Sea-entering river nitrogen and phosphorus diffusion simulation method and system for representing microtopographic features

Abstract

The invention discloses a sea-entering river nitrogen and phosphorus diffusion simulation method for representing microtopographic features. The method comprises the steps of data acquisition and preprocessing; side inflow nitrogen and phosphorus pollution density time sequence calculation; constructing a one-dimensional river nitrogen and phosphorus pollution diffusion model; constructing a two-dimensional sea surface nitrogen and phosphorus pollution diffusion model; key parameters are updated. Aiming at the problems of sparse observation data, difficulty in explicit representation of micro-topographic features, insufficient coupling of a land-river-bay process and difficulty in dynamically reflecting source and sink changes in nitrogen and phosphorus pollution diffusion simulation of a sea-entering river and a near-shore bay, under the condition of only depending on satellite water quality inversion data and a small amount of river monitoring data, the method provided by the invention is used for solving the problems that the observation data is sparse, the micro-topographic features are difficult to explicitly represent and a model is insufficient in coupling. The method comprises the following steps: depicting a multistage catchment area confluence process representing microtopographic features, converting land two-dimensional nitrogen and phosphorus pollution into one-dimensional river side inflow, and further coupling with a one-dimensional river and two-dimensional sea surface diffusion model to realize high-resolution space-time simulation and rolling updating of nitrogen and phosphorus pollution density distribution; and a refined technical support is provided for sea-entering nitrogen and phosphorus flux accounting and emission reduction management and control.

Description

Technical Field

[0001] This invention relates to the field of water quality monitoring technology, specifically to a method and system for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features. Background Technology

[0002] With global climate change and intensified human activities, nitrogen and phosphorus pollutants carried by rivers flowing into the sea are accumulating continuously, leading to increasingly prominent water environment problems such as eutrophication and red tides in coastal areas. Numerous studies have shown that nitrogen and phosphorus pollution from rivers flowing into the sea has become one of the key drivers of water quality deterioration and ecosystem degradation in coastal bays. my country's coastal waters are significantly affected by land-based pollution, with some bays experiencing prolonged periods of moderate to severe eutrophication. Nitrogen and phosphorus fluxes from rivers flowing into the sea exhibit significant high-frequency fluctuations and spatial heterogeneity. The state has clearly stated the need to strengthen process control of nitrogen and phosphorus pollution from rivers flowing into the sea, achieving refined assessment and dynamic early warning of pollutant fluxes.

[0003] Currently, the methods for simulating the diffusion of nitrogen and phosphorus pollutants in rivers flowing into the sea mainly fall into the following three categories:

[0004] (1) Hydrodynamic-water quality numerical model based on physical mechanisms:

[0005] Typical examples include MIKE, EFDC, and Delft3D, which simulate nitrogen and phosphorus density and flux in rivers and nearshore seas based on mass and momentum conservation equations, coupled with convection-dispersion equations and degradation and transformation processes. These models have clear physical mechanisms and are suitable for conducting scheme comparisons and scenario analyses.

[0006] (2) Data-driven spatiotemporal inversion and prediction model for water quality:

[0007] Using machine learning / deep learning methods such as random forest, support vector machine, LSTM, and convolutional neural network, and leveraging multi-source data including satellite remote sensing water quality inversion results, limited cross-sectional monitoring data, and meteorological and hydrological elements, we perform spatiotemporal interpolation, short-term prediction, or risk assessment of nitrogen and phosphorus density.

[0008] (3) A method based on the fusion of physical models and AI models:

[0009] By combining Kalman filtering, variational assimilation, and physical information neural networks, physical equations are coupled with AI models, and observational data or remote sensing inversion results are embedded into the solution process to correct model parameters and states, thereby improving the accuracy and stability of nitrogen and phosphorus diffusion simulation.

[0010] The current method has the following drawbacks:

[0011] (1) It is difficult to balance computational efficiency and multi-scale refinement:

[0012] Physical models require the construction of a grid for the study area and iterative solution of partial differential equations. To ensure computational accuracy, coarse spatial grids are often used, which makes it difficult to explicitly represent micro-topographic features such as slopes, ditches, and depressions around rivers flowing into the sea. If the resolution is forcibly increased, the computational load increases dramatically, making it difficult to meet the efficiency requirements of long-term series or multi-scenario simulations.

[0013] (2) Strong dependence on high-density, long-sequence monitoring data:

[0014] Traditional water quality models are highly sensitive to parameters such as roughness, dispersion coefficient, and source-sink terms, requiring extensive cross-sectional monitoring and flow observation for calibration. Data-driven models, on the other hand, require long-term high-frequency monitoring or training with multiple periods of remote sensing imagery. In situations where monitoring stations are sparse in rivers flowing into the sea and their surrounding areas, and nitrogen and phosphorus observation data are limited, it is difficult to guarantee model accuracy and generalization ability.

[0015] (3) Insufficient characterization of micro-topographic features and surface source inflow process:

[0016] Existing models often use administrative units or regular grids as computational units, which makes it difficult to reflect the control effect of micro-topography such as roads, embankments, ditches, and coastal depressions on runoff paths and nitrogen and phosphorus pollution accumulation processes. They often distribute pollution evenly to river nodes, resulting in distortion of the location and intensity of lateral inflows, which in turn affects the simulation accuracy of nitrogen and phosphorus density and flux into the sea along the river.

[0017] (4) Limited coupling and dynamic updating capabilities of land-river-bay processes:

[0018] Many studies only simulate single river sections or single bay areas, lacking a description of the continuous process from two-dimensional surface source diffusion in the catchment area to one-dimensional river channel diffusion to two-dimensional bay diffusion. Existing models often require recalibration or manual parameter adjustment after incorporating new satellite monitoring results, making it difficult to reflect in a timely manner the dynamic impacts of watershed land use change and sewage discharge structure adjustments on nitrogen and phosphorus fluxes into the sea and bay water quality. Summary of the Invention

[0019] This invention provides a method and system for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features. Based on geographic information and observational data of rivers flowing into the sea and nearshore sea surface areas, a three-stage technical approach is constructed: two-dimensional land surface nitrogen and phosphorus pollution simulation, one-dimensional river nitrogen and phosphorus pollution diffusion simulation, and two-dimensional sea surface nitrogen and phosphorus pollution diffusion simulation. By converting the two-dimensional land-based area source load into a one-dimensional lateral inflow along the river channel, and mapping the one-dimensional flux of the estuary into a two-dimensional sea-bound boundary of the bay, continuous simulation and spatiotemporal evolution analysis of nitrogen and phosphorus pollutant density distribution between the catchment area, rivers flowing into the sea, and the bay are achieved, solving the problems mentioned in the background art.

[0020] This invention provides the following technical solution: a method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea to characterize micro-topographic features, comprising:

[0021] S1. Data Acquisition and Preprocessing: Geographic information data and observation data covering rivers flowing into the sea and nearshore sea surface areas are acquired and preprocessed to divide the water catchment areas into multiple levels. The observation data includes water quality remote sensing and cross-sectional monitoring data reflecting the spatial distribution characteristics of nitrogen and phosphorus.

