Three-dimensional red tide drift risk early warning method and system based on set disturbance

Abstract

The invention discloses a three-dimensional red tide drift risk early warning method and system based on set disturbance, and relates to the technical field of physical oceanography, and the system comprises an ocean data summarization module, a red tide outbreak evaluation module, a red tide underwater three-dimensional drift module, and a red tide water surface multi-parameter disturbance drift module. A red tide risk early warning module and a red tide visual product manufacturing module. Performing dynamic evaluation by combining a biological model structure set to obtain a red tide outbreak early warning result, performing calculation by adopting a three-dimensional floating dynamic equation to obtain an upward floating speed of the red tide, performing calculation based on multi-parameter disturbance to obtain a red tide drift path, and performing spatial overlay analysis based on a red tide potential coverage area to obtain a red tide hazard risk grade result. And a visual product is generated. According to the method, the whole-process and three-dimensional early warning of the red tide from underwater growth and floating to water surface drifting is realized, and the accuracy and decision support value of red tide risk forecasting are remarkably improved through wave-tide-flow-ecology coupling analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of physical oceanography, and particularly to a three-dimensional red tide drift risk early warning method and system based on ensemble perturbation. BACKGROUND

[0002] Under the background of global climate change, extreme weather events occur frequently, and the risk of marine disasters in coastal areas is increasing. The abnormal changes in seawater temperature and salinity and other factors interact to cause the frequent occurrence of nearshore red tides, which poses a serious challenge to the sustainable development of coastal economy, ecological environment and aquaculture. In the prior art, numerical simulation of the drift transport of red tides in the sea is mainly based on the Lagrangian method to simulate and predict the drift trajectory of red tides in the horizontal direction.

[0003] In the prior art, the prediction of red tide drift trajectory mainly relies on the numerical simulation method based on Lagrangian particle tracking. This method simplifies the red tide biological population as passive tracer particles, and simulates their drift path in the horizontal direction under the driving of the given sea surface flow field, which has become the mainstream means of current operational prediction. However, such a scheme has inherent limitations: its module generally regards red tides as inert particles, completely ignoring the key influence of marine ecological environment on the growth, death and vertical migration of algal species, and also failing to take into account the horizontal displacement of many red tide algae caused by different layer flow during their upward process from a certain depth under water, resulting in insufficient biological accuracy and inherent trajectory deviation of the prediction; in addition, the existing module is generally insufficient in spatial resolution, which makes it difficult to effectively capture the essential role of small-scale physical processes such as turbulence and wave breaking in red tide diffusion and aggregation, resulting in over-smoothed simulation trajectories that cannot reflect the true diffusion pattern; more importantly, the existing scheme uses fixed empirical parameters for deterministic simulation, without fully considering the uncertainty and environmental sensitivity of the module parameters themselves. A single simulation can only give one possible deterministic trajectory, and when there are errors in the input flow field or deviations in the parameter setting, the prediction result may deviate significantly from the actual situation, lacking the probabilistic guidance information necessary for risk warning, thus significantly restricting the reliability and decision support value in business application. Therefore, it is a high application value research direction to predict the drift risk of red tides through wave-tide-flow-ecological coupling analysis. SUMMARY

[0004] The technical problem solved by the present application is that the existing red tide drift prediction technology regards red tides as passive tracers, without fully considering the wave-tide-flow-ecological coupling analysis method, and lacks comprehensive consideration of ecological environment such as temperature, salinity, nutrient salt and irradiance. It lacks the coupling description of the growth-death-germination dynamics of the whole red tide, resulting in large errors in dynamic evaluation of red tide concentration and outbreak risk.

[0005] The prior art is deficient in depicting the continuous process of the red tide in the underwater growth stage, floating upward from the self-development depth and experiencing three-dimensional transport to the surface layer, and cannot accurately estimate the displacement and arrival time caused by the upward floating, vertical diffusion and vertical shear;

[0006] In addition, the small-scale and sub-mesoscale processes such as small-scale turbulence, wave breaking, wind drag and Coriolis deflection are not fully considered, resulting in deviation of the drift path and diffusion range; the parameter setting is single and lacks uncertainty representation, which cannot suppress the systematic drift caused by forced error and is insufficient in prediction dispersion; at the same time, the association with sensitive target areas is weak, and it is difficult to form an operable graded risk warning based on the affected area ratio and the duration, thereby affecting the decision support for aquaculture and ecological protection.

[0007] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, a three-dimensional red tide drift risk warning system based on ensemble disturbance includes: a marine data collection module, a red tide outbreak assessment module, a red tide underwater three-dimensional drift module, a red tide surface multi-parameter disturbance drift module, a red tide risk warning module, and a red tide visualization product production module.

[0008] The marine data collection module is used to receive and collect a multi-source marine statistical data set, and the multi-source marine statistical data set is classified and stored by date.

[0009] The red tide outbreak assessment module is used to combine a biological model structure set for dynamic assessment to obtain a red tide concentration, and to make a red tide outbreak warning judgment based on the red tide concentration to obtain a red tide outbreak warning result.

[0010] The red tide underwater three-dimensional drift module is used to calculate the upward floating speed of the red tide through a three-dimensional floating dynamic equation.

[0011] The red tide surface multi-parameter disturbance drift module is used to calculate the drift path of the red tide.

[0012] The red tide risk warning module is used to perform spatial overlay analysis on the potential coverage area of the red tide and a preset sensitive target area to obtain a risk index, and to make a red tide hazard risk grade judgment based on the risk index to obtain a red tide hazard risk grade result.

[0013] The red tide visualization product production module is used to generate a visualization product.

[0014] As a preferred scheme of the three-dimensional red tide drift risk warning system based on ensemble disturbance, the marine data collection module is used to receive and collect a multi-source marine statistical data set, and the multi-source marine statistical data set is classified and stored by date.

[0015] The multi-source marine statistical data includes biological environmental parameter data, marine dynamics data, atmospheric forcing data, and geographical and risk assessment data;

[0016] The biological environmental parameter data includes nutrient concentration data, temperature data, salinity data, and irradiance data;

[0017] The ocean dynamics data includes ocean current velocity data, surface turbulence velocity data, and surface wave breaking velocity data;

[0018] The ocean current velocity data includes ocean current velocity component data in the x-direction, ocean current velocity component data in the y-direction, and ocean current velocity component data in the z-direction.

[0019] The surface turbulence velocity data includes turbulence velocity component data in the x-direction and turbulence velocity component data in the y-direction;

[0020] The surface wave breaking velocity data includes velocity component data in the x-direction of wave breaking and velocity component data in the y-direction of wave breaking;

[0021] The atmospheric forcing data includes wind speed data;

[0022] The wind speed data includes wind speed component data in the x-direction and wind speed component data in the y-direction.

[0023] The geographical and risk assessment data includes data on sensitive target areas;

[0024] The sensitive target area data includes marine aquaculture areas and ecological protection areas.

[0025] The marine aquaculture area includes the geographical coordinates of the aquaculture facilities, the species being cultured, and the scale of the aquaculture.

[0026] The ecological protection zone includes ecological boundaries and protected targets.

[0027] As a preferred embodiment of the three-dimensional red tide drift risk early warning system based on set perturbation described in this invention, the red tide outbreak assessment module is used to perform dynamic assessment by combining biological model structure set to obtain red tide concentration, and to make red tide outbreak early warning judgment based on red tide concentration to obtain red tide outbreak early warning result.

[0028] The processing logic for obtaining red tide concentration through dynamic assessment includes:

[0029] The optimal target model structure is selected from the set of biological model structures by using the maximum likelihood ratio test.

[0030] The biological model structure set includes basic line structure, death structure, nutrient structure, death and nutrient structure, and zero growth structure;

[0031] Based on the target model structure, the red tide concentration is calculated using the red tide growth and death equation.

