A method and system for analyzing the impact of urban expansion on vegetation carbon sequestration capacity

Abstract

The present invention discloses an analysis method and system for the impact of urban expansion on vegetation carbon fixation capacity, including calculating the core vegetation net primary productivity kNPP and the vegetation potential net primary productivity PNPP of the study area, calculating the impact HNPP of human activities on vegetation net primary productivity through the difference between kNPP and PNPP; constructing multiple annular buffer zones based on the geometric center of the built-up area of ​​the study area, obtaining a fitting curve of the buffer zone radius and the HNPP value in the buffer zone through generalized additive model fitting, identifying the impact distance DHNPP of urban expansion on vegetation carbon fixation capacity according to the change and inflection point of the curve; selecting factors related to urban development and environmental quality, establishing the relationship between each factor and DHNPP through the XGBoost model, and calculating the impact weight of each factor on DHNPP through SHAP analysis. The present invention introduces the new concept of the impact distance of urban expansion on vegetation carbon fixation capacity, quantifies the spatial impact range of urban expansion on the surrounding vegetation net primary productivity, and fills the research gap in this field.

Description

Technical Field

[0001] The present invention relates to the fields of remote sensing, geographic information, urban planning and ecology, and in particular to a method and system for analyzing the impact of urban expansion on the carbon fixation capacity of vegetation. Background Art

[0002] Urban expansion not only weakens the resilience of ecosystems, but also disrupts ecological balance. As the population increases, urban expansion further exacerbates the consumption of water resources and energy, as well as soil and air pollution. In addition, greenhouse gas emissions in urban environments can lead to global warming and heat island effects, extreme droughts and natural disasters, and even affect the living environment and human well-being. Therefore, exploring the impact of urban expansion on urban ecosystems is crucial for sustainable development.

[0003] In urban ecosystems, vegetation can absorb and fix carbon dioxide in the atmosphere, thus playing a key role in mitigating climate change and improving urban environmental quality. Net primary productivity (NPP) is the organic carbon content accumulated by vegetation photosynthesis per unit time and per unit area, excluding respiration, and is an important indicator for measuring the carbon sequestration capacity of vegetation. However, the current NPP product data lacks the characteristics of long-term series, high spatial resolution, and high precision, which makes it difficult to meet the research in urban ecosystems. In addition, during the process of urban expansion, changes in the boundaries of built-up areas will be affected by factors such as topography, urban development level, and urban development policies, and there is uncertainty. Therefore, it is necessary to analyze the impact of urban expansion on the carbon sequestration capacity of vegetation more comprehensively from the temporal and spatial dimensions, and explore the changing laws of the carbon sequestration capacity of vegetation with the change of urban-rural gradients. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a method for analyzing the impact of urban expansion on the carbon sequestration capacity of vegetation, comprising the following steps:

[0005] Step 1: Obtain satellite remote sensing data, long-term land use classification data and built-up area vector range of the study area, and calculate the monthly average temperature, monthly total precipitation, monthly total solar radiation data and kernel normalized difference vegetation index kNDVI;

[0006] Step 2, based on the CASA model, the land use classification data, monthly average temperature, monthly total precipitation, monthly total solar radiation data of the study area obtained in step 1, and kNDVI were used to calculate the core vegetation net primary productivity;

[0007] Step 3: Based on the monthly average temperature and monthly total precipitation data collected in step 1, the potential net primary productivity of vegetation is inverted through the Thornthwaite Memorial model, and the impact of human activities on the net primary productivity of vegetation HNPP is calculated by the difference between the net primary productivity of vegetation and the potential net primary productivity of vegetation;

[0008] Step 4: Use the built-up area vector range data of the study area obtained in step 1 to calculate the geometric center of the built-up area, construct multiple annular buffers based on the geometric center, and calculate the average HNPP value in each annular buffer. The fitting curve of the annular buffer radius and the HNPP value in the annular buffer is obtained by generalized additive model fitting. According to the changes and inflection points of the curve, the impact distance DHNPP of urban expansion on vegetation carbon sequestration capacity is identified;

[0009] Step 5: Select factors related to urban development and environmental quality, establish the relationship between each factor and DHNPP through the XGBoost model, and use SHAP analysis to calculate the impact weight of each factor on DHNPP.