[0022] S2. Time series calculation of nitrogen and phosphorus pollution density in lateral inflow: Based on multi-level catchment areas and observation data, the time series of lateral inflow flow and nitrogen and phosphorus pollution density of each river segment flowing into the sea are calculated.

[0023] S3. Construct a one-dimensional river nitrogen and phosphorus pollution diffusion model: Establish a one-dimensional river network and cross-sectional parameters for rivers flowing into the sea, and use the flow time series and nitrogen and phosphorus pollution density time series of each river segment flowing into the sea as boundary conditions to establish and solve the one-dimensional river nitrogen and phosphorus pollution diffusion equation, thereby obtaining the flow time series and nitrogen and phosphorus pollution density time series of the rivers flowing into the sea.

[0024] S4. Construct a two-dimensional sea surface nitrogen and phosphorus pollution diffusion model: Construct a two-dimensional computational grid for the sea area of ​​the study area, and use the inflow time series of rivers flowing into the sea and the inflow nitrogen and phosphorus pollution density time series as boundary conditions to establish and solve the two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation to obtain the sea surface nitrogen and phosphorus pollution density time series.

[0025] S5. Key Parameter Update: Based on geographic information data and observation data, the time series of nitrogen and phosphorus pollution density entering the sea and the time series of nitrogen and phosphorus pollution density on the sea surface are compared with the observation field to calibrate and dynamically update the key parameters of the model.

[0026] This invention addresses the problems in nitrogen and phosphorus pollution diffusion simulation of rivers flowing into the sea and nearshore bays, such as sparse observational data, difficulty in explicitly representing micro-topographic features, insufficient coupling of land-river-bay processes, and difficulty in dynamically reflecting source-sink changes. It provides a method for simulating the diffusion of nitrogen and phosphorus pollutants in rivers flowing into the sea by characterizing micro-topographic features. This method can effectively depict the confluence process of multi-level catchment areas with micro-topographic features, relying only on satellite water quality inversion data and a small amount of river monitoring data. It converts two-dimensional nitrogen and phosphorus pollution in the land into one-dimensional river side inflow and further couples it with one-dimensional river and two-dimensional sea surface diffusion models to achieve high-resolution spatiotemporal simulation and rolling updates of nitrogen and phosphorus pollution density distribution, providing refined technical support for the calculation of nitrogen and phosphorus fluxes into the sea and emission reduction management.

[0027] This invention also provides a system for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea by applying a method for characterizing micro-topographic features, comprising:

[0028] Data acquisition and preprocessing module: used to acquire geographic information data and observation data covering rivers flowing into the sea and nearshore sea surface areas, and to preprocess the data to divide the water catchment areas into multiple levels. The observation data includes water quality remote sensing and cross-sectional monitoring data reflecting the spatial distribution characteristics of nitrogen and phosphorus.

[0029] Side-flow nitrogen and phosphorus pollution density time series calculation module: used to calculate the side-flow flow time series and nitrogen and phosphorus pollution density time series for each river segment flowing into the sea based on multi-level catchment areas and observation data;

[0030] One-dimensional river nitrogen and phosphorus pollution diffusion model: By establishing a one-dimensional river network and cross-sectional parameters of rivers flowing into the sea, the flow time series and nitrogen and phosphorus pollution density time series of each river segment flowing into the sea are used as boundary conditions to establish and solve the one-dimensional river nitrogen and phosphorus pollution diffusion equation, and obtain the inflow time series and inflow nitrogen and phosphorus pollution density time series of the rivers flowing into the sea.

[0031] Two-dimensional sea surface nitrogen and phosphorus pollution diffusion model: By constructing a two-dimensional computational grid for the sea area of ​​the study area, and using the time series of inflow of rivers into the sea and the time series of nitrogen and phosphorus pollution density into the sea as boundary conditions, a two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation is established and solved to obtain the time series of nitrogen and phosphorus pollution density on the sea surface.

[0032] Key parameter update module: Based on geographic information data and observation data, it compares the time series of nitrogen and phosphorus pollution density entering the sea and the time series of nitrogen and phosphorus pollution density on the sea surface with the observation field, calibrates and dynamically updates the key parameters of the model.

[0033] The present invention has the following beneficial effects:

[0034] 1. This method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, which characterizes micro-topographic features, improves the ability to represent micro-topography: by dividing the water catchment area into multiple levels and extrapolating the lateral inflow, it depicts the nitrogen and phosphorus accumulation process of rivers flowing into the sea corresponding to micro-topography such as slopes, depressions, and ditches, thus making up for the problem that traditional coarse grid models are insufficient in characterizing the impact of micro-topography.

[0035] 2. This method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea by characterizing micro-topographic features improves the accuracy of integrated river-bay simulation: By simulating the diffusion of nitrogen and phosphorus pollution through the coupling of multi-level catchment areas, one-dimensional river channels, and two-dimensional sea surfaces, the migration and diffusion of nitrogen and phosphorus from land to estuary and then to the ocean are continuously tracked, thereby improving the fineness of the distribution of nitrogen and phosphorus pollution density in rivers flowing into the sea.

[0036] 3. The dynamic updating and management application capabilities of this simulation method for nitrogen and phosphorus diffusion in rivers flowing into the sea, which characterizes micro-topographic features, are enhanced: as new satellite images and monitoring data continue to be introduced, the surface source load and diffusion parameters can be continuously updated and corrected, enabling the model to be more adaptable to scenarios such as sewage discharge structure adjustment and land use change, and providing real-time support for nitrogen and phosphorus emission reduction and water quality management. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the steps in the method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea to characterize micro-topographic features, as described in an embodiment of the present invention.

[0038] Figure 2 This is a schematic diagram of the data preprocessing steps in Embodiment 1 of the present invention.

[0039] Figure 3 This is a schematic diagram of the time-series calculation steps for the side-flow nitrogen and phosphorus pollution density in Embodiment 1 of the present invention.

[0040] Figure 4 This is a schematic diagram of the one-dimensional river nitrogen and phosphorus pollution diffusion simulation steps in Embodiment 1 of the present invention.

[0041] Figure 5 This is a schematic diagram of the two-dimensional sea surface nitrogen and phosphorus pollution diffusion simulation steps in Embodiment 1 of the present invention.

[0042] Figure 6 This is a schematic diagram of the parameter calibration and dynamic update steps in Embodiment 1 of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] This method establishes a multi-level catchment system and combines it with GIS information such as land use and sewage outlets to simulate land pollution and generate side inflows of nitrogen and phosphorus pollution. Using this as a boundary, a one-dimensional river pollution diffusion model is constructed, outputting the time series of nitrogen and phosphorus density at river mouths. Then, using this as a boundary, a two-dimensional sea surface pollution diffusion model is constructed, outputting the spatiotemporal series of nitrogen and phosphorus pollution density. Finally, using satellite and station monitoring data, combined with Kalman filtering and parameter calibration, the model parameters from the preceding steps are adjusted, continuously iterating and updating the model results to enable dynamic updates and adaptability to changes in various external geographic information.

