Lake water volume estimation method based on geospatial data

Abstract

The invention discloses a lake water volume estimation method based on geospatial data, and belongs to the technical field of lake water volume estimation, and the method comprises the following steps: S1, based on digital elevation model data, calculating a slope value of each grid pixel on a lake boundary, and smoothing the slope value; s2, constructing a lake polygon buffer area, and extracting a lake boundary point set; s3, matching a nearest boundary point for each grid pixel in the lake; according to the method, boundary gradient data derived by a DEM is smoothed by adopting a 31-point moving average method, subsequent calculation errors caused by single-point elevation deviation are eliminated, and the stability of basic data is guaranteed; the basin elements are incorporated into lakebed deposition thickness calculation, the theoretical maximum water depth is accurately corrected, and logic is formed by fitting the actual terrain of the lake; and finally, through actual measurement verification, the water quantity estimation RE is less than 3%, the R2 is more than 0.99, the area estimation RE is less than 4%, the R2 is more than 0.97, and the Sehlin error verification error is less than 2%, so that the estimation precision is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of lake water volume estimation, and particularly relates to a lake water volume estimation method based on geographic spatial data. BACKGROUND

[0002] As the core carrier of global hydrological cycle and biogeochemical cycle, lakes play an irreplaceable role in climate regulation, groundwater recharge, freshwater ecosystem maintenance, and water supply for human production and life, and are the key water resource library for maintaining regional ecological security and sustainable development. In recent years, affected by factors such as accelerated glacier melting and water cycle disorder caused by global warming, the spatiotemporal variation of global lake water volume has become more and more significant, and some lakes have appeared shrinkage, salinization or expansion, which directly threatens the safety of freshwater resources and the stability of the ecological system.

[0003] However, there are many proven lakes in the world, and most of them are distributed in remote mountainous areas, plateaus and other areas with poor transportation. The existing lake water volume estimation technology still has some deficiencies, and it is difficult to meet the monitoring needs of large scale and high precision, which is embodied in the following two aspects:

[0004] 1. Limitations of field measurement method

[0005] Traditional lake water volume estimation relies on field measurement to obtain underwater topography (bathymetric map), and the mainstream technologies include laying in-situ monitoring stations and using unmanned aerial vehicles / unmanned ships to carry sonar or laser radar systems for detection. Although such methods can obtain high-precision data in local areas, they have inherent defects: on the one hand, due to the limitations of bad weather, complex terrain (such as lake bottom reefs and marshes) and equipment cost, field measurement is time-consuming and labor-intensive, and the economic investment is extremely high, making it difficult to cover large lakes (such as large lakes on the Qinghai-Tibet Plateau) or achieve batch monitoring of global lakes; on the other hand, field measurement is easily disturbed by external interference (such as wind waves and water turbidity), and it is difficult to reach some areas (such as deep water areas and remote lake areas), which may lead to insufficient data integrity and affect the overall accuracy of water volume estimation.

[0006] 2. Deficiencies of remote sensing and model estimation methods

[0007] In order to break through the limitations of field measurement, existing technologies have gradually developed lake water volume estimation technologies based on remote sensing data and mathematical models, which can be divided into "relative water volume estimation" and "absolute water volume estimation", but both have obvious deficiencies:

[0008] Relative water volume estimation techniques: These techniques aim to monitor the trend of lake water volume changes, but cannot obtain the total water volume (absolute water volume) and underwater topographic map of the lake. For example, Siyu Zhu et al. established a volume-area-height curve database by integrating global water depth maps to achieve monthly-scale lake water volume change monitoring. Fangfang Yao et al. combined satellite observations with climate hydrological data to reveal the global trend of large lake water volume decline. However, such techniques can only reflect the "water volume change amount" and cannot provide key information such as the actual total water volume and water depth distribution of the lake. Moreover, they are insufficient for small lakes and cannot support fine water resource management.

