A method for lake and reservoir stability zoning based on environmental factors

Abstract

The present invention discloses a lake and reservoir stability zoning method based on environmental factors, which belongs to the field of remote sensing and environment. The method comprises: obtaining regional remote sensing images containing complete target lakes and reservoirs in multiple years, and obtaining remote sensing images containing regional surface reflectivity in each year and water surface range vector data of target lakes and reservoirs in the region in each year through processing; obtaining classified raster images of multi-year shoreline complexity index of target lakes and reservoirs and classified raster images of multi-year average water surface temperature difference of target lakes and reservoirs respectively through natural break method; superimposing the two classified raster images, and obtaining the stability zoning result of target lakes and reservoirs through natural break method. The present invention extracts the distribution of multi-year shoreline complexity and water surface temperature difference of target lakes and reservoirs through satellite remote sensing images, effectively reflecting the changes in the hydrothermal environment of the lakes and reservoirs themselves, thereby achieving the effect of zoning monitoring of lake and reservoir stability.

Description

Technical Field

[0001] The invention belongs to the field of remote sensing and environment, and in particular relates to a lake stability zoning method based on environmental factors. Background Art

[0002] Water body zoning is an objective understanding of the goals of the ecosystem and a comprehensive examination of the relationship between natural laws and human activities. Through zoning, we can fully understand the characteristics of different regions and ensure regional environmental protection and sustainable resource development.

[0003] The changes in environmental factors in different areas of lakes and reservoirs determine their overall performance. In arid cities, the stability of the water environment comes from sufficient water, and significant fluctuations in water volume will seriously undermine the entire stable aquatic system. Lakes and reservoirs store surface runoff in the rainy season and can replenish rivers when rainfall is insufficient. The stability of lake and reservoir environments is greatly affected by changes in water volume, which seriously affects the surrounding ecosystems and environmental safety. At the same time, artificially constructed lake and reservoir dams cause varying degrees of interference to the water environment, resulting in different degrees of ecological and environmental impacts in different environmental areas. Lake and reservoir zoning belongs to a micro-scale area, and the zoning indicators considered include topography, aquatic biological population structure, etc. Lakes and reservoirs are different from rivers. Generally, the flow rate of lakes and reservoirs is very low, the hydraulic exchange time is long, and human interference factors are obvious.

[0004] At present, there are many studies on the changes in water volume and water quality of reservoirs or lakes over a long time series based on remote sensing images, or on the zoning of lake and reservoir basins, but there is a lack of methods for zoning lakes and reservoirs themselves. The defect of the existing methods is that they regard lakes and reservoirs as a whole, only focusing on the unified characteristics of each region of the lake and reservoir, and ignoring the differences in the changes of each part of the lake and reservoir itself. Without zoning lakes and reservoirs, it is not conducive to the supervision of lakes and reservoirs. Local changes in the water environment may be difficult to detect in the early stage, and can only be discovered when the entire lake and reservoir area is affected. This is not conducive to timely curbing the deterioration of the lake and reservoir environment; and when the lake and reservoir environment is suddenly changed by changes in certain factors, it may take a long time to find the inducing cause, and miss the best time for governance; in addition, without zoning lakes and reservoirs, it is impossible to scientifically plan the corresponding use according to the characteristics of each area of ​​the lake and reservoir, such as areas suitable for freshwater aquaculture, areas suitable for wetland protection, and areas suitable for nearby agricultural activities, etc., and it is impossible to effectively grasp the optimal operating state of the lake and reservoir. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the existing technology and propose a lake and reservoir stability zoning method based on environmental factors. By zoning the water environment of the lake and reservoir, the present invention can better monitor the changes in the reservoir area, try to avoid the deterioration of the overall lake and reservoir environment caused by local environmental changes, and improve the effect of lake and reservoir monitoring.

[0006] The embodiment of the present invention proposes a lake stability zoning method based on environmental factors, including:

[0007] Obtain regional remote sensing images that include the entire target lake reservoir in multiple years, calculate the surface reflectance of the regional remote sensing images in each year, and obtain remote sensing images that include the regional surface reflectance in each year;

[0008] Extract the water surface of the target lake or reservoir in the region of each year based on the remote sensing image containing the regional surface reflectivity in each year, and generate water surface range vector data of the target lake or reservoir in the region of each year;

[0009] Based on the water surface range vector data of the target lakes and reservoirs in the region in each year, a classified raster image of the multi-year shoreline complexity index of the target lakes and reservoirs is obtained through the natural discontinuity method;

[0010] Based on the remote sensing images containing the regional surface reflectivity in each year, a classified raster image of the multi-year average water surface temperature difference of the target lake reservoir is obtained through the natural discontinuity method;

[0011] The classified raster image of the multi-year shoreline complexity index of the target lake and the classified raster image of the multi-year average water surface temperature difference of the target lake are superimposed, and the stability zoning result of the target lake is obtained by the natural break method.

[0012] In a specific embodiment of the present invention, the multiple years are equally spaced.

[0013] In a specific embodiment of the present invention, the regional remote sensing images are collected in the same month of each year; if remote sensing images of the same month as other years are not collected in any year, remote sensing images of the months adjacent to the month are used in that year.

