Evaluation and attribution method for service change of farmland-wetland composite ecosystem

Abstract

The invention discloses an evaluation and attribution method for service change of a farmland-wetland composite ecosystem, relates to the technical field of environmental evaluation, and aims to solve the problem that the service change evaluation of the farmland-wetland composite ecosystem is inaccurate in an existing evaluation method. The method comprises the following steps: 1, acquiring land utilization / coverage data and driving force factor data; 2, model parameters are set, and output results about habitat quality, water yield, soil conservation, wind prevention and sand fixation and grain production are obtained; 3, establishing a comprehensive evaluation index system, wherein the system comprises three aspects: supply service, adjustment service and support service; 4, selecting five ecosystem service indexes, and performing comprehensive quantitative evaluation on the farmland-wetland composite ecosystem service; 5, analyzing space-time evolution characteristics by adopting a cold and hot spot pattern analysis method; and 6, analyzing the interpretation force of different driving factors on each ecosystem service through a generalized linear mixed effect model. The method is used for accurately evaluating the service change of the farmland-wetland composite ecosystem.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental assessment, and in particular to an assessment and attribution method for service changes in a farmland-wetland composite ecosystem. Background Art

[0002] The farmland-wetland complex ecosystem is an important ecosystem that maintains biodiversity, regulates climate, protects water resources, and provides a variety of ecosystem services. However, the dual impacts of climate change and human activities have exacerbated the pressure on the ecosystem. Climate warming and frequent extreme weather events have had a serious impact on the ecosystem, while irrational land use and overexploitation of water resources have further deteriorated the ecological environment, leading to the degradation of the service functions of the farmland-wetland complex ecosystem, seriously threatening its sustainable development. Therefore, quantitative analysis of the changes in farmland-wetland complex ecosystem services and their driving factors is crucial to promoting regional ecological construction and sustainable development.

[0003] In the current technical field, the assessment of farmland and wetland ecosystem services is often carried out independently, ignoring their interdependence and synergy as a whole, which makes it difficult to meet the needs of ecological protection and sustainable agricultural development. Therefore, the current technical problem that needs to be solved is the lack of a method to accurately assess the changes in farmland-wetland composite ecosystem services and attribute and analyze their driving factors. Summary of the invention

[0004] The present invention aims to solve the problem of inaccurate assessment of service changes in farmland-wetland composite ecosystems in existing assessment methods, and to provide an assessment and attribution method for service changes in farmland-wetland composite ecosystems.

[0005] The method for assessing and attributing changes in farmland-wetland composite ecosystem services of the present invention comprises the following steps:

[0006] Step 1: Obtain land use / cover data and driving factor data;

[0007] Step 2: Select an ecosystem service assessment model based on the research objectives and regional characteristics. Based on the data obtained in step 1, set the model parameters, including land use type, soil properties, meteorological conditions, hydrological data, digital elevation data, vegetation data, and snow cover data. By running simulations in the study area, output results on habitat quality, water yield, soil conservation, windbreak and sand fixation, and food production can be obtained.

[0008] Step 3: Establish a comprehensive evaluation indicator system, which includes three aspects: supply services, regulation services and support services;

[0009] Step 4: Starting from the three dimensions of provisioning services, regulating services, and supporting services, five ecosystem service indicators were selected to conduct a comprehensive quantitative assessment of farmland-wetland composite ecosystem services; ecosystem service indicators for provisioning services included grain yield and water yield, ecosystem service indicators for regulating services included soil conservation and windbreak and sand fixation, and ecosystem service indicators for supporting services included habitat quality;

[0010] Step 5: Based on the five ecosystem service indicators in step 4, the hot and cold spot pattern analysis method is used to comprehensively analyze the spatiotemporal evolution characteristics of farmland-wetland composite ecosystem services;

[0011] Step 6. Use a generalized linear mixed effects model to analyze the explanatory power of different driving factors on changes in ecosystem services, accurately identify and quantify the specific impact of each driving factor on changes in different ecosystem services, reveal the main factors affecting changes in ecosystem services, and complete the attribution analysis of changes in farmland-wetland complex ecosystem services.

