A method and system for precise accounting and potential assessment of marine carbon sink flux

Abstract

This invention relates to the field of marine carbon sink technology, specifically disclosing a method and system for accurate calculation and potential assessment of marine carbon sink flux. The invention collects multi-source heterogeneous marine data through an integrated air-water-shore system, and obtains a standardized dataset through noise reduction, interpolation, and normalization. It employs a spatially constrained improved K-means algorithm to finely divide the sea area into four types of carbon sink ecological scenarios. For the differentiated carbon sequestration mechanisms of different scenarios, a dedicated dynamic calculation model is constructed to measure the carbon sink flux across the entire area in real time. Simultaneously, a spatiotemporal coupling error correction model is established to correct multi-dimensional interference errors, controlling the calculation error to within 5%. This invention establishes a three-level hierarchical potential assessment system, quantifying the inherent, restoration and enhancement, and medium- to long-term carbon sink potential of the sea area and completing the level classification, ultimately achieving result visualization and dynamic encrypted storage of the entire data process. This invention can achieve high-precision calculation and potential assessment of marine carbon sink flux, adapting to the needs of marine ecological management and carbon sink resource development.

Description

Technical Field

[0001] This invention relates to the field of marine carbon sink technology, specifically to a method and system for accurate calculation and potential assessment of marine carbon sink flux. Background Technology

[0002] Marine carbon sinks are a core component of the global carbon cycle, playing a crucial supporting role in areas such as marine ecological protection and restoration, and blue carbon trading. Accurately calculating marine carbon flux and scientifically assessing marine carbon sink potential are fundamental to quantifying marine carbon sequestration capacity, formulating marine ecological management policies, and promoting the industrialization of blue carbon sinks.

[0003] Current marine carbon sequestration technologies generally suffer from several shortcomings: First, data sources are often limited, relying primarily on single remote sensing or fixed-point monitoring data without integrating multi-dimensional monitoring resources from the air, water, and shore. This results in insufficient data completeness and timeliness, making it difficult to adapt to the dynamic carbon sequestration needs of complex marine areas. Second, marine ecological zoning methods are often crude, relying heavily on manual experience-based zoning models that ignore the differentiated characteristics of marine vegetation, topography, and spatial location. This leads to highly subjective zoning results with poor spatial continuity, resulting in low adaptability of carbon sequestration models. Third, the accounting models are generally general but lack specificity, failing to differentiate between mangroves and seaweed. The carbon sequestration mechanisms in different scenarios such as seabeds, tidal flats, wetlands, and nearshore planktonic ecosystems are often calculated using fixed empirical coefficients, which cannot adapt to the dynamic changes in the marine environment and result in large calculation errors. Fourth, the spatiotemporal differences in the impact of disturbance factors such as tides, seasonal temperature differences, ocean current disturbances, and human activities are not fully considered, and there is a lack of systematic error correction mechanisms, making it difficult to guarantee the accuracy of the calculation results. Fifth, the carbon sequestration potential assessment has a single dimension, only measuring the current static carbon sequestration capacity, without taking into account the space for ecological restoration and improvement as well as the potential for medium- and long-term sustainable development. The assessment results are one-sided and cannot provide data support for the long-term management of marine ecology.

[0004] In summary, existing technologies suffer from shortcomings such as low data standardization, insufficient refinement of ecological zoning, poor adaptability of accounting models, lack of error correction systems, and limited potential assessment dimensions, making it difficult to achieve high-precision dynamic accounting and comprehensive potential assessment of marine carbon flux. Therefore, this invention proposes a method and system for accurate accounting and potential assessment of marine carbon flux. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for accurate calculation and potential assessment of marine carbon sink flux, which solves the technical problems of low accuracy, poor scenario adaptability, one-sided potential assessment, and inability to dynamically update existing marine carbon sink calculation methods. It realizes intelligent processing of the entire process, including standardized processing of multi-source data, intelligent and precise zoning of marine ecology, dynamic and accurate calculation by scenario, spatiotemporal dual-dimensional error correction, multi-dimensional hierarchical potential assessment, and dynamic visualization and storage of results, thereby significantly improving the accuracy and scientific nature of marine carbon sink calculation and assessment.