[0014] In a specific embodiment of the present invention, the water surface of the target lake in each year region is extracted based on the remote sensing image containing the regional surface reflectivity in each year, and the water surface range vector data of the target lake in each year region is generated, including:

[0015] 1) For any year's remote sensing image containing regional surface reflectance, the improved normalized difference water extraction MNDWI index and the automatic water surface extraction AWEIsh index of each grid are calculated respectively, and the grid images corresponding to the calculation results of the MNDWI index and the AWEIsh index of that year are obtained;

[0016] 2) According to the raster image obtained in step 1), the distribution of the MNDWI index value and the AWEIsh value in the corresponding raster image is counted to obtain the bimodal distribution map corresponding to each index; the median between the two peaks in each bimodal distribution map is used as the threshold of the corresponding index of the year;

[0017] 3) Based on the grid image of the AWEIsh index calculation result corresponding to that year, the AWEIsh index value corresponding to each grid in the image is compared with the AWEIsh index threshold of that year:

[0018] If the AWEIsh index value corresponding to any grid is less than the AWEIsh index threshold of that year, the grid is determined to be the initial water surface range; otherwise, the grid is determined to be the non-water surface range;

[0019] After all grids of the grid image corresponding to the AWEIsh index calculation result of that year are determined, all grids determined to be the initial water surface range are retained;

[0020] 4) Based on the grid image of the MNDWI index calculation result corresponding to the year, obtain the MNDWI index values ​​corresponding to all grids retained in step 3) and determine:

[0021] If the MNDWI index value corresponding to any grid retained in step 3) is less than the MNDWI index threshold of the year, the grid is determined to be a water surface range; otherwise, the grid is determined to be a non-water surface range;

[0022] After the determination is completed, all grids determined as water surface ranges in that year constitute the water surface range vector data of the target lakes and reservoirs in the region in that year;

[0023] 5) Repeat steps 1) to 4) to obtain the water surface area vector data of the target lake in the region for each year.

[0024] In a specific embodiment of the present invention, the classification grid image of the multi-year shoreline complexity index of the target lake reservoir is obtained by the natural discontinuity method based on the water surface range vector data of the target lake reservoir in each year, including:

[0025] 1) Using the water surface range vector data of the target lakes and reservoirs in the region of each year, the water surface area and perimeter of the target lakes and reservoirs of each year are counted; according to the water surface area and perimeter, based on the shoreline complexity index formula, the shoreline complexity index of the target lakes and reservoirs of each year is calculated;

[0026] 2) Create corresponding null value grid images based on the water surface range vector data of the target lakes and reservoirs in the region in each year, assign the shoreline complexity index value of the corresponding year to each grid in each null value grid image, and obtain a grid image reflecting the distribution of the shoreline complexity index value of the target lakes and reservoirs in each year;

[0027] 3) After sorting the grid images reflecting the distribution of the shoreline complexity index values ​​of the target lakes and reservoirs in each year by area from large to small, the shoreline complexity index value of the target lakes and reservoirs corresponding to each grid in the first raster image in the sequence is used as the current value, and then starting from the first raster image, the shoreline complexity index value of the target lakes and reservoirs in the subsequent raster images is used to replace the current value at the corresponding grid. After traversing all the raster images in the sequence, a raster image reflecting the shoreline complexity distribution of the target lakes and reservoirs for many years is obtained;

[0028] If there is a grid in the next grid image that exceeds the coverage of the current maximum area grid image, the grid of the next grid image is retained and the target lake shoreline complexity index of the grid is used as the current target lake shoreline complexity index value of the retained grid.

[0029] 4) Using the grid image reflecting the multi-year shoreline complexity distribution of the target lake, the shoreline complexity index value of each grid is classified according to the set shoreline complexity index classification number through the natural discontinuity method. After the classification is completed, the category value of the shoreline complexity index classification corresponding to each grid is used to replace the shoreline complexity index value of the grid to obtain the classified raster image of the multi-year shoreline complexity of the target lake.

[0030] In a specific embodiment of the present invention, the classification grid image of the multi-year average water surface temperature difference of the target lake reservoir is obtained by the natural discontinuity method based on the remote sensing image containing the regional surface reflectivity in each year, including:

[0031] 1) using the remote sensing images containing the regional surface reflectivity in each year, performing surface temperature inversion by atmospheric radiation transfer method to obtain a raster image of the regional surface temperature value in each year;

[0032] 2) using the water surface range vector data of the target lake in the region for each year to correspondingly clip the raster image of the regional surface temperature value for each year to obtain the raster image of the water surface temperature value of the target lake in each year;

[0033] 3) subtracting the average water surface temperature of all grids in the year from the water surface temperature value of each grid in the grid image of the water surface temperature value of the target lake reservoir in each year to obtain the water surface temperature difference value of each grid, and establishing a grid image of the distribution of the water surface temperature difference value in each year;

[0034] 4) superimposing the grid images of the water surface temperature difference distribution in each year, calculating the average value of the water surface temperature difference of the grid in each year for the grid at the same position as the final water surface temperature difference of the grid, and obtaining a grid image reflecting the distribution of the average water surface temperature difference of the target lake reservoir over many years;

[0035] 5) Using the grid image reflecting the distribution of the multi-year average water surface temperature difference of the target lake reservoir, the water surface temperature difference value of each grid is classified according to the set number of water surface temperature difference value classifications through the natural break method. After the classification is completed, the category value of the water surface temperature difference value classification corresponding to each grid is replaced by the water surface temperature difference value of the grid to obtain the classified grid image of the multi-year average water surface temperature difference of the target lake reservoir.

[0036] In a specific embodiment of the present invention, the classified raster image of the multi-year shoreline complexity index of the target lake and the classified raster image of the multi-year average water surface temperature difference of the target lake are superimposed, and the stability zoning result of the target lake is obtained by the natural discontinuity method, including:

[0037] The classified raster image of the multi-year shoreline complexity of the target lake and the classified raster image of the multi-year average surface temperature difference of the target lake are superimposed, and the category value of the shoreline complexity index classification corresponding to each grid and the category value of the water surface temperature difference classification corresponding to each grid are summed to obtain the total category value of each grid; the total category value of each grid is classified according to the set number of classifications of the target lake and reservoir stability through the natural discontinuity method, and after the classification is completed, the category value of each grid corresponding to the stability degree classification replaces the total category value of the grid, and the classified raster image of the multi-year stability of the target lake and reservoir is obtained, which is the stability zoning result of the target lake and reservoir.