[0012] Furthermore, the land use data and land cover data described in step 1 are derived from the Resources and Environment Data Center of the Chinese Academy of Sciences.

[0013] Furthermore, the driving factor data in step 1 include natural factors and socio-economic factors. Natural factors include precipitation, evaporation, temperature, vegetation coverage, slope and elevation, and the data are from "Hourly Observation Data of China's Ground Meteorological Stations", the National Qinghai-Tibet Plateau Data Center and the Geospatial Data Cloud. Socio-economic factors include population density and gross national product, which are from the Jilin Province Statistical Yearbook.

[0014] Furthermore, the ecosystem service assessment models described in step 2 are the InVEST model, the RWEQ model and the food yield revision model.

[0015] Furthermore, the calculation formula for habitat quality in step 4 is as follows:

[0016]

[0017] In the formula, Q xj represents the habitat quality of the j-type land use grid cell x, H j is the habitat suitability of the j-type land use grid unit, k is the half-saturation constant with a value of 15, and z is the normalized index with a value of 2.5.

[0018] Furthermore, the calculation formula for water production in step 4 is as follows:

[0019]

[0020] Where Yxj is the annual water yield (mm) of grid cell x on land use type j, AETxj is the annual average evapotranspiration of grid cell x on land use type j, and Px is the annual precipitation of grid cell x.

[0021] Furthermore, the calculation formula for soil conservation in step 4 is as follows:

[0022] SEDRET=PKLS x -USLE x +SEDR x (3)

[0023] In the formula, SEDRET x and SEDR x represents the soil retention and sediment retention of a certain land use type grid cell x, PKLS x It represents the potential soil loss of a certain land use type grid cell x, and USLEx represents the actual erosion of a certain land use type grid cell x, that is, the soil erosion under vegetation cover and soil and water conservation measures.

[0024] Furthermore, the calculation formula for windbreak and sand fixation in step 4 is as follows:

[0025] SR=S L潜 -S L (4)

[0026] Where SR is the amount of sand fixation (kg / m 2 ), S L潜 Indicates potential wind erosion (kg / m 2 ), S L Indicates the actual wind erosion (kg / m 2 ).

[0027] Furthermore, the calculation formula for grain production in step 4 is as follows:

[0028]

[0029] Where Gi is the agricultural and animal husbandry product output allocated to the i-th grid (kg / hm 2 a 1 ), G sum is the total output of agricultural and animal husbandry products (kg), NDVI i Represents the i-th grid value, NDVI sum It is the sum of the NDVI of regional agricultural and pastoral vegetation.

[0030] Furthermore, the calculation formula for the spatiotemporal evolution characteristics in step 5 is as follows:

[0031]

[0032] In the formula, N represents the total number of elements in the study area, n is the number of neighbors of each element, and w i is the weight of each factor, z i is the farmland-wetland complex ecosystem service value of each element.

[0033] Furthermore, the analysis method of the explanatory power of different driving factors on each ecosystem service in step 6 is as follows:

[0034] Step 1: Take the change in the services of each farmland-wetland composite ecosystem over 20 years as the dependent variable, and the driving factors under each type of driving factor as the independent variable. Determine the dependent variable and independent variable, as well as the format of input data, screen abnormal data, make the data conform to the normal distribution, and normalize the data for preprocessing. The specific formula is as follows:

[0035]

[0036] In the formula, β is the coefficient of the independent variable, σ X and σ Y are the standard deviations of the independent and dependent variables, respectively.

[0037] Step 2: According to the data characteristics and research objectives, set the parameters of the GLMM model, select logistic regression as the link function, expand the distribution of the dependent variable to make it suitable for the exponential distribution family; at the same time, set the specific forms of fixed effects and random effects, with socioeconomic factors and climate variables as fixed effects, and geographical location and soil type as random effects;

[0038] Step 3: Use statistical software to fit the GLMM model to obtain the explanatory power of each driving factor on different ecosystem services. The specific formula is as follows:

[0039]

[0040] In the formula, y i is the actual value of the change in the farmland-wetland complex ecosystem, is the predicted value of the model, It represents the mean of the change of farmland-wetland complex ecosystem, and N is the number of samples. 2 The value range is between 0 and 1. The closer it is to 1, the stronger the model's ability to explain the variables, and the closer it is to 0, the weaker the model's ability to explain the variables.

