Gridding identification method for preferential recovery area of large-water-surface submerged plants

Abstract

The invention discloses a gridding identification method for a large-water-surface submerged plant preferential recovery area. A lake and a reservoir are divided into a plurality of grid units; performing simulation calculation through a lake and reservoir two-dimensional hydrodynamic model to obtain gridding distribution of the flow velocity of a lake and reservoir water body; utilizing a remote sensing inversion model to obtain grid distribution of lake and reservoir water depth, suspended matter concentration and euphotic layer depth; the method comprises the following steps: for each grid of a lake and a reservoir, respectively calculating the product of index values of flow velocity, water depth, suspended matter concentration and euphotic layer depth parameters as a recoverability index of the submerged plants, dividing the difficulty levels of the restoration of the submerged plants in the grids based on the recoverability index, and identifying a preferential restoration area according to the order of the restoration difficulty levels. According to the method, remote sensing inversion and hydrodynamic simulation technologies are fused, the submerged plant preferential recovery areas of large water surfaces such as lakes and reservoirs can be rapidly and finely identified in a low-cost and large-range mode, the success rate of submerged plant recovery is increased, and the recovery cost of a water ecosystem is saved.

Description

Technical Field

[0001] This invention belongs to the field of water environment ecological restoration technology, specifically involving a grid-based identification method for priority restoration areas of submerged plants in large water surfaces. Background Technology

[0002] The decline of submerged plants in large water bodies such as lakes and reservoirs has led to a shift in aquatic ecosystems from a clear state dominated by submerged plants to a turbid state dominated by phytoplankton, exacerbating the frequent occurrence of algal blooms and eutrophication, and seriously threatening the health of large water body ecosystems.

[0003] Submerged plant restoration is crucial for improving the aquatic ecological environment quality of large bodies of water such as lakes and reservoirs. However, the restoration of submerged plants in large bodies of water is challenging, primarily because their growth is affected by various environmental factors such as water depth, flow velocity, turbidity, and light intensity. Furthermore, the significant differences in physiological characteristics among different submerged plants lead to unclear areas suitable for restoration. Currently, the determination of these areas relies mainly on lake and reservoir environmental conditions and expert experience to select easily restored zones. However, this method suffers from significant uncertainties, insufficient accuracy, and difficulty in widespread application. There is an urgent need to develop a grid-based, refined identification method for the restoration levels of submerged plants in large bodies of water to improve the success rate of restoration and reduce the cost of aquatic ecosystem restoration. Summary of the Invention

[0004] The purpose of this invention is to overcome the problems existing in the prior art and provide a gridded method for quickly identifying priority restoration areas for submerged plants in large water surfaces.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A grid-based identification method for priority restoration areas of submerged plants in large water surfaces includes: The lake and reservoir are divided into several grid units; The gridded distribution of water flow velocity in lakes and reservoirs was obtained through simulation calculation using a two-dimensional hydrodynamic model. Remote sensing surface temperature is obtained based on high spatial resolution remote sensing surface temperature products. A relationship model between lake and reservoir water depth and surface water temperature is constructed using measured depth and remote sensing surface temperature. The gridded distribution of lake and reservoir water depth is obtained by combining the remote sensing surface temperature of lake and reservoir with the relationship model. High spatial resolution optical remote sensing data is acquired. The optical remote sensing data is combined with the measured suspended matter concentration in the lake and reservoir to construct a remote sensing inversion model for the suspended matter concentration in the lake and reservoir. A fitting model for the measured suspended matter concentration and the euphotic depth is also established. Based on the suspended matter concentration remote sensing inversion model and the fitting model, the gridded distribution of suspended matter concentration and euphotic depth is obtained. For each grid in the lake / reservoir, the normalized values ​​of grid water depth, suspended solids concentration, and euphotic depth are calculated as the corresponding index values, and the complement of the normalized value of the flow velocity is calculated as the flow velocity index value. The product of the index values ​​of parameters such as flow velocity, water depth, suspended solids concentration, and euphotic depth is used as the submerged plant recoverability index. Based on the recoverability index, the grid of submerged plants is divided into levels of ease of recovery, and priority recovery areas are identified according to the order of ease of recovery.

[0006] In some embodiments of the present invention, the method of dividing the lake into several grid units is as follows: using the lake boundary as a constraint, the lake is divided into several grids by using a square grid.

[0007] In some embodiments of the present invention, the remote sensing surface temperature product is constructed based on the following: Acquire high spatial resolution remote sensing data with thermal infrared channels; Establish a model relating the thermal infrared band of the high spatial resolution remote sensing data to the Earth's surface temperature; By substituting the thermal infrared band data of the high spatial resolution remote sensing data for the required time period and region into the relational model, the corresponding land surface temperature product can be obtained.