[0006] To achieve the above objectives, the technical solution provided by this invention is: a method for accurate calculation and potential assessment of marine carbon sink flux, comprising the following steps: S1. Real-time acquisition and preprocessing of multi-source heterogeneous marine monitoring data: An integrated air-water-shore acquisition system is built, which simultaneously accesses four types of core data: satellite remote sensing data, underwater sensor real-time monitoring data, shore-based meteorological time-series data, and marine static ecological topography data. The data is then processed sequentially through outlier removal, discrete wavelet transform noise reduction, spatiotemporal joint interpolation completion, and spatiotemporal resolution and dimension unification to generate a standardized marine carbon sink basic dataset. S2. Intelligent and refined classification of marine ecological scenes: Construct a four-dimensional clustering feature vector that includes vegetation type, seawater depth, seabed sediment type, and carbon sink carrier density. Use an improved K-means unsupervised clustering algorithm that introduces spatial distance constraint weights to automatically divide the marine area into four types of ecological scenes: mangrove carbon sink area, seaweed bed carbon sink area, tidal wetland carbon sink area, and open nearshore planktonic carbon sink area. Generate refined ecological zoning results with coding, boundaries, and attributes. S3. Dynamic carbon flux calculation for different scenarios: For the differentiated carbon sequestration mechanisms of four ecological scenarios, a dedicated dynamic calculation model is constructed for each. The core parameters of the model are dynamically updated based on the pre-processed real-time monitoring data. The instantaneous carbon flux of each scenario is calculated hourly. The daily and monthly cumulative carbon flux is obtained by time-series integration and accumulation. The initial carbon flux of the target sea area is obtained after spatial superposition. S4. Spatiotemporal dynamic error correction of carbon sink flux: Construct a spatiotemporal coupled error correction model. Based on four major disturbance factors, namely tidal cycle, seasonal temperature difference, ocean current disturbance, and human activities, generate time-series dynamic correction coefficients and spatial differential correction coefficients respectively. Perform weighted correction on the initial carbon sink flux of the whole domain to obtain high-precision real-time carbon sink flux. S5. Multi-dimensional carbon sink potential stratified assessment: Based on the corrected accurate carbon sink flux, a three-level stratified potential assessment model is constructed to quantify and measure the inherent potential of the current marine area, the potential for ecological restoration and improvement, and the potential for medium- and long-term sustainability. The carbon sink potential is divided into four levels through a multi-indicator comprehensive scoring system. S6. Visualized Output and Dynamic Closed-Loop Storage of Results: Based on gridded carbon sink data, spatiotemporal heat maps and potential level zoning distribution maps are generated, supporting multi-scale time-series dynamic display; a hierarchical encryption and spatiotemporal database linkage architecture is adopted to classify and store the data throughout the entire process, and to realize dynamic iterative updates of data and accounting results on an hourly basis.

[0007] Furthermore, the four types of core data specifically include: Satellite remote sensing data includes monthly vegetation coverage, effective water area, and concentration of suspended particulate matter in the target sea area; Underwater sensors monitor real-time data including seawater temperature, salinity, dissolved oxygen concentration, plankton abundance, and dissolved inorganic carbon dioxide concentration in the water. Coastal meteorological time-series data, including wind speed, wind direction, cumulative rainfall, and sunshine duration; Static marine ecological topographic data includes seabed elevation, seabed type, native vegetation distribution type, and intensity of human development along the coast.

[0008] Furthermore, in step S1, the discrete wavelet transform denoising adopts a multi-layer signal hierarchical decomposition and threshold screening reconstruction method to extract the low-frequency effective feature components of the data, filter out high-frequency random noise generated by sensor jitter and short-term water flow disturbance, and retain the core trend of carbon sink time series changes; the spatial weight of the spatiotemporal joint interpolation completion is assigned according to the similarity of sea area distance, and the temporal weight is assigned according to the time interval decay, and the missing data is restored with high precision by combining spatially adjacent point data and consecutive temporal data.

[0009] Further, in step S2, the execution flow of the improved K-means unsupervised clustering algorithm includes: The four-dimensional clustering feature vectors are normalized to eliminate dimensional differences; four cluster centers corresponding to different carbon sink scenarios are initialized; the Euclidean distance between each grid cell and the cluster center is calculated iteratively, and the cluster centers are updated by integrating spatial neighborhood constraint weights; until the cluster centers converge and the partitioning results are stable, the ecological scenario is automatically partitioned, avoiding scattered and isolated partitions and ensuring that the partition space is contiguous and the boundaries are regular.

[0010] Furthermore, in step S3, the specific dynamic accounting model for each ecological scenario is as follows: For mangrove carbon sink areas and tidal wetland carbon sink areas, a dynamic flux accounting model coupling vegetation photosynthesis, soil deposition and water interface carbon exchange is constructed. The comprehensive carbon sink flux is dynamically calculated based on sunshine duration, water temperature, vegetation coverage, substrate type, tidal inundation duration and sea-air carbon concentration gradient. For seaweed bed carbon sink areas, a dynamic balance accounting model of growth-enrichment-loss is constructed. Based on the real-time growth rate of seaweed, combined with the carbon enrichment coefficient of water body and the water body scour loss coefficient, the effective net carbon sink flux is dynamically calculated. For open nearshore planktonic carbon sink areas, a dynamic accounting model for planktonic communities was constructed. Based on planktonic abundance, community carbon sequestration efficiency, carbon dioxide diffusion flux, and vertical transport efficiency of biological pumps, the net carbon sequestration flux of the water body was obtained by deducting the carbon release from biological respiration.

[0011] Furthermore, in step S4, the time-series dynamic correction coefficient integrates three types of interference factors: daily-scale tidal cycle fluctuations, seasonal-scale water temperature deviations, and monthly precipitation disturbances, to correct for fluctuations in intertidal carbon sequestration efficiency, the impact of temperature on biological carbon sequestration rates, and water carbon environment deviations caused by freshwater input, respectively; the spatial differentiation correction coefficient integrates three types of indicators: ocean current velocity, land-sea location differences, and intensity of human development along the coast, to correct for water carbon transport deviations, differences in the land-sea ecological base, and the weakening effect of human activities on carbon sequestration.

[0012] Furthermore, in step S4, a weighted coupling operation method of time-series correction coefficient and spatial correction coefficient is adopted to double correct the initial carbon sink flux, eliminate systematic and random errors caused by spatiotemporal environmental interference and human interference, and stabilize the overall accounting error within 5%.