[0041] Beneficial effects of the present invention:

[0042] This paper integrates multi-source data to construct an evaluation system suitable for composite ecosystem services, and selects five key service functions, namely habitat quality, water production, soil conservation, windbreak and sand fixation, and food production, to ensure a comprehensive and systematic evaluation of the comprehensive benefits of the composite ecosystem. This method comprehensively considers natural factors and socio-economic factors, and deeply analyzes the impact of different driving factors on the changes in various ecosystem services through a generalized linear mixed model, thereby providing a more accurate and practical method for evaluating and attributing changes in composite ecosystem services, effectively improving the accuracy of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the process of the present invention;

[0044] Figure 2 The spatiotemporal distribution and change diagram of the five services of the farmland-wetland complex ecosystem in Example 1 over the 20 years;

[0045] Figure 3 This is an analysis diagram of the driving forces of changes in farmland-wetland complex ecosystem services. DETAILED DESCRIPTION

[0046] The embodiments of the present invention are described in detail below. The following embodiments are implemented on the premise of the technical solution of the present invention, and detailed implementation schemes and specific operation processes are given, but the protection scope of the present invention is not limited to the following embodiments.

[0047] like Figure 1 As shown, the present invention proposes an assessment and attribution method for changes in farmland-wetland complex ecosystem services from the two aspects of assessment and attribution by collating relevant data.

[0048] Embodiment 1:

[0049] Take the analysis of spatiotemporal variation characteristics of ecosystem services and their driving factors in western Jilin as an example.

[0050] Step 1. Prepare basic data on farmland and wetlands in western Jilin in 2000, 2010 and 2020, including land use / cover data and driving factor data such as climate, soil, hydrology, digital elevation, and snow cover.

[0051] Step 2: According to the secondary classification, the land use data obtained in western Jilin was classified into paddy fields and dry land as farmland, and rivers, lakes, reservoirs, ponds and swamps as wetlands. Other required data were uniformly projected into WGS1984 Albers with a resolution of 500m using the RasterProject tool in the Arc GIS software. The specific steps are as follows:

[0052] (1) Model selection: Based on the research objectives and data characteristics, the InVEST model, RWEQ model and grain yield revision model were selected.

[0053] (2) Parameter setting: According to the collected basic data, the model parameters are set, including land use type, soil properties, meteorological conditions, hydrological data, digital elevation data, vegetation data and snow cover data.

[0054] (3) Simulation operation: The model is used to simulate the study area to obtain the output results of different farmland-wetland complex ecosystem services, including habitat quality, water yield, soil conservation, windbreak and sand fixation, and food production.

[0055] Through the above steps, an assessment model that accurately reflects the services of the farmland-wetland complex ecosystem is constructed, providing a data basis and scientific basis for the implementation of subsequent steps five and six.

[0056] Step 3: In this embodiment, in order to comprehensively evaluate the farmland-wetland composite ecosystem services, five ecosystem services were selected from three aspects for evaluation. The specific steps are as follows:

[0057] Supply services: including grain production and water production.

[0058] Regulation services: including soil conservation, wind and sand control.

[0059] Support services: including habitat quality, etc.

[0060] Through the above steps, a comprehensive and systematic ecosystem service assessment data can be established to provide standards for quantitative assessment.

[0061] Step 4. Based on multi-source data such as climate and land use, five service function indicators, including habitat quality, water yield, soil conservation, windbreak and sand fixation, and food production, are selected from the three levels of supply service, regulation service, and support service to evaluate the farmland-wetland complex ecosystem services.