[0038] In a specific embodiment of the present invention, the classification number of the target lake stability degree is the larger value of the classification number of the shoreline complexity index and the classification number of the water surface temperature difference.

[0039] The characteristics and beneficial effects of the present invention are:

[0040] (1) The present invention utilizes continuous remote sensing images, combines remote sensing technology with geographic information technology, and innovatively combines shoreline complexity with water surface temperature, thus revolutionizing existing lake and reservoir monitoring methods.

[0041] (2) The present invention uses continuous remote sensing images. Compared with the previous large-scale research methods on lakes and reservoirs, this method is refined to the lake and reservoir area itself. The historical remote sensing images store a large amount of information, which is convenient for exploring the historical changes of lakes and reservoirs. By comparing the changes over many years, the change patterns of lakes and reservoirs can be discovered, thereby providing strong data support for the stability zoning of lakes and reservoirs. Make full use of remote sensing images to reduce labor costs. At the same time, overcome the limitations of areas that are difficult to measure by manpower, thereby achieving comprehensive analysis of lake and reservoir areas and reasonable planning and detection.

[0042] (3) The present invention transforms the complexity of the shoreline from a simple line analysis to a surface analysis combined with the water surface range, thus breaking through the spatial limitations of the original shoreline research.

[0043] (4) The present invention adopts objective satellite data and natural break classification, and the obtained results are not subject to the subjective influence of scientific and technological personnel and are objective and accurate.

[0044] (5) The present invention uses remote sensing images, which will effectively reduce the cost and human resources of lake and reservoir detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is an overall flow chart of a lake stability zoning method based on environmental factors in an embodiment of the present invention.

[0046] Figure 2 The figure is a schematic diagram of the principle of a lake stability zoning method based on environmental factors in a specific embodiment of the present invention.

[0047] Figure 3 It is a schematic diagram of the effect of a lake stability zoning method based on environmental factors in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0048] The present invention proposes a lake stability zoning method based on environmental factors, which is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] This paper proposes a lake stability zoning method based on environmental factors. The overall process is as follows: Figure 1 As shown, the following steps are included:

[0050] 1) Obtain regional remote sensing images that contain the entire target lake reservoir in multiple years, calculate the surface reflectance of the regional remote sensing images in each year, and obtain remote sensing images that contain the regional surface reflectance in each year, where the intervals between adjacent years are the same.

[0051] In this embodiment, remote sensing images (such as Landsat series images) of the area to be studied in the same or similar months for many years are obtained from the official website of the Geospatial Data Cloud or the official website of USGS, wherein the area contains at least one complete lake reservoir, and a remote sensing image is obtained for each year, and the value range of the year is more than 3 years. The more years are used, the more detailed the results obtained. Since multispectral data is greatly affected by the weather at the time of shooting, if the remote sensing images available in the same month cannot be obtained, the remote sensing images within 2 months before and after the target month can be selected to make it as close to the target month as possible. The target month is mostly selected from the month with more stable temperature and sunny weather in the year. If the area of ​​the same year is not on the same image, it is necessary to splice multiple remote sensing images of the same period of the year into an image containing the complete area.

[0052] It should be noted that the lakes and reservoirs in the present invention include: the main part of lakes and reservoirs. There may be more than one lake or reservoir in the region. This method treats each lake or reservoir as an independent object. Any lake or reservoir in the region can be used as a target lake or reservoir.

[0053] Based on the regional remote sensing images acquired over the years, the regional surface reflectance in the regional remote sensing images of each year is calculated to obtain the remote sensing images containing the regional surface reflectance of each year.

[0054] In the embodiment of the present invention, the regional remote sensing image acquired in each year is preprocessed in ENVI software, and the remote sensing image containing the regional surface reflectance in each year is obtained by using the Radiometric Calibration and FLAASH Atmospheric Correction tools in the Toolbox.

[0055] 2) Using the remote sensing images containing regional surface reflectance for each year obtained in step 1), extract the water surface of the target lake or reservoir in the region for each year, and generate water surface range vector data (image) of the target lake or reservoir in the region for each year;

[0056] In this embodiment, the specific steps are as follows:

[0057] 2.1) In ENVI software, for any year of remote sensing images containing regional surface reflectance, the MNDWI (modified normalized difference water extraction index) index and the AWEIsh (automatic water surface extraction index) index of each grid are calculated respectively, and the raster images corresponding to the MNDWI index calculation results and the raster images corresponding to the AWEIsh index calculation results of that year are obtained.

[0058] 2.2) Using the grid image of the calculation results of the two indices obtained in step 2.1), the distribution of the grid values ​​calculated by the two indices in the corresponding raster image can be counted by the grid statistics tool to obtain the bimodal distribution map corresponding to each index. The median between the two peaks of each bimodal distribution map is taken as the threshold of the corresponding index of the year (the MNDWI threshold is 0-0.3, and the AWEIsh threshold is 0).

[0059] 2.3) Based on the grid image of the AWEIsh index calculation result corresponding to that year, compare the AWEIsh index value corresponding to each grid in the image with the AWEIsh index threshold of that year:

[0060] If the AWEIsh index value corresponding to any grid is less than the AWEIsh index threshold of that year, the grid is determined to be the initial water surface range; otherwise, the grid is determined to be the non-water surface range;

[0061] After all grids of the grid image corresponding to the AWEIsh index calculation result of that year are determined, all grids determined to be the initial water surface range are retained;

[0062] 2.4) Based on the grid image of the MNDWI index calculation result corresponding to the year, obtain the MNDWI index values ​​corresponding to all grids retained in step 2.3) and determine:

[0063] If the MNDWI index value corresponding to any grid retained in step 2.3) is less than the MNDWI index threshold of that year, the grid is determined to be a water surface range; otherwise, the grid is determined to be a non-water surface range;

[0064] After the determination is completed, all rasters determined to be water surface ranges in that year constitute the water surface range vector data of the target lakes and reservoirs in the region in that year.