[0008] In some embodiments of the present invention, the method for constructing the remote sensing inversion model of suspended matter concentration is as follows: Using the ratio of red / blue band reflectance in optical remote sensing data as the independent variable and the measured suspended matter concentration as the dependent variable, an exponential inversion model was established and the coefficients of the inversion model were calibrated. The inversion model with calibration coefficients was used as the remote sensing inversion model for suspended matter concentration.

[0009] In some embodiments of the present invention, the fitting model for the measured suspended matter concentration and the depth of the true light layer is constructed as follows: We constructed quantitative models for the relationship between the depth of the true light layer and the optical attenuation coefficient, as well as models for the relationship between the optical attenuation coefficient and the concentration of suspended matter. Using the optical attenuation coefficient as an intermediate parameter, we obtained a fitting model for the relationship between the depth of the true light layer and the concentration of suspended matter.

[0010] In some embodiments of the present invention, the high spatial resolution refers to a spatial resolution ≤30m.

[0011] In some embodiments of the present invention, a two-dimensional hydrodynamic model of a lake or reservoir is constructed using the hydrodynamic module of the EFDC model, and the gridded distribution of the water flow velocity in the lake or reservoir is simulated and calculated.

[0012] In some embodiments of the present invention, the method for identifying the priority recovery region is as follows: The recoverability index is divided into five equal parts according to the interval [0,1], and the grid area corresponding to the interval with the highest recovery level value is taken as the priority recovery area.

[0013] In some embodiments of the present invention, based on the historical growth spatial distribution of submerged plants in lakes and reservoirs, a spatiotemporal inversion model of the growth distribution of submerged plants is constructed using remote sensing data combined with the optical remote sensing characteristic index of submerged plants to obtain the long-term gridded spatial distribution of submerged plants in lakes and reservoirs. By utilizing the long-term temporal gridded spatial distribution of submerged plants in the lake and reservoir, the spatial distribution of high-frequency growth areas of submerged plants is obtained, and the identified priority recovery areas are verified.

[0014] In some embodiments of the present invention, empirical models are employed. MAI Optical remote sensing characteristic index of submerged plants:

[0015] In the formula, ; , , and The reflectances are for blue, green, red, and shortwave infrared band 1, respectively. , and These are the center wavelengths of the green, blue, and shortwave infrared band 1, respectively.

[0016] This invention integrates remote sensing inversion and hydrodynamic simulation technologies, enabling rapid, low-cost, and precise identification of priority restoration areas for submerged plants in large water bodies such as lakes and reservoirs. This improves the success rate of submerged plant restoration and saves on the cost of aquatic ecosystem restoration. Attached Figure Description

[0017] Figure 1 This is a flowchart for identifying priority restoration areas for submerged plants in large water surfaces.

[0018] Figure 2 This is a mesh partition of the hydrodynamic model of Lake A and a map showing the locations of its inflow and outflow rivers.

[0019] Figure 3 This is a verification image showing the results of the thermal infrared remote sensing inversion of the water depth of Lake A.

[0020] Figure 4 This is a distribution map of the light transmittance depth (a) and the recoverability index (b) of lake A.

[0021] Figure 5 This is a distribution map of growth limiting factors for submerged plants in Lake A over different periods.

[0022] Figure 6The data includes the recoverability index (a), priority recovery area (b), and historical distribution map (c) of submerged plants in Lake A. Detailed Implementation

[0023] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0024] Example 1 This embodiment takes a lake A as an example to illustrate a grid-based identification method for priority restoration areas of submerged plants in large water surfaces, which includes the following steps: 1. Acquisition of large-scale lake flow field based on hydrodynamic model. A two-dimensional hydrodynamic model of the lake is constructed to quantify the gridded distribution of water flow velocity.

[0025] 1.1. Model Selection: The hydrodynamic module of the EFDC (Environmental Fluid Dynamics Codes) model was used for hydrodynamic simulation. This model has demonstrated good applicability in multiple fields, including simulating the spatial extent and depth of water bodies, predicting river water quality in urban water cycles, and simulating the spatiotemporal distribution of cyanobacteria. Furthermore, the EFDC model has been widely applied to large water bodies such as rivers, reservoirs, lakes, wetlands, and estuaries.