[0062] (1) Habitat quality (HQ): Based on formula 1, the value of the habitat quality service index is calculated as follows:

[0063]

[0064] In the formula, Q xj represents the habitat quality of the j-type land use grid cell x, H j is the habitat suitability of the j-type land use grid unit, k is the half-saturation constant with a value of 15, and z is the normalized index with a value of 2.5.

[0065] When calculating habitat quality, the key is to correctly assess the habitat suitability and stress impact of each grid cell. Therefore, detailed sensitivity data and stress factor parameters are required. The specific parameters are as follows:

[0066] Sensitivity data: reflects the coercion intensity of different land use types or vegetation cover types to various threat sources. This study obtained the sensitivity data by referring to the InVEST model user manual, relevant literature and field investigations, as shown in Table 1. In this embodiment, the factors affecting the development of ecological land are defined as ecological threat factors, including: roads (road), industrial land (ind), construction land (urban), cultivated land (agr), saline-alkali land (silt), rural settlements (res), etc.

[0067] Intimidation factor: The intensity of intimidation includes the maximum impact distance (MAX_DIST), weight (WEIGHT) and decay coefficient (DECAY) of the threat factor, as shown in Table 2.

[0068] Table 1 Habitat sensitivity parameters of InVEST habitat quality module

[0069]

[0070] Table 2 InVEST habitat quality module threat factor impact distance and weight table

[0071]

[0072] (2) Water yield (WY): Water yield is estimated based on Formula 2 by estimating the precipitation for each grid cell and subtracting it from the actual evapotranspiration to estimate the total annual water yield:

[0073]

[0074] In the formula, Yxj is the annual water yield of grid cell x on a certain land use type j (mm), AETxj is the annual average evapotranspiration of grid cell x on land use type j, and Px is the annual precipitation of grid cell x. During the operation of the water production model, since the Z value is an adjustable parameter, a sensitivity test is required. By combining the data of the "Jilin Province Water Resources Bulletin", comparing and verifying the actual value and simulated value of water production in 2020, and adjusting the test many times, the seasonal factor Z value was finally selected as 3.5.

[0075] The biophysical parameter table (CSV) required by the InVEST water production module includes information such as plant root depth (root_depth) and crop evapotranspiration coefficient (Kc). For details, please see Table 3:

[0076] Table 3 Biophysical coefficients of InVEST water production module

[0077]

[0078] (3) Soil conservation (SC): The soil conservation amount was estimated based on Formula 3 and simulated based on the universal soil loss equation (USLE). The sum of erosion reduction and sediment retention was taken as the soil conservation amount.

[0079] SEDRET=PKLS x -USLEx+SEDR x (3)

[0080] In the formula, SEDRET x and SEDR x represents the soil retention and sediment retention of a certain land use type grid cell x, PKLS x It represents the potential soil loss of a certain land use type grid cell x, and USLEx represents the actual erosion of a certain land use type grid cell x, that is, the soil erosion under vegetation cover and soil and water conservation measures.

[0081] (4) Windbreak and sand fixation (WB): The windbreak and sand fixation amount is estimated based on the modified wind erosion (RWEQ) model. The value of potential soil wind erosion minus actual soil wind erosion is the windbreak and sand fixation amount. The actual wind erosion amount refers to the soil wind erosion amount when the surface is covered with vegetation in reality, and the potential wind erosion amount refers to the soil wind erosion amount when there is no vegetation. The calculation formula is:

[0082] SR=S L潜 -S L (4)

[0083] Where SR is the amount of sand fixation (kg / m 2 ), S L潜 Indicates potential wind erosion (kg / m 2 ), S L Indicates the actual wind erosion (kg / m 2 ).

[0084] (5) Food production (FP): Based on Formula 5, the food supply capacity is evaluated according to the agricultural and animal husbandry output per unit area of ​​different land use types. The calculation formula is as follows:

[0085]

[0086] Where Gi is the agricultural and animal husbandry product output allocated to the i-th grid (kg / hm 2 a 1 ), G sum is the total output of agricultural and animal husbandry products (kg), NDVI iRepresents the i-th grid value, NDVI sum It is the sum of the NDVI of regional agricultural and pastoral vegetation.