[0065] 2.5) Repeat steps 2.1) to 2.4) to obtain the water surface area vector data (image) of the target lake in the region for each year.

[0066] 3) Using the water surface range vector data of the target lake in each year obtained in step 2), a classified raster image of the multi-year shoreline complexity index of the target lake is obtained through the natural discontinuity method; the specific steps are as follows:

[0067] 3.1) Using the water surface range vector data of the target lakes and reservoirs in each year obtained in step 2), calculate the water surface area and perimeter of the target lakes and reservoirs in each year. According to the water surface area and perimeter, based on the shoreline complexity index formula, calculate the shoreline complexity index of the target lakes and reservoirs in each year;

[0068] 3.2) Create a corresponding null value raster image based on the water surface range vector data of the target lake or reservoir in the region in each year (the size of the null value raster image is equal to the image size of the lake or reservoir water surface range vector data in the corresponding year) and assign a shoreline complexity index value of the corresponding year to each grid in each null value raster image, so as to obtain a raster image reflecting the distribution of the shoreline complexity index value of the target lake or reservoir in each year; wherein, after assigning values ​​to the null value raster images of the same year, the shoreline complexity index of each grid is the same.

[0069] In this embodiment, ArcGIS software is used to process the water surface range vector data of the target lakes and reservoirs in the region for each year, and the water surface vector data of the target lakes and reservoirs in the region for each year is input. A fishing net of corresponding resolution is established based on the maximum water surface area of ​​all input years, and grids are selected according to the water surface range of each year. The null-value raster images of the water surface size of the target lakes and reservoirs are output for each year, and then the water surface shoreline complexity index value of each year is assigned to the null-value raster image of that year, so as to obtain a raster image reflecting the distribution of the shoreline complexity index value of the target lakes and reservoirs in each year.

[0070] Figure 2 The figure is a schematic diagram of the principle of a lake stability zoning method based on environmental factors in a specific embodiment of the present invention. Figure 2 The grids in each layer represent the pixels of the raster image, and the grids in the shaded part constitute the water surface range of the target lake in the layer. Figure 2 (a) is a raster layer of three years. The shadow part (i.e., the water surface part) of each year can be extracted by combining the NDWI and AWEIsh indexes, and the perimeter and area of ​​the shadow part can be obtained. Based on the perimeter and area, the shoreline complexity index of layer 1 can be calculated as 1.3, the shoreline complexity index of layer 2 as 2.7, and the shoreline complexity index of layer 3 as 4.0. The shoreline complexity index value of each year is assigned to the raster representing the water surface range of the corresponding year, and a raster image reflecting the shoreline complexity distribution of each year is obtained. It should be noted that in this embodiment, Figure 2 (a) The raster images extracted from the layers of the three years showing the distribution of coastline complexity in each year will only retain Figure 2 The shaded part in (a). In order to illustrate the relationship between the water surface ranges in three years, this embodiment adds extra blank grids to the water surface range images extracted in the years with smaller water surface ranges.

[0071] 3.3) After sorting the grid images reflecting the distribution of the shoreline complexity index values ​​of the target lakes and reservoirs in each year obtained in step 3.2) by area from large to small, the shoreline complexity index value of the target lakes and reservoirs corresponding to each grid in the first raster image in the sequence is used as the current value, and then starting from the first raster image, the shoreline complexity index value of the target lakes and reservoirs in the subsequent raster images is used to replace the current value at the corresponding grid. After traversing all the raster images in the sequence, a raster image reflecting the shoreline complexity distribution of the target lakes and reservoirs for many years is obtained; among them, if there is a grid in the subsequent raster image that exceeds the coverage of the current maximum area raster image, the grid of the subsequent raster image is retained and the shoreline complexity index of the target lakes and reservoirs of the grid is used as the current shoreline complexity index of the retained grid.

[0072] In this embodiment, the Update tool in ArcGIS software is used to input the raster image reflecting the distribution of the shoreline complexity index value of the target lake reservoir in each year, and after sorting the target lake reservoir water surface from large to small, the target lake reservoir shoreline complexity index value corresponding to each grid in the first raster image in the sequence is used as the current value, and then starting from the first raster image, the target lake reservoir shoreline complexity index value of the subsequent raster image is used to replace the current target lake reservoir shoreline complexity index value at the corresponding grid in sequence, until all raster images are traversed, and a raster image reflecting the multi-year shoreline complexity distribution of the target lake reservoir is output, and the water surface range of the target lake reservoir in the raster image is the maximum water surface range of the target lake reservoir after superposition in all input years.

[0073] In a specific embodiment of the present invention, Figure 2 The three raster images in (a) are sorted by area from large to small after extracting the water surface range. Figure 2 (b) shows the arrangement result, Figure 2 The grid value representing the water surface range in (b) is the shoreline complexity index value of the corresponding year. Then, the shoreline complexity index value of the year with a smaller water surface range is used to update the shoreline complexity index value corresponding to the year with a larger water surface range, and a grid image reflecting the shoreline complexity distribution of the target lake for many years is obtained. Figure 2 (c) as shown.

[0074] 3.4) Using the grid image obtained in step 3.3), classify the shoreline complexity index value of each grid by the natural break method to obtain a classified grid image of the shoreline complexity of the target lake for many years;

[0075] In this embodiment, the raster image obtained in step 3.3) is processed by ArcGIS software, and a raster image reflecting the distribution of the complexity of the shoreline of the target lake for many years is input. The shoreline complexity index of all the grids in the raster image is reclassified by the natural discontinuity method, and the shoreline complexity is divided into multiple categories and assigned corresponding category values. The larger the category value, the higher the complexity. In a specific embodiment of the present invention, the number of shoreline complexity categories is set to 5 by the Reclassify tool in ArcGIS software (the number of categories can be set according to the classification requirements. The more categories, the more refined the classification; 3-5 categories are generally recommended), and the corresponding category values ​​of each category are set to 1, 2, 3, 4, and 5. After the classification is completed, each grid replaces the corresponding shoreline index value with the category value to which it belongs, wherein the larger the shoreline complexity index value of the grid, the larger the category value after classification, indicating that the shoreline complexity of the corresponding range is greater and the lake habitat environment is richer; after the assignment of each grid category value is completed, the classified raster image of the long-term shoreline complexity of the target lake is output.