[0045] like Figure 1 As shown: The implementation methods include:

[0046] S1. Data Acquisition and Preprocessing: Geographic information data and observation data covering rivers flowing into the sea and nearshore sea surface areas are acquired and preprocessed to divide the watershed into multiple levels. Observation data includes water quality remote sensing and cross-sectional monitoring data reflecting the spatial distribution characteristics of nitrogen and phosphorus. Specifically, geographic information data includes surface elevation, land use type, nautical chart data, river centerline vector data, riverbank boundary vector data, and coastline vector data, vector data of drainage pipes, roads, and ditches along the riverbanks, and the location and discharge data of sewage outlets along the river. Observation data are obtained through satellite observation and station observation. The apparent reflectance of the water body is converted into a nitrogen and phosphorus pollutant density field through satellite observation data inversion. The monitoring sequence of nitrogen and phosphorus pollutant density and the corresponding flow, water level, and precipitation are obtained through hydrological station observation and river cross-sectional observation, and then fused to form a nitrogen and phosphorus pollutant density observation field.

[0047] Preferably, during preprocessing, the DEM undergoes a series of preprocessing steps, including depression filling, leveling, and micro-topography correction. Then, flow direction and flow rate are analyzed to extract primary catchment areas. Specifically, this can be achieved by combining the DEM with the main channel of the river flowing into the sea, dividing the river into several computational river segments and control sections. Using the control sections as outlets, primary catchment areas corresponding to each control section are delineated based on flow analysis. Virtual inflow points are then established along the riverbank. Based on information such as roads and pipelines, Thiessen polygons are constructed to form secondary sub-catchment areas representing micro-topography. Specifically, the intersection points of the river centerline, the boundary lines on both banks, the outlets of the drainage network along the riverbank, the road endpoints, and the riverbank embankments can be marked as generation points. Based on these generation points, Thiessen polygons are constructed and spatially superimposed and clipped with the primary catchment areas of the corresponding river segments, resulting in the intersections that yield the secondary catchment areas distributed along the riverbank.

[0048] S2. Calculation of Nitrogen and Phosphorus Pollution Density in Side Inflows: Based on multi-level catchment areas and observation data, the time series of side inflow discharge and nitrogen and phosphorus pollution density for each river segment flowing into the sea are calculated. Specifically, combining land use and satellite inversion results, nitrogen and phosphorus pollution density coefficients are assigned to each sub-catchment area. The hydrological response model is used to invert the time series of side flow and nitrogen and phosphorus mass flux of the corresponding river segment for that sub-catchment area, thereby obtaining the side inflow discharge and nitrogen and phosphorus density process of each river segment.

[0049] Specifically, one approach could be to first base the precipitation intensity time series P within the hydrological unit k of the secondary catchment area. k (t), combining the land use type and soil properties of hydrological unit k, effective rainfall calculation is performed on the original precipitation; then, the Horton model is used to describe the decay process of soil infiltration capacity of hydrological unit k over time, and the time series Q of lateral inflow of hydrological unit k to the corresponding river section flowing into the sea during precipitation is obtained. k(t); Nitrogen and phosphorus output coefficients are set for hydrological unit k, and combined with the nitrogen and phosphorus pollutant density field within hydrological unit k, the time series relationship of nitrogen and phosphorus mass flux and density is constructed to obtain the pollution density time series C of the side inflow total nitrogen and total phosphorus. j N (t), C j P (t).

[0050] Preferably, the time-series relationship between nitrogen and phosphorus mass flux and density includes:

[0051] The underlying surface within hydrological unit k is divided into different types based on land use, and a pre-defined total nitrogen load factor e per unit flow is given for each type. N,r 0 and the total phosphorus base load factor per unit flow e P,r 0 Based on the area occupied by each land use type in hydrological unit k, calculate the basic non-point source load factor e of total nitrogen and total phosphorus per unit flow, determined by the land use type, for hydrological unit k. N,k land and e P,k land ;

[0052] Calculate the total nitrogen and total phosphorus load factors per unit flow rate within hydrological unit k based on the number of sewage outlets within hydrological unit k. N,k out and e P,k out ;

[0053] Based on the nitrogen and phosphorus pollutant density fields obtained from water quality remote sensing inversion, the area-weighted average density C of total nitrogen and total phosphorus within hydrological unit k is calculated. N,k sat C P,k sat A correction factor was constructed by comparing the ratio of the density to the reference background density of the study area. The land use non-point source contribution and the sewage outlet point source contribution were summed and adjusted using a satellite correction factor to obtain the comprehensive unit flow total nitrogen load coefficient e for hydrological unit k. N , k and total phosphorus loading factor e P , k The nitrogen and phosphorus mass flux M was calculated. N,k (t), M P,k (t), construct the time series relationship between nitrogen and phosphorus mass flux and density.

[0054] S3. Construct a one-dimensional river nitrogen and phosphorus pollution diffusion model: Establish a one-dimensional river network and cross-sectional parameters for rivers flowing into the sea. Use the flow time series and nitrogen and phosphorus pollution density time series of each river segment flowing into the sea as boundary conditions to establish and solve the one-dimensional river nitrogen and phosphorus pollution diffusion equation, and obtain the inflow time series and inflow nitrogen and phosphorus pollution density time series of the rivers flowing into the sea. Specifically: Establish a one-dimensional river network and cross-sectional parameters on the river channel flowing into the sea. Based on the SWMM model, construct a one-dimensional nitrogen and phosphorus pollution diffusion equation. Use the side inflow as the source term and solve it in a coupled manner to obtain the linear density distribution of nitrogen and phosphorus on the river channel and the nitrogen and phosphorus flux time series at the river estuary.

[0055] Preferably, the feasible approach is to construct a one-dimensional water quality model of the river flowing into the sea based on the divided river segments, and use the one-dimensional unsteady flow equation under the SWMM dynamic wave solution framework. Combined with river hydrological parameters, the river flow rate Q at any position x and time t of any river segment can be obtained. ch The spatiotemporal distribution of (x,t) and the cross-sectional area A(x,t); the time series of the side inflow flow Q k (t), pollution density time series of side-flow total nitrogen and total phosphorus C j N (t), C j P (t) and precipitation data were introduced as mass source terms to construct a one-dimensional river nitrogen and phosphorus pollution diffusion equation, and the inflow time series and inflow nitrogen and phosphorus pollution density time series of rivers flowing into the sea were calculated.

[0056] S4. Constructing a Two-Dimensional Sea Surface Nitrogen and Phosphorus Pollution Diffusion Model: A two-dimensional computational grid is constructed for the study area. Using the time series of river inflows and the time series of nitrogen and phosphorus pollution densities flowing into the sea as boundary conditions, a two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation is established and solved to obtain the time series of nitrogen and phosphorus pollution densities. This simulates the diffusion and transport process of nitrogen and phosphorus on the sea surface near river mouths, yielding the two-dimensional spatiotemporal distribution of nitrogen and phosphorus density. Specifically: A two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation can be constructed using the time series of river inflows and the time series of nitrogen and phosphorus pollution densities flowing into the sea as boundary conditions, combined with dynamic parameters, through a two-dimensional shallow water equation and a mass conservation equation. The finite volume method is then used to calculate the time series of nitrogen and phosphorus pollution densities on the sea surface.

[0057] Preferred methods for establishing two-dimensional sea surface nitrogen and phosphorus pollution diffusion equations include:

[0058] The bottom Manning coefficient of the sea area is obtained, and the time-varying velocity field (u,v) of the sea area is calculated by the two-dimensional shallow water equations.