[0087] The spatiotemporal distribution of the five services of the farmland-wetland complex ecosystem and their changes over the past 20 years are shown in the figure. Figure 2 As shown (red represents increase, blue represents decrease). Between 2000 and 2020, the supply service and regulation service of the farmland-wetland complex ecosystem have been improved, while the support service has declined. Food production services have increased significantly in the past 20 years, and high-value areas are distributed in the southeastern and northern farming areas; at the same time, water yield and soil conservation services are increasing, and windbreak and sand fixation are decreasing. The overall habitat quality has declined, with low-value areas mainly distributed in construction land and saline-alkali wasteland, while good and excellent areas are mainly distributed in forests, grasslands, wetlands and other dominant land types.

[0088] Step 5. Based on the above five ecosystem service indicators (habitat quality, water production, soil conservation, windbreak and sand fixation, and food production), the hotspot pattern analysis method was adopted, and the hotspot analysis tool of Gi* statistics (Getis-Ord Gi*statistics) in ArcGIS software was used to visualize the ecosystem services, and the spatiotemporal evolution characteristics of the farmland-wetland composite ecosystem services in the study area were comprehensively analyzed. The specific calculation formula is as follows:

[0089]

[0090] In the formula, N represents the total number of elements in the study area, n is the number of neighbors of each element, and w i is the weight of each factor, z i is the farmland-wetland composite ecosystem service value of each element. Based on formula (6), the areas where the ecosystem service values ​​exceed their respective average values ​​are defined as "hot" spots, while the areas below the average values ​​are defined as "cold" spots. Hot and cold spot analysis can reveal areas where ecosystem services are increasing or degrading, help identify ecosystem areas that are sensitive to changes in ecosystem services and their impacts on ecological stability and sustainability, and facilitate the formulation of more scientific and effective management and protection strategies, thereby promoting the health and sustainable development of ecosystems.

[0091] Step 6: Attribution analysis is an important step to clarify the sources and influencing factors of changes in farmland-wetland complex ecosystem services. In order to analyze the main influencing factors and their contribution rates of each ecological service, the specific steps are as follows:

[0092] The generalized linear mixed model (GLMM) was used to analyze the changes in farmland-wetland complex ecosystem services. The explanatory power of different driving factors on each ecosystem service was analyzed through model fitting. The driving factor data included natural factors and socioeconomic factors. Natural factors included precipitation (PRE), evapotranspiration (EVP), temperature (TEM), vegetation cover (FVC), elevation (DEM), and slope; socioeconomic factors included population density (POP) and gross national product (GDP). By calculating the relative importance of the above factors, the relative contribution of each factor to the changes in different types of farmland-wetland complex ecosystem services was quantified.

[0093] This example uses the "lme4" package in the R4.3.2 software to run, and the process is as follows:

[0094] (1) The change in the service of each farmland-wetland composite ecosystem is taken as the dependent variable, and the driving factors under each type of driving factor are taken as the independent variable. The dependent variable and independent variable, as well as the format of the input data, are determined, abnormal data are screened, the data are made to conform to the normal distribution, and the data are normalized and preprocessed. The specific formula is as follows:

[0095]

[0096] In the formula, β is the coefficient of the independent variable, σ X and σ Y are the standard deviations of the independent and dependent variables, respectively.

[0097] (2) According to the data characteristics and research objectives, set the parameters of the GLMM model. Select an appropriate link function. This study chooses logistic regression to expand the distribution of the dependent variable to make it suitable for the exponential distribution family. At the same time, set the specific forms of fixed effects and random effects, such as socioeconomic factors, climate variables, etc. as fixed effects, and geographical location, soil type, etc. as random effects.

[0098] (3) Use statistical software to fit the GLMM model and obtain the explanatory power of each driving factor on different ecosystem services. The model's fitting effect and explanatory power are evaluated through residual analysis, variance interpretation and other indicators. The specific formula is as follows:

[0099]

[0100] In the formula, y i is the actual value of the change in the farmland-wetland complex ecosystem, is the predicted value of the model, It represents the mean of the change of farmland-wetland complex ecosystem, and N is the number of samples.2 The value range is between 0 and 1. The closer it is to 1, the stronger the model's ability to explain the variables, and the closer it is to 0, the weaker the model's ability to explain the variables.