[0009] Absolute water volume estimation techniques: These techniques focus on the total water volume and underwater topographic reconstruction of the lake, but generally have low precision, rely on a single data source, or ignore key natural processes. For example, Zhu, S. et al. developed a step-by-step water recession method (WRM) based on digital elevation model (DEM) to construct a digital lake bathymetric model (DLBM). Although this method does not require field data, it completely relies on the accuracy of DEM data. However, the elevation errors in DEM (such as deviations caused by snow cover in plateau regions) directly affect the accuracy of water volume estimation. Messager, M. et al. combined the HydroLAKES database with a geographic statistical model to estimate global lake volume. Although this method performs well at the global scale, it has large estimation errors for individual lakes, especially those affected by sediments. Moreover, it does not consider the impact of soil erosion and sedimentation processes on lake bottom topography, which is inconsistent with the actual formation process of the lake. In addition, existing auxiliary methods based on elevation-area relationship, machine learning, or regression analysis rely on a single data source (such as only using lake area or slope) and do not integrate natural elements at the basin scale (such as precipitation, soil properties, and basin slope). This results in poor adaptability of the model to different climate zones and terrain conditions, and insufficient stability of the estimation results.

[0010] In summary, existing lake water volume estimation techniques either have high costs and limited coverage, making them difficult to apply on a large scale, or ignore key processes such as basin sedimentation and rely on a single data source, resulting in insufficient precision. Moreover, most techniques cannot simultaneously obtain absolute water volume and underwater topographic maps, making it difficult to meet the needs of global lake resource fine management. SUMMARY

[0011] Problems to be solved

[0012] To address the problems raised in the existing background art, the present application provides a method for estimating the water volume of a lake based on geographic spatial data.

[0013] Technical solution

[0014] To solve the above problems, the present application adopts the following technical solution.

[0015] The application provides a method for estimating lake water volume based on geospatial data, and the steps are as follows:

[0016] Step S1, based on digital elevation model data, the slope values of each grid cell on the lake boundary are calculated, and the slope values are smoothed;

[0017] Step S2, a lake polygon buffer zone is constructed, and a lake boundary point set is extracted;

[0018] Step S3, a nearest boundary point is matched for each grid cell inside the lake;

[0019] Step S4, based on the lake area and the average slope value of the boundary points, the theoretical maximum water depth D of the lake is calculated Tmax ; combined with the basin area, the average slope of the basin, the average annual precipitation and the soil silt content, the lake bottom sediment thickness ST is calculated, and the actual maximum water depth D of the lake is obtained max ;

[0020] Step S5, based on the distance between each internal grid cell and its matched boundary point, the slope value of the boundary point, the theoretical maximum water depth and the actual maximum water depth, the depth value of each grid cell is calculated, and a lake underwater topographic map is generated;

[0021] Step S6, based on the depth value of each grid cell and the grid area, the total water volume of the lake is calculated;

[0022] The calculation formula of the actual maximum water depth D of the lake max is D max =D Tmax -ST.

[0023] Further, the calculation formula of the slope value in step S1 includes:

[0024]

[0025] Δz x is the elevation change rate in the x direction, and Δz y is the elevation change rate in the y direction;

[0026]

[0027] Δx and Δy are the size of the grid unit, z0 is the elevation of the center grid unit, and z1, z2,..., z8 are the elevations of the 8 adjacent grid units around the center grid unit.

[0028] Further, the formula for calculating the theoretical maximum water depth of the lake in step S4 is:

[0029]

[0030] wherein slope is the average slope value of the lake boundary points, A is the area size of the lake, and a and b are water depth parameters.

[0031] Further, the formula for calculating the thickness of the lake bottom sediment in step S4 is:

[0032] ST = D Tmax -D max = a · D A β · P · SILT · Slope

[0033] wherein Slope is the average slope value of the lake basin, D A is the area size of the lake basin, P is the average annual precipitation of the lake basin, SILT is the soil silt content, and a and b are sediment parameters.

[0034] Further, the formula for calculating the depth value of each grid cell in step S5 is:

[0035]

[0036] wherein L is the distance from the lake boundary point to the lake center point, D i is the depth of the cell, D max is the maximum depth of the lake, l is the nearest distance from the grid cell to the lake shore, a is the slope value of the lake boundary point corresponding to the grid cell point, D Tmax is the theoretical maximum depth of the lake.

[0037] Further, the formula for calculating the distance from the lake boundary point to the lake center point is:

[0038]

[0039] The formula for calculating the nearest distance from the grid cell to the lake shore is:

[0040]

[0041] wherein x I , y I are the plane coordinates of the internal grid point (I) of the lake, and x B , y B are the plane coordinates of the nearest lake boundary point (B) matched with the internal point (I).