[0059] Based on the depth-average volumetric density of nitrogen and phosphorus pollutant components at time t and location (x,y), the equivalent horizontal dispersion coefficients of pollutant components in the x and y directions, and the reaction term per unit volume, a depth-average mass conservation equation is designed to construct a two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation.

[0060] Possible methods for obtaining time series of sea surface nitrogen and phosphorus pollution density by solving the two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation include:

[0061] The sea area is divided into several computational units on a plane, and a mass conservation relationship is established for each unit within a finite volume framework.

[0062] The density field is updated using explicit, semi-implicit, or fully implicit time integration schemes. The density of all units in the entire sea area is updated sequentially throughout the simulation period to obtain the time series of nitrogen and phosphorus pollution density on the sea surface.

[0063] S5. Key Parameter Update: Based on geographic information data and observation data, the time series of nitrogen and phosphorus pollution density entering the sea and the time series of nitrogen and phosphorus pollution density on the sea surface are compared with the observation field to calibrate and dynamically update the key parameters of the model. Specifically, satellite monitoring data and river monitoring section data can be used to compare the one-dimensional simulated density of rivers and the two-dimensional simulated density of bays with the observation field. Through optimization solutions or data assimilation methods, parameters such as nitrogen and phosphorus pollution density coefficients, river roughness, and sea surface diffusion coefficients are assigned to each sub-catchment area. The key parameters are updated on a rolling basis when new observation data arrives, so that the nitrogen and phosphorus diffusion simulation results can dynamically perceive environmental changes.

[0064] To provide a detailed explanation of this method, the following specific embodiments are provided.

[0065] Example 1

[0066] A method for simulating the diffusion of nitrogen and phosphorus pollutants in rivers flowing into the sea, characterizing micro-topographic features, includes:

[0067] like Figure 2 As shown, the first step is data preprocessing, specifically:

[0068] Step 1.1 Multi-source data acquisition includes:

[0069] To acquire geographic information and observational data covering rivers flowing into the sea and nearshore sea surface areas.

[0070] Geographic information data includes surface elevation, land use, nautical chart data, river centerline vector data, riverbank boundary vector data, and coastline vector data, as well as vector data of drainage pipes, roads, and ditches along the riverbanks, and the location and discharge data of sewage outlets along the river.

[0071] The observational data primarily reflects the spatial distribution characteristics of nitrogen and phosphorus through water quality remote sensing and cross-sectional monitoring data, including satellite and station observations. Visible and near-infrared multispectral observation data from geostationary satellites, after atmospheric correction and water body extraction, are converted into a nitrogen and phosphorus pollutant density field using existing publicly available water quality inversion algorithms, such as the FAI model (Hu, C., 2009. A Novel Ocean Color Index to Detect Floating Algae in the Global Oceans. Remote Sensing of Environment, 113, 2118-2129). Hydrological station observations and river cross-sectional observations can obtain monitoring sequences of nitrogen and phosphorus pollutant density and corresponding flow, water level, and precipitation. These two data are then fused using Kalman filtering to form a nitrogen and phosphorus pollutant density observation field. Combining multi-source observational data with geographic information data provides a data foundation for subsequent micro-topographic analysis, catchment area delineation, and nitrogen and phosphorus pollutant density estimation.

[0072] Step 1.2 DEM and micro-topography processing includes:

[0073] The DEM (Digital Elevation Model) was processed by filling depressions and leveling, followed by flow direction and discharge analysis using the D8 algorithm. The main surface channels and secondary surface runoff paths were identified based on the cumulative runoff threshold. Using river vector data and the DEM, the main river channel flowing into the sea was divided into several computational river segments and control sections based on tributary inlets and estuaries. Using the control sections as outlets, and based on discharge analysis, the primary catchment areas corresponding to each control section were obtained, representing the catchment range of each river segment flowing into the sea under the overall topographic control.

[0074] The points where the river centerline, the boundary lines on both banks, the outlets of the drainage network along the river, the end points of roads, and the river embankments are marked as generation points. Based on the generation points, Thiessen polygons are constructed. The resulting polygons are spatially superimposed and clipped with the primary catchment areas of the corresponding river segments, and the intersections are used to obtain secondary catchment area units distributed along the river. Each secondary catchment area spatially belongs to a specific control section or a specific generation point, representing the catchment unit of a specific river segment of the river flowing into the sea under the combined effects of micro-topography and drainage facility operations.

[0075] Through the above DEM processing and multi-level catchment area division process, a multi-level catchment unit system is formed, providing a fine spatial discretization basis for subsequent model processing.

[0076] Then as Figure 3 The following steps are performed: Step 2. Time-series calculation of the inflow nitrogen and phosphorus pollution density from the side stream. Specifically:

[0077] For each hydrological unit divided into multi-level catchment areas, a response relationship between precipitation, runoff, and nitrogen and phosphorus pollution density is established, and the time series of flow rate and nitrogen and phosphorus pollution density of the corresponding lateral inflow for that hydrological unit are output.

[0078] Let the hydrological unit be numbered k, the time variable be t, and the time series of the lateral inflow discharge of the river segment corresponding to this hydrological unit be Q. k (t), the corresponding time series of the density of nitrogen and phosphorus pollutants flowing from the side is C. k N (t), C k P (t), where the superscripts N and P represent nitrogen and phosphorus, respectively.

[0079] First, based on the time series of precipitation intensity P within the secondary catchment area hydrological unit k. k (t), combining the land use type and soil properties of this unit, the proportion of impervious area, the interception amount of surface depressions, and infiltration parameters are determined to calculate the effective rainfall of the original precipitation. The Horton model is used to describe the decay process of soil infiltration capacity over time in this hydrological unit:

[0080]

[0081] Among them, f k (t) represents the infiltration rate of hydrological unit k at time t, f 0,k f is the initial infiltration rate of the unit under completely dry conditions. c,k For a long-term stable infiltration rate, k f,k This is the infiltration attenuation coefficient.

[0082] For the precipitation P within a given time step Δt k (t)Δt deducts the amount intercepted by the depression and the amount of infiltration. The effective rainfall P that generates surface runoff is obtained. k ∗ (t). Based on this, the effective rainfall is input into the linear reservoir-type runoff model, and the water storage capacity of hydrological unit k is defined as S. k (t), with area A k The confluence coefficient is K k Then we have:

[0083]

[0084] Among them, Q k (t) represents the lateral inflow discharge of the hydrological unit into the river segment at time t; the confluence coefficient K k The preferred length L of the main bus path of this unit k Average surface slope S 0,k , surface roughness n kand characteristic water depth h k The correlation can be approximated as:

[0085]

[0086] Among them, v k To represent the surface velocity under average runoff conditions in this hydrological unit, S 0,k For the average slope, n k The Manning coefficient, which reflects the roughness of the earth's surface, is h. k Characteristic water depth.

[0087] Based on the above hydrological calculations, the time series Q of the lateral inflow discharge of each secondary catchment unit to the corresponding river section flowing into the sea during the entire precipitation process can be obtained. k (t).

[0088] After obtaining the lateral inflow flow, nitrogen and phosphorus output coefficients are set for each hydrological unit, and the time-series relationship of nitrogen and phosphorus mass flux and density is constructed by combining the nitrogen and phosphorus pollutant density field within the unit.