[0010] Furthermore, in step 1, the satellite remote sensing data of the study area is used to calculate the monthly kernel normalized difference vegetation index, and the calculation formula is as follows:

[0011] (1)

[0012] Where kNDVI represents the kernel normalized difference vegetation index, Represents the near-infrared band of remote sensing images. Represents the red light band of satellite remote sensing images. is a length scale parameter, setting , then the formula of kNDVI is transformed into:

[0013] (2)

[0014] Where NDVI stands for Normalized Difference Vegetation Index.

[0015] Furthermore, the core vegetation net primary productivity kNPP in step 2 is calculated in the Monteith equation of the light energy utilization model by the core photosynthetically active radiation kAPAR absorbed by the vegetation and the actual light energy utilization rate of the land use type. Calculated, that is:

[0016] (3)

[0017] (4)

[0018] (5)

[0019] In the formula, Represents pixel x exist t Monthly nuclear vegetation net primary productivity; Represents pixel x exist t Nuclear photosynthetically active radiation absorbed by the moon; For pixels x exist t The actual light energy utilization rate of vegetation in a month depends on the land use, temperature and precipitation in the study area; Represents pixel x exist t Monthly photosynthetically active radiation; Represents pixel x exist t Total solar radiation for the month; Represents pixel x exist t The specific calculation method of the effective photosynthetic radiation absorption ratio of the nuclear vegetation in a month is as follows:

[0020] (6)

[0021] (7)

[0022] In the formula, and The maximum and minimum values ​​of the effective photosynthetic radiation absorption ratio FPAR of vegetation are respectively; the kernel vegetation coverage kFVC is calculated using the pixel binary model, and the lower and upper limits of the 95% confidence interval of the kernel normalized difference vegetation index kNDVI value in the study area are selected as the pure bare soil pixels. and pure vegetation pixels .

[0023] Furthermore, the Thornthwaite Memorial model in step 3 is derived by establishing the least square relationship between the measured vegetation data and the evapotranspiration, temperature and precipitation data, and the calculation formula is as follows:

[0024] (8)

[0025] (9)

[0026] (10)

[0027] Where PNPP is the potential net primary productivity of vegetation, e is a natural constant, v is the actual evaporation, r is the total annual precipitation, L is the annual average evapotranspiration,t The annual average temperature.

[0028] The difference between the core net primary productivity of vegetation and the potential net primary productivity of vegetation is used to calculate the impact of human activities on the net primary productivity of vegetation. The calculation formula is as follows:

[0029] (11)

[0030] Where HNPP represents the impact of human activities on vegetation net primary productivity, PNPP is the potential net primary productivity of vegetation, and kNPP is the core net primary productivity of vegetation.

[0031] when When , it means that human activities have a negative impact on the net primary productivity of vegetation; when This indicates that human activities have a positive impact on the net primary productivity of vegetation.

[0032] Furthermore, in step 4, the built-up area vector range is divided into multiple triangles, the area and centroid of each triangle are calculated, and the centroid of the triangle is weighted according to the proportion of the triangle area to the total area of ​​the built-up area vector range to obtain the geometric center of the built-up area vector range. The specific calculation formula is as follows:

[0033] (12)

[0034] Where P is the geometric center of the built-up area vector range, N is the number of triangles, For the i The area of ​​a triangle, For the i The centroid of a triangle.

