Method for dynamic tracking and prediction of carbon storage in coastal zone

Abstract

The present application relates to carbon storage monitoring technical field, specifically to a coastal carbon storage dynamic tracking and prediction method, comprising the following steps: S1, collecting multi-source monitoring data of coastal area and performing standardization alignment, constructing ecological factor cross-scale projection atlas; S2, identifying carbon storage jump events caused by vegetation degradation rate mutation, sediment carbon rebound or terrain sudden drop in the atlas, extracting corresponding micro-terrain disturbance structure and ecological disturbance factor combination, and constructing event structure trigger set; S3, inputting the trigger set into carbon storage structure transition prediction model, fusing ecological function area disturbance response parameters, and outputting carbon storage diffusion main path and jump section dynamic tracking graph in future period. The present application realizes accurate identification, dynamic diffusion path tracking and risk grade evaluation of coastal carbon storage jump events, and significantly improves the scientificity and application value of carbon storage migration prediction.

Description

Technical Field

[0001] This invention relates to the field of carbon storage monitoring technology, and in particular to a method for dynamic tracking and prediction of carbon storage in coastal zones. Background Technology

[0002] With the advancement of global carbon peaking and carbon neutrality goals, coastal zones, as typical blue carbon ecosystems, have become important targets for ecological management and carbon storage potential assessment in coastal areas. In particular, ecological functional areas such as mangroves, salt marshes, and sandbars exhibit significant spatiotemporal evolution characteristics under the influence of multiple factors such as tides, topography, and sediment disturbance. How to accurately identify the dynamic changes in carbon storage in these areas has become a core technical issue for promoting coastal ecological restoration, carbon storage enhancement, and risk early warning management.

[0003] Existing methods for studying carbon storage in coastal zones mainly rely on periodic monitoring or single-factor statistics, lacking the ability to integrate spatiotemporally heterogeneous data from multiple sources and failing to effectively identify the temporal characteristics of abrupt changes in carbon storage behavior. At the same time, there is a lack of predictive modeling tools for carbon storage diffusion pathways and structural transition risks caused by disturbance factors. Especially in the context of frequent ecosystem disturbances, existing methods are unable to characterize the migration patterns of carbon storage within functional zones, nor can they dynamically track potential structural transition segments, thus limiting in-depth analysis of carbon storage change mechanisms and precise management of high-risk areas. Summary of the Invention

[0004] This invention provides a method for dynamic tracking and prediction of carbon storage in coastal zones. By constructing a cross-scale projection map of ecological factors based on a unified time reference axis, it identifies carbon storage jump events driven by typical disturbances, extracts the combination of micro-topographic disturbance structure and ecological disturbance factors, establishes an event structure trigger set, and integrates coastal functional zone response parameters and graph attention mechanism to predict the main carbon storage diffusion path and jump risk level in future cycles, providing precise support for ecological restoration zoned policies and blue carbon resource management.

[0005] A method for dynamic tracking and prediction of coastal carbon storage includes the following steps: S1. Collect multi-source monitoring data in the coastal zone, including vegetation carbon storage time series data, sediment carbon content change sequence and tidal topographic elevation change record. Asynchronously align multi-source monitoring data at different sampling scales based on a unified time reference axis to construct a cross-scale projection map of ecological factors. S2, in the cross-scale projection map of ecological factors, based on anomalous indicators such as sudden changes in vegetation degradation rate, rebound of sedimentary carbon bulk density, or sudden appearance of topographic depression areas, carbon storage jump events are identified, and the corresponding micro-topographic disturbance structures and combinations of ecological disturbance factors are extracted to construct event structure trigger sets. S3 inputs the event structure trigger set into the carbon storage structure transition prediction model, combines the disturbance response parameters of different ecological functional zones in the coastal zone, and outputs a dynamic tracking map of the main carbon storage diffusion path and structural transition segments in the future cycle.

[0006] Optionally, S1 includes: S11 collects multi-source monitoring data in the coastal zone, including time-series data of vegetation carbon storage, sequence of sediment carbon content changes, and records of tidal topographic elevation changes, and performs standardization processing on the collected multi-source monitoring data. S12, based on a unified time reference axis, asynchronously aligns standardized multi-source monitoring data at different sampling scales, and combines their temporal and spatial characteristics to construct a cross-scale projection map of ecological factors.

[0007] Optionally, S11 includes: S111 involves collecting multi-source monitoring data related to carbon storage within the coastal zone, specifically including: Time series data of vegetation carbon storage This includes aboveground / underground biomass and biocarbon content obtained from quadrat and sample grid monitoring; Sediment carbon content variation sequence This includes the dry weight, bulk density, and percentage of organic carbon in the substrate sample. Tidal topographic elevation change record This includes shoreline elevation data, tidal level records, and remote sensing DEM; S112, Constructing a unified time reference axis ,in, For the i-th time slice, n is the number of time slices in the unified time reference axis; S113 uses the Z-Score normalization method to standardize the collected multi-source monitoring data.