[0026] 1.2. Grid Partitioning: Based on the basic geographic information of lake boundaries, square grids are used to partition the lakes. The grid resolution takes into account the lake size, computational accuracy, and computational efficiency, and the total number of grids is controlled to be less than 10,000. Vertical stratification is set according to the water depth of the lakes; shallow lakes use a single-layer structure, while deep lakes are vertically stratified as needed.

[0027] 1.3. Model Configuration: After importing the mesh, initial conditions, boundary conditions, and hydrodynamic parameters are set for the model. Initial conditions include the initial lake level and lakebed elevation; boundary conditions include the wind field and the flow rates of rivers flowing into and out of the lake. The data for the initial lake level and the flow rates of rivers flowing into and out of the lake are obtained from hydrological stations; the lakebed elevation data are obtained from field measurements; and the wind field data are obtained from the ERA5 dataset provided by the European Centre for Medium-Range Weather Forecasts (ECMWF). Hydrodynamic parameters include bottom roughness, windbreak coefficient, background vertical eddy viscosity, and background horizontal eddy viscosity. These parameters are set based on experience and iteratively optimized in conjunction with the model simulation results.

[0028] 1.4. Model Simulation: The maximum time step of the model simulation is set according to the CFL conditions calculated by explicit difference of the linear equations. The model outputs simulation results every 24 hours. A cold start model is used, and continuous simulation is performed for 30 days to ensure that the flow field reaches a steady state. Based on the simulated steady flow field, gridded spatial simulation results are output.

[0029] 2. Environmental feature remote sensing inversion based on Google Earth Engine (GEE). A remote sensing inversion model for environmental elements was constructed to obtain long-term spatiotemporal variations of submerged plants and aquatic ecological environment characteristics in large lakes.

[0030] 2.1. Construct a remote sensing inversion model for submerged plants in lakes and calculate the long-term spatiotemporal variations of submerged plants.

[0031] Based on the GEE platform, the historical spatial distribution of submerged plants was obtained, and an empirical model was used. MAI Characteristic indices of submerged plants: (1) In the formula, ; , , and The reflectances are for blue, green, red, and shortwave infrared band 1, respectively. , and These are the center wavelengths of the green, blue, and shortwave infrared band 1, respectively.

[0032] 2.2. Construct a remote sensing inversion model for lake water temperature and establish a model to calculate the relationship between lake water depth and water temperature.

[0033] Using high spatial resolution Landsat remote sensing data with a thermal infrared channel, and utilizing measured water depth and remote sensing water temperature data samples from sampling points, a model relating water depth and water temperature was established; Landsat 8 / 9 thermal infrared band ( TIRS The data in GEE (such as Landsat Collection 2 Level 2) already provides radiometrically calibrated preprocessed bands. ST_B10 Based on the GEE platform, Landsat 8 Collection 2 Level 2 surface temperature products (LANDSAT / LC08 / C02 / T1_L2 dataset) with no or few clouds were selected within the study area. This product already provides atmospherically corrected surface temperature bands. ST_ B10 Its value represents the brightness temperature of the Earth's surface at the sensor location, and its unit is Kelvin (K). The formula is as follows: (2) (3) Formula (2) is used for surface temperature calculation, where, T Surface water temperature (°C). ST_B10The surface brightness temperature value (K); Formula (3) is used to fit the linear regression. and T The relationship between water depth and surface water temperature is obtained, where, H Because of the water depth, T Water surface temperature, a and b The coefficients are to be determined. Based on the measured values ​​and water level data of the lake using ADEM, the image of April 14, 2013 (a similar date, clear weather, and low cloud cover) was selected for water temperature inversion, and 20 sampling points were randomly selected in Lake A for model verification.

[0034] 2.3. Construct a remote sensing inversion model for the concentration of suspended matter in lakes and calculate the gridded distribution of suspended matter concentration in lakes.

[0035] The model was calibrated using various empirical statistical methods (including linear regression, exponential fitting, and polynomial fitting). Based on the GEE platform, atmospherically corrected Landsat remote sensing data with no or few clouds were selected, and the suspended particulate matter concentration was retrieved using an exponential model with red / blue band ratio. (4) in, TSM Total suspended particulate matter, R The red band is the band. B It is the blue band.

[0036] 2.4. Construct a lake euphotic depth index model and calculate the gridded distribution of euphotic depth.