[0089] For a secondary catchment area hydrological unit k, its internal underlying surface is first divided according to land use type. Different types such as farmland, construction land, forest land, wetland, and water body are denoted as index r. Based on historical experience or relevant research, a basic load factor e for total nitrogen per unit flow is pre-assigned for each land use type. N,r 0 and the total phosphorus base load factor per unit flow e P,r 0 Let ω be the area proportion of land use type r in this hydrological unit. k,r The basic non-point source load factor obtained from land use composition is:

[0090]

[0091] Where e N,k land and e P,k land Let represent the basic non-point source load coefficients of total nitrogen and total phosphorus per unit flow, determined by land use, for hydrological unit k, respectively. Next, considering the point source contribution of coastal sewage outlets within this hydrological unit, let's assume there are several sewage outlets within unit k, indexed as m, with a design or representative discharge flow rate of Q. p,m The total nitrogen and total phosphorus emission densities are C p,m N C p,m P Then, the emissions of total nitrogen and total phosphorus from point sources during a representative period can be converted into equivalent unit flow load factors:

[0092]

[0093] in e represents the representative lateral inflow discharge of hydrological unit k, such as the multi-year average or the average lateral discharge of a typical event. N,k out and e P,k out These represent the total nitrogen and total phosphorus load coefficients per unit flow after point source emissions are converted. Furthermore, to reflect the spatial differences in current nitrogen and phosphorus pollution levels across different hydrological units, the area-weighted average density C of total nitrogen and total phosphorus within hydrological unit k is calculated based on the nitrogen and phosphorus pollutant density fields obtained from satellite remote sensing in Step 1. N,k sat C P,k sat and compared with the reference background density C of the study area N ref C P ref The ratio is used to construct the correction factor:

[0094]

[0095] Finally, the land use non-point source contribution and the sewage outlet point source contribution are summed and adjusted using satellite correction factors to obtain the comprehensive unit flow total nitrogen load coefficient and total phosphorus load coefficient for hydrological unit k:

[0096]

[0097] Where e N,k e represents the total nitrogen load factor per unit flow of hydrological unit k. P,k This represents the total phosphorus load factor per unit flow of hydrological unit k. At this point, the time series of total nitrogen and total phosphorus mass fluxes can be written as follows:

[0098]

[0099] It can be seen that M N,k (t), M P,k (t) explicitly depends on land use ratio ω k,r Discharge parameters Q of sewage outlet p,m C p,m N C p,m P And the density C obtained by satellite inversion N,k sat C P,k sat Then by

[0100]

[0101] The time series of total nitrogen and total phosphorus densities in the side inflow of hydrological unit k at time t are obtained:

[0102]

[0103] Among them, M N,k (t) and M P,k (t) represent the total nitrogen and total phosphorus mass fluxes flowing into the river channel through this hydrological unit at time t, respectively. From this, the time series of total nitrogen and total phosphorus pollutant densities from the side inflow can be obtained:

[0104]

[0105] Among them, C k N (t) represents the total nitrogen density of the lateral inflow into the river segment at time t for hydrological unit k, C k P (t) represents the corresponding total phosphorus density; when Q k When (t) is below a preset threshold or close to zero, C can be adjusted. k N (t) and C k P (t) is processed by smoothing or preserving background values ​​to avoid numerical instability.

[0106] Finally, all secondary catchment hydrological units under the same river segment j flowing into the sea are aggregated, and the lateral inflow discharge Q of each unit is calculated. k (t) The time series of lateral inflow discharge at the river segment scale is obtained by summing over time. j (t), and the time series C of total nitrogen and total phosphorus pollutant density in the lateral inflow at the river reach scale was calculated by weighting by flow rate. j N (t) and C j P (t) serves as the source term input for subsequent one-dimensional river nitrogen and phosphorus pollution diffusion simulations.

[0107] like Figure 4 As shown, Step 3, a one-dimensional simulation of nitrogen and phosphorus pollution diffusion in a river, is performed. Specifically:

[0108] Step 3.1 Construction of a one-dimensional water quality model based on SWMM includes:

[0109] Based on the river segments defined in Step 1, a one-dimensional water quality model of the river flowing into the sea is constructed. Let the coordinate of the river centerline along the flow direction be x, with x=0 upstream and x=L downstream, where L is the river segment distance and time is t. For any river segment's position x and time t, denote Q. ch (x,t) represents the river flow rate, A(x,t) represents the cross-sectional area of ​​the river, and C... s(x,t) represents the volume density of pollutant component s in the water body, where s=N represents total nitrogen and s=P represents total phosphorus.

[0110] Using the one-dimensional unsteady flow equations within the SWMM dynamic wave solution framework, and incorporating parameters such as channel roughness and riverbed elevation, Q is obtained. ch The spatiotemporal distribution of (x,t) and A(x,t).

[0111] One-dimensional diffusion reaction equations were established for nitrogen and phosphorus respectively, and the side inflow rate obtained in Step 2 and its nitrogen and phosphorus pollutant density were introduced as mass source terms. Let E s (x) is the one-dimensional equivalent dispersion coefficient of component s in the river channel, R s (C s (x,t) represents the unit volume reaction term of component s. For example, considering the combined effects of first-order decay and sedimentation, the mass conservation equation along the river channel can be written as:

[0112]

[0113] Among them, S s lat (x,t) represents the lateral nitrogen and phosphorus mass source terms from the watershed hydrological unit, indicating the mass flux of component s flowing into the waterway cross section per unit length and per unit time.

[0114] For any river segment j flowing into the sea, let the set of hydrological units of the secondary catchment area connected to it be denoted as K(j). Then the total lateral inflow discharge Q of this river segment at time t is... j (t) Flow rate Q through each hydrological unit k The superposition of (t) yields:

[0115]

[0116] Let s∈{N,P} represent total nitrogen and total phosphorus, respectively. Let the mass flux of unit k at time t be... Then the total lateral mass flux of river segment j is

[0117]

[0118] This defines the temporal sequence of lateral inflow density at the river reach scale.

[0119]

[0120] Where C j s (t) represents the density of total nitrogen or total phosphorus flowing alongside river segment j at time t, C s bg Let L be the background density of this component. Let the length of river segment j be L. jThen, the total lateral flow and mass flux can be evenly distributed along the length of the river segment to obtain the lateral flow source term and mass source term per unit river length:

[0121]

[0122] In a continuous one-dimensional equation, for any position x located within river segment j, i.e. The corresponding interval, the side mass source term S s lat (x,t) is taken as

[0123]

[0124] This represents the total nitrogen or total phosphorus mass flowing into the cross-section per unit river length and per unit time. When discretizing the equation using finite volume, the river segment j is divided into several spatial units, and the length of the i-th unit is denoted as Δx. j,i Then the side mass input received by the unit at time t is

[0125]

[0126] In the discrete mass conservation equation

[0127]

[0128] China and Israel M s,j,i lat The (t) form is explicitly represented, thereby realizing the coupling of the side inflow of each hydrological unit obtained from Step 2 with the one-dimensional river nitrogen and phosphorus transport-diffusion process.