[0035] Take the geometric center of the built-up area vector range as the center of the circle, according to km is the distance between them. km to km of multiple ring buffers, , The value is set according to actual needs. yes An integer multiple of , and calculates the average HNPP value in each ring buffer , fitting the ring buffer radius through the generalized additive model a and The nonlinear relationship between the radius of the circular buffer and the average HNPP value in the circular buffer is obtained. The specific calculation formula is as follows:

[0036] (13)

[0037] In the formula, represents the intercept of the fitted model; is a smooth function that reflects the radius a and The nonlinear relationship between represents the random error term, which is assumed to be normally distributed.

[0038] The generalized additive model is used to calculate the radius of each ring buffer. r The corresponding Fitting is performed, and the radius of the circular buffer is a Compared with the average HNPP value in the ring buffer The maximum slope of the fitting curve times as the slope threshold , when the ring buffer radius r HNPP value in the ring buffer After the maximum slope of the fitted curve appears, the first slope is less than the slope threshold The location is the distance DHNPP from which urban expansion affects the carbon sequestration capacity of vegetation.

[0039] Furthermore, in step 5, the factors related to urban development and environmental quality include: temperature, precipitation, total solar radiation, vegetation coverage, soil moisture, standardized precipitation evapotranspiration index, population, population growth rate, urban area ratio, urban population ratio, gross domestic product, urban area, fossil fuel CO2 emissions, night light intensity, and night light concentration index; all the above factors are unified to the same spatial and temporal resolution, outliers are directly deleted, and missing values ​​are processed by mean filling, and then the data is standardized to eliminate the influence of different dimensions on the subsequent training of the XGBoost model to obtain a data set; the data set is divided into a training set and a test set, the XGBoost model is trained using the training set, the relationship between each factor and DHNPP is established, the XGBoost model parameters are tuned using the cross-validation method to optimize the performance and generalization ability of the model, and the model is evaluated on the test set data, and the SHAP analysis method is used to calculate the influence weight of each factor on DHNPP. The specific calculation formula is as follows:

[0040] (14)

[0041] In the formula, Representation elements The impact weight on DHNPP, i.e., the factor SHAP value of S Indicates that it does not contain features A subset of X Represents the set of all elements; Representation SubsetS The number of elements in express The factorial of Representing a collection X The number of elements in express The factorial of Indicates that when the SHAP analysis model calculation process only uses a subset S Output when predicting the elements in ; Indicates when the SHAP analysis model uses a subset S Elements and elements in Output when making predictions.

[0042] The present invention also provides an analysis system for the impact of urban expansion on the carbon fixation capacity of vegetation, which is used to implement the analysis method for the impact of urban expansion on the carbon fixation capacity of vegetation as described above.

[0043] Furthermore, the method comprises a processor and a memory, wherein the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute the above-mentioned analysis method of the impact of urban expansion on the carbon sequestration capacity of vegetation.

[0044] Alternatively, it includes a readable storage medium having a computer program stored thereon, and when the computer program is executed, the above-mentioned method for analyzing the impact of urban expansion on the carbon sequestration capacity of vegetation is implemented.

[0045] Compared with the prior art, the present invention has the following advantages:

[0046] 1) The new concept of the distance effect of urban expansion on vegetation carbon sequestration capacity (DHNPP) was introduced to quantify the spatial impact range of urban expansion on the net primary productivity of surrounding vegetation, filling the research gap in this field;

[0047] 2) The CASA model, Thornthwaite Memorial model and generalized additive model (GAM) were combined to systematically calculate the nuclear vegetation net primary productivity (kNPP), potential net primary productivity (PNPP) and the impact of human activities on net primary productivity (HNPP), and the DHNPP was accurately identified through GAM model fitting;

[0048] 3) Using the XGBoost machine learning model and SHAP analysis method, we conducted an in-depth analysis of various factors related to the ecological environment and urban development, accurately calculated the impact weight of each factor on DHNPP, and improved the interpretability and analysis depth of the model;

[0049] 4) This paper deeply reveals the negative impact of human activity intensity on the carbon sequestration capacity of vegetation, and emphasizes the need to balance the relationship between economic development and ecological protection in the urbanization process. The research results provide a scientific basis for urban planning and ecological environment management, and can be used to guide urban green development strategies, optimize urban layout, and reduce the adverse effects of urban expansion on the ecological environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 This is a flow chart of a method for analyzing the impact of urban expansion on the carbon sequestration capacity of vegetation according to an embodiment of the present invention.