[0042] Further, the formula for calculating the total water volume of the lake in step S6 is:

[0043]

[0044] wherein V is the total water volume of the lake, D iis the depth corresponding to the i-th pixel, and Δx 2 is the area size of a single grid.

[0045] Further, the method further comprises generating a lake underwater topographic map by using geographic information system software.

[0046] Further, the smoothing processing adopts a 31-point moving average method.

[0047] Further, the basin data includes basin area, average slope, multi-year average precipitation, and soil silt content, which are used to improve the calculation accuracy of lake sediment thickness and actual maximum water depth.

[0048] Advantages

[0049] Compared with the prior art, the advantages of the present application are:

[0050] (1) The present application adopts a "31-point moving average method" to smooth the boundary slope data derived from DEM, eliminates the subsequent calculation error caused by single-point elevation deviation, and ensures the stability of the basic data; and by including the basin elements (area, average slope, multi-year precipitation, and soil silt content) into the calculation of lake bottom sediment thickness, the theoretical maximum water depth is accurately corrected, and the actual lake terrain formation logic is fitted; finally, through the actual measurement verification, the water quantity estimation RE<3%, R 2 >0.99, the area estimation RE<4%, R 2 >0.97, the verification error of Selin Co is less than 2%, and the accuracy is far superior to the existing single data dependent model, effectively improving the estimation accuracy.

[0051] Figure 1 Fig. 1 is a flowchart of the lake water quantity estimation method based on geographic spatial data of the present application;

[0052] Figure 2 Fig. 3 is the geographic information of three lakes used in the implementation process of the present application;

[0053] Figure 3 Fig. 4 is an evaluation comparison chart of the area and water quantity estimation results of three lakes at different depths in the implementation process of the present application. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments of the present application. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations.

[0055] Therefore, the following detailed description of embodiments of the application provided in the drawings is not intended to limit the scope of the application claimed, but merely represents selected embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the application.

[0056] Embodiments:

[0057] In specific implementation, three typical lakes in western China (Qinghai Lake, Nam Co, Ebinur Lake) are selected as research objects. The three lakes have significant differences in geographical location, area size and watershed characteristics (as shown in Table 1), can fully cover lake types under different climate zones and terrain conditions, and can effectively verify the universality of the method.

[0058] In this embodiment, the water conservancy census measured data (lake area, water depth, total water volume) are taken as the benchmark, and the deviation of the estimated results of the application from the measured data is compared to evaluate the accuracy of the method.

[0059]

[0060] Table 1

[0061] Before the specific experiment, the data required for the experiment need to be prepared, and all the data are preprocessed (such as denoising, cropping, coordinate unification) to ensure spatial consistency (the coordinate system adopts WGS84);

[0062] In specific implementation, the experimental data sources are as follows:

[0063] Digital Elevation Model (DEM) data: SRTM1 Arc-Second DEM (resolution 30m x 30m) is adopted, which is from the United States Geological Survey, and is used to calculate the slope of the lake boundary grid;

[0064] Lake boundary vector data: from the 2024 national water conservancy census data set, vector cropping is carried out through ArcGIS10.8 to obtain the accurate polygon boundary of the three lakes;

[0065] Watershed data:

[0066] Watershed boundary: extracted based on DEM data through ArcGIS hydrological analysis tool;

[0067] Average slope (Slope) of the watershed: the average value is obtained by calculating all the grid slopes in the watershed through DEM;

[0068] Multi-year average precipitation P (1980-2024): from the CRU global meteorological reanalysis data set, spatial resolution 0.5°x0.5°, after interpolation, it is matched to each watershed;

[0069] SILT: derived from HWSD global soil database, extracted the average value of soil silt content in the basin;

[0070] Measured verification data: derived from the measured data of the 2024 national water conservancy census, including the water surface area, total water volume and maximum water depth corresponding to different water depths of lakes.

[0071] Reference Figure 1 In specific implementation, the steps are as follows:

[0072] Step S1: Calculate the slope of the lake boundary grid:

[0073] Based on the vector data of the lake boundary and the DEM data, the DEM data is converted into slope data using Arcgis software, and the values are extracted to each boundary point corresponding grid; finally, the boundary point slope is smoothed by 30-point average using the code, and the final boundary point slope value for calculation is obtained.

