A method for evaluating water ecological service value based on long time series multi-source remote sensing data

Abstract

The application discloses a kind of water area ecological service value evaluation method based on long time sequence multi-source remote sensing data, relating to the technical field of ecological environment assessment, comprising the following steps: S1: obtaining the multi-source remote sensing data of research area for many years and social economy and ecological data, establish comprehensive database, S2: given multi-source remote sensing data is pretreated, S3: statistics year after year grain crop data, calculate unit area farmland ecosystem grain production service price, S4: arrange year after year water area ecological service value data, clean, S5: each kind of data is imported into model, generates water area ESV space-time variation chart and with economic development coordination degree space-time variation chart.The application has the advantages that the problems of imbalance between extraction precision and efficiency, lack of long time sequence data support and single-discipline evaluation limitations in the prior art are solved, more accurate, comprehensive and dynamic evaluation is realized, and a strong basis is provided for water area ecological protection and management.

Description

Technical Field

[0001] This invention relates to the field of ecological environment assessment technology, focusing on innovative methods for accurately assessing the value of aquatic ecosystem services using long-term multi-source remote sensing data, geographic information systems (GIS), and deep learning technologies. Background Technology

[0002] As global environmental change and regional sustainable development face increasingly severe challenges, the importance of aquatic ecosystems is becoming ever more prominent. However, aquatic ecosystems currently face numerous problems such as pollution, ecological degradation, and biodiversity loss. Accurately assessing the value of aquatic ecosystem services is crucial for the rational planning and management of water resources and for achieving coordinated economic and ecological development.

[0003] Existing assessment methods have several shortcomings. Some water extraction technologies struggle to balance accuracy, computational resource requirements, and processing efficiency, resulting in inaccurate and untimely assessments. Traditional methods often lack long-term data support, making it difficult to comprehensively reflect the spatiotemporal variations in the value of aquatic ecosystem services. Furthermore, single-discipline assessments cannot adequately consider the complexity and diversity of aquatic ecosystems, limiting the scientific rigor and comprehensiveness of the assessment results. Summary of the Invention

[0004] In view of the many shortcomings of current methods for assessing the value of aquatic ecosystem services, such as the imbalance between the accuracy and efficiency of water body extraction, the lack of long-term series data support, and the limitations of single-discipline assessment, this invention aims to provide a method for assessing the value of aquatic ecosystem services based on long-term series multi-source remote sensing data, so as to achieve a more accurate, comprehensive and dynamic assessment and provide a strong basis for aquatic ecosystem protection and management.

[0005] A method for assessing the value of aquatic ecosystem services based on long-term multi-source remote sensing data includes the following steps:

[0006] S1: Data Acquisition. Collaborate with satellite data providers to acquire long-term (e.g., 2000–2022) multi-source remote sensing data (such as Landsat and GF-1 imagery), while simultaneously collecting socio-economic data from local government departments, including information on food crops, GDP, and economic coordination, to build a basic database.

[0007] S2: Data preprocessing and water body extraction. For the given multi-source remote sensing data, after preprocessing, water bodies are extracted using NSWI combined with the Otsu thresholding method. After calculating NSWI, the threshold is determined by the Otsu method combined with a local adaptive strategy. A large number of sample points are randomly selected, and the true category is obtained through field or high-resolution image interpretation. The accuracy is evaluated using a confusion matrix.

[0008] S3: Calculate the value of ecosystem services. Calculate the total price of grain crops and the sown area, then substitute these into Equation 1 to calculate the price of grain production services per unit area of ​​farmland ecosystem. Obtain authoritative equivalents of ecosystem services for aquatic areas, and combine this with water area data to calculate the value of aquatic ecosystem services using Equation 2.

[0009]

[0010] S4: LSTM model construction and prediction. After importing data into Excel, clean and transform it, select key columns, normalize using formula 3, create a time series dataset according to formula 4, divide the training set and test set into 70% and 30% respectively, construct a two-layer LSTM model (select appropriate activation function, optimizer, and initial learning rate such as 0.001), train it using the learning rate scheduler and early stopping callback function, predict future values, restore it using inverse scaling formula 4, and evaluate the performance using formula 5.

[0011]

[0012] original_value=scaled_value×(max-min)+min(5)

[0013]

[0014] S5: Comprehensive analysis and decision support. Data is imported into the model, and techniques such as spatial interpolation and overlay analysis are used to generate spatiotemporal variation maps of water area ESV and the degree of coordination with economic development. Based on the analysis results, ecological protection and management strategies are formulated for the region. This method can be adapted and applied to other regions.