[0008] Optionally, S12 includes: S121 refers to standardized multi-source monitoring data, including vegetation carbon storage standard sequences. Sediment carbon content standard sequence Topographic elevation standard sequence The cubic spline interpolation method was used to interpolate the data to a unified time point. ; S122, on a unified time reference axis Above, in time slices Constructing an ecological factor map based on ,in, For a set of nodes, As an edge set, the connection weights of the edge set are defined based on the correlation between factors, ultimately yielding a cross-scale projection map of ecological factors. .

[0009] Optionally, S2 includes: S21, in the constructed cross-scale projection map of ecological factors In the middle, for each time slice map This study progressively analyzes the temporal abrupt change behavior of standardized ecological factor node values. By defining abrupt changes in vegetation degradation rate, rebounds in sedimentary carbon bulk density, or abnormal change patterns in topographic depressions as criteria for jump events, and combining this with a sliding window mechanism to calculate the change gradient between adjacent time slices, and setting threshold conditions, the study identifies potential carbon storage jump event time points. This forms a set of jump events. ; S22, for each identified jump event time point By retrospectively analyzing the changes in nodes of the cross-scale projection map of ecological factors before and after the occurrence, the corresponding vegetation and sediment disturbance characteristics were extracted, and the combination of ecological disturbance factors was determined. And by combining topographic elevation and slope changes, micro-topographic disturbance structures can be identified. By integrating hydrodynamic anomaly information, the structure is constructed in the form of a triplet. The event structure trigger set.

[0010] Optionally, S21 includes: S211, from the cross-scale projection map of ecological factors In the middle, for each time slice map Various standardized node values ​​were extracted to construct a time-varying sequence of ecological factors, including the vegetation carbon storage node value in the nth time slice. Node values ​​of sediment carbon content in the nth time slice Topographic elevation node values ​​in the nth time slice And generate factor time series, including time series of vegetation carbon storage node values. Time series of sediment carbon content node values Time series of tidal topographic elevation nodes ; S212 employs a sliding window mechanism to calculate the average rate of change (i.e., first-order difference) of the time series of each type of ecological factor within the window interval, which is used to measure the degree of ecological factor mutation. S213, Set thresholds for determining rate of change mutations, including thresholds for vegetation degradation rate mutations. Sediment carbon content rebound rate threshold Threshold for sudden drop in terrain elevation ,like If it is determined to be a candidate point for vegetation degradation jump, then it is considered as such. If it is determined to be a candidate point for sediment carbon bounce jump, then it is considered a candidate point. If it is determined to be a candidate point for a topographic depression jump, or a candidate point for a vegetation degradation jump, a candidate point for a sediment carbon rebound jump, or a candidate point for a topographic depression jump, then the time point is considered to be... Add potential jump event points to the jump event set. .

[0011] Optionally, S22 includes: S221, for the set of jump events any event time point in Set the forward backtracking window length to The backward observation window length is Extract the corresponding spectral node value trajectories within the time period before and after the event point, and construct a factor change path sequence, including the node value sequence of vegetation carbon storage. Node value sequence of sediment carbon content Node value sequence of terrain elevation ; S222: Extract the intensity of perturbation characteristic changes before and after the jump from the factor change path sequence, and construct causal triggering elements, including combinations of ecological perturbation factors. and micro-topographic disturbance structure Specifically, it includes: Ecological disturbance factor combination: The disturbance intensity is measured by the moving difference method, including the change in vegetation carbon storage corresponding to the jump event. Changes in sediment carbon content corresponding to jump events ,like If the vegetation degradation meets the disturbance criterion, then the vegetation degradation is considered to meet the disturbance criterion. If so, the sedimentary disturbance is considered to meet the disturbance determination criteria, where, , These are the threshold values ​​for vegetation carbon storage disturbance and sedimentary carbon content disturbance, respectively, representing the changes in vegetation degradation or sedimentary disturbance that meet the disturbance criteria. , Incorporating combinations of ecological disturbance factors ; Micro-topographic disturbance structure: Based on the nodal value sequence of topographic elevation, calculate the elevation change per unit time interval before and after a jump event. With multi-step moving average slope change rate If satisfied or If a sudden drop in terrain is detected at the current location, an anomalous tidal range term is introduced from the tidal data. ,like The presence of hydrodynamic disturbance was determined, among which, The threshold for terrain drop. The threshold for the average change in terrain slope. As the threshold for the magnitude of tidal anomalies, and based on the determination results of the presence of abrupt topographic drop structures and hydrodynamic disturbances at the current location, a micro-topographic disturbance structure corresponding to the jump event is constructed. ; S223 combines each jump event point with its corresponding ecological disturbance factor. and micro-topographic disturbance structure The event set is constructed by combining the three elements into a structure-triggered event set.