[0076] It should be noted that one of the means of stability division in this embodiment is to classify all grids whose shoreline complexity index values ​​in the raster image of step 3.3) are close to the shoreline complexity index mean area and whose total number of grids under the qualified shoreline complexity index values ​​are similar into one category through the natural discontinuity method, that is, as a general stable area of ​​shoreline complexity index distribution, which is usually regarded as an intermediate category (in this embodiment, it is set as the third category, and the corresponding shoreline complexity category value is 3); then the shoreline complexity index value range that is smaller than the shoreline complexity index value range of the general stable area is segmented into two intervals, so that the difference in the number of grids occupied by different shoreline complexity index values ​​in each interval is minimized: this embodiment In this example, the one with the smaller shoreline complexity index in the two intervals is set as the first category, i.e., the stable area (the shoreline complexity category value is set to 1), and the larger one is set as the second category, i.e., the relatively stable area (the shoreline complexity category value is set to 2); the shoreline complexity index value range greater than the shoreline complexity index value range of the general stable area is segmented into two intervals, so that the difference in the number of grids occupied by different shoreline complexity index values ​​in each interval is also minimized: in this embodiment, the one with the smaller shoreline complexity index in the two intervals is set as the fourth category, i.e., the relatively unstable area (the shoreline complexity category value is set to 4), and the larger one is set as the fifth category, i.e., the unstable area (the shoreline complexity category value is set to 5).

[0077] In another specific embodiment of the present invention, the shoreline complexity index is classified by natural discontinuity method (here divided into three categories), and the distribution of shoreline complexity classification in multiple years is obtained as follows: Figure 2 (d) shows that the category values ​​of categories 1, 2, and 3 are set to 1, 2, and 3 respectively. Figure 2 Each grid value in (d) is the shoreline complexity category value corresponding to the grid.

[0078] 4) Using the remote sensing images containing regional surface reflectivity for each year obtained in step 1), the classified raster images of the multi-year average water surface temperature difference of the target lake reservoir are obtained through the atmospheric radiation transfer method and the natural discontinuity method; the specific steps are as follows:

[0079] 4.1) Using the remote sensing images containing the regional surface reflectance in each year in step 1), the surface temperature is inverted by the atmospheric radiation transfer method to obtain a raster image of the regional surface temperature value in each year;

[0080] 4.2) Using the water surface range vector data of the target lake in each year obtained in step 2), the raster image of the regional surface temperature value in each year obtained in step 4.1) is correspondingly clipped to obtain the raster image of the water surface temperature value of the target lake in each year;

[0081] In a specific embodiment of the present invention, based on Figure 2The raster image shown in (a) can be used to invert the water surface temperature of each year to obtain the raster image of the regional surface temperature value of each year. Then, the raster image of the regional surface temperature value of each year obtained in step 4.1) is correspondingly cropped using the water surface range vector data of the target lake reservoir in the region in each year to obtain the raster image of the water surface temperature value of the target lake reservoir in each year as shown in Figure 2 (e), where Figure 2 The shaded grid values ​​in (e) are sample values ​​of the water surface temperature for that year only.

[0082] 4.3) Subtracting the average water surface temperature of all grids in the year from the water surface temperature value of each grid in the grid image of the water surface temperature value of the target lake in each year obtained in step 4.2) to obtain the water surface temperature difference of each grid, and establishing a grid image of the distribution of water surface temperature difference values ​​in each year;

[0083] In this embodiment, ArcGIS software is used to process the raster image of the target lake water surface temperature value of each year, the raster image of the target lake water surface temperature value of each year is input, the water surface temperature value of each grid in the raster image of each year is subtracted from the average water surface temperature of all grids in that year, and the raster image of the water surface temperature difference distribution of each year is output.

[0084] 4.4) The grid images of the water surface temperature difference distribution of each year obtained in step 4.3) are superimposed, and the average value of the water surface temperature difference of each year in the grid at the same position is calculated as the final water surface temperature difference of the grid (wherein, if any grid only exists in the grid images of the water surface temperature difference distribution of some years, only the average value of the grid in these years is calculated; if any grid only exists in the grid images of the water surface temperature difference distribution of a certain year, then the average value of the grid does not need to be calculated), and a grid image reflecting the distribution of the average water surface temperature difference of the target lake for many years is obtained;

[0085] In a specific embodiment of the present invention, grid images of the water surface temperature difference distribution in three years are superimposed, wherein the grids at the same position are averaged over many years to obtain a grid image reflecting the distribution of the average water surface temperature difference of the target lake for many years. Figure 2 (f) This image covers the maximum water surface of all years. Among them, since some grids in the grid image of the water surface temperature difference distribution in each year have negative water surface temperature differences, the final grid image reflecting the multi-year average water surface temperature difference distribution of the target lake also has negative water surface temperature differences corresponding to some grids. Figure 2Take the upper left corner grid in (f) as an example. The water surface temperature value of this grid in the corresponding year is 1, and the average water surface temperature of that year is 5. The initial water surface temperature difference value of this grid is -4. Since the corresponding position of this grid is not a water surface in other years, after the grid images of the water surface temperature difference distribution in all years are superimposed, the final water surface temperature difference value of this grid is still -4. Figure 2 Taking the central grid in (f) as an example, the water surface temperature values ​​of the grid in the corresponding years are 5, 2, and 5, and the average water surface temperature values ​​of the corresponding years are 5, 2, and 5. The initial water surface temperature difference values ​​of the grid in the three years are all 0. Therefore, after the grid images of the water surface temperature difference distribution of all years are superimposed, the final water surface temperature difference value of the grid is still 0.