[0074] The calculation formula of the original slope of the lake boundary point is:

[0075]

[0076] Δz x is the change rate of x-direction elevation, Δz y is the change rate of y-direction elevation;

[0077]

[0078] Δx and Δy are the size of the grid unit, z0 is the elevation of the center grid unit, z1, z2,..., z8 are the elevations of the 8 adjacent grid units around the center grid unit.

[0079] It should be noted that the direction of z1, z2,..., z8 is from west to east and from north to south.

[0080] Step S2: Extract the lake boundary point set:

[0081] Based on the lake polygon boundary, a 10m buffer zone is constructed in ArcGIS (to ensure that all grid points are covered), and then the boundary points within the buffer zone are extracted by the "feature to point" tool to form the lake boundary point set (B), with about 500-800 boundary points extracted for each lake.

[0082] Step S3, match the lake interior grid with the boundary point:

[0083] The "nearest neighbor analysis method" is adopted, and a program is written by Python to calculate the Euclidean distance from each grid pixel (I) in the lake interior to all boundary points (B), and match the nearest boundary point.

[0084] Step S4: First, the theoretical maximum depth of the lake is calculated based on the average slope of the boundary points of the lake and the area of the lake; then the thickness of the lake sediment is calculated in combination with the area, average slope, multi-year average precipitation of the watershed where the lake is located, and the silt content in the soil; and finally the actual maximum depth of the lake is calculated based on the two values.

[0085] The formula for calculating the theoretical maximum depth D of the lake is: Tmax

[0086] The formula for calculating the thickness ST of the lake sediment is: ST = D Tmax max = 0.5491 × D A 0.208 · P · SILT · Slope

[0087] The formula for calculating the actual maximum depth D of the lake is: max

[0088]

[0089] where slope is the average slope value of the lake boundary points, A is the size of the lake area, Slope is the average slope value of the watershed where the lake is located, D A is the size of the watershed where the lake is located, P is the multi-year average precipitation of the watershed where the lake is located, SILT is the transitional particle between sand and clay in the soil, which is the most easily eroded part of the soil.

[0090] Step S5: Using the code, based on the position information of each grid, the slope value of the boundary points, and the theoretical maximum depth of the lake and the actual maximum depth of the lake, the depth value corresponding to each grid is calculated. The formula for calculating the depth value of each grid point is:

[0091]

[0092] D max = max(D i )

[0093] where l is the distance from the lake interior grid pixel to the nearest boundary point of the lake shore, L is the distance from the lake boundary point (B) to the lake center point (O), D i is the depth value of the grid, D max is the maximum depth of the lake, l is the nearest distance of the grid pixel to the lake shore, a is the slope value of the lake boundary point (B) corresponding to the grid pixel point (I), D Tmax is the theoretical maximum depth of the lake.

[0094] ​​​Step S6: According to the area of each grid and the corresponding depth, the volume of the column corresponding to each grid is calculated, and the volume of the column corresponding to all the grids in the lake range is added, that is, the volume of the water body of the lake, that is, the total water volume of the lake.

[0095] Meanwhile, the maximum depth of the lake can also be adjusted as required, so that the corresponding water surface area and water volume under different water depths can be conveniently calculated, and the area-water volume curve and the area-water depth curve of the lake are obtained; the formula for calculating the water surface area and the water volume under different water depths is:

[0096]

[0097] wherein V W is the water volume of the lake below a certain depth, A W is the area of the water surface under a certain depth, D i is the depth corresponding to the i-th pixel, ΔD is a certain depth value, Δx 2 is the area size of a single grid.

[0098] Step S7: Based on the calculated depth value corresponding to each grid, an underwater topographic map of the lake is drawn by using Arcgis.

[0099] Step S8: Based on the measured results of the water conservancy census, the area and the water volume of the water body under different depths are estimated in combination with the calculated water depth corresponding to each grid inside the lake; and the estimation results are compared with the measured results of the water conservancy census, so as to obtain the accuracy results of estimating the area and the water volume of the lake under different depths in the present example.

[0100] In the specific implementation analysis, the accuracy results of the three groups of lakes are as shown in the following table 2:

[0101]

[0102]

[0103] Table 2

[0104] Referring to Figures 2-3 and table 2, the accuracy of the estimation results of the area and the water volume of the water body under different depths of different lakes is high (the RE of the estimation of the area of the water body under different depths of different lakes is less than 4%, and the R2 is greater than 0.97, the RE of the estimation of the water volume of the water body under different depths of different lakes is less than 3%, and the R2 is greater than 0.99), and in summary, the present application has high accuracy in estimating the water volume of the lake, the area and the water volume of the water body under different depths, etc., and provides a higher-precision calculation method for the estimation of the water volume of the lake, which has important significance for the management of the water resources of the lake in the actual application process.