[0032] The calculation formula for the red tide growth and death equation is as follows:

[0033] ;

[0034] in, C represents the rate of change of red tide concentration over time, where C represents the red tide concentration. Let F represent the growth term, m represent the death term, and F represent the growth term. g Indicates sediment germination term;

[0035] growth items The calculation formula is:

[0036] ;

[0037] Where G([N],T,S) represents the term affected by temperature, salinity and nutrients, G(E,T,S) represents the term affected by light, temperature and salinity, [N] represents nutrient concentration, T represents temperature, E represents irradiance and S represents salinity;

[0038] The formulas for calculating G([N],T,S) and G(E,T,S) are as follows:

[0039] ;

[0040] ;

[0041] Among them, K [N] [N] represents the half-saturation constant indicating nutritional limitation, [N] represents the nutrient concentration, and G represents the nutrient concentration. max (T,S) represents the maximum growth rate. This represents the basal metabolic rate, a g tanh represents the light absorption coefficient, and tanh represents the hyperbolic tangent function.

[0042] The calculation formula is:

[0043] ;

[0044] ;

[0045] ;

[0046] in, This indicates the effect of salinity. Indicates the effect of temperature. This represents the theoretical maximum growth rate at the optimal growth temperature and optimal growth salinity. Indicates the optimal growth temperature. The optimal salinity for growth is represented by b1, b2, b3, b4, b5, and F(T=5). The temperature effect term at 5°C is represented by a1, a2, a3, a4, and a4.

[0047] As a preferred embodiment of the three-dimensional red tide drift risk early warning system based on set perturbation described in this invention, the red tide outbreak early warning judgment is performed to obtain red tide outbreak early warning results, which include red tide outbreak low-level early warning, red tide outbreak medium-level early warning, red tide outbreak high-level early warning and red tide outbreak emergency early warning.

[0048] For each hour predicted in the future, the average concentration growth rate over the past 12 hours is calculated using the following formula:

[0049] R t =[(C t -C t-12 ) / C t-12 ×100%;

[0050] Among them, C t C represents the red tide concentration at the current moment. t-12 R represents the red tide concentration 12 hours ago, t represents the predicted concentration for each hour in the future. t This indicates the average concentration growth rate over the past 12 hours;

[0051] The current red tide concentration is compared with a preset concentration threshold, and the average concentration growth rate over the past 12 hours is also compared with a preset growth rate threshold. Specifically, this includes:

[0052] The preset concentration thresholds include potential hazard concentration thresholds and red tide confirmed outbreak concentration thresholds;

[0053] The preset growth rate thresholds include a potential outbreak growth rate threshold and a defined outbreak growth rate threshold;

[0054] When C t <C potential And R t <G potential At that time, a low-level warning for a red tide outbreak was issued;

[0055] When C t ≥C potential And R t ≥ Gpotential At that time, a medium-level red tide outbreak warning was issued;

[0056] When C potential ≤C t <C outbreak And R t ≥G outbreak At that time, a high-level warning for a red tide outbreak was issued;

[0057] When C t ≥C outbreak At that time, an emergency warning for a red tide outbreak was issued;

[0058] Among them, C potential C represents the potential hazard concentration threshold. outbreak This indicates the concentration threshold for a confirmed red tide outbreak, G. potential G represents the potential outbreak growth rate threshold. outbreak This indicates a specific threshold for the outbreak growth rate;

[0059] The system continuously monitors the dynamic changes in red tide concentration and the average concentration growth rate over the past 12 hours during the forecast period, and immediately triggers the corresponding warning when any level of warning conditions are met.

[0060] As a preferred embodiment of the three-dimensional red tide drift risk early warning system based on set perturbation described in this invention, the underwater three-dimensional red tide drift module is used to calculate the upward floating velocity of the red tide through the three-dimensional floating dynamic equation, and the processing logic includes:

[0061] The dynamic process of red tides floating to the sea surface as they grow underwater is calculated using the three-dimensional floating dynamic equations of red tides in seawater. The calculation expression is as follows:

[0062] ;

[0063] ;

[0064] in, W represents the speed of the ocean current. a K represents the upward floating velocity of the red tide, and K represents the diffusion coefficient. Z represents the direction and rate of the fastest change in concentration in all directions, Z represents the current depth, and Z1 represents the reference depth.

[0065] As a preferred embodiment of the three-dimensional red tide drift risk early warning system based on ensemble perturbation described in this invention, the red tide surface multi-parameter perturbation drift module is used to calculate the red tide drift path, and the processing logic includes:

[0066] The coordinates of red tide particles are calculated based on the Lagrange method and by adding a perturbation term.

[0067] The disturbance additions include ocean current drag coefficient P1, wind drag coefficient P2, and Coriolis effect motion deflection angle ϑ0.

[0068] The red tide drift path is obtained by statistically analyzing the coordinates of red tide particles. The formula for calculating the coordinates of red tide particles is as follows:

[0069] ;

[0070] Where, x i The x-coordinate and y-coordinate of red tide particle i in the planar coordinate system at time t represent the position of the red tide particle i at time t. i U represents the position of the red tide particle i in the planar coordinate system at time t, where u is the vertical coordinate. a v represents the component of the ocean current velocity in the x-direction. a u represents the component of the ocean current velocity in the y-direction. t v represents the turbulent velocity component in the x-direction. t w represents the turbulent velocity component in the y-direction. u w represents the velocity component in the x-direction of wave breaking. v u represents the velocity component in the y-direction of wave breaking. d V represents the wind speed component in the x-direction. d P1 represents the wind speed component in the y-direction, P2 represents the ocean current drag coefficient, and ϑ represents the wind drag coefficient. i ϑ0 represents the angle between the wind direction and the x-direction, and ϑ0 represents the deflection angle due to the Coriolis effect.

[0071] As a preferred embodiment of the three-dimensional red tide drift risk early warning system based on set perturbation described in this invention, the red tide risk early warning module is used to perform spatial overlay analysis on the potential red tide coverage area and the preset sensitive target area to obtain risk indicators, and to judge the red tide hazard risk level based on the risk indicators to obtain the red tide hazard risk level result.

[0072] The sensitive target area data includes marine aquaculture areas and ecological protection areas. The marine aquaculture areas include the geographical coordinates of aquaculture facilities, aquaculture species, and aquaculture scale.

[0073] The ecological protection zone includes ecological boundaries and protection targets;

[0074] The risk indicators include the percentage of sensitive target areas affected and the duration of the red tide impact;

[0075] The processing logic for spatial overlay analysis of potential red tide coverage areas and pre-defined sensitive target areas includes:

[0076] Based on the red tide drift path, the spatial range covered by the red tide mass within a future preset time period is extracted to form the potential red tide coverage area;

[0077] The spatial geometric intersection operation is performed between the potential red tide coverage area and the geographical range of the preset sensitive target area to calculate the percentage of the sensitive target area affected and the duration of the red tide impact.

[0078] The formula for calculating the percentage of the sensitive target area affected is as follows:

[0079] ;

[0080] Wherein, AIR represents the percentage of the sensitive target area affected, A I A represents the area of ​​intersection between the potential red tide coverage area and the sensitive target area. T This represents the total area of ​​the sensitive target region;

[0081] The duration of the red tide impact is the total number of hours during which the red tide mass continuously covers the same sensitive target area within the prediction period.

[0082] As a preferred embodiment of the three-dimensional red tide drift risk early warning system based on set perturbation described in this invention, the red tide hazard risk level is judged based on risk indicators to obtain the red tide hazard risk level result.

[0083] The red tide hazard risk level results include low risk, medium risk, high risk, and extremely high risk;

[0084] The affected area percentage (AIR) of the sensitive target area is compared with a preset area percentage threshold, and the red tide impact duration (DOI) is simultaneously compared with a preset red tide impact duration threshold. Specifically, this includes:

[0085] When AIR < 10% and DOI < 6 hours, it is considered low risk;

[0086] When AIR ≥ 10% and DOI ≥ 6 hours, or AIR < 10% and DOI ≥ 24 hours, or DOI ≥ 12 hours, it is judged as medium risk;

[0087] When AIR ≥ 30% and DOI ≥ 18 hours, or AIR ≥ 60% and DOI ≥ 12 hours, or AIR < 30% and DOI ≥ 48 hours, it is considered high risk;

[0088] When AIR ≥ 60% and DOI ≥ 36 hours, or AIR ≥ 80% 24 and DOI ≥ 24 hours, it is judged as extremely high risk;

[0089] DOI indicates the duration of red tide impact.