[0086] 4.5) Using the grid image obtained in step 4.4), classify the water surface temperature difference value of each grid by the natural break method to obtain a classified grid image of the multi-year average water surface temperature difference of the target lake;

[0087] In a specific embodiment of the present invention, the raster image obtained in step 4.4) is processed by ArcGIS software, and a raster image reflecting the distribution of the average temperature difference of the water surface of the target lake reservoir over the years is input. By using the Reclassify tool in ArcGIS software, the water surface temperature difference values ​​of all grids in the raster image are reclassified by the natural discontinuity method, and the water surface temperature difference values ​​are divided into multiple categories and assigned corresponding category values. The larger the category value, the larger the water surface temperature difference. In a specific embodiment of the present invention, the number of water surface temperature difference classifications is set to 3, and the category values ​​of the three categories are set to 1, 3, and 5, respectively. After the classification is completed, each grid replaces the corresponding water surface temperature difference value with the category value to which it belongs, wherein the larger the water surface temperature difference value of the grid, the larger the category value after classification, indicating that the average temperature difference of the water surface in the corresponding range is larger and more unstable. After the assignment of each grid category value is completed, the classified raster image of the average temperature difference of the water surface of the target lake reservoir over the years is output.

[0088] It should be noted that the second means of stability division in this embodiment is to classify all grids whose water surface temperature difference values ​​in the grid image of step 4.4) are close to the mean water surface temperature difference and whose total number of grids under the qualified water surface temperature difference values ​​are similar into one category through the natural discontinuity method, that is, as a generally stable area with multi-year temperature difference, and this category is usually regarded as an intermediate category (in this embodiment, it is set as the second category, and the corresponding category value is 3); then all grids whose water surface temperature difference values ​​are less than the temperature difference value range of the general stable area are regarded as the first category, that is, the stable area (the category value is set to 1); on the contrary, all grids whose water surface temperature difference values ​​are greater than the temperature difference value range of the general stable area are regarded as the third category, that is, the unstable area (the category value is set to 5), and the annual temperature difference in this area is higher than the average value, which is an area with large changes in water surface temperature difference.

[0089] In another specific embodiment of the present invention, the water surface temperature difference is divided into three categories based on the natural discontinuity method, and the distribution of the average water surface temperature difference is obtained as follows: Figure 2 As shown in (g), because the mean temperature difference of each grid in this embodiment is 0, and each value occupies only one grid, the temperature difference values ​​close to or equal to the mean -1, 0, 1 are classified into one category as Category 2, and the water surface temperature difference category value is 2; the three values ​​with temperature difference values ​​less than -1 are classified as Category 1, and the water surface temperature difference category value is 1; the remaining three values ​​with temperature difference values ​​greater than 1 are classified as Category 3, and the water surface temperature difference category value is 3. Figure 2 Each grid value in (g) is the water surface temperature difference category value corresponding to the grid.

[0090] 5) Sum the corresponding grid category values ​​of the classified raster image of the multi-year shoreline complexity of the target lake obtained in step 3) and the classified raster image of the multi-year average water surface temperature difference of the target lake obtained in step 4) to obtain the total category value of each grid, and reclassify all grids based on the total category value using the natural break method to obtain the stability zoning result of the target lake based on the shoreline change and temperature difference change.

[0091] In this embodiment, ArcGIS software is used for processing, and a classified raster image of the multi-year shoreline complexity of the target lake and a classified raster image of the multi-year average water surface temperature difference of the target lake are input. The two raster images are superimposed and analyzed, that is, the category values ​​of the corresponding grids are summed to obtain the total category value of each grid, and then the total category values ​​of all grids are classified by the natural break method, and a classified raster image of the multi-year stability of the target lake is output.

[0092] In a specific embodiment of the present invention, by using the Reclassify tool in ArcGIS software, the number of stability degree classifications is set to 5 (it should be noted that when the total category value is used to classify the stability degree, the number of stability degree classifications needs to be consistent with the larger value of the number of classifications set in step 3) and step 4), and the stability degree categories are replaced by integers 1-5, and the corresponding category values ​​of each category are set to 1, 2, 3, 4, and 5. After the classification is completed, each raster replaces the corresponding total category value with the category value to which it belongs, and the classified raster image of the multi-year stability of the target lake reservoir is output as the stability zoning result of the target lake reservoir.

[0093] It should be noted that the stability degree division corresponding to step 5) is derived from the classification results of the superimposed water surface temperature difference value and the classification results of the shoreline complexity. After the classification values ​​are superimposed, the overall stability of the area is determined. Similarly, the natural discontinuity method is used to classify all grids whose total category values ​​are close to the mean of the total category values ​​and whose total number of grids under the total category values ​​that meet the conditions is similar as one category, that is, as the final general stable area, which is usually regarded as the intermediate category (set as the third category in this embodiment); then the shoreline complexity index value range that is smaller than the shoreline complexity index value range of the general stable area is segmented into two intervals, so that the different shoreline complexity index values ​​in each interval occupy the grid area. The difference in the number of grids is minimal: in this embodiment, the category with the smaller shoreline complexity index in the two intervals is set as the first category, i.e., the stable area (the category value is set to 1), and the larger category is set as the second category, i.e., the relatively stable area (the category value is set to 2); the shoreline complexity index value range that is greater than the shoreline complexity index value range of the general stable area is segmented into two intervals, and the difference in the number of grids occupied by different shoreline complexity index values ​​in each interval is also minimized: in this embodiment, the category with the smaller shoreline complexity index in the two intervals is set as the fourth category, i.e., the relatively unstable area (the category value is set to 4), and the larger category is set as the fifth category, i.e., the unstable area (the category value is set to 5). There are five categories, each of which represents the degree of fluctuation of the superimposed area, that is, under the condition of the same water inflow, it can characterize the stability of the habitats in different areas of the lake. Category 1 is the area with the smallest environmental changes, and category 5 is the area with the largest environmental changes.