[0105] The above described embodiments only express the preferred embodiments of the present application, which are described in more detail and in more specifically, but can not be understood as the limitation of the patent scope of the present application. It should be noted that for the ordinary skilled in the art, several modifications, improvements and substitutions can be made without departing from the concept of the present application, which all belong to the protection scope of the present application.

Claims

1. A method for estimating lake water volume based on geospatial data, characterized in that, The steps are as follows: Step S1, based on digital elevation model data, calculating the slope value of each grid cell on the lake boundary, and smoothing the slope value; Step S2, constructing a lake polygon buffer zone and extracting a lake boundary point set; Step S3, matching a nearest boundary point for each grid cell inside the lake; Step S4, based on the lake area and the average slope value of the boundary point, the theoretical maximum water depth D of the lake is calculated Tmax ; combined with the basin area, the average slope of the basin, the average annual precipitation and the soil silt content, the lake bottom sediment thickness ST is calculated, and the actual maximum water depth D of the lake is obtained max ; Step S5, based on the distance between each internal grid cell and its matching boundary point, the boundary point slope value, the theoretical maximum water depth and the actual maximum water depth, calculating the depth value of each grid cell, and generating a lake underwater topographic map; Step S6, based on the depth value of each grid cell and the grid area, calculating the total water volume of the lake; The actual maximum water depth D of the lake max The calculation formula is: D max = D Tmax -ST.

2. The method for estimating lake water volume based on geospatial data according to claim 1, characterized in that: The calculation formula of the slope value in step S1 Comprises: Δz x is a rate of change of elevation in the x-direction, Δz y is a rate of change of elevation in the y-direction; Δx and Δy are the size of the grid unit, z0 is the elevation of the center grid unit, z1, z2,..., z8 are the elevations of the 8 adjacent grid units around the center grid unit.

3. The method of estimating lake water volume based on geospatial data according to claim 2, wherein: The formula for calculating the theoretical maximum water depth of the lake in step S4 is: Wherein slope is the average slope value of the lake boundary point, A is the size of the lake area, a and b are water depth parameters.

4. The method for estimating lake water volume based on geospatial data according to claim 3, wherein: The calculation formula of the lake bottom sediment thickness in step S4 is: ST = D Tmax - D max = a · D A β · P · Silt · Slope Wherein, Slope is the average slope value of the lake basin, Dx is the area size of the lake basin, P is the average annual precipitation of the lake basin, SILT is the soil silt content, and α and β are sediment parameters.

5. The method for estimating lake water volume based on geospatial data according to claim 1, wherein: The calculation formula for calculating the depth value of each grid cell in step S5 is: where L is the distance from the lake boundary point to the lake center point, D i is the depth of the pixel, D max is the maximum depth of the lake, l is the nearest distance from the grid pixel to the lake shore, a is the slope value of the lake boundary point corresponding to the grid pixel point, D Tmax is the theoretical maximum depth of the lake.

6. The method for estimating lake water volume based on geospatial data according to claim 5, wherein: The calculation formula of the distance from the lake boundary point to the lake center point is: The calculation formula of the nearest distance from the grid cell to the lake shore is: where x I , y I are the planar coordinates of the internal grid point (I), x B , y B are the planar coordinates of the nearest lake boundary point (B) matching the internal point (I).

7. The method for estimating lake water volume based on geospatial data according to claim 1, wherein: The calculation formula of the total water volume of the lake in step S6 is: where V is the total water volume of the lake, D i is the depth corresponding to the i-th pixel, Δx 2 is the size of a single grid.

8. The method for estimating lake water volume based on geospatial data according to claim 1, wherein: The method further comprises generating a lake underwater topographic map using geographic information system software.

9. The method for estimating lake water volume based on geospatial data according to claim 1, wherein: The smoothing process uses a 31-point moving average method.

10. The method for estimating lake water volume based on geospatial data according to claim 1, wherein: The basin data includes basin area, average slope, average annual precipitation and soil silt content, which is used to improve the calculation accuracy of the lake sediment thickness and the actual maximum water depth.

Images