[0101] The explanatory power of the driving factors is evaluated through residual analysis of the model. The specific steps are as follows:

[0102] 1) Fit the full model containing all independent variables (including random effects) and calculate its residual sum of squares (RSS full ):

[0103] y=β0+β1x1+β2x2+…+β p x p +μ+∈ (9)

[0104] Where y is the change in the farmland-wetland complex ecosystem, β0 is the intercept term, and β1, β2, …, β8 are the independent variables x1, x2, …, x p The corresponding regression coefficient, μ is the random effect, indicating the contribution from the random factors, ∈ is the error term, indicating the unexplained error in the model, and μ and ∈ obey the normal distribution.

[0105] 2) Fitting removes a certain independent variable x j The RSS of the simplified model (including random effects) is calculated reduced，j ), remove a certain independent variable x by fitting j Using a simplified model, we can assess the effect of each independent variable on the dependent variable individually while controlling for the effects of other independent variables and random effects. This approach helps determine the relative importance of each independent variable and understand the unique contribution of each independent variable to the variation in the dependent variable:

[0106] y=β0+β1x1+β2x2+…+β j-1 x j-1 +β j+1 x j+1 +μ+∈ (10)

[0107] Where y is the change in the farmland-wetland complex ecosystem, β0 is the intercept term, β1, β2, …, β j-1 , β j+1 , β p are the other independent variables x1, x2, …, x retained in the model j-1 , x j+1 , x p The corresponding regression coefficient, μ is the random effect, indicating the contribution from the random factors, ∈ is the error term, indicating the unexplained error in the model, and μ and ∈ obey the normal distribution.

[0108] 3) Calculate the explanatory power of driving factors:

[0109]

[0110] In the formula, RSS full To fit the full model including all independent variables, RSS reduced,j Remove an independent variable x for fitting j A simplified model of .

[0111] like Figure 3 The driving force analysis of the changes in farmland-wetland complex ecosystem services is shown in the figure. It can be seen that water yield and habitat quality are mainly affected by evaporation, windbreak and sand fixation are mainly affected by gross national product, food production is mainly affected by elevation, and soil conservation is most affected by slope.

Claims

1. A method for assessing and attributing changes in farmland-wetland complex ecosystem services, characterized in that: The method comprises the following steps: Step 1: Obtain land use / cover data and driving factor data; Step 2: Select an ecosystem service assessment model based on the research objectives and regional characteristics. Based on the data obtained in step 1, set the model parameters, including land use type, soil properties, meteorological conditions, hydrological data, digital elevation data, vegetation data, and snow cover data. By running simulations in the study area, output results on habitat quality, water yield, soil conservation, windbreak and sand fixation, and food production can be obtained. Step 3: Establish a comprehensive evaluation indicator system, which includes three aspects: supply services, regulation services and support services; Step 4: Starting from the three dimensions of provisioning services, regulating services, and supporting services, five ecosystem service indicators were selected to conduct a comprehensive quantitative assessment of farmland-wetland composite ecosystem services; ecosystem service indicators for provisioning services included grain yield and water yield, ecosystem service indicators for regulating services included soil conservation and windbreak and sand fixation, and ecosystem service indicators for supporting services included habitat quality; Step 5: Based on the five ecosystem service indicators in step 4, the hot and cold spot pattern analysis method is used to comprehensively analyze the spatiotemporal evolution characteristics of farmland-wetland composite ecosystem services; Step 6. Analyze the explanatory power of different driving factors on changes in ecosystem services through generalized linear mixed effects models, accurately identify and quantify the impact of each driving factor on changes in different ecosystem services, reveal the main factors affecting changes in ecosystem services, and complete the attribution analysis of changes in farmland-wetland complex ecosystem services.