[0094] In another specific embodiment of the present invention, Figure 2 (d) and Figure 2 (g) After the overlay analysis, the total category values ​​obtained are 2, 3, 4, 5, and 6, which occupy 1, 3, 2, 2, and 1 grids respectively. This embodiment divides the degree of stability into 3 categories by the natural break method. Since the mean value of the total category value is 3.9, which is close to 4, and the number of grids with a total category value of 4 is 2, and the number of grids occupied by the total category value 5 close to 4 is also 2, the total category values ​​4 and 5 are divided into one category and set as the second category, and its stability category value is 2; the value with a total category value less than 4 is set as the first category, and its stability category value is 1; the value with a total category value greater than 5 is set as the third category, and its stability category value is 3. Figure 2 Each grid value in (h) is the stability level category value corresponding to the grid.

[0095] The comprehensive classification of the multi-year average water surface temperature difference distribution changes and shoreline complexity distribution in each area of ​​the study area obtained in step 5) indicates that the larger the classification value, the greater the change in each factor. This can intuitively show the distribution characteristics of changes in different areas of the study area and intuitively show the stability of the study area. Since the shoreline changes are combined with the water surface range, the areas with large shoreline changes can be accurately judged. The superimposed temperature difference changes can better show the stability of each area, and can also determine the water surface range and area when the lake reservoir is operating optimally.

[0096] In a specific embodiment of the present disclosure, Guanting Reservoir is selected as the area to be studied. The main part of the reservoir area of ​​Guanting Reservoir spans Huailai County, Zhangjiakou City, Hebei Province and Yanqing District, Beijing City. The water surface coverage has not exceeded 100km in the past 21 years. 2 . Remote sensing data (Landsat5 and Landsat 8 series images) mainly concentrated in May in the past 21 years were obtained from the official website of Geospatial Data Cloud or the official website of USGS. The research area of ​​each year is on the same image. The effect of the lake stability zoning method based on environmental factors proposed in this embodiment is as follows Figure 3 shown.

[0097] Figure 3 (a) is a surface reflectance image of the Guanting Reservoir obtained in an embodiment of the present invention. A total of 21 images from 2000 to 2020 were obtained.

[0098] Figure 3 (b) is a raster image reflecting the distribution of the complexity of the shoreline of Guanting Reservoir over the years obtained in the embodiment of the present invention. The range of the raster image is the maximum water surface range of Guanting Reservoir in 21 years. In the figure, the colors from light to dark indicate that the reservoir shoreline complexity index (RSDI) is from large to small. The darkest color represents the smallest shoreline complexity index, which is 6.11; the lightest color represents the largest shoreline complexity index, which is 10.98.

[0099] Figure 3 (c) is a grid image reflecting the distribution of the average temperature difference of the water surface of Guanting Reservoir over many years obtained in the embodiment of the present invention. The temperature difference value of each grid represents the average of the temperature difference of the water surface over many years in the corresponding range. In the figure, the color from light to dark indicates that the temperature difference is from small to large. The darkest color represents the largest temperature difference value, which is 12.51°C; the lightest color represents the smallest temperature difference value, which is -3.8°C.

[0100] Figure 3(d) is the multi-year stability classification raster image of Guanting Reservoir obtained in the embodiment of the present invention. The stability of Guanting Reservoir is divided into 5 categories in the image, representing unstable areas to stable areas from deep to shallow. The raster image reflects the stability zoning result of Guanting Reservoir, and its range is the maximum water surface range of Guanting Reservoir in 21 years.