[0013] Furthermore, in step S5, the three-tiered potential assessment specifically includes: Current Status and Inherent Potential: Based on existing vegetation cover, water environment quality, and biological community structure, calculate the upper limit of steady-state annual carbon sequestration in the sea area without human intervention; Ecological restoration potential enhancement: Based on vegetation replanting coverage, water quality improvement, and ecological management efforts, a restoration-carbon increment response model is constructed to quantify the carbon sequestration enhancement potential brought about by artificial restoration. Medium- to long-term sustainable potential: Based on multi-year time-series carbon sink data, combined with climate warming scenarios, ecological protection plans, and human activity constraints, we iteratively predict the dynamic changes in carbon sinks over the next 5-20 years. By combining multiple indicators such as average carbon sink flux, potential increment, stability coefficient, and anti-interference ability, the carbon sink potential is divided into four levels: excellent, good, medium, and poor.

[0014] Furthermore, in step S6, the visualization output includes a spatiotemporal heat map of the entire carbon sink and a zoning distribution map of the potential level of ecological scenarios, supporting dynamic playback of time series at multiple scales such as daily, monthly, quarterly, and annual. Data storage adopts a hierarchical encrypted archiving mode for raw data, preprocessing parameters, accounting model coefficients, and result data. Through the linkage between the time series database and the spatial database, data traceability, historical backtracking, access control, and hourly dynamic updates are realized.

[0015] This invention also provides a system for accurate calculation and potential assessment of marine carbon sink flux, used to implement the above method, including: The data acquisition and preprocessing module is used to build an integrated air-water-shore acquisition system, access multi-source marine monitoring data, and complete anomaly removal, wavelet denoising, spatiotemporal interpolation, resolution and dimension unification preprocessing to output a standardized basic dataset. The ecological scene intelligent segmentation module is used to construct a four-dimensional clustering feature vector. By improving the K-means clustering algorithm, it completes the fine-grained segmentation of four types of marine ecological scenes and generates spatial segmentation results with attributes and codes. The scenario-specific carbon sequestration accounting module is used to match the exclusive dynamic accounting model for each ecological scenario, and dynamically update parameters based on real-time monitoring data to complete multi-scale carbon sequestration flux accounting and global flux overlay. The spatiotemporal error correction module is used to construct a spatiotemporal dual-dimensional correction model, generate dynamic correction coefficients, and complete high-precision error correction of the initial carbon sink flux. The multi-dimensional potential assessment module is used to calculate the carbon sink potential at three levels: current status, restoration and improvement, and medium- and long-term sustainability, and to complete the hierarchical assessment of potential levels. The visualization and dynamic storage module is used to realize the visualization rendering and time-series dynamic display of carbon sequestration results, while completing the hierarchical encrypted storage, dynamic iterative update and data traceability management of the entire process data.

[0016] The advantages of this invention compared to the prior art are: This invention constructs an integrated multi-source data acquisition and standardized preprocessing system encompassing air, water, and shore, integrating four types of core marine monitoring data. Through wavelet denoising, spatiotemporal joint interpolation, and unified resolution and dimensional processing, it solves the problems of traditional data being singular, lacking completeness, and having low standardization, providing a high-quality data foundation for high-precision carbon sink accounting.

[0017] This invention employs an improved K-means clustering algorithm with spatial neighborhood constraints to achieve intelligent and refined zoning of marine ecological scenarios. It overcomes the shortcomings of traditional manual zoning, such as strong subjectivity, fragmented zoning, and chaotic boundaries, ensuring that the zoning space is contiguous and the boundaries are regular, providing accurate scenario support for scenario-based differentiated accounting.

[0018] This invention constructs a dedicated dynamic accounting model for four types of marine ecological scenarios, abandoning fixed empirical coefficients and updating all parameters dynamically in real time. It accurately matches the carbon sequestration, loss, and exchange mechanisms of different ecosystems. Compared with traditional general accounting models, it has stronger scenario adaptability, higher accounting accuracy, and is more in line with the dynamic changes of marine carbon sinks.

[0019] This invention employs a spatiotemporal dual-dimensional coupled error correction system, comprehensively covering multi-dimensional interference factors such as tides, temperature differences, precipitation, ocean currents, and human activities, to achieve dual error correction in both time and space, thereby stabilizing the accounting error within 5% and significantly improving the accuracy and reliability of marine carbon sink accounting.

[0020] This invention constructs a three-tiered carbon sink potential assessment system that takes into account the existing inherent capacity, artificial restoration potential, and medium- to long-term sustainable potential. It combines multi-indicator weighted scoring to achieve quantitative classification of potential levels, solving the problems of single assessment dimensions and one-sided results in traditional assessments. The assessment results are more comprehensive and scientific, and are adapted to the needs of long-term marine ecological management and carbon sink resource development. Attached Figure Description

[0021] Figure 1 This is a flowchart of a method for accurate calculation and potential assessment of marine carbon sink flux according to the present invention.

[0022] Figure 2 This is a system block diagram of a marine carbon flux accurate accounting and potential assessment system according to the present invention.

[0023] Figure 3 This is a flowchart of the integrated air-water-shore data preprocessing process.

[0024] Figure 4 This is an improved flowchart for dividing the K-means ecosystem. Detailed Implementation

[0025] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present invention.

