Web3-based ecological value evaluation method and system

Abstract

The invention discloses an ecological value assessment method and system based on web3, and relates to the technical field of ecological value assessment, and the method comprises the steps: collecting multi-dimensional ecological data, and carrying out the decentralized storage of the multi-dimensional ecological data in a web3 platform; calling data on the platform to execute multi-scale spatio-temporal analysis and prediction; establishing an ecological value dynamic trend model according to a prediction result, and identifying key driving factors; and generating an ecological value evaluation result by using the model and the key factor, and uploading the result to a platform for sharing management. The technical problems that an existing ecological value evaluation method cannot respond to dynamic changes of the ecological environment at high frequency in real time, and centralized potential safety hazards exist in data storage and sharing are solved, and decentralized safe storage and sharing of multi-dimensional data of the ecological environment are achieved through a Web3 platform; and through multi-scale space-time correlation analysis and dynamic trend modeling, it is ensured that an evaluation result is kept consistent with actual ecological evolution, and the real-time performance and credibility of ecological value evaluation are improved.

Description

Technical Field

[0001] This invention relates to the field of ecological value assessment technology, specifically to a web3-based ecological value assessment method and system. Background Technology

[0002] Traditional ecological value assessment methods often suffer from drawbacks such as limited data acquisition, static assessment models, and a lack of real-time and dynamic capabilities. On the one hand, data collection relies heavily on localized, periodic monitoring methods, making it difficult to comprehensively and promptly capture subtle changes and complex dynamics in the ecological environment, leading to discrepancies between assessment results and actual conditions. On the other hand, assessment models are typically based on fixed parameters and assumptions, failing to adapt to the real-time evolution of ecosystems and thus failing to accurately reflect the dynamic changes in ecological value across different temporal and spatial scales. Furthermore, traditional assessment systems face numerous obstacles in data sharing and management. Data silos between different departments and institutions are severe, hindering information flow. Centralized storage and management also present challenges related to data security and privacy protection, limiting the comprehensiveness and accuracy of ecological value assessments. Summary of the Invention

[0003] This application provides a Web3-based ecological value assessment method and system to address the technical problems of existing ecological value assessment methods being unable to respond to dynamic changes in the ecological environment in a high-frequency and real-time manner, and the centralized security risks of data storage and sharing. It achieves decentralized and secure storage and sharing of multi-dimensional ecological environment data using the Web3 platform, and ensures that the assessment results are consistent with the actual ecological evolution through multi-scale spatiotemporal correlation analysis and dynamic trend modeling, thereby improving the real-time performance and credibility of ecological value assessment.

[0004] The first aspect of this application provides a web3-based method for ecological value assessment. The method includes: collecting multi-dimensional ecological environment data, including climate, species diversity, and soil quality data, using multiple remote sensing devices and environmental monitoring equipment; storing the multi-dimensional ecological environment data in a decentralized manner through a web3 platform; retrieving the decentralized data on the web3 platform and performing multi-scale spatiotemporal correlation analysis and prediction, including trend analysis at multiple time scales and spatial ecological evolution analysis based on geographic information; establishing a dynamic trend model of ecological value based on the multi-scale spatiotemporal correlation analysis and prediction results, which is used to analyze the impact of long-term changes in the ecosystem on ecosystem service functions and identify key ecological change drivers; establishing an ecological value assessment result using the dynamic trend model and the key ecological change drivers; and uploading the ecological value assessment result to the web3 platform for shared management.

[0005] The second aspect of this application provides a web3-based ecological value assessment system, comprising: a data acquisition module: collecting multi-dimensional ecological environment data through multiple remote sensing devices and environmental monitoring devices, including climate, species diversity, and soil quality data, and storing the multi-dimensional ecological environment data in a decentralized manner through a web3 platform; an analysis and prediction module: calling the decentralized stored data on the web3 platform and performing multi-scale spatiotemporal correlation analysis and prediction of the data, including trend analysis at multiple time scales and spatial ecological evolution analysis based on geographic information; a factor identification module: establishing a dynamic trend model of ecological value based on the multi-scale spatiotemporal correlation analysis and prediction results, the dynamic trend model of ecological value being used to analyze the impact of long-term changes in the ecosystem on ecosystem service functions and identify key ecological change driving factors; and a sharing and management module: establishing ecological value assessment results using the dynamic trend model of ecological value and the key ecological change driving factors, and uploading the ecological value assessment results to the web3 platform for sharing and management.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: Multi-dimensional ecological and environmental data, including climate, species diversity, and soil quality data, are collected through multiple remote sensing devices and environmental monitoring equipment. This data is then stored decentralizedly via a Web3 platform. The Web3 platform is used to retrieve this decentralized data and perform multi-scale spatiotemporal correlation analysis and prediction. This analysis includes trend analysis at multiple time scales and spatial ecological evolution analysis based on geographic information. An ecological value dynamic trend model is established based on the multi-scale spatiotemporal correlation analysis prediction results. This model analyzes the impact of long-term ecosystem changes on ecosystem service functions and identifies key ecological change drivers. An ecological value assessment result is then established using the ecological value dynamic trend model and the key ecological change drivers, and uploaded to the Web3 platform for shared management. This achieves the technical effect of using the Web3 platform to realize decentralized and secure storage and sharing of multi-dimensional ecological and environmental data, ensuring that the assessment results are consistent with actual ecological evolution through multi-scale spatiotemporal correlation analysis and dynamic trend modeling, and improving the real-time performance and reliability of ecological value assessment. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 This is a schematic diagram of the web3-based ecological value assessment method provided in the embodiments of this application.

[0009] Figure 2 This is a schematic diagram of the structure of a web3-based ecological value assessment system provided in an embodiment of this application.