[0052] Figure 2 This is a spatial distribution map of the kernel normalized difference vegetation index (kNDVI) in the study area of ​​the embodiment of the present invention.

[0053] Figure 3 This is the spatial distribution map of the net primary productivity (kNPP) of the core vegetation in the study area of ​​the embodiment of the present invention.

[0054] Figure 4 This is a spatial distribution map of the impact of human activities on vegetation net primary productivity (HNPP) in the study area of ​​the embodiment of the present invention.

[0055] Figure 5 This is a schematic diagram of a circular buffer zone divided with the geometric center of the built-up area as the center in the study area of ​​an embodiment of the present invention.

[0056] Figure 6 The impact distance (DHNPP) of urban expansion on vegetation carbon sequestration capacity identified by the embodiment of the present invention.

[0057] Figure 7 The importance of the factors related to urban development and environmental quality on the distance of influence of urban expansion on vegetation carbon sequestration capacity (DHNPP) in the embodiment of the present invention is ranked. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in combination with the drawings and embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] Example 1

[0060] like Figure 1 As shown, an embodiment of the present invention provides a method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity, comprising the following steps:

[0061] Step 1: Obtain satellite remote sensing data, long-term land use classification data and built-up area vector range of the study area, and calculate the monthly average temperature, monthly total precipitation, monthly total solar radiation data and kernel normalized difference vegetation index kNDVI.

[0062] This embodiment uses the land use classification data of the China Land Cover Dataset (CLCD) study area from 2000 to 2020 to match the vegetation light energy utilization rate of different land use types. The vector range of the built-up area in the study area is obtained for statistics in subsequent steps. The meteorological station data of the monthly average temperature, monthly total precipitation, and monthly total solar radiation are collected month by month. First, the collected meteorological station data is converted into a point vector format, and then converted into raster data with a spatial resolution of 250m through thin disk spline interpolation.

[0063] The monthly Kernel Normalized Difference Vegetation Index (kNDVI) was calculated using satellite remote sensing data of the study area. The calculation formula is as follows:

[0064] (1)

[0065] In the formula, Represents the near-infrared band of remote sensing images. Represents the red light band of satellite remote sensing images. is a length scale parameter, setting , then the formula of kNDVI is transformed into:

[0066] (2)

[0067] Where NDVI stands for Normalized Difference Vegetation Index.

[0068] The spatial distribution of kNDVI in the study area is shown in Figure 2 shown.

[0069] Step 2: Based on the CASA model, the land use classification data, monthly mean temperature, monthly total precipitation, monthly total solar radiation data of the study area obtained in step 1 and kNDVI were used to calculate the core vegetation net primary productivity.

[0070] The CASA model is a light energy utilization model used to estimate the net primary productivity of vegetation. The data required for calculation include kNDVI grid data and temperature, precipitation, and total solar radiation grid data. Kernel Net Primary Productivity (kNPP) is calculated in the Monteith equation of the light energy utilization model by using the kernel photosynthetically active radiation kAPAR absorbed by vegetation and the actual light energy utilization rate of land use types. Calculated, that is:

[0071] (3)

[0072] (4)

[0073] (5)

[0074] In the formula, Represents pixel x exist t The net primary productivity of vegetation in the month ; Represents pixel x exist t Nuclear photosynthetically active radiation absorbed by the moon ; For pixels x exist t Actual light energy utilization rate of vegetation in January , whose value depends on the land use, temperature and precipitation in the study area; Represents pixel x exist t Monthly photosynthetically active radiation ; Represents pixel x exist t Total solar radiation per month ; The constant 0.5 means that vegetation is at a wavelength of 0.38-0.71 The ratio of effective solar radiation that can be used by the location to the total solar radiation; Represents pixel x exist t The specific calculation method of the effective photosynthetic radiation absorption ratio of the nuclear vegetation in a month is as follows:

[0075] (6)

[0076] (7)

[0077] In the formula, and are the maximum and minimum values ​​of the vegetation effective photosynthetic radiation absorption ratio FPAR, which are 0.95 and 0.001 respectively in this embodiment; the kernel vegetation coverage kFVC is calculated using the pixel binary model, and the 95% confidence interval lower and upper limits of the kernel normalized difference vegetation index kNDVI value in the study area are selected as the pure bare soil pixel and pure vegetation pixels .

[0078] The spatial distribution of kNPP in the study area is shown in Figure 3 shown.

[0079] Step 3: Based on the monthly average temperature and monthly total precipitation data collected in step 1, the potential net primary productivity of vegetation is inverted through the Thornthwaite Memorial model, and the impact of human activities on the net primary productivity of vegetation HNPP is calculated by the difference between the core net primary productivity of vegetation and the potential net primary productivity of vegetation.

[0080] The Thornthwaite Memorial model is derived by establishing the least squares relationship between measured vegetation data and evapotranspiration, temperature and precipitation data. The calculation formula is as follows:

[0081] (8)

[0082] (9)

[0083] (10)

[0084] Where PNPP is the potential net primary productivity of vegetation, e is a natural constant, v is the actual evapotranspiration (mm), r is the total annual precipitation (mm), L is the annual average evapotranspiration (mm), t is the annual average temperature (℃).

[0085] The impact of human activities on vegetation net primary productivity is calculated by the difference between PNPP and kNPP. The calculation formula is as follows:

[0086] (11)

[0087] Where HNPP represents the impact of human activities on vegetation net primary productivity, PNPP is the potential net primary productivity of vegetation, and kNPP is the core net primary productivity of vegetation.

[0088] when When , it means that human activities have a negative impact on the net primary productivity of vegetation; when This indicates that human activities have a positive impact on vegetation net primary productivity. The spatial distribution of HNPP in the study area is shown in the figure. Figure 4 shown.

[0089] Step 4: Use the built-up area vector range data of the study area obtained in step 1 to calculate the geometric center of the built-up area, construct multiple annular buffers based on the geometric center, and calculate the average HNPP value in each annular buffer. The fitting curve of the annular buffer radius and the HNPP value in the annular buffer is obtained by fitting the generalized additive model. The impact distance DHNPP of urban expansion on the carbon sequestration capacity of vegetation is identified according to the changes and inflection points of the curve.

[0090] Calculate the geometric center of the built-up area vector range of the study area. Divide the built-up area vector range into multiple triangles, calculate the area and centroid of each triangle, and weight the triangle centroid according to the proportion of the triangle area to the total area of ​​the built-up area vector range to obtain the geometric center of the built-up area vector range. The specific calculation formula is as follows:

[0091] (12)

[0092] Where P is the geometric center of the built-up area vector range, N is the number of triangles, For the i The area of ​​a triangle, For the i The centroid of a triangle.

[0093] like Figure 5 As shown in the figure, with the geometric center P as the center, multiple circular buffers with radii ranging from 1km to 50km are established at intervals of 1km, and the average HNPP value in each circular buffer is calculated. Fitting the ring buffer radius via a generalized additive model (GAM) a and The nonlinear relationship between the radius of the circular buffer and the average HNPP value in the circular buffer is obtained. The specific calculation formula is as follows:

[0094] (13)

[0095] In the formula, represents the intercept of the fitted model; is a smooth function that reflects the radius a and The nonlinear relationship between represents the random error term, which is assumed to be normally distributed.