[0026] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0027] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0028] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0029] The following detailed description, in conjunction with the accompanying drawings, provides a method and system for accurate calculation and potential assessment of marine carbon sink flux according to the present invention.

[0030] Combined with appendix Figure 1-4 The specific implementation process of the method and system for accurate calculation and potential assessment of marine carbon sink flux of the present invention is as follows: A method for accurate accounting and potential assessment of marine carbon sink fluxes includes the following steps: S1. Real-time acquisition and preprocessing of multi-source heterogeneous marine monitoring data An integrated air-water-shore data acquisition system was established, simultaneously accessing four types of core monitoring data: satellite remote sensing data, underwater real-time sensor monitoring data, shore-based meteorological time-series data, and marine static ecological topography data. Specifically, the satellite remote sensing data includes monthly vegetation cover, effective water area, and suspended particulate matter concentration in the target sea area; the underwater real-time sensor monitoring data includes seawater temperature, salinity, dissolved oxygen concentration, plankton abundance, and dissolved inorganic carbon dioxide concentration; the shore-based meteorological time-series data includes wind speed, wind direction, cumulative rainfall, and sunshine duration; and the marine static ecological topography data includes seabed elevation, seabed type, native vegetation distribution type, and coastal human development intensity.

[0031] The accessed multi-source raw data are subjected to standardized preprocessing in sequence, including outlier removal, discrete wavelet transform denoising, spatiotemporal joint interpolation completion, and spatiotemporal resolution and dimension unification processing, and finally a standardized marine carbon sink basic dataset is generated.

[0032] During the outlier removal phase, reasonable threshold ranges for each monitored physical quantity are pre-set. The monitored value will exceed this range. Abnormal samples are identified and removed directly, eliminating invalid data caused by human error or equipment malfunction.

[0033] In the discrete wavelet transform denoising stage, the original time-series observation signal X(t) is decomposed into multi-scale hierarchical components, and the decomposition formula is as follows: ; in, It is a low-frequency approximation component used to preserve the core trends of carbon sink time-series changes; The high-frequency detail components at each layer mainly include random noise caused by sensor jitter and short-term water flow disturbances. By applying threshold filtering to the high-frequency components, invalid high-frequency noise is filtered out, and the denoised signal is reconstructed. While eliminating noise interference, it fully preserves the true temporal variation characteristics of carbon sinks.

[0034] In the spatiotemporal joint interpolation completion stage, to address data loss issues caused by cloud cover, equipment disconnection, and signal interference, a spatiotemporal weighted interpolation algorithm is employed to achieve high-precision restoration of missing data. The formula for calculating the completed value of the grid point to be estimated is as follows: ; in, The spatial weights decrease as the spatial Euclidean distance increases. The time weighting decreases as the time interval increases. The effective monitoring samples around the grid to be estimated are used. This algorithm simultaneously fuses spatially adjacent point data with continuous temporal data from previous and subsequent times, achieving higher data reconstruction accuracy compared to simple temporal or spatial interpolation.

[0035] After interpolation, all data are mapped to a unified standard of 30m spatial grid and 1h time step. The physical quantities of monitoring with different dimensions and scales are normalized to generate a standardized marine carbon sink basic dataset, which provides standardized data support for subsequent ecological scenario division and carbon sink flux accounting.

[0036] S2, Intelligent and Refined Division of Marine Ecological Scenes A four-dimensional clustering feature vector containing vegetation type, seawater depth, seabed sediment type, and carbon sink carrier density is constructed. An improved K-means unsupervised clustering algorithm with spatial distance constraint weights is adopted to achieve fully automatic and refined division of marine ecological scenes.

[0037] First, construct the four-dimensional feature vector of each grid cell. To eliminate the influence of differences in the dimensions of features, min-max normalization is performed on the feature data of each dimension. The normalization formula is as follows: ; Secondly, cluster centers are initialized for four types of carbon sink scenarios, corresponding to mangrove carbon sink areas, seaweed bed carbon sink areas, tidal flat wetland carbon sink areas, and open nearshore planktonic carbon sink areas, respectively. To overcome the shortcomings of traditional K-means clustering, which relies solely on feature distance, is prone to fragmented and isolated partitions, and has poor spatial continuity, this invention adds a spatial neighborhood constraint term to construct a grid. To the cluster center The formula for the combined distance is: ; in, The Euclidean distance is the normalized feature vector. For spatial neighborhood penalty terms, when the grid Most of the neighboring grids do not belong to the category hour, Increasing the value suppresses the generation of scattered partitions; This is the spatial constraint weight, used to balance the influence weights of feature distance and spatial constraints.

[0038] Minimize the total distance of the global grid through iteration. The cluster centers are continuously updated until they converge and the partitioning results stabilize, completing the fully automated partitioning of the marine ecological scene. After partitioning, a unique scene code is assigned to each grid. Simultaneously calculate the area of ​​each ecological zone. Core attribute parameters such as carrier density and basic carbon conversion efficiency are used to generate refined ecological zoning results with unique codes, precise spatial boundaries and complete attributes, which are then sent to the scenario-specific carbon sink accounting module.