[0012] Optionally, S3 includes: S31, based on the constructed structure-triggered event set triplet, combined with the division results of different ecological functional zones (salt marsh, mangrove, sandbar) of the coastal zone, introduces the disturbance response parameter matrix, establishes a carbon storage structure transition prediction model, integrates the temporal sequence of jump events, identifies the potential path evolution trend of carbon storage spatial migration after disturbance triggering, and outputs a carbon storage transition response probability map. S32 integrates the output of the carbon storage structure transition prediction model with the functional area topographic evolution map and historical evolution trajectory map to simulate the carbon storage migration trend of each disturbance source area within a specified future period. By dynamically tracking the carbon storage diffusion path caused by different disturbance types and superimposing the carbon storage spatial gradient change value, the main carbon storage diffusion path map and structural transition section distribution map are drawn, and the transition risk level assessment results are output.

[0013] Optionally, S31 includes: S311, Pre-structure Triggered Event Set Triple Each event is assigned to a corresponding set of coastal ecological functional zones based on its spatial location. Including salt marshes, mangroves, and sandbars, construct the disturbance response parameter matrix R; S312, using the time series of perturbation events Construct a directed transition graph for the node sequence , where the set of nodes For each jump event edge set This represents the perturbation propagation path established in chronological order, with each edge... The weight is ; S313 will have a directed transition diagram. By jointly embedding with the perturbation response parameter matrix R, a carbon storage transition response probability map is constructed. ,in, Functional area The probability of a carbon storage transition occurring in the future cycle. The probability output of carbon storage transitions is obtained by weighting through a graph attention mechanism.

[0014] Optionally, S32 includes: S321 outputs the probability of carbon storage transitions in each functional region. Regional topographic evolution map Historical carbon storage trajectory map Spatial matching was performed, and carbon storage distribution maps were overlaid. ; S322, Based on the perturbation type j and the carbon storage change value of each pixel in the fused map, a path tracking map induced by the perturbation is constructed; S323, through main path extraction and segmented analysis, plots the main paths of structural transitions and high-risk transition segments. The main path extraction employs the maximum cumulative response path algorithm, and the transition risk level is determined. Determined by weighting the transition probability with the magnitude of carbon storage change, when When, it indicates low risk, when When, it is indicated as medium risk, when When indicated, it signifies high risk.

[0015] The beneficial effects of this invention are: This invention, by constructing a cross-scale projection map of ecological factors under a unified time reference axis, achieves standardized fusion and dynamic alignment of data with different sampling frequencies, spatial scales, and monitoring types. It breaks through the limitations of traditional carbon storage assessment methods that rely on single factors or fixed-period data, and provides a data foundation for the accurate identification of anomalous jump events and spatial causal analysis.

[0016] This invention proposes an event structure trigger set construction mechanism, which combines the mutation signals of multi-source factors such as vegetation degradation, sediment rebound and topographic disturbance with their corresponding micro-topographic disturbance structures and ecological disturbance factors to form a triplet representation. Furthermore, it introduces a disturbance response parameter matrix and a directed graph of structural transitions, thereby realizing the graphical modeling of the carbon storage response mechanism in coastal ecological functional zones and improving the structural accuracy and interpretability of carbon storage transition trend prediction.

[0017] This invention, by integrating topographic evolution maps, historical trajectory maps, and predicted transition results, constructs a carbon storage and diffusion path tensor and risk level assessment mechanism. It can dynamically simulate the disturbance propagation process and identify the main diffusion path and high-risk transition sections, providing a scientific basis and path visualization support for determining priority areas for coastal ecological restoration and maintaining carbon sink stability. It has important practical value for ecological protection and carbon management. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the prediction method flow according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the construction of a cross-scale map of ecological factors according to an embodiment of the present invention. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Those skilled in the art may employ other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0021] like Figures 1-2 As shown, the method for dynamic tracking and prediction of coastal carbon storage includes the following steps: S1. Collect multi-source monitoring data in the coastal zone, including vegetation carbon storage time series data, sediment carbon content change sequence and tidal topographic elevation change record. Asynchronously align multi-source monitoring data at different sampling scales based on a unified time reference axis to construct a cross-scale projection map of ecological factors. S2, in the cross-scale projection map of ecological factors, based on anomalous indicators such as sudden changes in vegetation degradation rate, rebound of sedimentary carbon bulk density, or sudden appearance of topographic depression areas, carbon storage jump events are identified, and the corresponding micro-topographic disturbance structures and combinations of ecological disturbance factors are extracted to construct event structure trigger sets. S3 inputs the event structure trigger set into the carbon storage structure transition prediction model, combines the disturbance response parameters of different ecological functional zones in the coastal zone, and outputs a dynamic tracking map of the main carbon storage diffusion path and structural transition segments in the future cycle.