[0037] True light layer depth ( D eu The euphotic depth refers to the depth at which photosynthetically active radiation (PAR) in a water body decays to 1% of its surface value. This depth characterizes the maximum depth at which photosynthesis can occur in water. Surface PAR data were acquired using remote sensing data from the GEE platform and MODIS. A certain quantitative relationship exists between the euphotic depth and the light attenuation coefficient. Using an underwater light attenuation relationship model (with a mean contribution of 82.4%), the euphotic depth was obtained by inverting suspended matter concentration data using remote sensing. The formula is as follows: (5) (6) (7) Formula (5) is used to express the quantitative relationship between the depth of the true light layer and the light attenuation coefficient. The depth of the true light layer is often expressed as 1% of the surface depth.C The value is a constant 4.605; Formula (6) is used to calculate the diffuse attenuation coefficient; Combining Formula (5) and Formula (6), we get Formula (7), which is the calculation formula for estimating the depth of the true light layer using suspended matter; where, The depth of the true light layer (m). The diffuse attenuation coefficient (m) -1 ), z represents the depth (m) from the lake surface to the underwater surface. Irradiance at depth z (µmol·m -2 ·s -1 ), Initial irradiance at the water surface (µmol·m -2 ·s -1 ).

[0038] 3. Calculation of the recoverability index of submerged plants based on raster overlay. A normalized environmental characteristic index is used to calculate the recoverability index of submerged plants based on raster overlay, and the gridding method is used to determine the recoverability level of submerged plants.

[0039] 3.1. Normalized lake environmental characteristic indices, including water depth index, suspended matter index, euphotic depth index, and hydrodynamic index, are normalized using the following formulas for grid cell normalization: (8) (9) in, X’ A normalized lake environmental characteristic index, X max 、X min These are the maximum and minimum values ​​of the time series data, respectively; Formula (8) is used for the water depth index ( WDI ), suspended matter index ( SMI ) and hydrodynamic index calculation ( HI The greater the water depth, the higher the concentration of suspended matter, and the stronger the hydrodynamic force, the more difficult it is for submerged plants to grow; Formula (9) is used for the euphotic depth index ( DEI According to calculations, the deeper the euphotic layer, the easier it is for submerged plants to grow. In areas where the water depth is greater than the euphotic layer depth, the euphotic layer depth index is set to 0. The index is restricted to the range of [0,1]. The higher the index, the easier it is for submerged plants to grow, and the lower the index, the more difficult it is for submerged plants to grow.

[0040] 3.2. The recoverability index of submerged plants is calculated using the following formula: (10) In the formula, RIThe resilience index of submerged plants; SMI The suspended solids concentration index; WDI This refers to the water depth index; DEI The true light layer depth index; HI This refers to the hydrodynamic index.

[0041] 4. Locate priority restoration areas for submerged plants, identify growth limiting factors for submerged plants, and determine priority restoration areas by gridding.

[0042] 4.1. Identification of Limiting Factors for Submerged Plant Growth: After normalizing the lake's water environment characteristic indices, the limiting level was determined by the value of each cell. A raster overlay analysis was performed on the suspended solids concentration index, water depth index, euphotic depth index, and hydrodynamic index; the index with the smallest value in each cell was identified as the limiting factor.

[0043] 4.2. Grid-based determination of the recoverability level of submerged plants: A recovery level assessment model for submerged plants in lakes was constructed. The recovery index of submerged plants was calculated based on four indices: hydrodynamics, water depth, suspended solids concentration, and euphotic depth. The recovery difficulty level of submerged plant recovery areas in lake A was divided into five levels. The interval [0,1] was divided into five equal parts. The recovery level was determined according to the value. The recovery level was divided into five levels: I [0.8,1], II [0.6,0.8], III [0.4,0.6], IV [0.2,0.4], and V [0,0.2].

[0044] 4.3. Priority areas for the restoration of submerged plants: Areas with high recoverability indices were identified as priority restoration areas. Based on the grading results in section 4.2, Grade I areas were the easiest to restore, corresponding to high index values; Grade V areas were the most difficult to restore, corresponding to low index values. Further, using the inversion method in section 2.1, historical distribution data of submerged plants were obtained to verify the rationality of the priority restoration area location. Specifically: Based on historical distribution data of submerged plants, the spatial distribution of high-frequency growth areas was obtained and compared with areas of high recoverability index for verification. The results are as follows: Figure 6 As shown, the priority restoration area identified in this invention is basically consistent with the historical high-frequency growth area. This demonstrates that this invention can quickly and accurately identify the priority restoration area of ​​submerged plants in large water surfaces.

[0045] Furthermore, such as Figure 5 As shown, the factor with the smallest index value is identified as the limiting factor. By combining the limiting factors for the growth of submerged plants, the environmental characteristics of the areas where submerged plants are given priority for restoration can be identified, providing a reference for the restoration of submerged plants.