[0090] As a preferred embodiment of the three-dimensional red tide drift risk early warning system based on set perturbation described in this invention, the red tide visualization product creation module is used to generate visualization products, and its processing logic includes:

[0091] The output data from the marine data aggregation module, red tide outbreak assessment module, red tide underwater three-dimensional drift module, red tide surface multi-parameter disturbance drift module, and red tide risk early warning module are imported and integrated into the GIS platform.

[0092] Visualization products are generated using Python scripts, including red tide drift trajectory maps, underwater red tide 3D distribution maps, biochemical environmental parameter maps, risk assessment thematic maps, and time series animations.

[0093] Secondly, the three-dimensional red tide drift risk early warning method based on ensemble perturbation includes the following steps:

[0094] Step S1: Receive and aggregate multi-source oceanographic data sets;

[0095] Step S2: Combine the biological model structure set for dynamic evaluation to obtain the red tide concentration, and make a red tide outbreak warning judgment based on the red tide concentration to obtain the red tide outbreak warning result;

[0096] Step S3: Calculate the upward floating velocity of the red tide using the three-dimensional floating dynamic equation;

[0097] Step S4: Calculate the red tide drift path;

[0098] Step S5: Perform spatial overlay analysis on the potential red tide coverage area and the preset sensitive target area to obtain risk indicators. Based on the risk indicators, determine the red tide hazard risk level and obtain the red tide hazard risk level result.

[0099] Step S6: Generate a visual product.

[0100] The beneficial effects of this invention are as follows: This invention employs a wave-tide-current-ecology coupled analysis method to achieve dynamic prediction of underwater red tide biomass, overcoming the deficiency of existing Lagrange drift methods in predicting red tide biomass; it introduces a three-dimensional floating dynamic equation, solving the problem that traditional Lagrange drift methods cannot consider the underwater floating of red tides, enabling the system to accurately predict the three-dimensional movement trajectory of red tides from the underwater outbreak point to the water surface; it introduces small-scale turbulent processes and wave breaking effects, solving the problem that traditional Lagrange drift methods cannot consider the influence of marine environmental changes on phase drift at multiple spatial scales; and it introduces a random perturbation parameterization scheme, solving the problems of single parameter settings and failure to consider drift deviations caused by parameter uncertainty in traditional Lagrange drift methods. This enables the system to identify the risk of underwater red tide outbreaks earlier and more accurately, improves the prediction accuracy of when, where, and at what concentration the red tide will reach the water surface, significantly improves the prediction accuracy of the horizontal drift trajectory of the red tide, and provides more accurate initial conditions for the horizontal drift prediction of surface red tides. This enhances the spatiotemporal accuracy of the overall early warning and provides a more solid biological basis for subsequent drift prediction and risk warning. Attached Figure Description

[0101] Figure 1 A schematic diagram of the basic process of a three-dimensional red tide drift risk early warning system based on set perturbation provided in an embodiment of the present invention;

[0102] Figure 2 A probability diagram of red tide algal blooms provided in one embodiment of the present invention;

[0103] Figure 3 The red tide drift trajectory is provided in one embodiment of the present invention. Detailed Implementation

[0104] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0105] Example 1, referring to Figure 1 As an embodiment of the present invention, a three-dimensional red tide drift risk early warning system based on set perturbation is provided, including: a marine data aggregation module, a red tide outbreak assessment module, a red tide underwater three-dimensional drift module, a red tide surface multi-parameter perturbation drift module, a red tide risk early warning module, and a red tide visualization product production module;

[0106] The marine data aggregation module is used to receive multi-source marine data and calculate a set of marine environmental elements;

[0107] The red tide outbreak assessment module is used to dynamically assess the red tide concentration by combining a set of biological model structures, and then to make a red tide outbreak early warning judgment based on the red tide concentration, thereby obtaining the red tide outbreak early warning result;

[0108] The red tide underwater three-dimensional drift module is used to calculate the upward floating speed of the red tide through three-dimensional floating dynamic equations;

[0109] The red tide surface multi-parameter disturbance drift module is used to calculate the red tide drift path;

[0110] The red tide risk early warning module is used to perform spatial overlay analysis on the potential red tide coverage area and the preset sensitive target area to obtain risk indicators. Based on the risk indicators, the red tide hazard risk level is judged to obtain the red tide hazard risk level result.

[0111] The Red Tide Visualization Product Creation Module is used to generate visualization products.

[0112] This invention provides a highly applicable and reliable three-dimensional red tide drift risk early warning method and system based on ensemble perturbation. Specifically, this invention employs a wave-tide-current-ecology coupled analysis method, introducing the three-dimensional floating mechanism of red tides underwater. It fully considers the dynamic process of red tides growing underwater and floating to the sea surface, overcoming the limitation of traditional Lagrange drift methods that cannot consider the underwater floating of red tides. By introducing small-scale turbulent processes and wave breaking, it adds the displacement influence process caused by small-scale turbulent motion and wave breaking to the horizontal drift, solving the problem that traditional Lagrange drift methods cannot consider the impact of marine environmental changes on phase drift at multiple spatial scales. Through a random perturbation parameterization scheme, it directly perturbs important parameters using random physical methods to form ensemble members, while maintaining the internal consistency of the physical parameterization scheme. This enables accurate prediction of red tide drift risk.

[0113] The marine data aggregation module is used to receive and aggregate multi-source marine statistical data sets, classify and store the multi-source marine statistical data sets by date;

[0114] Multi-source ocean statistical data includes biological environmental parameter data, ocean dynamics data, atmospheric forcing data, and geographic and risk assessment data;

[0115] Biological environmental parameter data include nutrient concentration data, temperature data, salinity data, and irradiance data;

[0116] Ocean dynamics data includes ocean current velocity data, surface turbulence velocity data, and surface wave breaking velocity data;

[0117] Ocean current velocity data includes ocean current velocity components in the x-direction, ocean current velocity components in the y-direction, and ocean current velocity components in the z-direction.

[0118] Surface turbulent velocity data includes turbulent velocity components in the x-direction and turbulent velocity components in the y-direction;

[0119] Surface wave breaking velocity data includes velocity component data in the x-direction of wave breaking and velocity component data in the y-direction of wave breaking;

[0120] Atmospheric forcing data includes wind speed data;

[0121] Wind speed data includes wind speed components in the x-direction and wind speed components in the y-direction;

[0122] Geographic and risk assessment data includes data on sensitive target areas;

[0123] Data on sensitive target areas includes marine aquaculture areas and ecological protection areas.

[0124] Marine aquaculture areas include the geographical coordinates of aquaculture facilities, the types of aquaculture, and the scale of aquaculture.

[0125] Ecological protection zones include ecological boundaries and protected targets.

[0126] Among them, the marine data aggregation module receives and aggregates multi-source marine statistical data sets and stores them in categories by date. It realizes the unified integration and management of biological environmental parameters (nutrient concentration, temperature, salinity, irradiance), marine dynamic data (ocean currents, turbulence, and surface wave breaking velocities in various directions), atmospheric forcing data (including wind speed components), and geographical and risk assessment data (covering specific information on mariculture areas and ecological protection areas). This provides a comprehensive, multi-source, and structured data foundation for three-dimensional red drift risk early warning, effectively supporting the system's comprehensive analysis capabilities and early warning accuracy.

[0127] The red tide outbreak assessment module is used to dynamically assess the red tide concentration by combining a set of biological model structures, and then to make a red tide outbreak early warning judgment based on the red tide concentration, thereby obtaining the red tide outbreak early warning result;

[0128] The processing logic for obtaining red tide concentration through dynamic assessment includes:

[0129] The optimal target model structure is selected from the set of biological model structures by using the maximum likelihood ratio test.

[0130] The biological model structure set includes basic line structure, death structure, nutrient structure, death and nutrient structure, and zero growth structure;

[0131] Based on the target model structure, the red tide concentration is calculated using the red tide growth and death equation.