[0010] Figure labeling: Data acquisition module 11, analysis and prediction module 12, factor identification module 13, sharing and management module 14. Detailed Implementation

[0011] This application provides a Web3-based ecological value assessment method and system to address the technical problems of existing ecological value assessment methods being unable to respond to dynamic changes in the ecological environment in a high-frequency and real-time manner, and the centralized security risks of data storage and sharing. It achieves decentralized and secure storage and sharing of multi-dimensional ecological environment data using the Web3 platform, and ensures that the assessment results are consistent with actual ecological evolution through multi-scale spatiotemporal correlation analysis and dynamic trend modeling, thereby improving the real-time performance and credibility of ecological value assessment.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] Example 1, as Figure 1 As shown, this application provides a web3-based ecological value assessment method, the method comprising: Multi-dimensional ecological and environmental data are collected through multiple remote sensing devices and environmental monitoring devices. The multi-dimensional ecological and environmental data includes climate, species diversity, and soil quality data. The multi-dimensional ecological and environmental data is stored in a decentralized manner through the web3 platform.

[0014] In one embodiment, the ecological environment is first continuously observed using multiple remote sensing satellite devices and ground environmental monitoring devices deployed within the target ecological area. The remote sensing devices acquire large-scale climate element data, such as temperature, precipitation, wind speed, and humidity; the environmental monitoring devices collect soil quality data, including soil moisture content, pH, organic matter content, and heavy metal indicators. Simultaneously, automated or semi-automated sampling devices are used to monitor and identify regional species diversity, for example, by acquiring species distribution and quantity characteristics through infrared imaging, acoustic sensing, and drone patrols. The collected multi-dimensional ecological environment data undergoes preliminary preprocessing on edge computing nodes, including outlier removal, format standardization, temporal and spatial coordinate alignment, and necessary data compression, forming structured data packets. Outlier removal is performed using statistical analysis methods such as Z-score or IQR algorithms; format standardization is performed using data conversion tools such as the Pandas library and OpenRefine; temporal and spatial coordinate alignment is performed using a geographic information system platform and a time synchronization protocol (such as NTP); and data compression is performed using Huffman coding or Gzip compression algorithms. Subsequently, the data packets are signed and encrypted using encryption algorithms, and metadata such as data source identifiers, collection timestamps, and geographic coordinates are attached. The data is then stored in a decentralized manner through a Web3 platform. This decentralized storage can be based on distributed storage networks such as IPFS, Filecoin, or Arweave, ensuring that the data is redundantly backed up on multiple nodes to prevent data loss due to single points of failure. At the same time, the uniqueness and immutability of the data are guaranteed through blockchain hash registration, thereby achieving secure and reliable storage and sharing of ecological and environmental data, providing a reliable data foundation for subsequent multi-scale spatiotemporal analysis and value assessment.

[0015] In one possible implementation, the multi-dimensional ecological environment data is stored in a decentralized manner through a web3 platform, including: Before implementing decentralized storage, data verification of the multi-dimensional ecological environment data is performed. The data verification includes generating an anomaly trust identifier based on the data source, performing data interaction and data consistency authentication, and establishing data verification results. After correcting the multi-dimensional ecological environment data based on the data verification results, decentralized distributed storage management is then implemented.

[0016] Preferably, before writing the collected multi-dimensional ecological environment data into a decentralized storage network, the data needs to be verified for integrity and reliability to ensure the accuracy and traceability of subsequent analysis and evaluation results. Specifically, firstly, an anomaly trust identifier is generated based on the data collection source. Different types of data collection devices differ in data collection frequency, accuracy, and transmission methods. Each type of monitoring device has a predefined trust level; for example, remote sensing satellite data may have a higher trust level, while low-cost ground sensor data may have a lower one. For each data source, the system assigns an initial trust level based on its historical performance, such as data accuracy, equipment calibration frequency, and maintenance frequency, using normalized weighting. This anomaly trust identifier typically ranges from 0 to 1. When data enters the verification process, this identifier is compared with the digital signature of the collection node to determine the reliability of the data source. If an anomaly is detected or the trust level is too low, an alert is triggered, and the data is marked as anomalous. Subsequently, the current batch of data is compared with historical stored data. Through hash verification, random sampling comparison, and cross-node interactive verification, it is confirmed whether the data has been tampered with or lost during transmission. Simultaneously, data consistency authentication is performed, checking the continuity and logical rationality of the data in both temporal and spatial dimensions. For temporal consistency, the changes in each data point are checked to ensure they conform to ecological patterns; for example, drastic fluctuations in soil moisture within a short period are considered anomalies. For spatial consistency, changes in a region are assessed to ensure they align with its actual geographical environment. For instance, temperature or precipitation within the same region typically exhibits a relatively smooth spatial distribution; if a local data point differs significantly from its surroundings, it is considered anomaly. The changes in a species' habitat area within a certain timeframe are also examined to ensure they conform to its habitat preferences; if a species migrates drastically to areas with unsuitable physical conditions in a short period, the data is considered anomaly. After data interaction and consistency authentication are completed, a data verification result is generated, recording the confidence level of each data item, anomaly markers, and correction suggestions. Based on the verification results, the system corrects abnormal data by using interpolation compensation with nearby spatiotemporal data, using statistical models for outlier regression correction, or directly removing severely distorted data to ensure the integrity and reliability of the overall dataset.Finally, the corrected ecological and environmental data is repackaged in a structured format, with added verification digests, correction records, and cryptographic signatures. The data is then sharded and stored on multiple nodes through networks such as IPFS or Filecoin to establish redundant backups. The data hashes and metadata are recorded using a blockchain ledger to achieve data immutability, traceability, and trusted sharing. In this way, the entire data stream has undergone strict source verification, interaction verification, and continuity assurance before entering decentralized storage, providing a high-quality data foundation for subsequent dynamic assessment of ecological value.