[0022] S1 includes: S11 collects multi-source monitoring data in the coastal zone, including time-series data of vegetation carbon storage, sequence of sediment carbon content changes, and records of tidal topographic elevation changes. The collected multi-source monitoring data is then standardized to ensure that they are aligned on the same time reference axis. S12, based on a unified time reference axis, asynchronously aligns standardized multi-source monitoring data at different sampling scales, and combines their temporal and spatial characteristics to construct a cross-scale projection map of ecological factors.

[0023] S11 includes: S111 involves collecting multi-source monitoring data related to carbon storage within the coastal zone, specifically including: Time series data of vegetation carbon storage This includes aboveground / underground biomass and biocarbon content obtained from quadrat and sample grid monitoring; Sediment carbon content variation sequence This includes the dry weight, bulk density, and percentage of organic carbon in the substrate sample. Tidal topographic elevation change record This includes shoreline elevation data, tidal level records, and remote sensing DEM; S112, Constructing a unified time reference axis ,in, For the i-th time slice, n is the number of time slices in the unified time reference axis; S113, the Z-Score normalization method is used to standardize the collected multi-source monitoring data, which is expressed as follows: ; in, To standardize multi-source monitoring data, X represents the original multi-source monitoring data. This represents the average of multi-source monitoring data over the historical sampling period. This represents the standard deviation of multi-source monitoring data over the historical sampling period.

[0024] S12 includes: S121 refers to standardized multi-source monitoring data, including vegetation carbon storage standard sequences. Sediment carbon content standard sequence Topographic elevation standard sequence The cubic spline interpolation method was used to interpolate the data to a unified time point. , represented as: ; in, For the corresponding time point The interpolated estimate, These are the interpolation coefficients. Let q be the starting sampling time point of the q-th segment, where q is the order index of the interpolation term. To and The next adjacent original time point; S122, on a unified time reference axis Above, in time slices Constructing an ecological factor map based on ,in, A set of nodes, representing a point in time. Various ecological factors, , A standardized estimate of vegetation carbon storage at the nth unified time point. A standardized estimate of sediment carbon content at the nth unified time point. For the standardized estimate of tidal topographic elevation at the nth unified time point, As an edge set, the connection weights of the edge set are defined based on the correlation between factors, ultimately yielding a cross-scale projection map of ecological factors. , represented as: ; in, The first A cross-scale projection map of ecological factors for each time slice, where N is the number of time slices on a unified time axis; Connection weights are represented as: ; in, For time slices The correlation edge weights between factors i and j , They are time points respectively The values ​​of the i-th and j-th factors, , These are the means of the corresponding factors.

[0025] S2 includes: S21, in the constructed cross-scale projection map of ecological factors In the middle, for each time slice map This study progressively analyzes the temporal abrupt change behavior of standardized ecological factor node values. By defining abrupt changes in vegetation degradation rate, rebounds in sedimentary carbon bulk density, or abnormal change patterns in topographic depressions as criteria for jump events, and combining this with a sliding window mechanism to calculate the change gradient between adjacent time slices, and setting threshold conditions, the study identifies potential carbon storage jump event time points. This forms a set of jump events. ; S22, for each identified jump event time point By retrospectively analyzing the changes in nodes of the cross-scale projection map of ecological factors before and after the occurrence, the corresponding vegetation and sediment disturbance characteristics were extracted, and the combination of ecological disturbance factors was determined. And by combining topographic elevation and slope changes, micro-topographic disturbance structures can be identified. By integrating hydrodynamic anomaly information, the structure is constructed in the form of a triplet. The event structure trigger set.