[0132] The formula for calculating the red tide growth and mortality equation is:

[0133] ;

[0134] in, C represents the rate of change of red tide concentration over time, where C represents the red tide concentration. Let F represent the growth term, m represent the death term, and F represent the growth term. g Indicates sediment germination term;

[0135] growth items The calculation formula is:

[0136] ;

[0137] Where G([N],T,S) represents the term affected by temperature, salinity and nutrients, G(E,T,S) represents the term affected by light, temperature and salinity, [N] represents nutrient concentration, T represents temperature, E represents irradiance and S represents salinity;

[0138] The formulas for calculating G([N],T,S) and G(E,T,S) are as follows:

[0139] ;

[0140] ;

[0141] Among them, K [N] [N] represents the half-saturation constant indicating nutritional limitation, [N] represents the nutrient concentration, and G represents the nutrient concentration. max (T,S) represents the maximum growth rate. This represents the basal metabolic rate, a g tanh represents the light absorption coefficient, and tanh represents the hyperbolic tangent function.

[0142] The calculation formula is:

[0143] ;

[0144] ;

[0145] ;

[0146] in, This indicates the effect of salinity. Indicates the effect of temperature. This represents the theoretical maximum growth rate at the optimal growth temperature and optimal growth salinity. Indicates the optimal growth temperature. The optimal salinity for growth is represented by b1, b2, b3, b4, b5, and F(T=5). The temperature effect term at 5°C is represented by a1, a2, a3, a4, and a4.

[0147] In specific embodiments, considerable uncertainty exists regarding biological aspects; therefore, five potential biological model structures were considered (Table 1). The first four structures are nested: a more complex model structure simplifies to a simpler one when certain parameter values ​​are set to zero. The first structure is the baseline case, where net growth is only a function of temperature, light, and salinity. The next two structures, based on the baseline model, add the possibilities of mortality and nutrient dependence, respectively. The fourth structure considers both nutrient and mortality rates. The fifth structure sets out a balance between growth and mortality, resulting in zero net growth at any time across the entire region. This highly precise structure does not strictly conform to the aforementioned nested set. It is used as a limiting case to understand the impact of net growth on the scale and spatiotemporal characteristics of red tides.

[0148] Table 1. Structures of five potential biological models

[0149] Structure Growth rate determinants Death Base line T, S, E No Death T, S, E Yes Nutrient salt T, S, E, N No Death and nutrient salt T, S, E, N Yes Zero growth NA NA

[0150] T = Temperature, S = Salinity, E = Irradiance, N = Nutrients (nitrates), NA = Not used.

[0151] Each of these biological model structures constitutes a different hypothesis about the factors controlling the initiation and development of red tide blooms. The specific steps are as follows:

[0152] For each hypothetical biological model structure, optimize the model-data fit within the range of parameter uncertainties most sensitive to the module;

[0153] The importance of using the maximum likelihood ratio test to assess changes in the optimal model and data fit;

[0154] If the improvement obtained by adding one or more additional factors is statistically significant, then a simpler modular structure is rejected and a more complex modular structure is chosen instead.

[0155] The simulation results are evaluated and diagnosed to gain a deeper understanding of the dynamics of algal blooms;

[0156] The formula for calculating the red tide growth and mortality equation is:

[0157] ;

[0158] in, C represents the rate of change of red tide concentration over time, where C represents the red tide concentration. Let F represent the growth term, m represent the death term, and F represent the growth term. g Indicates sediment germination term;

[0159] growth items The calculation formula is:

[0160] ;

[0161] Wherein, G([N],T,S) represents the terms affected by temperature, salinity, and nutrients, and G(E,T,S) represents the terms affected by light, temperature, and salinity. [N] represents nutrient concentration, T represents temperature, E represents irradiance, and S represents salinity. [N] emphasizes the concentration value of nutrients. S and T are physical parameters (salinity and humidity), and their values ​​directly correspond to the variables themselves. They do not need to be additionally labeled as "concentration", so they are not enclosed in boxes.

[0162] The formulas for calculating G([N],T,S) and G(E,T,S) are as follows:

[0163] ;

[0164] ;

[0165] Among them, K [N] [N] represents the half-saturation constant indicating nutritional limitation, [N] represents the nutrient concentration, and G represents the nutrient concentration. max (T,S) represents the maximum growth rate. This represents the basal metabolic rate, a g tanh represents the light absorption coefficient, and tanh represents the hyperbolic tangent function.

[0166] The calculation formula is:

[0167] ;

[0168] ;

[0169] ;

[0170] in, This indicates the effect of salinity. Indicates the effect of temperature. This represents the theoretical maximum growth rate at the optimal growth temperature and optimal growth salinity. Indicates the optimal growth temperature. The optimal salinity for growth is represented by b1, which is the first constant (b1=-0.000513), b2, which is the second constant (b2=0.016), b3, which is the third constant (b3=-0.0867), b4, which is the fourth constant (b4=0.382), and b5, which is the fifth constant (b5=0.0343). F(T=5) represents the temperature effect term at 5°C, where a1 is the first parameter constant (a1=0.0000882), a2 is the second parameter constant (a2=0.00808), a3 is the third parameter constant (a3=0.220), and a4 is the fourth parameter constant (a4=0.872).

[0171] The biomass change of red tide can be calculated using the above equation. The initial red tide concentration is set to 0. Based on data from physical-biological-chemical environmental factors such as temperature, salinity, light intensity, and nutrients, the hourly red tide concentration for the next 144 hours can be calculated using the above equation. In the first 12 hours, its concentration is calibrated as C. 12 The potential for a red tide outbreak is assessed by analyzing the changes in red tide concentration every 12 hours. If the rate of change in concentration over the next 12 hours is compared to C... 12 If the concentration increases by more than 50%, it is considered an outbreak; otherwise, no outbreak is considered.

[0172] Red tide outbreak warning judgment is carried out to obtain red tide outbreak warning results, which include red tide outbreak low-level warning, red tide outbreak medium-level warning, red tide outbreak high-level warning and red tide outbreak emergency warning;

[0173] For each hour predicted in the future, the average concentration growth rate over the past 12 hours is calculated using the following formula:

[0174] R t =[(C t -C t-12 ) / C t-12 ]×100%;

[0175] Among them, C t C represents the red tide concentration at the current moment. t-12 R represents the red tide concentration 12 hours ago, t represents the predicted concentration for each hour in the future. t This indicates the average concentration growth rate over the past 12 hours;

[0176] The current red tide concentration is compared with a preset concentration threshold, and the average concentration growth rate over the past 12 hours is also compared with a preset growth rate threshold. Specifically, this includes:

[0177] The preset concentration thresholds include potential hazard concentration thresholds and confirmed red tide outbreak concentration thresholds;

[0178] The preset growth rate thresholds include the potential outbreak growth rate threshold and the explicit outbreak growth rate threshold;

[0179] When C t <C potential And R t <G potential At that time, a low-level warning for a red tide outbreak was issued;

[0180] When C t ≥C potential And R t ≥ Gpotential At that time, a medium-level red tide outbreak warning was issued;

[0181] When C potential ≤C t <C outbreak And R t ≥G outbreak At that time, a high-level warning for a red tide outbreak was issued;

[0182] When C t ≥C outbreak At that time, an emergency warning for a red tide outbreak was issued;

[0183] Among them, C potential C represents the potential hazard concentration threshold. outbreak This indicates the concentration threshold for a confirmed red tide outbreak, G. potential G represents the potential outbreak growth rate threshold. outbreak This indicates a specific threshold for the outbreak growth rate;

[0184] The system continuously monitors the dynamic changes in red tide concentration and the average concentration growth rate over the past 12 hours during the forecast period, and immediately triggers the corresponding warning when any level of warning conditions are met.

[0185] In a specific embodiment, based on the above judgment and evaluation, it specifically includes initial conditions and simulation stage, red tide outbreak judgment criteria, and dynamic updates and early warning;

[0186] The initial conditions and simulation phase include initial concentration setting, environmental element input, and concentration calculation;

[0187] The initial concentration settings include: at the start of the simulation, setting the initial concentration of the target red tide algae species to a small but non-zero background concentration, where the initial concentration is set to 10. -6 The cell / L setting is intended to more realistically reflect the prevalence of red tide algae in natural water bodies and to avoid early growth biases that can result from simulating from scratch.

[0188] Environmental inputs include: using real-time or predicted physical-biological-chemical environmental data such as temperature, salinity, light intensity, dissolved oxygen, and seawater pH as inputs to the red tide outbreak assessment module.

[0189] Concentration calculation includes: Based on the above multi-factor coupled numerical equation, the red tide outbreak assessment module can calculate the concentration change of the target red tide algae species hourly over the next 144 hours.