[0017] The web3 platform calls decentralized storage data to perform multi-scale spatiotemporal correlation analysis and prediction of the data. The multi-scale spatiotemporal correlation analysis and prediction includes trend analysis at multiple time scales and spatial ecological evolution analysis based on geographic information.

[0018] In one embodiment, after multi-dimensional ecological and environmental data is decentralized, the system can obtain the required dataset on the Web3 platform by calling the data interface of the distributed storage node, and perform multi-scale spatiotemporal correlation analysis and prediction. Specifically, firstly, a multi-timescale trend analysis framework is constructed in the time dimension, and ecological indicators such as climate, soil, and species diversity are segmented and modeled according to different time scales (such as day, month, season, and year). Short-term fluctuations, medium-term periodic changes, and long-term evolution trends are fitted and predicted respectively, and potential time-driven factors are identified through trend superposition and interactive causal analysis. Subsequently, in the spatial dimension, the acquired data is spatially aligned and gridded using a Geographic Information System (GIS) to form a unified spatial dataset. Based on this dataset, the system performs adaptive-scale spatial analysis to identify key spatial factors and significantly related areas in the ecological evolution process. Then, through hierarchical weighted fusion of time prediction results and spatial prediction results, data coupling across time scales and spatial resolution is achieved, thereby obtaining multi-scale spatiotemporal correlation analysis and prediction results. This multi-scale spatiotemporal correlation analysis prediction result can not only reflect the evolution trend of the ecosystem at different time scales, but also reveal the differential change patterns in geospatial areas, providing a scientific basis and highly reliable data support for the subsequent construction of dynamic trend models of ecological value and identification of key driving factors.

[0019] In one possible implementation, decentralized data storage is invoked on the web3 platform to perform multi-scale spatiotemporal correlation analysis and prediction of the data, including: An integrated multi-timescale trend analysis framework is activated, which can perform independent trend analysis at different time scales. Short-, medium-, and long-term trend analysis of the data is performed using the multi-timescale trend analysis framework to establish time-scale mapping results. Cross-causal analysis is performed on the time-scale mapping results to identify interactive influencing factors. After weight compensation of the time analysis results using the interactive influencing factors, time prediction results are established. Based on the time prediction results, multi-scale spatiotemporal correlation analysis prediction results are established.

[0020] Preferably, to capture the dynamic evolution trend of ecological and environmental data, an integrated multi-timescale trend analysis framework is first activated. This framework pre-constructs multiple time windows, such as daily, weekly, monthly, quarterly, and annual scales, and can use Long Short-Term Memory (LSTM) networks to establish independent trend models for the same batch of ecological and environmental data at different time scales. After activating the multi-timescale trend analysis framework, decentralized storage data, including climate change data, soil quality indicators, and species diversity monitoring data, is retrieved from the Web3 platform. Subsequently, this data is input into the multi-timescale trend analysis framework. At the short-term scale, a lightweight single-layer or two-layer LSTM network is used to capture rapid fluctuations and short-term nonlinear changes, identifying short-term fluctuation characteristics. In addition, methods such as fast moving average, exponential smoothing, and STL time series decomposition can also be used to identify short-term fluctuation characteristics. At medium-term scales, multi-layer stacked LSTMs or bidirectional LSTMs (BiLSTMs) can be used to simultaneously consider forward and backward time dependencies, characterizing periodic fluctuations and medium-term trend evolution. In addition, Fourier analysis, seasonal regression models (SARIMA), and wavelet analysis can be employed to characterize medium-term evolution trends. At long-term scales, attention-based enhanced LSTMs (Attention-LSTMs) or variational LSTMs (VLSTMs) can be used to handle long-term dependency problems and extract long-term structural change patterns. Furthermore, ARIMA, Prophet models, and recurrent convolutional networks can be used to identify long-term structural change patterns. Through the above analysis at different time scales, the system establishes a time-scale mapping of time analysis results, that is, forming corresponding trend curves and numerical prediction results at each scale. Subsequently, the short-term, medium-term, and long-term time analysis results are aligned and standardized onto the same time axis. In this process, short-term prediction results are aggregated using a sliding window, while medium-term and long-term results are mapped to the same time dimension through interpolation or downsampling to ensure comparability and computability across scales. Pearson correlation coefficients and Spearman rank correlation coefficients are then calculated for the index results at different time scales to identify combinations of variables with significant correlations. For nonlinear relationships, mutual information methods are introduced for supplementary verification. For these combinations of variables with significant correlations, Granger causality tests can be used to detect whether a variable at one time scale has a predictive effect on a variable at another scale. For nonlinear or high-dimensional variable relationships, PC algorithms based on conditional independence, causal discovery networks, or time-series-based structural equation modeling methods can be used to improve the accuracy of causal identification.Then, the selected potential causal relationships are cross-validated. This involves repeated analysis across different time windows and training data subsets to eliminate random relationships and retain causal associations that are stable across multiple scales and datasets. This establishes interactive influencing factors that have significant impacts at multiple scales, such as the effect of long-term climate trends on short-term soil moisture fluctuations, or the feedback of species community changes on medium-term ecological balance. After identifying these interactive influencing factors, the system incorporates them into the temporal analysis results for weight compensation. This involves adjusting the weight distribution of the original trend results through weighted regression, ensuring that the temporal prediction results more accurately reflect the combined effects across different time scales. Finally, based on the weight-compensated temporal prediction results, the system integrates them with spatial scale prediction results to establish a complete multi-scale spatiotemporal correlation analysis prediction result. This ensures the scientific validity and robustness of the temporal trend prediction and provides reliable data support for the subsequent construction of a dynamic trend model of ecological value.