[0026] S21 includes: S211, from the cross-scale projection map of ecological factors In the middle, for each time slice map Various standardized node values ​​were extracted to construct a time-varying sequence of ecological factors, including the vegetation carbon storage node value in the nth time slice. Node values ​​of sediment carbon content in the nth time slice Topographic elevation node values ​​in the nth time slice And generate factor time series, including time series of vegetation carbon storage node values. Time series of sediment carbon content node values Time series of tidal topographic elevation nodes , represented as: ; ; ; in, The first Vegetation carbon storage node values ​​in each time slice The first Node values ​​of sediment carbon content in each time slice The first Terrain elevation node values ​​in each time slice; S212 employs a sliding window mechanism to calculate the average rate of change (i.e., first-order difference) for the time series of each type of ecological factor within the window interval. This rate of change is used to measure the degree of ecological factor mutation and is expressed as follows: ; ; ; Where w is the length of the sliding window. , , These represent the average rates of change of vegetation carbon storage, sediment carbon content, and topographic elevation corresponding to the nth time slice, respectively. , , These are the corresponding values ​​for the start time slice of the window; S213, Set thresholds for determining rate of change mutations, including thresholds for vegetation degradation rate mutations. Sediment carbon content rebound rate threshold Threshold for sudden drop in terrain elevation ,like If it is determined to be a candidate point for vegetation degradation jump, then it is considered as such. If it is determined to be a candidate point for sediment carbon bounce jump, then it is considered a candidate point. If it is determined to be a candidate point for a topographic depression jump, or a candidate point for a vegetation degradation jump, a candidate point for a sediment carbon rebound jump, or a candidate point for a topographic depression jump, then the time point is considered to be... Add potential jump event points to the jump event set. ; ; ; ; in, , These represent the mean and standard deviation of the vegetation change rate within the sliding window, respectively. This is the abnormal offset coefficient. , These represent the mean and standard deviation of the sediment change rate within the sliding window, respectively. This is the rebound strength coefficient. , These represent the mean and standard deviation of the rate of change of elevation within the sliding window, respectively. This represents the sensitivity coefficient during a sudden drop.

[0027] S22 includes: S221, for the set of jump events any event time point in Set the forward backtracking window length to The backward observation window length is Extract the corresponding spectral node value trajectories within the time period before and after the event point, and construct a factor change path sequence, including the node value sequence of vegetation carbon storage. Node value sequence of sediment carbon content Node value sequence of terrain elevation , represented as: ; ; ; in, , , These are the time points of the jump event. The previous The corresponding factor node values ​​for each time slice, , , These are the time points of the jump event. After that The corresponding factor node values ​​for each time slice; S222: Extract the intensity of perturbation characteristic changes before and after the jump from the factor change path sequence, and construct causal triggering elements, including combinations of ecological perturbation factors. and micro-topographic disturbance structure Specifically, it includes: Ecological disturbance factor combination: The disturbance intensity is measured by the moving difference method, including the change in vegetation carbon storage corresponding to the jump event. Changes in sediment carbon content corresponding to jump events ,like If the vegetation degradation meets the disturbance criterion, then the vegetation degradation is considered to meet the disturbance criterion. If so, the sedimentary disturbance is considered to meet the disturbance determination criteria, where, , These are the threshold values ​​for vegetation carbon storage disturbance and sedimentary carbon content disturbance, respectively, representing the changes in vegetation degradation or sedimentary disturbance that meet the disturbance criteria. , Incorporating combinations of ecological disturbance factors , represented as: ; ; in, , These represent the vegetation carbon storage node values ​​at the time point of the i-th jump event and the time slice preceding it, respectively. , These are the sediment carbon content node values ​​for the i-th jump event time point and its previous time slice, respectively. ; ; in, , These represent the average vegetation carbon storage and sedimentary carbon content within the sliding time window, respectively. , These are the standard deviations of the corresponding indicators. , These are the corresponding threshold control coefficients; Micro-topographic disturbance structure: Based on the nodal value sequence of topographic elevation, calculate the elevation change per unit time interval before and after a jump event. With multi-step moving average slope change rate If satisfied or If a sudden drop in terrain is detected at the current location, an anomalous tidal range term is introduced from the tidal data. ,like The presence of hydrodynamic disturbance was determined, among which, The threshold for terrain drop. The threshold for the average change in terrain slope. As the threshold for the magnitude of tidal anomalies, and based on the determination results of the presence of abrupt topographic drop structures and hydrodynamic disturbances at the current location, a micro-topographic disturbance structure corresponding to the jump event is constructed. , represented as: ; ; in, The elevation value at the time of the jump event. Its elevation value at the previous time point. The number of jumps before the current jump event point Elevation values ​​at each time point; ; ; ; in, This represents the mean value of the topographic elevation change in the undisturbed region. The standard deviation of the topographic elevation variation. The coefficient for determining sudden drop. This represents the average historical rate of change of the landslide slope. The corresponding standard deviation is... This is the coefficient for determining slope changes. This represents the average rise and fall in historical tide data. The standard deviation of historical tide levels. This is a factor for determining tidal anomalies; ; in, This is the terrain descent disturbance substructure corresponding to the jump event. This is the hydrodynamic perturbation substructure corresponding to the jump event; S223 combines each jump event point with its corresponding ecological disturbance factor. and micro-topographic disturbance structure The event set is constructed by combining the three elements into a structure-triggered event set.