[0129] Based on observations and calculations of inflows from various lateral sides, inflow flow and density boundary conditions are given upstream of each tributary or control section, while free outflow boundary conditions are adopted downstream of the estuary, forming a one-dimensional nitrogen and phosphorus pollutant transport and diffusion model based on SWMM.

[0130] Step 3.2 Model Solving and Result Output

[0131] The above one-dimensional nitrogen and phosphorus pollutant transport and diffusion model is solved by runaway discretization. Each river segment is divided into several computational units along the river channel coordinate x, and the volume of the i-th computational unit of the j-th river segment is denoted as . The corresponding average densities of total nitrogen and total phosphorus are C N,j,i (t), C P,j,i (t). A stepwise approach with a time step size of Δt is used to discretize the above partial differential equation in the sense of finite volume, and the components are... The law of conservation of mass can be written as:

[0132]

[0133] Among them, F s,j,i in (t) and F s,j,i out (t) represent the convective-dispersive mass flux flowing into and out of the computing unit from the upstream side at time t, respectively, which can be provided by the hydrodynamic module as flow rate Q. ch (x,t) and discrete diffusion flux are calculated; M s,j,i lat (t) represents the lateral mass input from the watershed hydrological unit to the corresponding spatial unit, and its value is determined by the lateral inflow Q of the river segment corresponding to that spatial unit. j (t) and its nitrogen and phosphorus densities C j N (t), C j P (t) is obtained by allocating according to the time step and spatial length. The total nitrogen and total phosphorus densities C of each computational unit at each time step can be obtained by iteratively solving the above ordinary differential equation using an explicit or semi-implicit time scheme. N,j,i (t) and C P,j,i The spatiotemporal distribution of (t).

[0134] At the river estuary, the coordinates of the estuary section are recorded as follows: The hydrodynamic module provides the outbound flow rate for... The water quality module provides the total nitrogen and total phosphorus densities as follows: Then the total nitrogen and total phosphorus fluxes into the sea at time t can be further calculated:

[0135]

[0136] Where, Φ N (t) and Φ P (t) represent the total nitrogen and total phosphorus mass fluxes entering the bay through the estuary per unit time, respectively. This is achieved by adjusting Q over the entire simulation period. out (t), Q out N (t), Q out P (t) and its corresponding Φ N (t), Φ P By performing continuous calculations (t), we can obtain the inflow sequence of rivers flowing into the sea and the flux sequence of nitrogen and phosphorus pollutants into the sea on a time scale, providing boundary conditions for subsequent two-dimensional simulation of nitrogen and phosphorus diffusion in the bay.

[0137] like Figure 5 As shown, next is Step 4. Two-dimensional simulation of nitrogen and phosphorus pollution diffusion on the sea surface, specifically:

[0138] Step 4.1 Construction of a two-dimensional water pollution diffusion model:

[0139] A two-dimensional sea surface nitrogen and phosphorus pollution diffusion model is developed by using the outlet section of rivers flowing into the sea as the boundary input for nearshore waters.

[0140] Let the horizontal region of the research sea area be . The horizontal coordinates of any point are (x, y), and the time variable is t. The water depth field H(x, y) of the study area is obtained from nautical charts and bathymetry data, and the bottom Manning coefficient nb(x, y) is determined by the seabed sediment type.

[0141] From the two-dimensional shallow water equations, we can calculate the horizontal velocity components u(x,y,t), v(x,y,t), and the water surface elevation ζ(x,y,t). The bottom friction term in the momentum equation of the shallow water equations can be written as:

[0142]

[0143] in, The horizontal velocity vector. Let H(x,y), u(x,y,t0), and v(x,y,t0) be the modulus, ρ be the water density, and g be the gravitational acceleration. By inputting the initial values ​​H(x,y), u(x,y,t0), and v(x,y,t0), the time-varying velocity field (u,v) over the entire region can be calculated.

[0144] Regarding water quality, for each type of pollutant component Let C represent nitrogen and phosphorus respectively. s (x, y, t) is the average volume density of the water at position (x, y) at time t. Let K... x,s (x,y,t) and K y,s (x, y, t) are the equivalent horizontal dispersion coefficients of component s in the x and y directions, respectively, and R s (C s (x,y,t) represents the reaction term per unit volume, which may include combined effects such as biological consumption, sedimentation, and first-order decay. Therefore, the water depth-average mass conservation equation can be written as:

[0145] middle HC, the amount of pollutants stored per unit volume s The rate of change over time; , For convective transport fluxes along the x and y directions; , Let be the horizontal diffusion flux, where:

[0146]

[0147] β x,s ,β y,sTo match the bottom roughness n b (x,y) and dimensionless coefficients.

[0148] HR s (C s The term represents the internal reaction of the water body and is expressed in a first-order decay form. , where λ s S is the attenuation coefficient of component s; s riv (x,y,t) represents the external mass source term caused by the river flowing into the sea, which is used to couple the one-dimensional estuary outflow in Step 3.

[0149] The estuary input is determined by the output of Step 3. Let Q be the outflow from the estuary section. out (t), the densities of total nitrogen and total phosphorus are respectively:

[0150]

[0151] The corresponding mass flux into the sea is:

[0152]

[0153] Where L out Φ represents the arc length coordinates of the river mouth section. s (t) represents the total mass flux of component s that enters the sea through the estuary per unit time.

[0154] To express the estuary flux in a two-dimensional equation, a small region near the estuary is selected as the estuarine mixing zone. Its area is denoted as

[0155]

[0156] Define indicator functions

[0157]

[0158] The external mass source term of the river flowing into the sea can be written as:

[0159]

[0160] That is, the area source term, which is uniformly distributed per unit area within the estuarine mixing zone, is zero in other sea areas. Thus, at any time step [t, t+Δt], the integral has...

[0161]

[0162] This ensures that the quality flux of rivers flowing into the sea and the two-dimensional source term are strictly conserved.

[0163] Open boundary ∂Ω outside the sea area openAbove, using background density boundary conditions or radial boundary conditions, let:

[0164]

[0165] Where C s bg This represents the background density in the open ocean.

[0166] Step 4.2 Model Solving and Result Output

[0167] The above equations were numerically discretized and solved to obtain the spatiotemporal evolution sequences of total nitrogen and total phosphorus density on the sea surface.

[0168] Divide the region Ω into N on the plane. cell There are 1 calculation unit, and let the horizontal area of ​​the p-th unit be ΔA. p The water depth is H p The average density of component s in this unit is C. s,p (t), the average flow velocity of the hydrodynamic field in this unit is u. p (t),v p (t), with a dispersion coefficient of K x,s,p (t),K y,s,p (t).

[0169] Within a finite volume framework, a mass conservation relationship is established for each element:

[0170]

[0171] Where ∂p is the set of all edges in cell p; F s,p,e adv (t) represents the convective mass flux through edge e, calculated from the normal velocity at the edge and the upstream and downstream densities using an upwind scheme; F s,p,e diff (t) represents the diffuse mass flux through edge e, calculated based on the density gradient of adjacent cells and the equivalent dispersion coefficient; S s,p riv (t) represents the river surface source term corresponding to element p, when the element center is located in the estuary mixing zone Ω. m At that time, take

[0172]

[0173] otherwise .