[0021] In one possible implementation, a multi-scale spatiotemporal correlation analysis prediction result is established based on the time prediction result, including: After spatially aligning the data from decentralized storage accessed on the Web3 platform, a spatial dataset is established. Adaptive scaling analysis is then performed on the spatial dataset, utilizing spatial hotspot distribution and spatial clustering analysis to identify key spatial factors influencing ecological evolution. Spatial autocorrelation identification is performed using the adaptive scaling analysis results to establish significant spatial correlations. Spatial prediction results are then established using the key spatial factors, the significant spatial correlations, and the adaptive scaling analysis results. Finally, multi-scale spatiotemporal correlation analysis prediction results are established based on the spatial prediction results and the temporal prediction results.

[0022] Optionally, after accessing the decentralized ecological and environmental data stored on the Web3 platform, the data from different sources are first spatially aligned. Remote sensing data, sensor monitoring data, and other ecological sampling data are projected onto the same coordinate system, and resolution and scale are standardized to ensure the consistency of various data in geospatial space. After alignment, a unified spatial dataset is established, which includes both spatial coordinates and time labels for subsequent core spatial analysis. Subsequently, based on the spatial distribution characteristics of ecological factors within the study area, the scale of the analysis units is dynamically adjusted to capture effective signals at different spatial levels. Then, through spatial hotspot distribution detection (such as Getis-Ord Gi* statistics) and spatial clustering analysis (such as DBSCAN clustering or Ripley's K function), geographically significant concentrated areas of ecological change are identified, thereby extracting key spatial factors affecting ecological evolution, such as soil degradation zones, high-density species areas, or anomalous climate hotspots. Subsequently, based on the adaptive scaling analysis results, spatial autocorrelation identification is performed. In this process, Moran's I, Geary's C, or local spatial autocorrelation methods can be used to quantify the correlation between spatial factors, establish significant spatial correlations, and reveal the linkage effects of certain ecological variables in spatially adjacent areas. For example, there is a significant spatial correlation between humidity anomalies and vegetation cover decline. Then, the key spatial factors, significant spatial correlations, and adaptive scaling analysis results are summarized to establish spatial prediction results. These spatial prediction results are then hierarchically weighted and fused with temporal prediction results to establish multi-scale spatiotemporal correlation analysis prediction results, providing comprehensive data support for subsequent dynamic modeling of ecological value.

[0023] In one possible implementation, a multi-scale spatiotemporal correlation analysis prediction result is established based on the spatial prediction result and the temporal prediction result, including: After parsing the spatial prediction results and the temporal prediction results, a hierarchical weighted fusion is performed. The hierarchical weighted fusion includes: performing hierarchical weight analysis under spatial and temporal resolution differences using multi-resolution spatiotemporal data fusion to establish a first fusion result; performing spatiotemporal locality analysis on the parsed results, configuring local weights using the spatiotemporal locality analysis results, and establishing a second fusion result; and completing the hierarchical weighted fusion based on the first fusion result and the second fusion result to establish a multi-scale spatiotemporal correlation analysis prediction result.

[0024] Optionally, after obtaining the spatial and temporal prediction results, the system performs differential analysis on the spatial and temporal prediction results based on multi-resolution spatiotemporal data fusion technology. Since spatial and temporal data differ significantly in resolution and coverage—for example, spatial data may cover large-scale geographical areas, while temporal data may emphasize high-frequency short-term changes—the system establishes a hierarchical weighted analysis model, assigning different weights to data at different resolution levels. Specifically, for large-scale prediction tasks such as long-term climate change, the weight of spatial data is prioritized to ensure the model accurately reflects macro-ecological trends; while in short-term climate events or rapid ecological disturbance predictions, the weight of time-series data is increased to make the results more sensitive to high-frequency fluctuations. The first fusion result is then formed through weighting, reflecting the hierarchical weighting effect at different resolutions. Subsequently, a spatiotemporal locality analysis is performed on the first fusion result. This analysis is based on local statistics and clustering tests to identify local data patterns that exhibit significant changes within a specific time period or geographical area. For these areas of local change, the contribution of the corresponding data in that area is increased through a local weighting mechanism; that is, higher weights are assigned to them in the weighted model. In this way, the prediction accuracy of local ecosystem changes can be enhanced. For example, in the case of a sudden drop in soil moisture in a hotspot area, the impact of this information in the prediction results can be amplified by weighting, thereby improving the model's responsiveness to local ecological dynamics and forming a second fusion result, reflecting the adjustment role of spatiotemporal local weighting. Finally, the first and second fusion results are synthesized hierarchically. Through iterative optimization and weighted normalization strategies, the overall hierarchical weighted fusion is completed, ultimately establishing a multi-scale spatiotemporal correlation analysis prediction result. This multi-scale spatiotemporal correlation analysis prediction result can accurately reflect the long-term evolution trend at a large scale and sensitively capture local short-term changes, providing comprehensive data support with both macro and micro perspectives for the construction of subsequent ecological value dynamic trend models.

[0025] An ecological value dynamic trend model is established based on the prediction results of multi-scale spatiotemporal correlation analysis. This model is used to analyze the impact of long-term changes in the ecosystem on ecosystem service functions and to identify key ecological change drivers.

[0026] In one embodiment, after obtaining the prediction results from multi-scale spatiotemporal correlation analysis, the temporal and spatial prediction results are input into a temporal modeling framework to dynamically model ecological elements such as climate change, species diversity, and soil quality at different time scales. Subsequently, ecological value functions are used to map and couple ecosystem service functions, such as carbon sequestration capacity, water conservation, and habitat provision, with the aforementioned temporal change results, quantitatively measuring the positive or negative impact of changes in ecological elements on service functions. To ensure the model's dynamic adaptability, intelligent optimization algorithms, such as particle swarm optimization, genetic algorithms, or Bayesian optimization, are introduced to continuously adjust the service function weights based on real-time feedback from the prediction results, enabling the model to reflect the value change characteristics of the ecosystem at different evolutionary stages. During the modeling process, the system extracts factors with high contributions to the model results from the input multidimensional data through factor analysis and correlation tests, defining them as key ecological change drivers, such as the impact of long-term climate trends on vegetation cover or the constraint effect of soil degradation on species habitat. Ultimately, the established dynamic trend model of ecological value can not only dynamically track the overall trend of ecological value changes over time and space, but also accurately identify the core driving factors that affect the evolution and service functions of the ecosystem, providing a scientific basis and reliable data support for subsequent ecological value assessment and sharing.