[0028] S3 includes: S31, based on the constructed structure-triggered event set triplet, combined with the division results of different ecological functional zones (salt marsh, mangrove, sandbar) of the coastal zone, introduces the disturbance response parameter matrix, establishes a carbon storage structure transition prediction model, integrates the temporal sequence of jump events, identifies the potential path evolution trend of carbon storage spatial migration after disturbance triggering, and outputs a carbon storage transition response probability map. S32 integrates the output of the carbon storage structure transition prediction model with the functional area topographic evolution map and historical evolution trajectory map to simulate the carbon storage migration trend of each disturbance source area within a specified future period. By dynamically tracking the carbon storage diffusion path caused by different disturbance types and superimposing the carbon storage spatial gradient change value, the main carbon storage diffusion path map and structural transition section distribution map are drawn, and the transition risk level assessment results are output.

[0029] S31 includes: S311, Pre-structure Triggered Event Set Triple Each event is assigned to a corresponding set of coastal ecological functional zones based on its spatial location. The disturbance response parameter matrix R is constructed, including salt marshes, mangroves, and sandbars, specifically including: (1) Constructing a sample set of disturbance response for functional zones: Collect data from historical ecological monitoring data for each functional zone. The response records before and after the disturbance event form sample pairs. ,in, This represents the factor response value in the t-th period before the disturbance. denoted as the factor response value in the t-th period after the disturbance, j is the disturbance type index (abrupt change in vegetation carbon storage, abrupt change in sediment carbon content, sudden drop in topographic elevation, dramatic change in slope, tidal anomaly, and enhanced scouring and deposition intensity), and k is the functional zone index; (2) Calculate the disturbance response intensity: For each disturbance sample set, the normalized change ratio is used as the response intensity. The disturbance response parameter matrix is ​​ultimately formed. , represented as: ; in, To prevent small constants with a denominator of 0, This represents the average response intensity of disturbance type j in functional region k; S312, using the time series of perturbation events Construct a directed transition graph for the node sequence , where the set of nodes For each jump event edge set This represents the perturbation propagation path established in chronological order, with each edge... The weight is , represented as: ; in, Cosine similarity among combinations of ecological disturbance factors. The similarity between micro-topographic disturbance structures. , This is the adjustment coefficient; S313 will have a directed transition diagram. By jointly embedding with the perturbation response parameter matrix R, a carbon storage transition response probability map is constructed. ,in, Functional area The probability of a carbon storage transition occurring in the future cycle. The probability output of carbon storage transitions is obtained through a weighted graph attention mechanism, and is expressed as: ; in, Let i be the attention weight of the perturbation factor i on functional region k. Here, M is the Sigmoid activation function, and M is the number of perturbation factors included in the perturbation factor combination.

[0030] S32 includes: S321 outputs the probability of carbon storage transitions in each functional region. Regional topographic evolution map Historical carbon storage trajectory map Spatial matching was performed, and carbon storage distribution maps were overlaid. , represented as: ; in, The value of the carbon storage fusion spectrum corresponding to functional region k. , These are the fusion weighting coefficients for the historical trajectory map and the carbon storage distribution map, respectively. S322, based on the perturbation type j and the carbon storage change value of each pixel in the fused map, a path tracking map induced by the perturbation is constructed, represented as: ; in, For the carbon storage-diffusion path tensor induced by perturbation type j in functional region k, at position... The value at that location, The transition probability modulation function is denoted by Sigmoid. For functional area k in position The gradient of the carbon storage distribution map at that location. The influence weighting coefficient for disturbance type j; S323, through main path extraction and segmented analysis, plots the main paths of structural transitions and high-risk transition segments. The main path extraction employs the maximum cumulative response path algorithm, and the transition risk level is determined. Determined by weighting the transition probability with the magnitude of carbon storage change, when When, it indicates low risk, when When, it is indicated as medium risk, when When this is the case, it indicates a high risk, as shown below: ; ; in, This represents the main carbon storage and diffusion pathway in functional region k. To disrupt the propagation path, , As a weighting factor, This represents the average carbon storage gradient value within the region.