[0190] The criteria for judging a red tide outbreak include the calculation of the basic growth rate and the assessment of graded early warning systems;

[0191] The red tide outbreak assessment module employs a phased, multi-indicator approach to determine and evaluate the risk of red tide outbreaks, rather than relying on a single indicator. This assessment focuses on the absolute level of red tide concentration and its relative growth rate, and introduces the concept of a sliding time window.

[0192] The basic growth rate calculation includes: for each hour (t) in the future forecast, calculating the average concentration growth rate over the past 12 hours, using the following formula:

[0193] R t =[(C t -C t-12 ) / C t-12 ]×100%;

[0194] Among them, C t C represents the red tide concentration at the current moment. t-12 R represents the red tide concentration 12 hours ago, t represents the predicted concentration for each hour in the future. t This indicates the average concentration growth rate over the past 12 hours;

[0195] The outbreak assessment thresholds include: The red tide outbreak assessment module will set two key thresholds to determine the outbreak status of red tides. These thresholds can be configured according to the biological characteristics of different red tide algae species and actual hazard standards: Potential outbreak growth rate threshold G potential Clearly define the outbreak growth rate threshold G outbreak Potential hazard concentration threshold C potential Confirming the outbreak concentration threshold C for red tides outbreak ;

[0196] The tiered early warning assessment is based on hourly calculations of red tide concentration and growth rate. The red tide outbreak assessment module will conduct a tiered assessment:

[0197] When C t <C potential And R t <G potential At that time, a low-level red tide outbreak warning will be issued. The low-level red tide outbreak warning is indicated by a blue warning, which means that the red tide is at the background level or in a slow growth stage and the risk is low.

[0198] When C t ≥C potential And R t ≥Gpotential At this time, a medium-level red tide outbreak warning will be issued. The medium-level red tide outbreak warning is a yellow warning, which means that the red tide concentration has reached a certain level and is showing a rapid growth trend, with a potential risk of outbreak. At this time, monitoring and early warning need to be strengthened.

[0199] When C potential ≤C t <C outbreak And R t ≥G outbrea At k, a high-level red tide outbreak warning is issued. The high-level red tide outbreak warning is an orange warning, indicating that the red tide concentration has increased significantly and is showing strong explosive growth. It is very likely to reach the outbreak level in the short term, and the emergency response mechanism needs to be activated.

[0200] When C t ≥C outbreak When a red tide outbreak occurs, an emergency red tide warning will be issued. A red tide warning indicates that the red tide has reached or exceeded the critical concentration for a confirmed outbreak, posing a real hazard. Immediate response measures are required.

[0201] This system will continuously monitor the red tide concentration and growth rate during the forecast period. Once any warning level is reached, the system will immediately trigger the corresponding warning information. This dynamic assessment mechanism can provide more timely and detailed information on red tide outbreak risks, providing management departments and aquaculture farmers with a scientific basis for decision-making.

[0202] The red tide underwater 3D drift module is used to calculate the upward floating velocity of the red tide through 3D floating dynamic equations. The processing logic includes:

[0203] The dynamic process of red tides floating to the sea surface as they grow underwater is calculated using the three-dimensional floating dynamic equations of red tides in seawater. The calculation expression is as follows:

[0204] ;

[0205] ;

[0206] in, W represents the speed of the ocean current. a K represents the upward floating velocity of the red tide, and K represents the diffusion coefficient. Z represents the direction and rate of the fastest change in concentration in all directions, Z represents the current depth, and Z1 represents the reference depth.

[0207] In a specific embodiment, newly germinating cells swim upwards at a rate of meters per day until they reach a depth of 1% average light level (Z1 = 21 meters). During this ascent, it is assumed that no cells die. Above this depth, to match the observed vertical distribution of the red tide, the vertical upstream velocity is specified as W at z = 0, where the natural boundary condition is W. a =0. For stations where abundance was measured at both depths, the ratio of average surface abundance to average abundance at 10 meters was 2:1, and there was a strong correlation between abundance measurements at the two depths for the same station (correlation coefficient of 0.71 after logarithmic transformation). The study found that by attenuating swimming speed, for a wide range of parameter values, the ratio of average surface abundance to average 10-meter abundance ranged between 1.75:1 and 2.25:1.

[0208] This innovative approach introduces a three-dimensional floating dynamic equation for red tides underwater, considering the complex dynamic process by which red tide algae gradually gain buoyancy as they grow (increase in biomass) and float upwards to the sea surface along with the vertical movement of the water. Through comprehensive modeling of algal buoyancy, vertical water velocity, and water density stratification, the system accurately simulates the vertical drift path and time of red tides from deep underwater to the surface. The introduction of the three-dimensional floating dynamic equation overcomes the limitation of traditional Lagrange drift methods in failing to account for the underwater floating of red tides, enabling the system to accurately predict the three-dimensional movement trajectory of red tides from their underwater outbreak point to the surface. This significantly improves the accuracy of predicting when, where, and at what concentration red tides will reach the surface, providing more accurate initial conditions for predicting the horizontal drift of surface red tides, thereby enhancing the overall spatiotemporal accuracy of early warning.

[0209] The red tide surface multi-parameter disturbance drift module is used to calculate the red tide drift path. The processing logic includes:

[0210] The coordinates of red tide particles are calculated based on the Lagrange method and by adding a perturbation term.

[0211] The disturbance additions include the ocean current drag coefficient P1, the wind drag coefficient P2, and the Coriolis effect deflection angle ϑ0;

[0212] The red tide drift path is obtained by statistically analyzing the coordinates of red tide particles. The formula for calculating the coordinates of red tide particles is as follows:

[0213] ;

[0214] Where, x i The x-coordinate and y-coordinate of red tide particle i in the planar coordinate system at time t represent the position of the red tide particle i at time t. i U represents the position of the red tide particle i in the planar coordinate system at time t, where u is the vertical coordinate. a v represents the component of the ocean current velocity in the x-direction.a u represents the component of the ocean current velocity in the y-direction. t v represents the turbulent velocity component in the x-direction. t w represents the turbulent velocity component in the y-direction. u w represents the velocity component in the x-direction of wave breaking. v u represents the velocity component in the y-direction of wave breaking. d V represents the wind speed component in the x-direction. d P1 represents the wind speed component in the y-direction, P2 represents the ocean current drag coefficient, and ϑ represents the wind drag coefficient. i ϑ0 represents the angle between the wind direction and the x-direction, and ϑ0 represents the deflection angle due to the Coriolis effect.

[0215] In a specific embodiment, the uncertainty of parameter values ​​in the parameterization scheme of the red tide surface multi-parameter perturbation drift module is one of the sources of red tide drift path forecast error and a common type of error. To address the error caused by the uncertainty of the physical process parameterization scheme, a stochastic perturbation parameterization scheme was developed. This method directly perturbs important parameters in the red tide surface multi-parameter perturbation drift module using stochastic physical methods to construct ensemble members, while maintaining the internal consistency of the physical parameterization scheme. This perturbation is considered a more physically plausible method, addressing the source of uncertainty in the red tide surface multi-parameter perturbation drift module. This method can improve the reliability and probabilistic forecasting skill of the ensemble forecasting system to a certain extent. According to the formula: ;

[0216] The red tide surface multi-parameter perturbation drift module has three key parameters: the current velocity drag coefficient, the wind drag coefficient, and the motion deviation angle caused by the Coriolis force. This study introduces a stochastic perturbation parameterization scheme to perturb these three key parameters, representing the uncertainties in the red tide drift process. The ensemble forecast members typically consist of a control member and multiple ensemble members derived from the perturbation of the control member. Calibrating the parameter and variable values ​​of the control member is the first step in constructing the ensemble members. The current drag coefficient in the red tide surface multi-parameter perturbation drift module is typically between 0.8 and 0.95, and the wind drag coefficient is generally between 0.005 and 0.02. The deviation between the red tide drift direction and the wind direction ranges from 2° to 35°. The parameter values ​​β of the ensemble control members, the standard deviation σ of the parameter perturbation, and the parameter perturbation range were determined through sensitivity experiments, as shown in Table 2.