[0174] In the time direction, using discrete moments Let represent the nth time step, where the density field is updated using an explicit, semi-implicit, or fully implicit time integration scheme. For example, in a first-order explicit scheme, we have:

[0175]

[0176] in F s,p,e adv,n F s,p,e diff,n S s,p riv,n At time t n The calculated convection flux, diffuse flux, and river source term are then used. When using a semi-implicit or fully implicit scheme, the right-hand side terms can be moved to t. n+1 At time, C is obtained by solving the matrix. s,p n+1 To improve numerical stability, this invention does not limit the specific solver.

[0177] By advancing the timeline as described above, the density C of all units across the entire sea area is updated sequentially throughout the simulation period. N,p n C P,p n This allows us to obtain the spatiotemporal evolution sequence of nitrogen and phosphorus on a two-dimensional sea surface.

[0178] like Figure 6 As shown, proceed to Step 5. Parameter Calibration and Dynamic Update, specifically:

[0179] Based on the two-dimensional sea surface nitrogen and phosphorus pollutant simulation results obtained in Step 4, a background field is constructed. Then, combined with measured water quality and satellite inversion data, data assimilation is performed through Kalman filtering to obtain a spatiotemporally continuous "true value" grid field. Based on this true value field, the key parameters in Steps 2-4 are calibrated and updated to form a closed-loop mechanism of observation-assimilation-calibration-re-forecasting.

[0180] All parameters are unified and aggregated into a parameter vector θ. Given a set of parameters θ, a new model configuration is formed through parameter calibration. After re-running Steps 2-4, an updated spatiotemporal sequence of nitrogen and phosphorus pollution density is obtained. By continuously executing the cycle of model forward modeling → Kalman filtering → truth value construction → parameter calibration, the model parameters are gradually adjusted in multiple rounds of time series iterations, continuously performing closed-loop correction and optimization to capture the dynamic impact of environmental changes on water pollution diffusion in land, rivers, and oceans.

[0181] In summary, this application utilizes a side-flow nitrogen and phosphorus pollution density inversion technique. Targeting the micro-topographic features such as slopes, depressions, and ditches surrounding rivers flowing into the sea, it employs high-resolution DEMs for grooving and correction. Multi-level catchment areas are constructed using D8 confluence analysis and Thiessen polygons along the coast. Combined with nitrogen and phosphorus pollution monitoring data, the temporal series of nitrogen and phosphorus pollution densities in the side-flow of each river segment are inverted. Through a nitrogen and phosphorus convection-dispersion simulation technique coupled with the catchment area, one-dimensional river, and two-dimensional bay, a one-dimensional nitrogen and phosphorus pollution diffusion model of the river channel flowing into the sea is constructed, using the catchment area's side-flow nitrogen and phosphorus pollution time series as the boundary. The time series of nitrogen and phosphorus pollution density at the estuary is output. Then, using this as the boundary, a two-dimensional nitrogen and phosphorus pollution density diffusion model is constructed and solved. Ultimately, a continuous mass conservation simulation of the two-dimensional catchment area, one-dimensional river channel, and two-dimensional sea surface is achieved, yielding the spatiotemporal distribution sequences of nitrogen and phosphorus in the river channel and sea surface. By using satellite and limited monitoring data-driven parameter calibration and dynamic update technology, the model parameters are jointly calibrated and rolled up using satellite monitoring and river cross-section monitoring data. This enables accurate simulation under sparse observation conditions and allows for automatic correction as sewage discharge structure, geographical environment, and hydrodynamic conditions change.

[0182] Furthermore, a two-dimensional catchment area-one-dimensional lateral inflow mapping method was constructed to characterize micro-topographic features. By constructing micro-topographically corrected DEMs, multi-level catchment areas along the river, and Thiessen polygons, the two-dimensional nitrogen and phosphorus non-point source loads in the land area were inverted into lateral inflow and nitrogen and phosphorus mass flux sequences at the river scale through time-area response and load coefficients. This formed a two-dimensional-one-dimensional pollution load mapping framework suitable for scenarios with complex micro-topography and sparse observation, improving the problems of neglecting the impact of micro-topography and coarsely processing lateral source terms in traditional methods.

[0183] An integrated simulation method for nitrogen and phosphorus pollutant diffusion in rivers and bays flowing into the sea under sparse observation conditions is proposed. This method adopts a full-process technical route, which includes micro-topographic multi-level catchment area division and lateral inflow inversion, one-dimensional river convection-diffusion simulation, two-dimensional bay convection-diffusion simulation, and satellite / monitoring-driven dynamic calibration. Under the condition of relying only on satellite data and a small number of cross-sectional monitoring, it outputs the spatiotemporal distribution of nitrogen and phosphorus density and inflow flux in rivers and bays flowing into the sea. This provides an efficient, detailed and dynamically updatable integrated technical solution for the calculation of nitrogen and phosphorus loads into the sea and emission reduction management.

[0184] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0185] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterized by: include: S1. Data Acquisition and Preprocessing: Geographic information data and observation data covering rivers flowing into the sea and nearshore sea surface areas are acquired and preprocessed to divide the water catchment areas into multiple levels. The observation data includes water quality remote sensing and cross-sectional monitoring data reflecting the spatial distribution characteristics of nitrogen and phosphorus. S2. Time series calculation of nitrogen and phosphorus pollution density in lateral inflow: Based on multi-level catchment areas and observation data, the time series of lateral inflow flow and nitrogen and phosphorus pollution density of each river segment flowing into the sea are calculated. S3. Construct a one-dimensional river nitrogen and phosphorus pollution diffusion model: Establish a one-dimensional river network and cross-sectional parameters for rivers flowing into the sea, and use the flow time series and nitrogen and phosphorus pollution density time series of each river segment flowing into the sea as boundary conditions to establish and solve the one-dimensional river nitrogen and phosphorus pollution diffusion equation, and obtain the flow time series and nitrogen and phosphorus pollution density time series of the rivers flowing into the sea. S4. Construct a two-dimensional sea surface nitrogen and phosphorus pollution diffusion model: Construct a two-dimensional computational grid for the sea area of ​​the study area, and use the time series of the inflow of rivers into the sea and the time series of nitrogen and phosphorus pollution density into the sea as boundary conditions to establish and solve the two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation to obtain the time series of nitrogen and phosphorus pollution density on the sea surface. S5. Key Parameter Update: Based on geographic information data and observation data, the time series of nitrogen and phosphorus pollution density entering the sea and the time series of nitrogen and phosphorus pollution density on the sea surface are compared with the observation field to calibrate and dynamically update the key parameters of the model.

2. The method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, as described in claim 1, is characterized by: Geographic information data includes surface elevation, land use type, nautical chart data, river centerline vector data, riverbank boundary vector data, and coastline vector data, as well as vector data of drainage pipes, roads, and ditches along the riverbanks, and the location and discharge data of sewage outlets along the river. Observational data is obtained through satellite observation and station observation. The apparent reflectance of water bodies is converted into a nitrogen and phosphorus pollutant density field through satellite observation data inversion. Monitoring sequences of nitrogen and phosphorus pollutant density and corresponding flow, water level, and precipitation are obtained through hydrological station observation and cross-sectional observation of the river, and then integrated to form a nitrogen and phosphorus pollutant density observation field.