[0096] Due to urban expansion, the impact of human activities on the carbon sequestration capacity of vegetation will change, and the impact will be weakened after a certain distance away from the urban built-up area, and the carbon sequestration capacity of vegetation will eventually stabilize. r The corresponding Fitting is performed, and the radius of the circular buffer is a Compared with the average HNPP value in the ring buffer The maximum slope of the fitting curve times as the slope threshold In this embodiment The value is 0.2. When the radius of the ring buffer HNPP value in the ring buffer After the maximum slope of the fitted curve appears, the first slope is less than the slope threshold The location is the distance DHNPP from which urban expansion affects the carbon sequestration capacity of vegetation. Figure 6 shown.

[0097] Step 5: Select factors related to urban development and environmental quality, establish the relationship between each factor and DHNPP through the XGBoost model, and use SHAP analysis to calculate the impact weight of each factor on DHNPP.

[0098] The factors related to urban development and environmental quality selected in this embodiment include: temperature (T), precipitation (P), total solar radiation (R), vegetation cover (FVC), soil moisture (SM), standardized precipitation evapotranspiration index (SPEI), population (POP), population growth rate (PGR), urban area ratio (UR), urban population ratio (UP), gross domestic product (GDP), urban area (BUA), fossil fuel CO2 emissions (FE), night light intensity (NLI), night light aggregation index (NLAI). In order to ensure the consistency of the data, all the above data are unified to the same spatial and temporal resolution. For missing values ​​and outliers, the data preprocessing method can be used to fill the mean or directly delete. In order to eliminate the influence of different dimensions on the subsequent training of the XGBoost model, the data is standardized to obtain a data set. The data set is divided into a training set and a test set. The XGBoost model is trained using the training set data to establish the relationship between each factor and DHNPP. The XGBoost model parameters are tuned using the cross-validation method to optimize the performance and generalization ability of the model, and the model is evaluated on the test set data. The SHAP analysis method is used to calculate the impact weight of each factor on DHNPP. The specific calculation formula is as follows:

[0099] (14)

[0100] In the formula, Representation elements The impact weight on DHNPP, i.e., the factor SHAP value of S Indicates that it does not contain features A subset of X Represents the set of all elements; Representation Subset S The number of elements in express The factorial of Representing a collection X The number of elements in express The factorial of Indicates that when the SHAP analysis model calculation process only uses a subset S Output when predicting the elements in ; Indicates when the SHAP analysis model uses a subset S Elements and elements in Output when making predictions.

[0101] The SHAP value can quantify the contribution of each feature to the prediction result. In the results of the SHAP analysis in this embodiment, the night light intensity (NLI) has the greatest influence on DHNPP. Figure 7 shown.

[0102] Example 2

[0103] Based on the same inventive concept, the present invention also provides an analysis system for the impact of urban expansion on the carbon sequestration capacity of vegetation, including a processor and a memory, the memory being used to store program instructions, and the processor being used to call the program instructions in the memory to execute the above-mentioned analysis method for the impact of urban expansion on the carbon sequestration capacity of vegetation.

[0104] Example 3

[0105] Based on the same inventive concept, the present invention also provides an analysis system for the impact of urban expansion on the carbon sequestration capacity of vegetation, including a readable storage medium, on which a computer program is stored. When the computer program is executed, an analysis method for the impact of urban expansion on the carbon sequestration capacity of vegetation as described above is implemented.

[0106] In specific implementation, the method proposed in the technical solution of the present invention can be implemented by technical personnel in this field using computer software technology to realize automatic operation process. System devices for implementing the method, such as computer-readable storage media storing the corresponding computer program of the technical solution of the present invention and computer equipment running the corresponding computer program, should also be within the scope of protection of the present invention.

[0107] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific implementation cases described, or replace them with similar methods, without departing from the spirit of the present invention or exceeding the scope defined by the appended claims.