[0217] Table 2 Basic Settings for Ensemble Perturbation Parameters

[0218] P1 represents the ocean current drag coefficient, and P2 represents the wind drag coefficient. They reflect the deceleration of red tide particles under the influence of various factors such as gravity, wave action, buoyancy, and turbulence during the peak of red tide outbreaks.

[0219] This scheme innovatively introduces the displacement effects of small-scale turbulent motion and wave breaking in the multi-parameter disturbance drift module of red tide surface. By coupling a high-resolution turbulence model or employing a parameterized scheme, it accurately simulates the random diffusion and mixing effects of turbulence on red tide particles. Simultaneously, it considers the intense agitation and displacement of surface water and red tide algae by wave breaking in the nearshore region, converting this into an additional horizontal displacement component. The introduction of small-scale turbulent processes and wave breaking overcomes the limitation of traditional Lagrange drift methods in considering the impact of marine environmental changes on phase drift at multiple spatial scales. By finely characterizing these small-scale dynamic processes, this patent can more accurately simulate the local diffusion and irregular movement of red tide algae, significantly improving the prediction accuracy of red tide horizontal drift trajectories, especially the prediction reliability under complex sea conditions such as nearshore areas and large waves.

[0220] The red tide risk early warning module is used to perform spatial overlay analysis on the potential red tide coverage area and the preset sensitive target area to obtain risk indicators. Based on the risk indicators, the red tide hazard risk level is judged to obtain the red tide hazard risk level result.

[0221] Data on sensitive target areas includes marine aquaculture areas and ecological protection areas. Marine aquaculture areas include the geographical coordinates of aquaculture facilities, aquaculture species, and aquaculture scale.

[0222] Ecological protection zones include ecological boundaries and protected targets;

[0223] Risk indicators include the percentage of sensitive target areas affected and the duration of red tide impact;

[0224] The processing logic for spatial overlay analysis of potential red tide coverage areas and pre-defined sensitive target areas includes:

[0225] Based on the red tide drift path, the spatial range covered by the red tide mass within a future preset time period is extracted to form the potential red tide coverage area;

[0226] The spatial geometric intersection of the potential red tide coverage area and the preset sensitive target area is performed to calculate the percentage of the sensitive target area affected and the duration of the red tide impact.

[0227] The formula for calculating the percentage of sensitive target areas affected is:

[0228] ;

[0229] Wherein, AIR represents the percentage of the sensitive target area affected, A I A represents the area of ​​intersection between the potential red tide coverage area and the sensitive target area. T This represents the total area of ​​the sensitive target region;

[0230] The duration of red tide impact is the total number of hours that a red tide mass continuously covers the same sensitive target area during the forecast period.

[0231] In a specific embodiment, based on the red tide drift path and trajectory obtained from ensemble prediction, the red tide risk early warning module calculates the area that the red tide mass may cover in the next 120 hours. By spatially overlaying these potential coverage areas with the geographical ranges of preset marine aquaculture areas (including the geographical coordinates, aquaculture species, and aquaculture scale of various aquaculture facilities) and ecological protection areas (including their boundaries and protection targets), the likelihood and extent of sensitive target areas being exposed to the influence of red tides are quantified. Specifically:

[0232] Based on the red tide drift path and trajectory obtained from ensemble prediction, the red tide risk early warning module calculates the area that the red tide mass may cover in the next 120 hours. By spatially overlaying these potential coverage areas with the geographical extent of pre-defined marine aquaculture areas (including the geographical coordinates, aquaculture species, and scale of various aquaculture facilities) and ecological protection zones (including their boundaries and protected targets), the module quantifies the probability and degree of exposure of sensitive target areas to the red tide. This includes calculating the area proportion of each aquaculture area within the red tide impact zone, the shortest distance from the red tide front, and the time and duration of the red tide passing through specific protected areas.

[0233] For different sensitive target areas, the red tide risk early warning module will conduct assessments based on their specific vulnerability characteristics. For marine aquaculture areas, vulnerability assessments may consider factors such as the sensitivity of farmed species to red tide toxins, farming density, the disaster resistance of aquaculture facilities, and the economic affordability of farmers. For ecological protection areas, vulnerability assessments will consider factors such as the ecological importance of the protected objects, the stability of the community structure, and the ecological recovery capacity against red tides.

[0234] The risk level of red tide hazard is determined based on risk indicators, and the result of the red tide hazard risk level is obtained.

[0235] The red tide hazard risk level results include low risk, medium risk, high risk, and extremely high risk;

[0236] The affected area percentage (AIR) of sensitive target areas is compared with a preset area percentage threshold, and the red tide impact duration (DOI) is also compared with a preset red tide impact duration threshold. Specifically, this includes:

[0237] When AIR < 10% and DOI < 6 hours, it is considered low risk;

[0238] When AIR ≥ 10% and DOI ≥ 6 hours, or AIR < 10% and DOI ≥ 24 hours, or DOI ≥ 12 hours, it is judged as medium risk;

[0239] When AIR ≥ 30% and DOI ≥ 18 hours, or AIR ≥ 60% and DOI ≥ 12 hours, or AIR < 30% and DOI ≥ 48 hours, it is considered high risk;

[0240] When AIR ≥ 60% and DOI ≥ 36 hours, or AIR ≥ 80% 24 and DOI ≥ 24 hours, it is judged as extremely high risk;

[0241] DOI indicates the duration of red tide impact.

[0242] In a specific embodiment, based on the above content and combining the results of risk exposure assessment and risk vulnerability assessment, the red tide risk early warning module will establish a graded risk assessment system. By setting scientific risk indicators and thresholds, the red tide risk faced by marine aquaculture areas and ecological protection areas will be quantified and classified into levels (low risk, medium risk, high risk, and extremely high risk).

[0243] Among them, Area Impact Ratio (AIR) is the percentage of the area that overlaps with the total area of ​​the potential red tide coverage area and the sensitive target (aquaculture area or protected area).

[0244] The following is an example of risk level assessment for a single sensitive target (an aquaculture area or an ecological protection zone). In practical applications, the threshold can be finely calibrated and adjusted based on the specific sea area, aquaculture species, the importance of the protected target, historical red tide event data, and the potential harm of red tide algae species (even if not directly reflected in this assessment).

[0245] When AIR < 10% and DOI < 6 hours, it is considered low risk. At this time, the impact of red tide is slight or far away, and the risk is low. However, vigilance is still required, and the latest forecast should be reviewed regularly.

[0246] When AIR ≥ 10% and DOI ≥ 6 hours, or AIR < 10% and DOI ≥ 24 hours, or DOI ≥ 12 hours, it is judged as medium risk. At this time, the red tide mass may have a certain impact on some sensitive areas, or may have an indirect impact due to long-term proximity. Continuous monitoring and preparation are required.

[0247] When AIR ≥ 30% and DOI ≥ 18 hours, or AIR ≥ 60% and DOI ≥ 12 hours, or AIR < 30% and DOI ≥ 48 hours, it is considered high risk. At this time, the red tide mass will have a medium to long-term impact on a large area of ​​sensitive areas, or a very long-term impact on local areas, posing a significant threat and requiring key prevention measures.

[0248] When AIR ≥ 60% and DOI ≥ 36 hours, or AIR ≥ 80% 24 and DOI ≥ 24 hours, it is judged as extremely high risk. At this time, the red tide mass will cover the sensitive area over a large area and for a long time, causing extremely high and continuous physical (such as blockage, hypoxia) or potential biotoxic threats to production and ecosystems.

[0249] Taking into account the dynamic changes in red tide drift and the marine environment, the red tide risk warning module will support real-time or near real-time updates. The risk assessment will be dynamically adjusted as new monitoring data or more accurate drift predictions become available.

[0250] Furthermore, the red tide risk early warning module will support multi-scenario analysis, providing decision-makers with multi-faceted risk management insights. In addition to simulating the potential risks to sensitive targets posed by different red tide intensities, drift paths, or durations, the red tide risk early warning module specifically introduces a worst-case assessment based on adjustments to ensemble perturbation parameters.

[0251] The adjustment of ensemble perturbation parameters includes: by adjusting the perturbation range or coefficient of key physical parameters in the aforementioned random perturbation parameterization scheme, setting the three main perturbation coefficients to their maximum or minimum values ​​respectively, to simulate extremely unfavorable marine environmental conditions or red tide biological characteristics.