3. The method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, as described in claim 1 or 2, is characterized by: Preprocessing includes: Combining the DEM, the river channel is divided into several calculated river sections and control sections along the main trunk of the river flowing into the sea. With the control sections as the outlet, the primary catchment areas corresponding to each control section are divided based on the flow analysis. The points where the river centerline, the boundary lines on both banks, the outlet of the drainage network along the river, the end of the road, and the river embankment are marked as generation points. Based on the generation points, Thiessen polygons are constructed and spatially superimposed and clipped with the primary catchment areas of the corresponding river sections. The intersections are used to obtain the secondary catchment areas distributed along the riverbank.

4. The method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, as described in claim 3, is characterized by: The calculations of the time series of S2 side inflow and nitrogen and phosphorus pollution density include: Precipitation intensity time series P within hydrological unit k of the secondary catchment area k (t), combined with the land use type and soil properties of hydrological unit k, to calculate the effective rainfall of the original precipitation; The Horton model was used to describe the decay process of soil infiltration capacity over time for hydrological unit k, resulting in the time series Q of lateral inflow discharge for the corresponding river section flowing into the sea during precipitation for each secondary catchment area. k (t); Nitrogen and phosphorus output coefficients are set for hydrological unit k, and combined with the nitrogen and phosphorus pollutant density field within hydrological unit k, the time series relationship of nitrogen and phosphorus mass flux and density is constructed to obtain the pollution density time series C of the side inflow total nitrogen and total phosphorus. j N (t), C j P (t).

5. The method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, as described in claim 4, is characterized by: The time-series relationships for nitrogen and phosphorus mass fluxes and densities were constructed as follows: The underlying surface within hydrological unit k is divided into different types based on land use, and a pre-defined total nitrogen load factor e per unit flow is given for each type. N,r 0 and the total phosphorus base load factor per unit flow e P,r 0 Based on the area occupied by each land use type in hydrological unit k, calculate the basic non-point source load factor e of total nitrogen and total phosphorus per unit flow, determined by the land use type, for hydrological unit k. N,k land and e P,k land ; Calculate the total nitrogen and total phosphorus load factors per unit flow rate within hydrological unit k based on the number of sewage outlets within hydrological unit k. N,k out and e P,k out ; Based on the nitrogen and phosphorus pollutant density fields obtained from water quality remote sensing inversion, the area-weighted average density C of total nitrogen and total phosphorus within hydrological unit k is calculated. N,k sat C P,k sat A correction factor was constructed by comparing the ratio of the density to the reference background density of the study area. The land use non-point source contribution and the sewage outlet point source contribution were summed and adjusted using a satellite correction factor to obtain the comprehensive unit flow total nitrogen load coefficient e for hydrological unit k. N , k and total phosphorus loading factor e P , k The nitrogen and phosphorus mass flux M was calculated. N,k (t), M P,k (t), construct the time series relationship between nitrogen and phosphorus mass flux and density.

6. The method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, as described in claim 3, is characterized by: The one-dimensional river nitrogen and phosphorus pollution diffusion equations established and solved in S3 include: A one-dimensional water quality model of the river flowing into the sea is constructed based on the divided river segments. The one-dimensional unsteady flow equation under the SWMM dynamic wave solution framework is adopted. Combined with the river hydrological parameters, the river flow rate Q at any position x and time t of any river segment is obtained. ch The spatiotemporal distribution of (x,t) and the cross-sectional area A(x,t); The time series of the side inflow flow Q k (t), pollution density time series of side-flow total nitrogen and total phosphorus C j N (t), C j P (t) and precipitation data were introduced as mass source terms to construct a one-dimensional river nitrogen and phosphorus pollution diffusion equation, and the inflow time series and inflow nitrogen and phosphorus pollution density time series of rivers flowing into the sea were calculated.

7. The method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, as described in claim 1, is characterized by: The process of establishing and solving a two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation to obtain the sea surface nitrogen and phosphorus pollution density time series includes: using the inflow time series of rivers flowing into the sea and the inflow nitrogen and phosphorus pollution density time series as boundary conditions, combined with dynamic parameters, constructing a two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation through a two-dimensional shallow water equation and a mass conservation equation, and using the finite volume method to calculate the sea surface nitrogen and phosphorus pollution density time series.

8. The method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, as described in claim 7, is characterized by: The establishment of two-dimensional sea surface nitrogen and phosphorus pollution diffusion equations includes: The bottom Manning coefficient of the sea area is obtained, and the time-varying velocity field (u,v) of the sea area is calculated by the two-dimensional shallow water equations. Based on the depth-average volumetric density of nitrogen and phosphorus pollutant components at time t and location (x,y), the equivalent horizontal dispersion coefficients of pollutant components in the x and y directions, and the reaction term per unit volume, a depth-average mass conservation equation is designed to construct a two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation.

9. The method for simulating nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, as described in claim 8, is characterized in that: Solving the two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation yields the time series of sea surface nitrogen and phosphorus pollution density, including: The sea area is divided into several computational units on a plane, and a mass conservation relationship is established for each unit within a finite volume framework. The density field is updated using explicit, semi-implicit, or fully implicit time integration schemes. The density of all units in the entire sea area is updated sequentially throughout the simulation period to obtain the time series of nitrogen and phosphorus pollution density on the sea surface.

10. A simulation system for nitrogen and phosphorus diffusion in rivers flowing into the sea, characterizing micro-topographic features, employing the simulation method for nitrogen and phosphorus diffusion in rivers flowing into the sea as described in any one of claims 1-9, characterized in that, include: Data acquisition and preprocessing module: used to acquire geographic information data and observation data covering rivers flowing into the sea and nearshore sea surface areas, and to preprocess the data to divide the water catchment areas into multiple levels. The observation data includes water quality remote sensing and cross-sectional monitoring data reflecting the spatial distribution characteristics of nitrogen and phosphorus. Side-flow nitrogen and phosphorus pollution density time series calculation module: used to calculate the side-flow flow time series and nitrogen and phosphorus pollution density time series for each river segment flowing into the sea based on multi-level catchment areas and observation data; One-dimensional river nitrogen and phosphorus pollution diffusion model: By establishing a one-dimensional river network and cross-sectional parameters of rivers flowing into the sea, the flow time series and nitrogen and phosphorus pollution density time series of each river segment flowing into the sea are used as boundary conditions to establish and solve the one-dimensional river nitrogen and phosphorus pollution diffusion equation, and obtain the inflow time series and inflow nitrogen and phosphorus pollution density time series of the rivers flowing into the sea. Two-dimensional sea surface nitrogen and phosphorus pollution diffusion model: By constructing a two-dimensional computational grid for the sea area of ​​the study area, and using the time series of inflow of rivers into the sea and the time series of nitrogen and phosphorus pollution density into the sea as boundary conditions, a two-dimensional sea surface nitrogen and phosphorus pollution diffusion equation is established and solved to obtain the time series of nitrogen and phosphorus pollution density on the sea surface. Key parameter update module: Based on geographic information data and observation data, it compares the time series of nitrogen and phosphorus pollution density entering the sea and the time series of nitrogen and phosphorus pollution density on the sea surface with the observation field, calibrates and dynamically updates the key parameters of the model.

Images