[0031] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0032] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for dynamic tracking and prediction of coastal carbon storage, characterized in that, Includes the following steps: S1. Collect multi-source monitoring data in the coastal zone, including vegetation carbon storage time series data, sediment carbon content change sequence and tidal topographic elevation change record. Asynchronously align multi-source monitoring data at different sampling scales based on a unified time reference axis to construct a cross-scale projection map of ecological factors. S2, in the cross-scale projection map of ecological factors, based on anomalous indicators such as sudden changes in vegetation degradation rate, rebound of sedimentary carbon bulk density, or sudden appearance of topographic depression areas, carbon storage jump events are identified, and the corresponding micro-topographic disturbance structures and combinations of ecological disturbance factors are extracted to construct event structure trigger sets. S3 inputs the event structure trigger set into the carbon storage structure transition prediction model, combines the disturbance response parameters of different ecological functional zones in the coastal zone, and outputs a dynamic tracking map of the main carbon storage diffusion path and structural transition segments in the future cycle.

2. The method for dynamic tracking and prediction of coastal carbon storage according to claim 1, characterized in that, S1 includes: S11 collects multi-source monitoring data in the coastal zone, including time-series data of vegetation carbon storage, sequence of sediment carbon content changes, and records of tidal topographic elevation changes, and performs standardization processing on the collected multi-source monitoring data. S12, based on a unified time reference axis, asynchronously aligns standardized multi-source monitoring data at different sampling scales, and combines their temporal and spatial characteristics to construct a cross-scale projection map of ecological factors.

3. The method for dynamic tracking and prediction of coastal carbon storage according to claim 2, characterized in that, S11 includes: S111 involves collecting multi-source monitoring data related to carbon storage within the coastal zone, specifically including: Time series data of vegetation carbon storage This includes aboveground / underground biomass and biocarbon content obtained from quadrat and quadrat monitoring; Sediment carbon content variation sequence This includes the dry weight, bulk density, and percentage of organic carbon in the substrate sample. Tidal topographic elevation change record This includes shoreline elevation data, tidal level records, and remote sensing DEM; S112, Constructing a unified time reference axis ,in, For the i-th time slice, n is the number of time slices in the unified time reference axis; S113 uses the Z-Score normalization method to standardize the collected multi-source monitoring data.

4. The method for dynamic tracking and prediction of coastal carbon storage according to claim 3, characterized in that, S12 includes: S121 refers to standardized multi-source monitoring data, including vegetation carbon storage standard sequences. Sediment carbon content standard sequence Topographic elevation standard sequence The cubic spline interpolation method was used to interpolate the data to a unified time point. ; S122, on a unified time reference axis Above, in time slices Constructing an ecological factor map based on ,in, For a set of nodes, As an edge set, the connection weights of the edge set are defined based on the correlation between factors, ultimately yielding a cross-scale projection map of ecological factors. .

5. The method for dynamic tracking and prediction of coastal carbon storage according to claim 4, characterized in that, S2 includes: S21, in the constructed cross-scale projection map of ecological factors In the middle, for each time slice map This study progressively analyzes the temporal abrupt change behavior of standardized ecological factor node values. By defining abrupt changes in vegetation degradation rate, rebounds in sedimentary carbon bulk density, or abnormal change patterns in topographic depressions as criteria for jump events, and combining this with a sliding window mechanism to calculate the change gradient between adjacent time slices, and setting threshold conditions, the study identifies potential carbon storage jump event time points. This forms a set of jump events. ; S22, for each identified jump event time point By retrospectively analyzing the changes in nodes of the cross-scale projection map of ecological factors before and after the occurrence, the corresponding vegetation and sediment disturbance characteristics were extracted, and the combination of ecological disturbance factors was determined. And by combining topographic elevation and slope changes, micro-topographic disturbance structures can be identified. By integrating hydrodynamic anomaly information, the structure is constructed in the form of a triplet. The event structure trigger set.

6. The method for dynamic tracking and prediction of coastal carbon storage according to claim 5, characterized in that, S21 includes: S211, from the cross-scale projection map of ecological factors In the middle, for each time slice map Various standardized node values ​​were extracted to construct a time-varying sequence of ecological factors, including the vegetation carbon storage node value in the nth time slice. Node values ​​of sediment carbon content in the nth time slice Topographic elevation node values ​​in the nth time slice And generate factor time series, including time series of vegetation carbon storage node values. Time series of sediment carbon content node values Time series of tidal topographic elevation nodes ; S212 employs a sliding window mechanism to calculate the average rate of change for each type of ecological factor time series within the window interval, which is used to measure the degree of ecological factor mutation. S213, Set thresholds for determining rate of change mutations, including thresholds for vegetation degradation rate mutations. Sediment carbon content rebound rate threshold Threshold for sudden drop in terrain elevation ,like If it is determined to be a candidate point for vegetation degradation jump, then it is considered as such. If it is determined to be a candidate point for sediment carbon bounce jump, then it is considered a candidate point. If it is determined to be a candidate point for a topographic depression jump, or a candidate point for a vegetation degradation jump, a candidate point for a sediment carbon rebound jump, or a candidate point for a topographic depression jump, then the time point is considered to be... Add potential jump event points to the jump event set. .