[0039] S3, Dynamic Carbon Sequestration Calculation by Scenario To address the differentiated carbon sequestration mechanisms across four ecological scenarios, a dedicated dynamic accounting model is constructed for each scenario. This model abandons traditional fixed empirical coefficients, and all core parameters are dynamically updated based on S1 standardized real-time monitoring data. Instantaneous carbon flux is calculated grid-by-grid and hourly for each scenario. Through temporal integration and spatial overlay, the initial carbon flux for the entire target sea area is obtained. This invention defines the instantaneous carbon flux per grid as… The unit is gC·m -2 ·h -1 The specific accounting models for each scenario are as follows: (1) Mangrove and tidal wetland carbon sink areas (S=1): The total net carbon sink flux is calculated using a coupled model of three mechanisms: vegetation photosynthesis, soil deposition, and sea-air interface carbon exchange. The formula is as follows: ; in, The potential photosynthetic rate of vegetation is dynamically driven by real-time sunshine duration, sea temperature, and vegetation cover. The conversion coefficient of photosynthetic carbon to biomass; The soil carbon deposition rate dynamically changes with the duration of tidal inundation and the amount of vegetation litter input. The net carbon exchange flux at the ocean-air interface is calculated in real time from the CO2 concentration difference between the water body and the atmosphere and the gas exchange coefficient.

[0040] (2) Carbon sink zone of seaweed bed (S=2): Considering the dynamic balance between carbon fixation by seaweed and carbon loss due to ocean current erosion, a growth-loss balance model is constructed, and the formula is: ; in, This represents the real-time growth rate of seaweed. The carbon enrichment and conversion coefficient; The carbon loss due to water scouring is positively correlated with local ocean current velocity, accurately characterizing the organic carbon loss caused by ocean current disturbance.

[0041] (3) Open nearshore planktonic carbon sink area (S=3): Based on the marine biological pump carbon sequestration mechanism, after deducting the carbon release from planktonic community respiration, a net carbon sequestration flux model is constructed, and the formula is: ; in, For planktonic abundance; The carbon sequestration efficiency per unit of phytoplankton; Carbon release from community respiration; It provides input flux for CO2 diffusion in water bodies, comprehensively covering the entire process mechanism of carbon sequestration, loss, and input in phytoplankton ecosystems.

[0042] For each grid across the entire domain, a corresponding dedicated accounting model is automatically matched based on the scenario code to obtain the instantaneous carbon sink flux F of a single grid. The daily cumulative carbon sink flux is then calculated through time-series integration. Monthly cumulative carbon flux The calculation formula is: ; The initial carbon sink flux of the entire target sea area is obtained by area-weighted superposition of fluxes from all grids. The calculation formula is: ; in, For the area of ​​a single grid, This represents the instantaneous carbon sink flux of each grid in the corresponding scenario.

[0043] S4. Spatiotemporal dynamic error correction of carbon sequestration flux A spatiotemporal coupled error correction model is constructed. Based on four core disturbance factors—tidal cycle, seasonal temperature difference, ocean current disturbance, and human activities—time-series dynamic correction coefficients and spatially differentiated correction coefficients are generated respectively. This model performs dual-weighted correction on the initial carbon sink flux across the entire region, eliminating systematic and random errors to obtain high-precision real-time carbon sink flux. The core correction formula is: ; in, This is the time correction factor. This is the spatial correction factor.

[0044] Time correction factor Integrating three types of disturbance factors—daily-scale tidal cycle fluctuations, seasonal-scale water temperature deviations, and monthly precipitation disturbances—this method is used to correct for fluctuations in intertidal carbon sequestration efficiency, the impact of temperature on biological carbon sequestration rates, and aquatic carbon environment deviations caused by freshwater input. The calculation formula is as follows: ; in, It is a tidal cycle correction factor that changes dynamically with tide level and submersion-exposure status. This is the temperature deviation correction factor, calculated from the ratio of real-time sea temperature to the multi-year average sea temperature; As a precipitation disturbance factor, it characterizes the disturbance effect of freshwater input on the carbon cycle environment of aquatic bodies; For the corresponding weight coefficients, satisfying .

[0045] Spatial Differentiation Correction Coefficient Integrating three types of indicators—ocean current velocity, land-sea locational differences, and intensity of coastal human development—this method is used to correct for biases in water carbon transport, differences in the land-sea ecological base, and the weakening impact of anthropogenic carbon sequestration. The calculation formula is as follows: ; In the formula, This is a correction factor for ocean current transport. For land and sea location factors; Human activities are a source of disturbance; the higher the intensity of human activities such as aquaculture, sewage discharge, and land reclamation, the more significant the impact. The smaller the value, the more accurately it characterizes the weakening effect of human interference on carbon sequestration; For the corresponding weight coefficients, satisfying .

[0046] By using spatiotemporal dual-dimensional coupling correction, the accounting deviations caused by natural environment and human interference can be completely eliminated, and the overall carbon sequestration accounting error can be stably controlled within 5%, greatly improving the accuracy of the accounting results.

[0047] S5. Multi-dimensional tiered assessment of carbon sequestration potential Based on the corrected high-precision carbon sink flux A three-tiered potential assessment model was constructed to quantify the inherent potential of the current marine environment, the potential for ecological restoration and improvement, and the potential for medium- and long-term sustainability. A multi-indicator comprehensive scoring system was used to complete the four-level classification of carbon sink potential, thereby achieving a comprehensive and multi-level assessment of carbon sink potential.