[0252] Worst-case assessments include: under these extreme parameter settings, the red tide risk warning module will rerun ensemble predictions to generate a set of drift paths and concentration distribution scenarios that are most likely to maximize the red tide drift range, have the longest duration of impact, or pose the greatest threat to sensitive targets.

[0253] Decision support includes: This worst-case scenario assessment helps decision-makers identify which areas will face the highest risk under the most adverse conditions, thereby enabling the development of more resilient contingency plans and risk management strategies, such as evacuating aquaculture products in advance, reinforcing protective facilities, or adjusting production plans to minimize the potential losses caused by red tides.

[0254] The red tide visualization product creation module is used to generate visualization products, and its processing logic includes:

[0255] The output data from the marine data aggregation module, red tide outbreak assessment module, red tide underwater three-dimensional drift module, red tide surface multi-parameter disturbance drift module, and red tide risk early warning module are imported and integrated into the GIS platform.

[0256] Visualization products are generated using Python scripts, including red tide drift trajectory maps, underwater red tide 3D distribution maps, biochemical environmental parameter maps, risk assessment thematic maps, and time series animations.

[0257] In a specific embodiment, an efficient and intuitive red tide visualization product creation module is constructed to transform complex data from red tide drift prediction, risk assessment, and ecological environment simulation into easily understandable and decision-making graphical information. The core function of this module is to leverage the powerful data processing and plotting capabilities of GIS technology and the Python programming language to achieve automated visualization and archiving of daily results data.

[0258] The output data from the marine data aggregation module, red tide outbreak assessment module, red tide underwater 3D drift module, red tide surface multi-parameter disturbance drift module, red tide risk early warning module, and red tide visualization product creation module are uniformly imported into the GIS platform for management. Simultaneously, basic geographic information data, such as land-sea boundaries, water depth, mariculture area boundaries, ecological protection zone boundaries, and hydrological station locations, are integrated. Python scripts are used to preprocess the raw data output from the red tide visualization product creation module, performing cleaning, format conversion, interpolation, and gridding to ensure the data meets GIS visualization requirements. For example, discrete particle trajectory data is transformed into continuous drift path layers, or gridded ecological parameters are converted into contour maps.

[0259] Red tide drift trajectory map: Displays the predicted drift path of a red tide mass in real-time or near real-time, including the horizontal drift trajectory, the probability distribution area predicted by the ensemble, and the location of the red tide at key time points. Red tide concentration or outbreak probability can be represented by color intensity or contour lines.

[0260] 3D distribution map of underwater red tide: Combining vertical drift prediction, it shows the growth, vertical ascent process of underwater red tide and its distribution at different depths in three dimensions.

[0261] Biochemical environmental parameter map: Shows the spatial distribution of key environmental factors simulated in ecological models, such as seawater temperature, salinity, nutrient concentration, chlorophyll a concentration, etc., as well as the predicted distribution of red tide mortality.

[0262] Risk assessment thematic map: In the form of a thematic map, it intuitively displays the spatial distribution of red tide risk levels in mariculture areas and ecological protection areas calculated based on the red tide risk early warning module, and can overlay red tide drift trajectory to clearly indicate the affected areas and extent.

[0263] Time-series animation: To illustrate the dynamic process of red tide drift and risk evolution, time-series animation is created to dynamically display the occurrence, development, drift, and disappearance of red tides, as well as the dynamic evolution of risk levels.

[0264] Red tide drift trajectory map, underwater red tide 3D distribution map, biochemical environmental parameter map, risk assessment thematic map and time series animation

[0265] Daily automated generation: The system sets daily scheduled tasks to automatically run Python scripts, retrieve the latest forecast data from various modules, and generate a series of standardized visualization products.

[0266] Archived by Date: All generated images, map files (such as PNG, JPEG, and PDF formats), and related GIS layer files (such as Shapefile and GeoJSON) will be stored according to the date of publication, creating a clear archive management system for easy later review, comparison, and trend analysis. For example, a separate folder will be generated under each date, containing all visualization outputs for that date.

[0267] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0268] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A three-dimensional red tide drift risk early warning system based on ensemble perturbation, characterized in that, include: The module includes: marine data aggregation module, red tide outbreak assessment module, red tide underwater 3D drift module, red tide surface multi-parameter disturbance drift module, red tide risk early warning module, and red tide visualization product creation module. The marine data aggregation module is used to receive and aggregate multi-source marine statistical data sets, and to classify and store the multi-source marine statistical data sets by date. The red tide outbreak assessment module is used to perform dynamic assessment by combining a set of biological model structures to obtain the red tide concentration, and to make a red tide outbreak early warning judgment based on the red tide concentration to obtain the red tide outbreak early warning result. The red tide underwater three-dimensional drift module is used to calculate the upward floating speed of the red tide through three-dimensional floating dynamic equations; The red tide surface multi-parameter disturbance drift module is used to calculate the red tide drift path; The red tide risk early warning module is used to perform spatial overlay analysis on the potential red tide coverage area and the preset sensitive target area to obtain risk indicators, and to judge the red tide hazard risk level based on the risk indicators to obtain the red tide hazard risk level result. The red tide visualization product creation module is used to generate visualization products.

2. The three-dimensional red tide drift risk early warning system based on ensemble perturbation as described in claim 1, characterized in that: The marine data aggregation module is used to receive and aggregate multi-source marine statistical data sets, and to classify and store the multi-source marine statistical data sets by date. The multi-source marine statistical data includes biological environmental parameter data, marine dynamics data, atmospheric forcing data, and geographical and risk assessment data; The biological environmental parameter data includes nutrient concentration data, temperature data, salinity data, and irradiance data; The ocean dynamics data includes ocean current velocity data, surface turbulence velocity data, and surface wave breaking velocity data; The ocean current velocity data includes ocean current velocity component data in the x-direction, ocean current velocity component data in the y-direction, and ocean current velocity component data in the z-direction. The surface turbulence velocity data includes turbulence velocity component data in the x-direction and turbulence velocity component data in the y-direction; The surface wave breaking velocity data includes velocity component data in the x-direction of wave breaking and velocity component data in the y-direction of wave breaking; The atmospheric forcing data includes wind speed data; The wind speed data includes wind speed component data in the x-direction and wind speed component data in the y-direction. The geographical and risk assessment data includes data on sensitive target areas; The sensitive target area data includes marine aquaculture areas and ecological protection areas; The marine aquaculture area includes the geographical coordinates of the aquaculture facilities, the species being cultured, and the scale of the aquaculture. The ecological protection zone includes ecological boundaries and protected targets.

3. The three-dimensional red tide drift risk early warning system based on ensemble perturbation as described in claim 1, characterized in that: The red tide outbreak assessment module is used to perform dynamic assessment by combining a set of biological model structures to obtain the red tide concentration, and to make a red tide outbreak early warning judgment based on the red tide concentration to obtain the red tide outbreak early warning result. The processing logic for obtaining red tide concentration through dynamic assessment includes: The optimal target model structure is selected from the set of biological model structures by using the maximum likelihood ratio test. The biological model structure set includes basic line structure, death structure, nutrient structure, death and nutrient structure, and zero growth structure; Based on the target model structure, the red tide concentration is calculated using the red tide growth and death equation. The calculation formula for the red tide growth and death equation is as follows: ; in, C represents the rate of change of red tide concentration over time, where C represents the red tide concentration. Let F represent the growth term, m represent the death term, and F represent the growth term. g Indicates sediment germination term; growth items The calculation formula is: ; Where G([N],T,S) represents the term affected by temperature, salinity and nutrients, G(E,T,S) represents the term affected by light, temperature and salinity, [N] represents nutrient concentration, T represents temperature, E represents irradiance and S represents salinity; The formulas for calculating G([N],T,S) and G(E,T,S) are as follows: ; ; Among them, K [N] [N] represents the half-saturation constant indicating nutritional limitation, [N] represents the nutrient concentration, and G represents the nutrient concentration. max (T,S) represents the maximum growth rate. This represents the basal metabolic rate, a g tanh represents the light absorption coefficient, and tanh represents the hyperbolic tangent function. The calculation formula is: ； ; ; in, This indicates the effect of salinity. Indicates the effect of temperature. This represents the theoretical maximum growth rate at the optimal growth temperature and optimal growth salinity. Indicates the optimal growth temperature. The optimal salinity for growth is represented by b1, b2, b3, b4, b5, and F(T=5). The temperature effect term at 5°C is represented by a1, a2, a3, a4, and a4.