[0015] Compared with existing technologies, the advantages of this invention are: it provides a method for assessing the ecological service value of aquatic waters based on long-term multi-source remote sensing data, which solves the problems of imbalance between water body extraction accuracy and efficiency, lack of long-term data support, and limitations of single-discipline assessment in existing technologies, and achieves a more accurate, comprehensive and dynamic assessment, providing a strong basis for aquatic ecological protection and management. Attached Figure Description

[0016] Figure 1 A flowchart illustrating the method for assessing the ecological service value of aquatic waters based on long-term multi-source remote sensing data.

[0017] Figure 2 A graph showing the change in the ecological service value (ESV) of aquatic waters.

[0018] Figure 3 Spatial distribution map of the coordination between aquatic ecosystem services and economic development.

[0019] Figure 4 The loss curve during the training process of the prediction model.

[0020] Figure 5 This is the ESV prediction map for region 5 obtained from the prediction model. Detailed Implementation

[0021] Reference Figure 1-5 A method for assessing the value of aquatic ecosystem services based on long-term multi-source remote sensing data includes the following steps:

[0022] S1: Acquire multi-source remote sensing data for a specific study area from 2000 to 2022, including Landsat satellite imagery and GF-1 imagery. Simultaneously, collect information on major local food crops, economic data, and ecological data to establish a comprehensive database.

[0023] S2: Preprocess the given multi-source remote sensing data. Water bodies are extracted using NSWI combined with the Otsu thresholding method. After calculating the NSWI, the image is segmented using the Otsu thresholding method combined with an adaptive strategy. A large number of sample points are randomly selected, and the true categories are obtained through field or high-resolution image interpretation. The confusion matrix is ​​calculated to ensure the accuracy of water body extraction.

[0024] S3: Statistically analyze historical grain crop data, integrate price information from multiple channels to calculate the total price, combine GIS and field surveys to measure the sown area, calculate the price of grain production services per unit area of ​​farmland ecosystem, and calculate the price of grain production services per unit area of ​​farmland ecosystem using Equation 1.

[0025]

[0026] Where Ea is the price of food production services per unit area of ​​farmland ecosystem (yuan·hm²). -2 );

[0027] Ni represents the total price of grain crops in year i (in yuan);

[0028] Mi represents the sown area of ​​grain crops in year i (hm²) 2 );

[0029] This allows us to obtain an authoritative equivalent of the aquatic ecosystem service value, which is then combined with water area data, and the aquatic ecosystem service value is calculated using Equation 3:

[0030]

[0031] Among them, ESV represents the total value of aquatic ecosystem services in the study area (in yuan);

[0032] Ek is the value equivalent of the kth aquatic ecosystem service function;

[0033] Eij represents the area (hm²) of the j-th type of water body in region i. 2 ).

[0034] Then, the ecological service value of the water body is calculated based on the water area data, and different water body types and ecological service functions are distinguished.

[0035] S4: Organize and clean historical data on aquatic ecosystem service value, and map the data to a specific range using a normalization method, employing Equation 3 for normalization:

[0036]

[0037] Where value is the original data value, min is the minimum value in the dataset, max is the maximum value in the dataset, and scaled_value is the normalized data value.

[0038] Then, create the time series dataset according to Equation 4:

[0039]

[0040] Where data is the normalized data sequence, Xi is the i-th input sequence, including data from three consecutive time steps, and Yi is the i-th target value, i.e., the data from the next time step after the input sequence. The training and test sets are divided into 70% and 30% portions, respectively. A two-layer LSTM model is constructed, with appropriate parameters selected and compiled. During training, a learning rate scheduler and an early stopping callback function are used to train and predict future values, which are then restored according to the following formula:

[0041] original_value=scaled_value×(max-min)+min(5)

[0042] After training, predict the training and test sets, reverse scale to restore the predicted values, and calculate metrics such as mean squared error to evaluate performance. Equation 6 is used to evaluate performance.

[0043]

[0044] Where n is the number of samples, Yi is the actual value, and predi is the predicted value. The model predicts the ESV value for the next 5 years and provides visualization and uncertainty analysis.

[0045] S5: Import various types of data into the model, and use techniques such as spatial interpolation and overlay analysis to generate spatiotemporal variation maps of water area ESV and spatiotemporal variation maps of the degree of coordination with economic development. Based on the analysis results, formulate ecological protection and management strategies for the study area.