2. The method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 1, characterized in that: The land use data and land cover data mentioned in step 1 are from the Resources and Environment Data Center of the Chinese Academy of Sciences.

3. The method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 1 or 2, characterized in that: The driving factor data in step 1 include natural factors and socio-economic factors; natural factors include precipitation, evaporation, temperature, vegetation coverage, slope and elevation, and socio-economic factors include population density and gross national product.

4. The method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 1, characterized in that: The ecosystem service assessment models described in step 2 are the InVEST model, the RWEQ model and the revised food production model.

5. The method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 1, characterized in that: The calculation formula for habitat quality in step 4 is as follows: In the formula, Q xj represents the habitat quality of the j-type land use grid cell x, H j is the habitat suitability of the j-type land use grid unit, k is the half-saturation constant with a value of 15, and z is the normalized index with a value of 2.

5.

6. A method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 1 or 5, characterized in that: The calculation formula for water production in step 4 is as follows: Where Yxj is the annual water yield (mm) of grid cell x on land use type j, AETxj is the annual average evapotranspiration of grid cell x on land use type j, and Px is the annual precipitation of grid cell x. Furthermore, the calculation formula for soil conservation in step 4 is as follows: SEDRET=PKLS x -USLE x +SEDR x (3) In the formula, SEDRET x and SEDR x represents the soil retention and sediment retention of a certain land use type grid cell x, PKLS x It represents the potential soil loss of a certain land use type grid cell x, and USLEx represents the actual erosion of a certain land use type grid cell x, that is, the soil erosion under vegetation cover and soil and water conservation measures.

7. The method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 6, characterized in that: The calculation formula for windbreak and sand fixation in step 4 is as follows: SR=S L潜 -S L (4) Where SR is the amount of sand fixation (kg / m 2 ), S L潜 Indicates potential wind erosion (kg / m 2 ), S L Indicates the actual wind erosion (kg / m 2 ).

8. The method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 7, characterized in that: The calculation formula for grain production in step 4 is as follows: Where Gi is the agricultural and animal husbandry product output allocated to the i-th grid (kg / hm 2 a 1 ), G sum is the total output of agricultural and animal husbandry products (kg), NDVI i Represents the i-th grid value, NDVI sum It is the sum of the NDVI of regional agricultural and pastoral vegetation.

9. The method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 1, characterized in that: The calculation formula of the spatiotemporal evolution characteristics in step 5 is as follows: In the formula, N represents the total number of elements in the study area, n is the number of neighbors of each element, and w i is the weight of each factor, z i is the farmland-wetland complex ecosystem service value of each element.

10. The method for assessing and attributing changes in farmland-wetland complex ecosystem services according to claim 9, characterized in that: The analysis method of the explanatory power of different driving factors on each ecosystem service in step 6 is: Step 1: Take the change in the services of each farmland-wetland composite ecosystem over 20 years as the dependent variable, and the driving factors under each type of driving factor as the independent variable. Determine the dependent variable and independent variable, as well as the format of input data, screen abnormal data, make the data conform to the normal distribution, and normalize the data for preprocessing. The specific formula is as follows: In the formula, β is the coefficient of the independent variable, σ X and σ Y are the standard deviations of the independent and dependent variables, respectively. Step 2: According to the data characteristics and research objectives, set the parameters of the GLMM model, select logistic regression as the link function, expand the distribution of the dependent variable to make it suitable for the exponential distribution family; at the same time, set the specific forms of fixed effects and random effects, with socioeconomic factors and climate variables as fixed effects, and geographical location and soil type as random effects; Step 3: Use statistical software to fit the GLMM model to obtain the explanatory power of each driving factor on different ecosystem services. The specific formula is as follows: In the formula, y i is the actual value of the change in the farmland-wetland complex ecosystem, is the predicted value of the model, It represents the mean of the change of farmland-wetland complex ecosystem, and N is the number of samples. 2 The value range is between 0 and 1. The closer it is to 1, the stronger the model's ability to explain the variables, and the closer it is to 0, the weaker the model's ability to explain the variables.

Images