[0027] In one possible implementation, a dynamic trend model of ecological value is established based on the prediction results of multi-scale spatiotemporal correlation analysis, including: Based on the prediction results of the multi-scale spatiotemporal correlation analysis, ecological change modeling is performed at different time scales of the time series prediction model to generate time series modeling results; an ecological value function is constructed, and the ecological service function is used to integrate the time prediction results of the time series modeling results with the ecological service function. The ecological value dynamic trend model is established by dynamically adjusting the weight of the service function through particle swarm optimization.

[0028] Preferably, after obtaining the multi-scale spatiotemporal correlation analysis prediction results, the system inputs the short-term, medium-term, and long-term prediction results from the multi-scale spatiotemporal correlation analysis prediction results into the time series prediction model, respectively. The time series prediction model can employ long short-term memory networks, recurrent neural networks, or temporal convolutional networks to capture dynamic change patterns at different time scales. Through this multi-level modeling approach, rapid fluctuations can be identified in the short term, periodic evolution characteristics can be captured in the medium term, and structural trends can be revealed in the long term, thereby generating complete time series modeling results. After obtaining the time-series modeling results, a pre-constructed ecological value function is acquired. This ecological value function is a weighted result of the response function values ​​between each ecological element and the ecosystem service function. These response functions are constructed through linear regression or quadratic polynomial regression using historical observation data. The weights used for weighting are determined by the corresponding ecosystem service functions. For example, based on the statistical proportion of each ecological element's contribution to the target ecosystem service function, the initial weights for carbon sink factors can be set to 0.35, water conservation factors to 0.25, soil conservation factors to 0.20, and biodiversity habitat factors to 0.20. Ecosystem service functions include, but are not limited to, carbon sink, water conservation, soil conservation, and biodiversity habitat supply. The ecological value function transforms the time-series prediction results of various ecological elements into dynamic changes in ecosystem service functions by weighted fusion. For example, soil moisture content has a direct positive correlation with water conservation function, vegetation cover is positively correlated with carbon sink function, and species diversity affects biodiversity habitat supply. To ensure that the ecological value function can dynamically reflect the changes in the ecosystem over time, a particle swarm optimization algorithm is introduced to dynamically adjust the service function weights. Specifically, the system first initializes a particle swarm, with each particle representing a weighted combination of ecosystem service functions. Then, iteratively updates the particles based on a fitness function, defined as the objective function to minimize the difference between model predictions and actual observations, typically the mean absolute error or mean squared error. During iteration, particles continuously adjust their search direction and step size based on their individual historical best solutions and the swarm's global best solution, gradually converging to the optimal weight distribution. Through particle swarm optimization, the system can adaptively adjust the weight ratios of each ecosystem service function in the model, ensuring high prediction accuracy and adaptability across different time scales. Finally, after the above modeling and optimization, the system establishes a dynamic trend model of ecosystem value. This model not only outputs a dynamic curve of ecosystem value changing over time but also updates the weight distribution of service functions in real time, ensuring that the assessment results are consistent with actual changes in the ecological environment. This provides reliable support for subsequent identification of key ecological change drivers and ecosystem value assessment.

[0029] In one possible implementation, key drivers of ecological change are identified, including: Based on the data extraction layer of the dynamic trend model of ecological value, ecological correlation factors are extracted to establish an ecological correlation factor set; the factor analysis layer is activated to perform feature dimensionality reduction of the ecological correlation factor set and establish the first factor criticality; the correlation analysis layer is activated to perform correlation analysis of the ecological correlation factor set and establish the second factor criticality; the first factor criticality and the second factor criticality are used to screen driving factors and establish key ecological change driving factors.

[0030] Optionally, after constructing the dynamic trend model of ecological value, to further identify the key driving factors influencing ecosystem evolution, the model's data extraction layer first analyzes the input multi-dimensional ecological and environmental data to extract ecologically relevant factors highly correlated with changes in ecosystem service functions and values. These factors include climate variables, soil variables, biodiversity indicators, and anthropogenic disturbances. Climate variables can be temperature, precipitation, and humidity; soil variables can be organic matter content, water content, and pH; biodiversity indicators can be species abundance and community structure characteristics; and anthropogenic disturbances can be land use change and pollution emissions. By extracting these factors, an ecologically relevant factor set can be established, providing a foundation for subsequent analysis. Subsequently, the factor analysis layer is activated to perform feature dimensionality reduction. Specifically, the system uses methods such as principal component analysis, linear discriminant analysis, or autoencoder neural networks integrated in the factor analysis layer to perform dimensionality reduction on the ecologically relevant factor set to eliminate redundancy and multicollinearity, extracting core factors that can explain the main change characteristics. After dimensionality reduction, a first factor criticality is established based on the loading or explanatory power of each factor in the principal components, reflecting the importance of the factor in the overall feature expression. Next, the correlation analysis layer is activated to perform correlation analysis on the set of ecologically related factors. During this process, methods such as Pearson correlation coefficient, Spearman rank correlation, and mutual information are used to quantitatively assess the correlation strength and directionality between different factors. Based on the analysis results, a second factor criticality is established to reflect the degree of influence of factors in the causal chain and variable linkage of the ecosystem. Finally, the first and second factor criticalities are weighted and fused to screen the set of ecologically related factors, removing factors with low weights or high redundancy, and retaining factors that show high importance in both feature expression and correlation. This establishes a set of key ecological change driving factors. This set of key ecological change driving factors can accurately identify the core factors that play a decisive role in the dynamic evolution of ecological value, thus providing a precise basis for ecological value assessment and decision support.