Claims

1. A lake stability zoning method based on environmental factors, characterized in that: include: Obtain regional remote sensing images that include the entire target lake reservoir in multiple years, calculate the surface reflectance of the regional remote sensing images in each year, and obtain remote sensing images that include the regional surface reflectance in each year; Extract the water surface of the target lake or reservoir in the region of each year based on the remote sensing image containing the regional surface reflectivity in each year, and generate water surface range vector data of the target lake or reservoir in the region of each year; Based on the water surface range vector data of the target lakes and reservoirs in the region in each year, a classified raster image of the multi-year shoreline complexity index of the target lakes and reservoirs is obtained through the natural discontinuity method; Based on the remote sensing images containing the regional surface reflectivity in each year, a classified raster image of the multi-year average water surface temperature difference of the target lake reservoir is obtained through the natural discontinuity method; The classified raster image of the multi-year shoreline complexity index of the target lake and reservoir and the classified raster image of the multi-year average water surface temperature difference of the target lake and reservoir are superimposed, and the stability zoning result of the target lake and reservoir is obtained by the natural discontinuity method; Wherein, the classified raster image of the multi-year shoreline complexity index of the target lake reservoir is obtained by the natural discontinuity method according to the water surface range vector data of the target lake reservoir in the region in each year, including: 1) Using the water surface range vector data of the target lakes and reservoirs in the region of each year, the water surface area and perimeter of the target lakes and reservoirs of each year are counted; according to the water surface area and perimeter, based on the shoreline complexity index formula, the shoreline complexity index of the target lakes and reservoirs of each year is calculated; 2) Create corresponding null value grid images based on the water surface range vector data of the target lakes and reservoirs in the region in each year, assign the shoreline complexity index value of the corresponding year to each grid in each null value grid image, and obtain a grid image reflecting the distribution of the shoreline complexity index value of the target lakes and reservoirs in each year; 3) After sorting the grid images reflecting the distribution of the shoreline complexity index values ​​of the target lakes and reservoirs in each year by area from large to small, the shoreline complexity index value of the target lakes and reservoirs corresponding to each grid in the first raster image in the sequence is used as the current value, and then starting from the first raster image, the shoreline complexity index value of the target lakes and reservoirs in the subsequent raster images is used to replace the current value at the corresponding grid. After traversing all the raster images in the sequence, a raster image reflecting the shoreline complexity distribution of the target lakes and reservoirs for many years is obtained; If there is a grid in the next grid image that exceeds the coverage of the current maximum area grid image, the grid of the next grid image is retained and the target lake shoreline complexity index of the grid is used as the current target lake shoreline complexity index value of the retained grid. 4) Using the grid image reflecting the multi-year shoreline complexity distribution of the target lake, the shoreline complexity index value of each grid is classified according to the set shoreline complexity index classification number through the natural discontinuity method. After the classification is completed, the shoreline complexity index value of each grid is replaced by the category value of the shoreline complexity index classification corresponding to each grid, so as to obtain the classified grid image of the multi-year shoreline complexity of the target lake; The classified raster images of the multi-year average water surface temperature difference of the target lake reservoir are obtained by the natural discontinuity method based on the remote sensing images containing the regional surface reflectivity in each year, including: 1) using the remote sensing images containing the regional surface reflectivity in each year, performing surface temperature inversion by atmospheric radiation transfer method to obtain a raster image of the regional surface temperature value in each year; 2) using the water surface range vector data of the target lake in the region for each year to correspondingly clip the raster image of the regional surface temperature value for each year to obtain the raster image of the water surface temperature value of the target lake in each year; 3) subtracting the average water surface temperature of all grids in the year from the water surface temperature value of each grid in the grid image of the water surface temperature value of the target lake reservoir in each year to obtain the water surface temperature difference value of each grid, and establishing a grid image of the distribution of the water surface temperature difference value in each year; 4) superimposing the grid images of the water surface temperature difference distribution in each year, calculating the average value of the water surface temperature difference of the grid in each year for the grid at the same position as the final water surface temperature difference of the grid, and obtaining a grid image reflecting the distribution of the average water surface temperature difference of the target lake reservoir over many years; 5) Using the grid image reflecting the distribution of the multi-year average water surface temperature difference of the target lake reservoir, the water surface temperature difference value of each grid is classified according to the set number of water surface temperature difference value classifications through the natural break method. After the classification is completed, the category value of the water surface temperature difference value classification corresponding to each grid is replaced by the water surface temperature difference value of the grid to obtain the classified grid image of the multi-year average water surface temperature difference of the target lake reservoir.

2. The lake and reservoir stability zoning method according to claim 1 is characterized in that: The multiple years are equally spaced.

3. The lake stability zoning method according to claim 1 is characterized in that: The regional remote sensing images are collected in the same month of each year; if remote sensing images of the same month as other years are not collected in any year, remote sensing images of the months adjacent to the said month will be used in that year.

4. The lake and reservoir stability zoning method according to claim 1 is characterized in that: The extracting of the target lake water surface in each year region according to the remote sensing image containing the regional surface reflectivity in each year, and generating the target lake water surface range vector data in each year region, include: 1) For any year's remote sensing image containing regional surface reflectance, the improved normalized difference water extraction MNDWI index and the automatic water surface extraction AWEIsh index of each grid are calculated respectively, and the grid images corresponding to the calculation results of the MNDWI index and the AWEIsh index of that year are obtained; 2) According to the raster image obtained in step 1), the distribution of the MNDWI index value and the AWEIsh value in the corresponding raster image is counted to obtain the bimodal distribution map corresponding to each index; the median between the two peaks in each bimodal distribution map is used as the threshold of the corresponding index of the year; 3) Based on the grid image of the AWEIsh index calculation result corresponding to that year, the AWEIsh index value corresponding to each grid in the image is compared with the AWEIsh index threshold of that year: If the AWEIsh index value corresponding to any grid is less than the AWEIsh index threshold of that year, the grid is determined to be the initial water surface range; otherwise, the grid is determined to be the non-water surface range; After all grids of the grid image corresponding to the AWEIsh index calculation result of that year are determined, all grids determined to be the initial water surface range are retained; 4) Based on the grid image of the MNDWI index calculation result corresponding to the year, obtain the MNDWI index values ​​corresponding to all grids retained in step 3) and determine: If the MNDWI index value corresponding to any grid retained in step 3) is less than the MNDWI index threshold of the year, the grid is determined to be a water surface range; otherwise, the grid is determined to be a non-water surface range; After the determination is completed, all grids determined as water surface ranges in that year constitute the water surface range vector data of the target lakes and reservoirs in the region in that year; 5) Repeat steps 1) to 4) to obtain the water surface area vector data of the target lake in the region for each year.

5. The lake and reservoir stability zoning method according to claim 1 is characterized in that: The classified raster image of the multi-year shoreline complexity index of the target lake and reservoir and the classified raster image of the multi-year average water surface temperature difference of the target lake and reservoir are superimposed, and the stability zoning result of the target lake and reservoir is obtained by the natural discontinuity method, including: The classified raster image of the multi-year shoreline complexity of the target lake and the classified raster image of the multi-year average surface temperature difference of the target lake are superimposed, and the category value of the shoreline complexity index classification corresponding to each grid and the category value of the water surface temperature difference classification corresponding to each grid are summed to obtain the total category value of each grid; the total category value of each grid is classified according to the set number of classifications of the target lake and reservoir stability through the natural discontinuity method, and after the classification is completed, the category value of each grid corresponding to the stability degree classification replaces the total category value of the grid, and the classified raster image of the multi-year stability of the target lake and reservoir is obtained, which is the stability zoning result of the target lake and reservoir.

6. The lake and reservoir stability zoning method according to claim 1 is characterized in that: The number of classifications of the target lake stability degree is the larger value of the number of classifications of the shoreline complexity index and the number of classifications of the water surface temperature difference.

Images