Claims

1. A grid-based identification method for priority restoration areas of submerged plants in large water surfaces, characterized in that, include: The lake and reservoir are divided into several grid units; The gridded distribution of water flow velocity in the lake and reservoir was obtained by simulating and calculating using a two-dimensional hydrodynamic model. Remote sensing surface temperature is obtained based on high spatial resolution remote sensing surface temperature products. A relationship model between lake and reservoir water depth and surface water temperature is constructed using measured depth and remote sensing surface temperature. The gridded distribution of lake and reservoir water depth is obtained by combining the remote sensing surface temperature of lake and reservoir with the relationship model. High spatial resolution optical remote sensing data is acquired. The optical remote sensing data is combined with the measured suspended matter concentration in the lake and reservoir to construct a remote sensing inversion model for the suspended matter concentration in the lake and reservoir. A fitting model for the measured suspended matter concentration and the euphotic depth is also established. Based on the suspended matter concentration remote sensing inversion model and the fitting model, the gridded distribution of suspended matter concentration and euphotic depth is obtained. For each grid in the lake / reservoir, the normalized values ​​of grid water depth, suspended solids concentration, and euphotic depth are calculated as the corresponding index values, and the complement of the normalized value of the flow velocity is calculated as the flow velocity index value. The product of the index values ​​of parameters such as flow velocity, water depth, suspended solids concentration, and euphotic depth is used as the submerged plant recoverability index. Based on the recoverability index, the grid of submerged plants is divided into levels of difficulty for recovery, and priority recovery areas are identified according to the order of recovery difficulty.

2. The method according to claim 1, characterized in that, The method of dividing the lake into several grid units is as follows: using the lake boundary as a constraint, the lake is divided into several grids by using a square grid.

3. The method according to claim 1, characterized in that, The remote-sensed surface temperature product is constructed based on the following method: Acquire high spatial resolution remote sensing data with thermal infrared channels; Establish a model relating the thermal infrared band of the high spatial resolution remote sensing data to the Earth's surface temperature; By substituting the thermal infrared band data of the high spatial resolution remote sensing data for the required time period and region into the relational model, the corresponding land surface temperature product can be obtained.

4. The method according to claim 1, characterized in that, The method for constructing the remote sensing inversion model for suspended matter concentration is as follows: Using the ratio of red / blue band reflectance in optical remote sensing data as the independent variable and the measured suspended matter concentration as the dependent variable, an exponential inversion model was established and the coefficients of the inversion model were calibrated. The inversion model with calibration coefficients was used as the remote sensing inversion model for suspended matter concentration.

5. The method according to claim 1, characterized in that, The fitting model for the measured suspended matter concentration and the depth of the eurythmic layer is constructed as follows: We constructed quantitative models for the relationship between the depth of the true light layer and the optical attenuation coefficient, as well as models for the relationship between the optical attenuation coefficient and the concentration of suspended matter. Using the optical attenuation coefficient as an intermediate parameter, we obtained a fitting model for the relationship between the depth of the true light layer and the concentration of suspended matter.

6. The method according to claim 1, characterized in that, The term "high spatial resolution" refers to a spatial resolution ≤ 30m.

7. The method according to claim 1, characterized in that, Using the hydrodynamic module of the EFDC model, a two-dimensional hydrodynamic model of the lake and reservoir is constructed, and the gridded distribution of water flow velocity in the lake and reservoir is simulated and calculated.

8. The method according to claim 1, characterized in that, The method for identifying the priority recovery area is as follows: The recoverability index is divided into five equal parts according to the interval [0,1], and the grid area corresponding to the interval with the highest recovery level value is taken as the priority recovery area.

9. The method according to claim 1, characterized in that, The method also includes, based on the historical growth spatial distribution of submerged plants in lakes and reservoirs, using remote sensing data combined with the optical remote sensing characteristic index of submerged plants, constructing a spatiotemporal inversion model of the growth distribution of submerged plants, and obtaining the long-term time-series gridded spatial distribution of submerged plants in lakes and reservoirs. By utilizing the long-term temporal gridded spatial distribution of submerged plants in the lake and reservoir, the spatial distribution of high-frequency growth areas of submerged plants is obtained, and the identified priority recovery areas are verified.

10. The method according to claim 9, characterized in that, The empirical model MAI was used as the optical remote sensing characteristic index for submerged plants. ; In the formula, ; , , and The reflectances are for blue, green, red, and shortwave infrared band 1, respectively. , and These are the center wavelengths of the green, blue, and shortwave infrared band 1, respectively.

Images