[0048] Current inherent carbon sequestration potential The average annual carbon sequestration capacity, representing the long-term stable output of the marine area based on existing vegetation cover, water quality, and biological community structure without any human intervention, is calculated using the following formula: ; in, This is the multi-year average corrected carbon flux; The effective carbon sink area of ​​the sea area; It is the ecosystem stability coefficient, used to reflect the ability of the marine ecological environment to resist disturbance.

[0049] Ecological restoration enhances potential The carbon sequestration increment in the sea area after artificial restoration measures such as vegetation replanting, water quality improvement, and ecological management is represented by the following formula: ; in, To simulate carbon sequestration capacity under restoration scenarios, the carbon sequestration potential of artificial ecological restoration is precisely quantified by substituting parameters such as the increase in vegetation coverage, water quality improvement level, and ecological management intensity into the corresponding scenario calculation model.

[0050] Medium- to long-term sustainable carbon sink potential This method is used to predict the dynamic changes in carbon sinks in marine areas over the next 5-20 years. It is based on a multivariate time-series prediction model, and the formula is as follows: ; in, Historical carbon sequestration time series data; Input data for future climate scenarios; As a constraint on ecological protection planning; This serves as a parameter for controlling the intensity of human activities, enabling dynamic prediction of medium- and long-term carbon sequestration potential under the coupling of multiple factors.

[0051] A multi-indicator weighted comprehensive scoring system is constructed to quantitatively rate the carbon sequestration potential of marine areas. The comprehensive score calculation formula is as follows: ; in, This represents the theoretical maximum carbon flux in this sea area; As a measure of future ecological risk; , , , These are the weighting coefficients for each indicator. Based on the threshold range of the comprehensive score, the marine carbon sequestration potential is divided into four levels: excellent, good, medium, and poor, completing a stratified quantitative assessment.

[0052] S6. Visualized Output and Dynamic Closed-Loop Storage of Results With gridded high-precision carbon sink flux Using the potential comprehensive score as the core data, spatial raster rendering technology is used to generate a spatiotemporal heat map of carbon sinks across the entire region and a regional distribution map of potential levels in ecological scenarios. It supports dynamic playback and visualization of time series at multiple time scales, including daily, monthly, quarterly, and annual, and intuitively presents the spatiotemporal distribution characteristics and potential level differences of marine carbon sinks.

[0053] A spatial-temporal joint database was built, employing a hierarchical encryption and categorized archiving storage model to achieve closed-loop storage of data throughout the entire process. The stored content covers all dimensions of results, including raw monitoring data, preprocessing intermediate parameters, calculation model coefficients, initial carbon sink flux, corrected accurate carbon sink flux, and three-tiered carbon sink potential data. Hierarchical access control was also implemented to ensure data security and confidentiality.

[0054] The system automatically iterates and executes the entire process of S1-S5 in a 1-hour cycle, realizing the dynamic updating of monitoring data, accounting parameters, and evaluation results. It also supports historical data backtracking and full-process data traceability, providing complete, accurate, and dynamic data support for marine carbon sink measurement, ecological restoration effectiveness assessment, and blue carbon sink development and utilization.

[0055] Secondly, the present invention provides a system for accurate calculation and potential assessment of marine carbon flux, used to implement the aforementioned method for accurate calculation and potential assessment of marine carbon flux, including: The data acquisition and preprocessing module is used to build an integrated air-water-shore acquisition system, connecting with four types of data sources: satellite remote sensing, underwater sensing, shore-based meteorology, and ecological topography, to complete the real-time access of multi-source heterogeneous data; through outlier removal, discrete wavelet denoising, spatiotemporal joint interpolation completion, and spatiotemporal resolution and dimension unification preprocessing, it outputs a standardized marine carbon sink basic dataset, providing standardized data support for subsequent accounting and assessment.

[0056] The intelligent ecological scene segmentation module is used to construct a four-dimensional clustering feature vector based on vegetation type, water depth, substrate type, and carbon sink carrier density. Through an improved K-means clustering algorithm with spatial constraints, it completes the refined and automated segmentation of four types of marine ecological scenes, generating ecological segmentation results with unique codes, accurate spatial boundaries, and complete attribute information, thus realizing the intelligent segmentation of differentiated carbon sink accounting scenarios.

[0057] The scenario-specific carbon sequestration accounting module is used to automatically match the exclusive dynamic accounting model for each ecological scenario based on the grid scenario code. It dynamically updates the core parameters of the model based on real-time standardized monitoring data, and completes the refined accounting of hourly instantaneous carbon sequestration flux and multi-time period cumulative carbon sequestration flux. The initial carbon sequestration flux of the entire target sea area is obtained through spatial superposition.

[0058] The spatiotemporal error correction module is used to construct a spatiotemporal dual-dimensional coupled error correction model. Based on natural environmental and human interference factors, it generates time-series and spatial dynamic correction coefficients respectively, performs dual weighted correction on the initial carbon sink flux, eliminates the accounting errors caused by various spatiotemporal interferences, and outputs high-precision real-time carbon sink flux data.

[0059] The multi-dimensional potential assessment module is used to calculate the inherent carbon sink potential of the current marine area, the carbon sink potential enhanced by ecological restoration, and the medium- and long-term sustainable carbon sink potential based on high-precision carbon sink flux data. It completes the four-level classification of carbon sink potential through a multi-indicator weighted scoring system, realizing a comprehensive, hierarchical, and quantitative assessment.