Claims

1. A method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity, characterized in that: The following steps are involved: Step 1: Obtain satellite remote sensing data, long-term land use classification data and built-up area vector range of the study area, and calculate the monthly average temperature, monthly total precipitation, monthly total solar radiation data and kernel normalized difference vegetation index kNDVI; Step 2, based on the CASA model, the land use classification data, monthly average temperature, monthly total precipitation, monthly total solar radiation data of the study area obtained in step 1, and kNDVI were used to calculate the core vegetation net primary productivity; Step 3: Based on the monthly average temperature and monthly total precipitation data collected in step 1, the potential net primary productivity of vegetation is inverted through the Thornthwaite Memorial model, and the impact of human activities on the net primary productivity of vegetation HNPP is calculated by the difference between the net primary productivity of vegetation and the potential net primary productivity of vegetation; Step 4: Use the built-up area vector range data of the study area obtained in step 1 to calculate the geometric center of the built-up area, construct multiple annular buffers based on the geometric center, and calculate the average HNPP value in each annular buffer. The fitting curve of the annular buffer radius and the HNPP value in the annular buffer is obtained by generalized additive model fitting. According to the changes and inflection points of the curve, the impact distance DHNPP of urban expansion on vegetation carbon sequestration capacity is identified; Step 5: Select factors related to urban development and environmental quality, establish the relationship between each factor and DHNPP through the XGBoost model, and use SHAP analysis to calculate the impact weight of each factor on DHNPP.

2. The method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity according to claim 1, characterized in that: In step 1, the satellite remote sensing data of the study area is used to calculate the monthly kernel normalized difference vegetation index. The calculation formula is as follows: （1） Where kNDVI represents the kernel normalized difference vegetation index, Represents the near-infrared band of remote sensing images. Represents the red light band of satellite remote sensing images. is a length scale parameter, setting , then the formula of kNDVI is transformed into: （2） Where NDVI stands for Normalized Difference Vegetation Index.

3. The method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity according to claim 1, characterized in that: In step 2, the core vegetation net primary productivity is calculated in the Monteith equation of the light energy utilization model by the core photosynthetically active radiation kAPAR absorbed by the vegetation and the actual light energy utilization rate of the land use type. Calculated, that is: （3） （4） （5） In the formula, Represents pixel x exist t Monthly nuclear vegetation net primary productivity; Represents pixel x exist t Nuclear photosynthetically active radiation absorbed by the moon; For pixels x exist t The actual light energy utilization rate of vegetation in a month depends on the land use, temperature and precipitation in the study area; Represents pixel x exist t Monthly photosynthetically active radiation; Represents pixel x exist t Total solar radiation for the month; Represents pixel x exist t The specific calculation method of the effective photosynthetic radiation absorption ratio of the nuclear vegetation in a month is as follows: （6） （7） In the formula, and They are the maximum and minimum values ​​of the vegetation effective photosynthetic radiation absorption ratio FPAR; The kernel vegetation coverage kFVC was calculated using the pixel binary model, and the lower and upper limits of the 95% confidence interval of the kernel normalized difference vegetation index kNDVI value in the study area were selected as the pure bare soil pixels. and pure vegetation pixels .

4. The method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity according to claim 1, characterized in that: The Thornthwaite Memorial model in step 3 is derived by establishing the least squares relationship between the measured vegetation data and the evapotranspiration, temperature and precipitation data. The calculation formula is as follows: （8） （9） （10） Where PNPP is the potential net primary productivity of vegetation, e is a natural constant, v is the actual evaporation, r is the total annual precipitation, L is the annual average evapotranspiration, t The annual average temperature.

5. The method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity according to claim 4, characterized in that: In step 3, the difference between the core net primary productivity of vegetation and the potential net primary productivity of vegetation is used to calculate the impact of human activities on the net primary productivity of vegetation. The calculation formula is as follows: （11） In the formula, HNPP represents the impact of human activities on vegetation net primary productivity, PNPP is the potential net primary productivity of vegetation, and kNPP is the core net primary productivity of vegetation; when When , it means that human activities have a negative impact on the net primary productivity of vegetation; when This indicates that human activities have a positive impact on the net primary productivity of vegetation.