[0031] An ecological value assessment result is established using the aforementioned ecological value dynamic trend model and the aforementioned key ecological change driving factors, and the ecological value assessment result is uploaded to the web3 platform for shared management.

[0032] In one embodiment, after constructing an ecological value dynamic trend model and identifying key ecological change drivers, this model is used to analyze stored data, outputting the dynamic trends of ecological indicators over time and space. Examples include climate factor change curves, long-term soil quality evolution trajectories, and species diversity fluctuations. These trends serve as core inputs for ecological value calculation, providing fundamental data support for subsequent value assessment. During this process, key ecological change drivers are used to adjust the sensitivity or weight of each ecological indicator in the value calculation, ensuring the assessment highlights the core factors truly driving ecological evolution. For instance, under a global warming trend, the weights of vegetation cover and soil moisture are increased to accurately reflect their significant contributions to carbon sequestration and water conservation services, generating ecological value assessment results with dynamic response characteristics. Finally, this ecological value assessment result is uploaded to a Web3 platform for sharing and management under decentralized storage and blockchain verification mechanisms, ensuring data transparency, immutability, and reliable multi-party use, thereby providing a scientific and reliable decision-making basis for environmental governance, ecological compensation, and green finance.

[0033] In one possible implementation, ecological value assessment results are established, including: Configure a verification time-series node on the web3 platform. The verification time-series node receives data from multiple upload nodes to establish a verification dataset. The verification dataset is used to evaluate and verify the ecological value assessment results, and a verification deviation is established. Feedback incentives are generated based on the verification deviations and then used for implementation.

[0034] Preferably, to ensure the credibility and traceability of the ecological value assessment results, the system is configured with a dedicated verification time-series node on the Web3 platform. This verification time-series node, supported by the blockchain consensus mechanism, can receive data from multiple upload nodes. This data may include observation results from different remote sensing satellites, environmental monitoring equipment, or third-party research institutions. The system aggregates data from multiple sources to form a verification dataset and verifies the data source, timestamp, and signature information to ensure the legality and integrity of the data. After the verification dataset is established, it is compared and verified with the assessment results output by the ecological value dynamic trend model. The comparison process includes calculating the differences of key ecological indicators, checking the consistency of trend curves, and analyzing the correlation of spatial distribution. Through these comparisons, the gap between the model's prediction results and the actual collected data can be quantified, forming a verification bias. The verification bias includes not only the overall error value but also the local bias distribution at different time scales and spatial regions, providing a precise reference for subsequent model adjustments. After obtaining the verification bias, a feedback incentive mechanism is generated based on this verification bias. Specifically, when a node's uploaded data effectively reduces the overall verification bias, that node will receive corresponding incentives through a smart contract, such as token rewards or increased credit score. If a node's data is identified as abnormal or deviates significantly from the majority of data, its weight may be reduced or it may even be penalized. In this way, the system forms a dynamic incentive and constraint mechanism within the decentralized network, encouraging all parties to provide high-quality, reliable ecosystem data. Ultimately, this feedback incentive not only drives continuous optimization of data collection and sharing but also promotes the continuous correction and iteration of the ecosystem value assessment model, enabling the model to maintain high adaptability and accuracy to changes in the actual ecosystem over the long term.

[0035] In one possible implementation, uploading the ecological value assessment results to a web3 platform for shared management also includes: The web3 platform performs multi-dimensional ecological environment data update monitoring and generates update records; after the update records meet the preset difference threshold, an evaluation update instruction is generated, and the ecological value assessment result is updated using the evaluation update instruction.

[0036] Optionally, the system is configured with a data update monitoring mechanism on the Web3 platform for multi-dimensional ecological and environmental data. This mechanism listens to data changes on decentralized storage nodes, periodically or in real-time checks new data uploaded from remote sensing devices, environmental monitoring devices, and third-party nodes, and compares it with existing stored data. The comparison process includes not only numerical difference detection but also timestamp verification, spatial consistency checks, and data quality verification, thereby generating update records containing update frequency, update magnitude, and update area. When the detected update record meets a preset difference threshold, it indicates that the difference between the newly uploaded data and the existing data is sufficient to affect the ecological value assessment results. For example, if the soil moisture in a certain area decreases by more than a set percentage compared to the previous record, or if the species diversity index fluctuates significantly in the short term, the system will trigger the update logic. At this time, the platform automatically generates an assessment update instruction through a smart contract. This instruction carries the triggering conditions, the range of differences, and the category of ecological service functions that need to be updated. After generating the assessment update instruction, the system sends it to the ecological value dynamic trend model. Upon receiving the instruction, the model re-accesses the latest multi-dimensional data, updates the relevant time trend analysis and spatial evolution analysis, and incorporates weight adjustments for key ecological driving factors to generate a new ecological value assessment result. The updated result is then re-uploaded to the Web3 platform for blockchain hash registration and distributed storage, ensuring the transparency of the update process and the immutability of the result. Through this mechanism, the ecological value assessment result can be automatically updated dynamically when significant changes occur in ecological and environmental data, thus ensuring that the assessment result remains highly consistent with the actual state of the ecosystem and providing real-time and accurate data support for environmental governance, ecological compensation, and green finance.