[0060] The visualization and dynamic storage module is used to visualize and render the results of carbon sink flux and potential level, generate multi-scale spatiotemporal distribution maps and support time-series dynamic display; at the same time, through the spatiotemporal linkage database, it completes the full-process data hierarchical encryption archiving and hourly dynamic iterative updates, realizes data traceability, historical backtracking and access control, and ensures the integrity, security and timeliness of the results data.

[0061] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for accurate accounting and potential assessment of marine carbon sink flux, characterized in that, Includes the following steps: S1. Establish an integrated air-water-shore data acquisition system. This system is used to acquire data from satellites, underwater sensors, and shore-based monitoring stations. It simultaneously receives satellite remote sensing data, underwater real-time monitoring data, and shore-based meteorological time-series data from this system, and then imports static marine ecological topography data to obtain four types of core data. The data is then processed sequentially through outlier removal, discrete wavelet transform denoising, spatiotemporal joint interpolation completion, and spatiotemporal resolution and dimension unification to generate a standardized marine carbon sink basic dataset. S2. Construct a four-dimensional clustering feature vector that includes vegetation type, seawater depth, seabed sediment type, and carbon sink carrier density. An improved K-means unsupervised clustering algorithm is adopted, which introduces spatial neighborhood distance constraint weights and superimposes a spatial neighborhood penalty term when calculating the distance between grid cells and cluster centers to suppress scattered and isolated partitions. The sea area is automatically divided into four ecological scenarios: mangrove carbon sink area, seaweed bed carbon sink area, tidal wetland carbon sink area, and open nearshore planktonic carbon sink area. Refined ecological partitioning results with coding, boundaries, and attributes are generated. S3. For the differentiated carbon sequestration mechanisms of the four types of ecological scenarios, a dedicated dynamic accounting model is constructed to adapt to the carbon sequestration mechanism of each ecological scenario and can dynamically update the internal parameters based on real-time monitoring data. The core parameters of the model are dynamically updated based on the pre-processed real-time monitoring data to calculate the instantaneous carbon sink flux of each scenario on an hourly basis. The daily and monthly cumulative carbon sink flux is obtained by time-series integration and accumulation. After spatial superposition, the initial carbon sink flux of the entire target sea area is obtained. S4. Construct a spatiotemporal coupled error correction model. Based on four major disturbance factors—tidal cycle, seasonal temperature difference, ocean current disturbance, and human activity—tidal cycle and seasonal temperature difference are classified as temporal disturbance factors to generate temporal dynamic correction coefficients, while ocean current disturbance and human activity are classified as spatial disturbance factors to generate spatial differential correction coefficients. The temporal dynamic correction coefficients are calculated by combining tidal cycle fluctuations and seasonal temperature differences, while the spatial differential correction coefficients are calculated by combining ocean current disturbance and human activity intensity. A weighted coupling method of multiplying the temporal dynamic correction coefficients and spatial differential correction coefficients is adopted to generate the temporal dynamic correction coefficients and spatial differential correction coefficients respectively. The initial carbon sink flux of the entire region is weighted and corrected to obtain a high-precision real-time carbon sink flux. S5. Based on the corrected accurate carbon sink flux, a three-level stratified potential assessment model is constructed to quantitatively measure the inherent potential of the current marine area, the potential for ecological restoration and improvement, and the medium- and long-term sustainable potential. The carbon sink potential is classified into four levels through a multi-indicator comprehensive scoring system. S6 generates spatiotemporal heat maps and potential level zoning distribution maps based on gridded carbon sink data, supporting multi-scale time-series dynamic display; adopts a spatiotemporal database linkage architecture with hierarchical encryption and mutual calling and linkage between time-series databases and spatial databases to classify and store data throughout the entire process, and realizes dynamic iterative updates of data and accounting results on an hourly basis.

2. The method for accurate calculation and potential assessment of marine carbon sink flux according to claim 1, characterized in that: The four types of core data specifically include: Satellite remote sensing data includes monthly vegetation coverage, effective water area, and concentration of suspended particulate matter in the target sea area; Underwater sensors monitor real-time data including seawater temperature, salinity, dissolved oxygen concentration, plankton abundance, and dissolved inorganic carbon dioxide concentration in the water. Coastal meteorological time-series data, including wind speed, wind direction, cumulative rainfall, and sunshine duration; Static marine ecological topographic data includes seabed elevation, seabed type, native vegetation distribution type, and intensity of human development along the coast.

3. The method for accurate calculation and potential assessment of marine carbon sink flux according to claim 1, characterized in that: In step S1, the discrete wavelet transform noise reduction adopts a multi-layer signal hierarchical decomposition and threshold screening reconstruction method to extract the low-frequency effective feature components of the data, filter out high-frequency random noise generated by sensor jitter and short-term water flow disturbance, and retain the core trend of carbon sink time series changes; the spatial weight of the spatiotemporal joint interpolation completion is assigned according to the similarity of sea area distance, and the temporal weight is assigned according to the time interval decay, and the missing data is restored with high precision by combining spatially adjacent point data and consecutive temporal data.

4. The method for accurate calculation and potential assessment of marine carbon sink flux according to claim 1, characterized in that: In step S2, the execution flow of the improved K-means unsupervised clustering algorithm includes: The four-dimensional clustering feature vectors are normalized to eliminate dimensional differences; four cluster centers corresponding to different carbon sink scenarios are initialized; the Euclidean distance between each grid cell and the cluster center is calculated iteratively, and the cluster centers are updated by integrating spatial neighborhood constraint weights; until the cluster centers converge and the partitioning results are stable, the ecological scenario is automatically partitioned, avoiding scattered and isolated partitions and ensuring that the partition space is contiguous and the boundaries are regular.