7. The method for dynamic tracking and prediction of coastal carbon storage according to claim 6, characterized in that, S22 includes: S221, for the set of jump events any event time point in Set the forward backtracking window length to The backward observation window length is Extract the corresponding spectral node value trajectories within the time period before and after the event point, and construct a factor change path sequence, including the node value sequence of vegetation carbon storage. Node value sequence of sediment carbon content Node value sequence of terrain elevation ; S222: Extract the intensity of perturbation characteristic changes before and after the jump from the factor change path sequence, and construct causal triggering elements, including combinations of ecological perturbation factors. and micro-topographic disturbance structure Specifically, it includes: Ecological disturbance factor combination: The disturbance intensity is measured by the moving difference method, including the change in vegetation carbon storage corresponding to the jump event. Changes in sediment carbon content corresponding to jump events ,like If the vegetation degradation meets the disturbance criterion, then the vegetation degradation is considered to meet the disturbance criterion. If so, the sedimentary disturbance is considered to meet the disturbance determination criteria, where, , These are the threshold values ​​for vegetation carbon storage disturbance and sedimentary carbon content disturbance, respectively, representing the changes in vegetation degradation or sedimentary disturbance that meet the disturbance criteria. , Incorporating combinations of ecological disturbance factors ; Micro-topographic disturbance structure: Based on the nodal value sequence of topographic elevation, calculate the elevation change per unit time interval before and after a jump event. With multi-step moving average slope change rate If satisfied or If a sudden drop in terrain is detected at the current location, an anomalous tidal range term is introduced from the tidal data. ,like The presence of hydrodynamic disturbance was determined, among which, The threshold for terrain drop. The threshold for the average change in terrain slope. As the threshold for the magnitude of tidal anomalies, and based on the determination results of the presence of abrupt topographic drop structures and hydrodynamic disturbances at the current location, a micro-topographic disturbance structure corresponding to the jump event is constructed. ; S223 combines each jump event point with its corresponding ecological disturbance factor. and micro-topographic disturbance structure The event set is constructed by combining the three elements into a structure-triggered event set.

8. The method for dynamic tracking and prediction of coastal carbon storage according to claim 7, characterized in that, S3 includes: S31, based on the constructed structure-triggered event set triplet, combined with the results of different ecological functional zones of the coastal zone, introduces the disturbance response parameter matrix, establishes a carbon storage structure transition prediction model, integrates the temporal sequence of jump events, identifies the potential path evolution trend of carbon storage spatial migration after disturbance triggering, and outputs a carbon storage transition response probability map. S32 integrates the output of the carbon storage structure transition prediction model with the functional area topographic evolution map and historical evolution trajectory map to simulate the carbon storage migration trend of each disturbance source area within a specified future period. By dynamically tracking the carbon storage diffusion path caused by different disturbance types and superimposing the carbon storage spatial gradient change value, the main carbon storage diffusion path map and structural transition section distribution map are drawn, and the transition risk level assessment results are output.

9. The method for dynamic tracking and prediction of coastal carbon storage according to claim 8, characterized in that, S31 includes: S311, Pre-structure Triggered Event Set Triple Each event is assigned to a corresponding set of coastal ecological functional zones based on its spatial location. Including salt marshes, mangroves, and sandbars, construct the disturbance response parameter matrix R; S312, using the time series of perturbation events Construct a directed transition graph for the node sequence , where the set of nodes For each jump event edge set This represents the perturbation propagation path established in chronological order, with each edge... The weight is ; S313 will have a directed transition diagram. By jointly embedding with the perturbation response parameter matrix R, a carbon storage transition response probability map is constructed. ,in, Functional area The probability of a carbon storage transition occurring in the future cycle. The probability output of carbon storage transitions is obtained by weighting through a graph attention mechanism.

10. The method for dynamic tracking and prediction of coastal carbon storage according to claim 9, characterized in that, S32 includes: S321 outputs the probability of carbon storage transitions in each functional region. Regional topographic evolution map Historical carbon storage trajectory map Spatial matching was performed, and carbon storage distribution maps were overlaid. ; S322, Based on the perturbation type j and the carbon storage change value of each pixel in the fused map, a path tracking map induced by the perturbation is constructed; S323, through main path extraction and segmented analysis, plots the main paths of structural transitions and high-risk transition segments. The main path extraction employs the maximum cumulative response path algorithm, and the transition risk level is determined. Determined by weighting the transition probability with the magnitude of carbon storage change, when When, it indicates low risk, when When, it is indicated as medium risk, when When indicated, it signifies high risk.

Images