4. The three-dimensional red tide drift risk early warning system based on ensemble perturbation as described in claim 3, characterized in that: Red tide outbreak warning judgment is carried out to obtain red tide outbreak warning results, which include red tide outbreak low-level warning, red tide outbreak medium-level warning, red tide outbreak high-level warning and red tide outbreak emergency warning; For each hour predicted in the future, the average concentration growth rate over the past 12 hours is calculated using the following formula: R t =[(C t -C t-12 ) / C t-12 ]×100%; Among them, C t C represents the red tide concentration at the current moment. t-12 R represents the red tide concentration 12 hours ago, t represents the predicted concentration for each hour in the future. t This indicates the average concentration growth rate over the past 12 hours; The current red tide concentration is compared with a preset concentration threshold, and the average concentration growth rate over the past 12 hours is also compared with a preset growth rate threshold. Specifically, this includes: The preset concentration thresholds include potential hazard concentration thresholds and red tide confirmed outbreak concentration thresholds; The preset growth rate thresholds include a potential outbreak growth rate threshold and a defined outbreak growth rate threshold; When C t <C potential And R t <G potential At that time, a low-level warning for a red tide outbreak was issued; When C t ≥C potential And R t ≥ Gpotential At that time, a medium-level red tide outbreak warning was issued; When C potential ≤C t <C outbreak And R t ≥G outbreak At that time, a high-level warning for a red tide outbreak was issued; When C t ≥C outbreak At that time, an emergency warning for a red tide outbreak was issued; Among them, C potential C represents the potential hazard concentration threshold. outbreak This indicates the concentration threshold for a confirmed red tide outbreak, G. potential G represents the potential outbreak growth rate threshold. outbreak This indicates a specific threshold for the outbreak growth rate; The system continuously monitors the dynamic changes in red tide concentration and the average concentration growth rate over the past 12 hours during the forecast period, and immediately triggers the corresponding warning when any level of warning conditions are met.

5. The three-dimensional red tide drift risk early warning system based on ensemble perturbation as described in claim 1, characterized in that: The red tide underwater three-dimensional drift module is used to calculate the upward floating speed of the red tide through three-dimensional floating dynamic equations. The processing logic includes: The dynamic process of red tides floating to the sea surface as they grow underwater is calculated using the three-dimensional floating dynamic equations of red tides in seawater. The calculation expression is as follows: ； ; in, W represents the speed of the ocean current. a K represents the upward floating velocity of the red tide, and K represents the diffusion coefficient. Z represents the direction and rate of the fastest change in concentration in all directions, Z represents the current depth, and Z1 represents the reference depth.

6. The three-dimensional red tide drift risk early warning system based on ensemble perturbation as described in claim 1, characterized in that: The red tide surface multi-parameter disturbance drift module is used to calculate the red tide drift path, and the processing logic includes: The coordinates of red tide particles are calculated based on the Lagrange method and by adding a perturbation term. The disturbance additions include ocean current drag coefficient P1, wind drag coefficient P2, and Coriolis effect motion deflection angle ϑ0. The red tide drift path is obtained by statistically analyzing the coordinates of red tide particles. The formula for calculating the coordinates of red tide particles is as follows: ; Where, x i The x-coordinate and y-coordinate of red tide particle i in the planar coordinate system at time t represent the position of the red tide particle i at time t. i U represents the position of the red tide particle i in the planar coordinate system at time t, where u is the vertical coordinate. a v represents the component of the ocean current velocity in the x-direction. a u represents the component of the ocean current velocity in the y-direction. t v represents the turbulent velocity component in the x-direction. t w represents the turbulent velocity component in the y-direction. u w represents the velocity component in the x-direction of wave breaking. v u represents the velocity component in the y-direction of wave breaking. d V represents the wind speed component in the x-direction. d P1 represents the wind speed component in the y-direction, P2 represents the ocean current drag coefficient, and ϑ represents the wind drag coefficient. i ϑ0 represents the angle between the wind direction and the x-direction, and ϑ0 represents the deflection angle due to the Coriolis effect.

7. The three-dimensional red tide drift risk early warning system based on ensemble perturbation as described in claim 1, characterized in that: The red tide risk early warning module is used to perform spatial overlay analysis on the potential red tide coverage area and the preset sensitive target area to obtain risk indicators; The risk level of red tide hazard is determined based on risk indicators, and the result of the red tide hazard risk level is obtained. The sensitive target area data includes marine aquaculture areas and ecological protection areas. The marine aquaculture areas include the geographical coordinates of aquaculture facilities, aquaculture species, and aquaculture scale. The ecological protection zone includes ecological boundaries and protection targets; The risk indicators include the percentage of sensitive target areas affected and the duration of the red tide impact; The processing logic for spatial overlay analysis of potential red tide coverage areas and pre-defined sensitive target areas includes: Based on the red tide drift path, the spatial range covered by the red tide mass within a future preset time period is extracted to form the potential red tide coverage area; The spatial geometric intersection operation is performed between the potential red tide coverage area and the geographical range of the preset sensitive target area to calculate the percentage of the sensitive target area affected and the duration of the red tide impact. The formula for calculating the percentage of the sensitive target area affected is as follows: ; Wherein, AIR represents the percentage of the sensitive target area affected, A I A represents the area of ​​intersection between the potential red tide coverage area and the sensitive target area. T This represents the total area of ​​the sensitive target region; The duration of the red tide impact is the total number of hours during which the red tide mass continuously covers the same sensitive target area within the prediction period.

8. The three-dimensional red tide drift risk early warning system based on ensemble perturbation as described in claim 7, characterized in that: The risk level of red tide hazard is determined based on risk indicators, and the result of the red tide hazard risk level is obtained. The red tide hazard risk level results include low risk, medium risk, high risk, and extremely high risk; The affected area percentage (AIR) of the sensitive target area is compared with a preset area percentage threshold, and the red tide impact duration (DOI) is simultaneously compared with a preset red tide impact duration threshold. Specifically, this includes: When AIR < 10% and DOI < 6 hours, it is considered low risk; When AIR ≥ 10% and DOI ≥ 6 hours, or AIR < 10% and DOI ≥ 24 hours, or DOI ≥ 12 hours, it is judged as medium risk; When AIR ≥ 30% and DOI ≥ 18 hours, or AIR ≥ 60% and DOI ≥ 12 hours, or AIR < 30% and DOI ≥ 48 hours, it is considered high risk; When AIR ≥ 60% and DOI ≥ 36 hours, or AIR ≥ 80% 24 and DOI ≥ 24 hours, it is judged as extremely high risk; DOI indicates the duration of red tide impact.

9. The three-dimensional red tide drift risk early warning system based on ensemble perturbation as described in claim 1, characterized in that: The red tide visualization product creation module is used to generate visualization products, and its processing logic includes: The output data from the marine data aggregation module, red tide outbreak assessment module, red tide underwater three-dimensional drift module, red tide surface multi-parameter disturbance drift module, and red tide risk early warning module are imported and integrated into the GIS platform. Visualization products are generated using Python scripts, including red tide drift trajectory maps, underwater red tide 3D distribution maps, biochemical environmental parameter maps, risk assessment thematic maps, and time series animations.

10. A three-dimensional red tide drift risk early warning method based on ensemble perturbation, as described in claims 1-9, characterized in that, Includes the following steps: Step S1: Receive and aggregate multi-source oceanographic data sets; Step S2: Combine the biological model structure set for dynamic evaluation to obtain the red tide concentration, and make a red tide outbreak warning judgment based on the red tide concentration to obtain the red tide outbreak warning result; Step S3: Calculate the upward floating velocity of the red tide using the three-dimensional floating dynamic equation; Step S4: Calculate the red tide drift path; Step S5: Perform spatial overlay analysis on the potential red tide coverage area and the preset sensitive target area to obtain risk indicators. Based on the risk indicators, determine the red tide hazard risk level and obtain the red tide hazard risk level result. Step S6: Generate a visual product.

Images