[0037] Example 2, based on the same inventive concept as the web3-based ecological value assessment method in the aforementioned examples, such as... Figure 2As shown, this application provides a web3-based ecological value assessment system. The system and method embodiments in this application are based on the same inventive concept. The ecological value assessment system includes: a data acquisition module 11: collecting multi-dimensional ecological environment data through multiple remote sensing devices and environmental monitoring devices, including climate, species diversity, and soil quality data, and storing the multi-dimensional ecological environment data in a decentralized manner through a web3 platform; an analysis and prediction module 12: calling the decentralized stored data on the web3 platform and performing multi-scale spatiotemporal correlation analysis and prediction, including trend analysis at multiple time scales and spatial ecological evolution analysis based on geographic information; a factor identification module 13: establishing a dynamic trend model of ecological value based on the multi-scale spatiotemporal correlation analysis and prediction results, which is used to analyze the impact of long-term changes in the ecosystem on ecosystem service functions and identify key ecological change driving factors; and a sharing and management module 14: establishing ecological value assessment results using the dynamic trend model of ecological value and the key ecological change driving factors, and uploading the ecological value assessment results to the web3 platform for sharing and management.

[0038] Furthermore, the data acquisition module 11 also includes: Before implementing decentralized storage, data verification of the multi-dimensional ecological environment data is performed. The data verification includes generating an anomaly trust identifier based on the data source, performing data interaction and data consistency authentication, and establishing data verification results. After correcting the multi-dimensional ecological environment data based on the data verification results, decentralized distributed storage management is then implemented.

[0039] Furthermore, the analysis and prediction module 12 also includes: An integrated multi-timescale trend analysis framework is activated, which can perform independent trend analysis at different time scales. Short-, medium-, and long-term trend analysis of the data is performed using the multi-timescale trend analysis framework to establish time-scale mapping results. Cross-causal analysis is performed on the time-scale mapping results to identify interactive influencing factors. After weight compensation of the time analysis results using the interactive influencing factors, time prediction results are established. Based on the time prediction results, multi-scale spatiotemporal correlation analysis prediction results are established.

[0040] Furthermore, the analysis and prediction module 12 also includes: After spatially aligning the data from decentralized storage accessed on the Web3 platform, a spatial dataset is established. Adaptive scaling analysis is then performed on the spatial dataset, utilizing spatial hotspot distribution and spatial clustering analysis to identify key spatial factors influencing ecological evolution. Spatial autocorrelation identification is performed using the adaptive scaling analysis results to establish significant spatial correlations. Spatial prediction results are then established using the key spatial factors, the significant spatial correlations, and the adaptive scaling analysis results. Finally, multi-scale spatiotemporal correlation analysis prediction results are established based on the spatial prediction results and the temporal prediction results.

[0041] Furthermore, the analysis and prediction module 12 also includes: After parsing the spatial prediction results and the temporal prediction results, a hierarchical weighted fusion is performed. The hierarchical weighted fusion includes: performing hierarchical weight analysis under spatial and temporal resolution differences using multi-resolution spatiotemporal data fusion to establish a first fusion result; performing spatiotemporal locality analysis on the parsed results, configuring local weights using the spatiotemporal locality analysis results, and establishing a second fusion result; and completing the hierarchical weighted fusion based on the first fusion result and the second fusion result to establish a multi-scale spatiotemporal correlation analysis prediction result.

[0042] Furthermore, the factor recognition module 13 also includes: Based on the prediction results of the multi-scale spatiotemporal correlation analysis, ecological change modeling is performed at different time scales of the time series prediction model to generate time series modeling results; an ecological value function is constructed, and the ecological service function is used to integrate the time prediction results of the time series modeling results with the ecological service function. The ecological value dynamic trend model is established by dynamically adjusting the weight of the service function through particle swarm optimization.

[0043] Furthermore, the factor recognition module 13 also includes: Based on the data extraction layer of the dynamic trend model of ecological value, ecological correlation factors are extracted to establish an ecological correlation factor set; the factor analysis layer is activated to perform feature dimensionality reduction of the ecological correlation factor set and establish the first factor criticality; the correlation analysis layer is activated to perform correlation analysis of the ecological correlation factor set and establish the second factor criticality; the first factor criticality and the second factor criticality are used to screen driving factors and establish key ecological change driving factors.

[0044] Furthermore, the shared management module 14 also includes: Configure a verification time-series node on the web3 platform. The verification time-series node receives data from multiple upload nodes to establish a verification dataset. The verification dataset is used to evaluate and verify the ecological value assessment results, and a verification deviation is established. Feedback incentives are generated based on the verification deviations and then used for implementation.

[0045] Furthermore, the shared management module 14 also includes: The web3 platform performs multi-dimensional ecological environment data update monitoring and generates update records; after the update records meet the preset difference threshold, an evaluation update instruction is generated, and the ecological value assessment result is updated using the evaluation update instruction.

[0046] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0047] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0048] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A Web3-based ecological value assessment method, characterized in that, The method includes: Multi-dimensional ecological and environmental data are collected through multiple remote sensing devices and environmental monitoring devices. The multi-dimensional ecological and environmental data includes climate, species diversity, and soil quality data. The multi-dimensional ecological and environmental data is stored in a decentralized manner through the web3 platform. The web3 platform calls decentralized storage data to perform multi-scale spatiotemporal correlation analysis and prediction of the data. The multi-scale spatiotemporal correlation analysis and prediction includes trend analysis at multiple time scales and spatial ecological evolution analysis based on geographic information. An ecological value dynamic trend model is established based on the prediction results of multi-scale spatiotemporal correlation analysis. The ecological value dynamic trend model is used to analyze the impact of the long-term change trend of the ecosystem on the ecosystem service function and identify key ecological change driving factors. An ecological value assessment result is established using the aforementioned ecological value dynamic trend model and the aforementioned key ecological change driving factors, and the ecological value assessment result is uploaded to the web3 platform for shared management.