[0046] As is known from common technical knowledge, this invention can be implemented through other embodiments that do not depart from its spirit or essential characteristics. Therefore, the disclosed embodiments described above are merely illustrative and not exhaustive. All modifications within the scope of this invention or its equivalents are included in this invention.

Claims

1. A method for assessing the ecological service value of aquatic waters based on long-term multi-source remote sensing data, characterized in that, Includes the following steps: S1: Data Acquisition; This includes multi-source remote sensing data and socio-economic data, thereby constructing a basic database; S2: Preprocess the given data, then use NSWI combined with Otsu thresholding to extract water bodies. After calculating NSWI, use Otsu's method combined with a local adaptive strategy to determine the threshold and randomly select multiple sample points. S3: Statistically analyze historical grain crop data, integrate price information from multiple channels to calculate the total price, combine GIS and field surveys to measure the sown area, calculate the price of grain production services per unit area of ​​farmland ecosystem, obtain the equivalent value of aquatic ecosystem services applicable to the study area, calculate the value of aquatic ecosystem services based on water body area data, and distinguish different water body types and ecosystem service functions. S4: Organize historical water ecosystem service value data, clean it, use normalization methods to map the data to a specific range, construct a time series dataset, divide it into training and test sets, construct a two-layer LSTM model, select appropriate parameters and compile it, use a learning rate scheduler and early stopping callback function during training, predict the training and test sets after training, reverse scale to restore the predicted values, calculate the mean square error and other indicators to evaluate the performance, predict the ESV value for the next 5 years based on the model, and perform visualization and uncertainty analysis. S5: Import various types of data into the model to generate a spatiotemporal variation map of water area ESV and a spatiotemporal variation map of the degree of coordination with economic development. Based on the analysis results, formulate ecological protection and management strategies for the study area.

2. The method for assessing the ecological service value of aquatic waters based on long-term multi-source remote sensing data according to claim 1, characterized in that: In step S1, the multi-source remote sensing data collected includes data from 2000 to 2022, including Landsat series satellite imagery and GF-1 imagery. The collected socio-economic data includes information on major local food crops, GDP, economic coordination, and ecological data.

3. The method for assessing the ecological service value of aquatic waters based on long-term multi-source remote sensing data according to claim 1, characterized in that: In step S2, the true category is obtained through field or high-resolution image interpretation, the confusion matrix is ​​calculated, and the accuracy is evaluated using the confusion matrix.

4. The method for assessing the ecological service value of aquatic waters based on long-term multi-source remote sensing data according to claim 1, characterized in that: The price of food production services per unit area of ​​farmland ecosystem is calculated using Equation 1. Among them, E a Price of food production services per unit area of ​​farmland ecosystem, in yuan·hm² -2 ; N i The total price of grain crops in year i is expressed in yuan. M i The sown area of ​​grain crops in year i is expressed in hectares (hm²). 2 ; This allows us to obtain an authoritative equivalent of the aquatic ecosystem service value, which is then combined with water area data, and the aquatic ecosystem service value is calculated using Equation 3: Wherein, ESV represents the total ecosystem service value of the aquatic area in the study area, in yuan; E k The value equivalent of the kth aquatic ecosystem service function; E ij Let be the area of ​​the j-th type of water body in region i, in hectares. 2 .

5. The method for assessing the ecological service value of aquatic waters based on long-term multi-source remote sensing data according to claim 1, characterized in that: In step S4, normalization is performed using Equation 3: Where value is the original data value, min is the minimum value in the dataset, max is the maximum value in the dataset, and scaled_value is the normalized data value; Then, create the time series dataset according to Equation 4: Where data is the normalized data sequence, Xi is the i-th input sequence, which includes data from three consecutive time steps, and Yi is the i-th target value, which is the data from the next time step after the input sequence; Then, the training and test sets are divided, and a two-layer LSTM model is constructed. This model is trained using a learning rate scheduler and an early stopping callback function to predict future values, and then restored according to the following formula: Finally, use Equation 6 to evaluate the performance: Where n is the number of samples, Y i For actual values, pred i These are predicted values.

6. The method for assessing the ecological service value of aquatic waters based on long-term multi-source remote sensing data according to claim 1, characterized in that: In step S5, spatial interpolation and overlay analysis techniques are used to generate a spatiotemporal variation map of the value of aquatic ecosystem services and a spatiotemporal variation map of the degree of coordination with economic development.

Images