cross-section line. This represents the distance from the i-th phase coastline to the baseline along the same cross-section line. This represents the difference between the j-th and i-th years of the coastline along this cross-section.

5. The evaluation method for shoreline change-tidal flat erosion and deposition synergy according to claim 1, characterized in that: In step S3, the enhanced water body index method is used to extract water body information and enhance the water body information. The formula is as follows: ； Green represents the green light band, which is the second band in TM and ETM+ satellite imagery and the third band in OLI imagery; NIR represents the near-infrared band, which is the fourth band in TM and ETM+ imagery and the fifth band in OLI; SWIR1 represents the shortwave infrared band 1, which is the fifth band in TM and ETM+ imagery and the sixth band in OLI.

6. The assessment method for shoreline change-tidal flat erosion and deposition synergy according to claim 1, characterized in that: In step S3, convolution filtering is used for filtering. Assuming the original image is f(i,j), an M×N window is opened in the upper left corner of the original image. Then, an M×N order filtering function K(m,n) is selected. This function is then used to weight the information within the window. Finally, the processing result is superimposed onto the original image to form the filtered image Q(i,j), represented as: ； Where Q(i,j) represents the result of filtering the image at position (i,j), f(i,j) represents the initial preimage at position (i,j), K(m,n) represents the filtering kernel matrix function, and m and n represent the parameters of the matrix function in the x and y directions, respectively.

7. The evaluation method for shoreline change-tidal flat erosion and deposition synergy according to claim 1, characterized in that: In step S3, the optimal threshold is obtained using Otsu's method, and the image is binarized. Let the number of image pixels be N, the grayscale range be [0, K], the image grayscale level be L, and the number of pixels with grayscale level i be... Then the probability of grayscale value i appearing is: ； in, This means that a pixel is randomly selected from the entire image, and the gray value of this pixel is the probability of i. express Summation; The pixels in the image are divided into class A and class B according to a grayscale threshold t. Class A consists of pixels with grayscale values ​​between [0, t], and class B consists of pixels with grayscale values ​​between [t+1, K]. The probabilities of class A and class B are as follows: ； The mean gray values ​​for class A and class B are as follows: ； The average gray level of the entire image is: ； Define the between-class variance as: ； in, and Let A and B represent the probabilities of A and B, respectively. and Let A and B represent the mean gray values ​​respectively; let t be in the range [0,K], and increment by 1 step, when The t value corresponding to the maximum value is the optimal threshold.

8. The evaluation method for shoreline change-tidal flat erosion and deposition synergy according to claim 1, characterized in that: In step S3, noise reduction is performed using opening and closing operations in morphological filtering. The opening operation based on morphological filtering first applies an erosion operation to strip away the bright areas, and then uses a dilation operation to restore the main outline. The closing operation first dilates and then erodes, which is used to remove isolated noise points in the water area.

9. The assessment method for the synergistic effect of shoreline change and tidal flat erosion and deposition according to claim 1, characterized in that: In step S3, for the denoised binarized image, the Laplacian operator is used to obtain continuous water edge information. The binarized image is then processed by Laplacian filtering in the software ENVI to detect edge information in the image, and instantaneous water edge data of a certain scene is obtained, thus completing the extraction of the water edge.

10. The evaluation method for shoreline change-tidal flat erosion and deposition synergy according to claim 1, characterized in that: In step S4, the instantaneous tide level result at the time of remote sensing image imaging is simulated using TPXO and assigned to the waterline obtained in step S3 to obtain contour data. The contour data is then imported into ArcGIS to construct an irregular triangular grid. The irregular triangular grid data is then converted into raster data using ArcGIS's function of automatically generating irregular triangular grids from features, thus constructing a digital elevation model of the tidal flat. When the exposed tidal flat is identified, a true color band combination of Landsat series satellite imagery, namely red, green, and blue bands, is used to highlight the ground features of the study area.

Images