2. The ecological value assessment method based on Web3 as described in claim 1, characterized in that, Accessing decentralized storage data on the Web3 platform to perform multi-scale spatiotemporal correlation analysis and prediction, including: Activate an integrated multi-timescale trend analysis framework that can perform independent trend analysis at different time scales; The multi-timescale trend analysis framework is used to perform short-, medium-, and long-term trend analysis of the data, and to establish time-scale mapping time analysis results. Cross-causal analysis was performed on the time-scale mapping results to identify interactive influencing factors. After weighting the time analysis results using the aforementioned interactive influence factors, a time prediction result is established, and a multi-scale spatiotemporal correlation analysis prediction result is established based on the time prediction result.

3. The ecological value assessment method based on Web3 as described in claim 2, characterized in that, Based on the time prediction results, a multi-scale spatiotemporal correlation analysis prediction result is established, including: After spatially aligning the data from decentralized storage accessed on the web3 platform, a spatial dataset is created. Adaptive scaling analysis was performed on the aforementioned spatial dataset, and key spatial factors influencing ecological evolution were identified by utilizing spatial hotspot distribution and spatial clustering analysis. Spatial autocorrelation identification is performed using the results of adaptive scaling analysis to establish spatially significant correlations. Spatial prediction results are established using the key spatial factors, the significant spatial correlations, and the adaptive scaling analysis results. Based on the spatial prediction results and the temporal prediction results, a multi-scale spatiotemporal correlation analysis prediction result is established.

4. The ecological value assessment method based on Web3 as described in claim 3, characterized in that, Based on the spatial prediction results and the temporal prediction results, a multi-scale spatiotemporal correlation analysis prediction result is established, including: After parsing the spatial prediction results and the temporal prediction results, a hierarchical weighted fusion is performed, which includes: By using multi-resolution spatiotemporal data fusion to perform hierarchical weight analysis under spatial and temporal resolution differences, a first fusion result is established. Spatiotemporal locality analysis is performed on the analysis results, and local weighting weights are configured using the spatiotemporal locality analysis results to establish a second fusion result; Based on the first fusion result and the second fusion result, perform hierarchical weighted fusion to establish multi-scale spatiotemporal correlation analysis and prediction results.

5. The ecological value assessment method based on Web3 as described in claim 1, characterized in that, A dynamic trend model of ecological value is established based on the prediction results of multi-scale spatiotemporal correlation analysis, including: Based on the multi-scale spatiotemporal correlation analysis prediction results, ecological change modeling is performed at different time scales of the time series prediction model to generate time series modeling results. An ecological value function is constructed, and the ecological service functions are integrated with the time prediction results of time series modeling using the ecological value function. The ecological value dynamic trend model is established by dynamically adjusting the weight of service functions through particle swarm optimization.

6. The ecological value assessment method based on Web3 as described in claim 1, characterized in that, Identify key drivers of ecological change, including: Based on the data extraction layer of the dynamic trend model of ecological value, ecological correlation factors are extracted, and a set of ecological correlation factors is established. Activate the factor analysis layer, perform feature dimensionality reduction of the ecological association factor set, and establish the criticality of the first factor; Activate the correlation analysis layer, perform correlation analysis on the ecologically related factor set, and establish the criticality of the second factor; The criticality of the first factor and the criticality of the second factor are used to screen driving factors and establish key ecological change driving factors.

7. The ecological value assessment method based on Web3 as described in claim 1, characterized in that, The multi-dimensional ecological environment data is stored in a decentralized manner through the web3 platform, including: Before implementing decentralized storage, data verification of the multi-dimensional ecological environment data is performed. The data verification includes generating an anomaly trust identifier based on the data source, performing data interaction and data consistency authentication, and establishing data verification results. After correcting the multi-dimensional ecological environment data based on the data verification results, decentralized distributed storage management is executed.

8. The ecological value assessment method based on Web3 as described in claim 1, characterized in that, Establish ecological value assessment results, including: Configure a verification timing node on the web3 platform. The verification timing node receives data from upload nodes from multiple parties and establishes a verification dataset. The ecological value assessment results are evaluated and verified using the aforementioned validation dataset, and validation bias is established. Feedback incentives are generated based on the verification deviations, and then used for verification.

9. The ecological value assessment method based on Web3 as described in claim 1, characterized in that, Uploading the ecological value assessment results to the web3 platform for shared management also includes: The web3 platform is used to monitor and update multi-dimensional ecological and environmental data, generating update records. Once the updated record meets the preset difference threshold, an evaluation update instruction is generated, and the ecological value evaluation result is updated using the evaluation update instruction.

10. A web3-based ecological value assessment system, characterized in that, The ecological value assessment system is used to execute the web3-based ecological value assessment method according to any one of claims 1 to 9, and the ecological value assessment system includes: Data acquisition module: Collects multi-dimensional ecological and environmental data through multiple remote sensing devices and environmental monitoring devices. The multi-dimensional ecological and environmental data includes climate, species diversity, and soil quality data. The multi-dimensional ecological and environmental data is stored in a decentralized manner through the web3 platform. Analysis and prediction module: It calls decentralized storage data on the web3 platform and performs multi-scale spatiotemporal correlation analysis and prediction of the data. The multi-scale spatiotemporal correlation analysis and prediction includes trend analysis at multiple time scales and spatial ecological evolution analysis based on geographic information. Factor identification module: Based on the prediction results of multi-scale spatiotemporal correlation analysis, an ecological value dynamic trend model is established. The ecological value dynamic trend model is used to analyze the impact of the long-term change trend of the ecosystem on the ecosystem service function and identify key ecological change driving factors. Shared Management Module: Utilizes the aforementioned dynamic trend model of ecological value and the aforementioned key ecological change driving factors to establish ecological value assessment results, and uploads the ecological value assessment results to the web3 platform for shared management.

Images