5. The method for accurate calculation and potential assessment of marine carbon sink flux according to claim 1, characterized in that: In step S3, each of the four types of ecological scenarios corresponds to a dedicated dynamic accounting model. The specific dedicated dynamic accounting model for each ecological scenario is as follows: For mangrove carbon sink areas and tidal wetland carbon sink areas, a dynamic flux accounting model coupling vegetation photosynthesis, soil deposition and water interface carbon exchange is constructed. The comprehensive carbon sink flux is dynamically calculated based on sunshine duration, water temperature, vegetation coverage, substrate type, tidal inundation duration and sea-air carbon concentration gradient. Constructing growth zones for seaweed beds as carbon sink areas Enrichment The loss dynamic balance accounting model is based on the real-time growth rate of seaweed and combined with the water body carbon enrichment coefficient and water body scour loss coefficient to dynamically calculate the effective net carbon flux. For open nearshore planktonic carbon sink areas, a dynamic accounting model for planktonic communities was constructed. Based on planktonic abundance, community carbon sequestration efficiency, carbon dioxide diffusion flux, and vertical transport efficiency of biological pumps, the net carbon sequestration flux of the water body was obtained by deducting the carbon release from biological respiration.

6. The method for accurate calculation and potential assessment of marine carbon sink flux according to claim 1, characterized in that: In step S4, the temporal dynamic correction coefficient further integrates three types of interference factors—daily-scale tidal cycle fluctuations, seasonal-scale water temperature deviations, and monthly precipitation disturbances—on the basis of tidal cycles and seasonal temperature differences, respectively correcting for fluctuations in intertidal carbon sequestration efficiency, the impact of temperature on biological carbon sequestration rates, and aquatic carbon environment deviations caused by freshwater input. The spatial differentiation correction coefficient further integrates three types of indicators—ocean current velocity, land-sea location differences, and coastal human development intensity—on the basis of ocean current disturbances and human activities, correcting for aquatic carbon transport deviations, differences in land-sea ecological bases, and the weakening effect of carbon sequestration by human activities.

7. The method for accurate calculation and potential assessment of marine carbon sink flux according to claim 1, characterized in that: Step S5, the three-tiered potential assessment specifically includes: Current Status and Inherent Potential: Based on existing vegetation cover, water environment quality, and biological community structure, calculate the upper limit of steady-state annual carbon sequestration in the sea area without human intervention; Ecological restoration potential enhancement: Based on vegetation replanting coverage, water quality improvement, and ecological management efforts, a restoration-carbon increment response model is constructed to quantify the carbon sequestration enhancement potential brought about by artificial restoration. Medium- to long-term sustainable potential: Based on multi-year time-series carbon sink data, combined with climate warming scenarios, ecological protection plans, and human activity constraints, we iteratively predict the dynamic changes in carbon sinks over the next 5-20 years. By combining multiple indicators such as average carbon sink flux, potential increment, stability coefficient, and anti-interference ability, the carbon sink potential is divided into four levels: excellent, good, medium, and poor.

8. The method for accurate calculation and potential assessment of marine carbon sink flux according to claim 1, characterized in that: In step S6, the visualization output includes a spatiotemporal heat map of carbon sequestration across the entire region and a zoning distribution map of ecological scenario potential levels, supporting dynamic playback of time series at multiple scales such as daily, monthly, quarterly, and annual. The data storage adopts a hierarchical and encrypted archiving mode for raw data, preprocessing parameters, accounting model coefficients, and result data. Through the linkage of time-series database and spatial database, it realizes data traceability, historical backtracking, access control, and hourly dynamic updates.

9. A system for accurate calculation and potential assessment of marine carbon sink flux, characterized in that, To implement the method according to any one of claims 1-8, comprising: The data acquisition and preprocessing module is used to build an integrated air-water-shore acquisition system, access multi-source marine monitoring data, and complete anomaly removal, wavelet denoising, spatiotemporal interpolation, resolution and dimension unification preprocessing to output a standardized basic dataset. The ecological scene intelligent segmentation module is used to construct a four-dimensional clustering feature vector. By improving the K-means clustering algorithm, it completes the fine-grained segmentation of four types of marine ecological scenes and generates spatial segmentation results with attributes and codes. The scenario-specific carbon sequestration accounting module is used to match the exclusive dynamic accounting model for each ecological scenario, and dynamically update parameters based on real-time monitoring data to complete multi-scale carbon sequestration flux accounting and global flux overlay. The spatiotemporal error correction module is used to construct a spatiotemporal dual-dimensional correction model, generate dynamic correction coefficients, and complete high-precision error correction of the initial carbon sink flux. The multi-dimensional potential assessment module is used to calculate the carbon sink potential at three levels: current status, restoration and improvement, and medium- and long-term sustainability, and to complete the hierarchical assessment of potential levels. The visualization and dynamic storage module is used to realize the visualization rendering and time-series dynamic display of carbon sequestration results, while completing the hierarchical encrypted storage, dynamic iterative update and data traceability management of the entire process data.

Images