6. The method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity according to claim 1, characterized in that: In step 4, the built-up area vector range is divided into multiple triangles, and the area and centroid of each triangle are calculated. The centroid of the triangle is weighted according to the proportion of the triangle area to the total area of ​​the built-up area vector range to obtain the geometric center of the built-up area vector range. The specific calculation formula is as follows: （12） Where P is the geometric center of the built-up area vector range, N is the number of triangles, For the i The area of ​​a triangle, For the i The centroid of a triangle.

7. The method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity according to claim 6, characterized in that: In step 4, the geometric center of the built-up area vector range is taken as the center of the circle. km is the distance between them. km to km of multiple ring buffers, , The value is set according to actual needs. yes An integer multiple of , and calculates the average HNPP value in each ring buffer , fitting the ring buffer radius through the generalized additive model a and The nonlinear relationship between the radius of the circular buffer and the average HNPP value in the circular buffer is obtained. The specific calculation formula is as follows: （13） In the formula, represents the intercept of the fitted model; is a smooth function that reflects the radius a and The nonlinear relationship between represents the random error term, which is assumed to be normally distributed; The generalized additive model is used to calculate the radius of each ring buffer. r The corresponding Fitting is performed, and the radius of the circular buffer is a Compared with the average HNPP value in the ring buffer The maximum slope of the fitting curve times as the slope threshold , when the ring buffer radius r HNPP value in the ring buffer After the maximum slope of the fitted curve appears, the first slope is less than the slope threshold The location is the distance DHNPP from which urban expansion affects the carbon sequestration capacity of vegetation.

8. The method for analyzing the impact of urban expansion on vegetation carbon sequestration capacity according to claim 1, characterized in that: The factors related to urban development and environmental quality selected in step 5 include: temperature, precipitation, total solar radiation, vegetation coverage, soil moisture, standardized precipitation evapotranspiration index, population, population growth rate, urban area share, urban population ratio, gross domestic product, urban area, fossil fuel CO2 emissions, night light intensity, and night light concentration index; all the above factors are unified to the same spatial and temporal resolution, outliers are directly deleted, and missing values ​​are processed by mean filling, and then the data is standardized to eliminate the influence of different dimensions on the subsequent training of the XGBoost model to obtain a data set; the data set is divided into a training set and a test set, and the XGBoost model is trained using the training set to establish the relationship between each factor and DHNPP. The XGBoost model parameters are tuned using the cross-validation method to optimize the performance and generalization ability of the model, and the model is evaluated on the test set data. The SHAP analysis method is used to calculate the influence weight of each factor on DHNPP. The specific calculation formula is as follows: （14） In the formula, Representation elements The impact weight on DHNPP, i.e., the factor SHAP value of S Indicates that it does not contain features A subset of X Represents the set of all elements; Representation Subset S The number of elements in express The factorial of Representing a collection X The number of elements in express The factorial of Indicates that when the SHAP analysis model calculation process only uses a subset S Output when predicting the elements in ; Indicates when the SHAP analysis model uses a subset S Elements and elements in Output when making predictions.

9. An analysis system for the impact of urban expansion on vegetation carbon sequestration capacity, characterized in that: It includes a processor and a memory, the memory is used to store program instructions, and the processor is used to call the program instructions in the memory to execute an analysis method for the impact of urban expansion on the carbon sequestration capacity of vegetation as described in any one of claims 1-8.

10. An analysis system for the impact of urban expansion on vegetation carbon sequestration capacity, characterized in that: It includes a readable storage medium, on which a computer program is stored. When the computer program is executed, an analysis method for the impact of urban expansion on the carbon sequestration capacity of vegetation as described in any one of claims 1 to 8 is implemented.

Images