Urban industrial park land utilization efficiency measurement and influence evaluation method

Abstract

The invention belongs to the technical field of urban land utilization and urban industrial park construction, and provides an urban industrial park land utilization efficiency measurement and influence evaluation method, which comprises the steps of basic evaluation database collection, current park data calculation, variable setting, staggered double difference evaluation model construction, land utilization efficiency measurement and influence evaluation analysis. The urban industrial park net carbon emission and the urban industrial park land utilization efficiency are evaluated and compared through carbon budget accounting, the global super-efficiency-based epsilon measurement model and the staggered difference model, data information is further analyzed and mined, key factors influencing the urban industrial park land utilization efficiency are clarified, and the urban industrial park net carbon emission and the urban industrial park land utilization efficiency are further analyzed and mined. And high-precision and dynamic evaluation of the land utilization efficiency of the urban industrial park is facilitated.

Description

Technical Field

[0001] This invention relates to the field of urban land use and urban industrial park construction technology, and in particular to a method for measuring the land use efficiency and assessing the impact of urban industrial parks. Background Technology

[0002] As a key platform for the integrated development of new industrialization and new urbanization, urban industrial parks are accelerating towards a high-quality development stage, becoming the main engine of local economic development. Urban industrial parks should not only strengthen their leading industries and cultivate emerging industries, but also play a demonstrative role in intensive land use, efficient energy utilization, and the application of green technologies. To improve the construction efficiency of urban industrial parks, it is necessary to assess whether their output value reaches the preset annual total output value and to pay attention to their land use efficiency, which can promote the transformation of parks from industrial clusters to innovation sources and green and low-carbon demonstration zones. However, the redundancy and over-construction of urban industrial parks have led to serious encroachment on carbon sinks such as agriculture and forest land, resulting in land vacancy and low land use efficiency, increasing carbon emissions and hindering the achievement of dual-carbon goals. In recent years, the transformation of urban industrial parks has encountered two major challenges: on the one hand, regional industrial coordination has weakened the unique advantages of urban industrial parks in pioneering experiments; on the other hand, the constraints of land quota bottlenecks have become more pronounced.

[0003] Currently, the assessment of land use efficiency in urban industrial parks lacks dynamic and high-precision considerations. In particular, in the context of the net-zero carbon target, there is a lack of a scientific, reasonable and accurate assessment method for land use efficiency in urban industrial parks, which further makes it difficult to clarify the relationship between land use in urban industrial parks and the achievement of net-zero carbon targets such as air pollution and carbon emissions. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide a method for measuring and assessing the land use efficiency of urban industrial parks, thereby addressing the current lack of dynamic and high-precision methods for assessing the land use efficiency of urban industrial parks.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] A method for measuring land use efficiency and assessing the impact of urban industrial parks includes:

[0007] Collect data on urban industrial parks, environmental data, historical land use efficiency data, and driving factor data to obtain a basic evaluation database;

[0008] The net carbon emissions of the target park's urban industrial park are calculated based on the aforementioned assessment database, and the land use efficiency of the target park's urban industrial park is calculated using the ε-metric model based on global superefficiency, to obtain the current park data.

[0009] Define the dependent variable, independent variable, and control variable;

[0010] Based on the assessment database and the current park data, an alternating difference-in-differences assessment model is constructed using the dependent variable, the independent variable, and the control variable.

[0011] The land use efficiency of the industrial park under evaluation is assessed using the staggered double difference evaluation model, and the land use efficiency measurement results of the urban industrial park are obtained.

[0012] The staggered difference-in-differences evaluation model was subjected to benchmark comparison, robustness testing, and dual-effect analysis to obtain the impact assessment results.

[0013] The present invention discloses the following technical effects:

[0014] This invention provides a method for measuring and assessing the land use efficiency of urban industrial parks. By using carbon budget accounting, an ε-measurement model based on global superefficiency, and an alternating difference model, it solves the problem of the current lack of dynamic and high-precision methods for assessing the land use efficiency of urban industrial parks. It enables the assessment and comparison of net carbon emissions and land use efficiency of urban industrial parks, as well as the identification of key factors affecting the land use efficiency of urban industrial parks. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A schematic diagram illustrating the process of measuring and assessing the land use efficiency of urban industrial parks, provided in an embodiment of the present invention.

[0017] Figure 2 This is a flowchart illustrating the method for measuring and assessing the land use efficiency of urban industrial parks, as provided in an embodiment of the present invention. Figure 2 (a) Data preparation section, Figure 2 (b) is the dynamic measurement of land use efficiency. Figure 2 (c) is the impact assessment section;

[0018] Figure 3 The assessment results of the average net carbon emission development trend of an industrial park in a certain city from 2006 to 2020 are provided for embodiments of the present invention.

[0019] Figure 4The assessment results of the development trend of land use efficiency in an industrial park in a certain city from 2006 to 2020 are provided for embodiments of the present invention.

[0020] Figure 5 The parallel trend test and dynamic effect analysis results of a robustness test for an industrial park in a certain city provided in an embodiment of the present invention;

[0021] Figure 6 The results of a placebo trial analysis for robustness testing of an industrial park in a certain city, as provided in an embodiment of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The purpose of this invention is to provide a method for measuring and assessing the land use efficiency of urban industrial parks, thereby addressing the current lack of dynamic and high-precision methods for assessing the land use efficiency of urban industrial parks.

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] Figure 1 This is a schematic diagram of the urban industrial park land use efficiency measurement and impact assessment process provided in an embodiment of the present invention, such as... Figure 1 As shown, this invention provides a method for measuring and assessing the land use efficiency of urban industrial parks, including:

[0026] Step 100: Collect data on urban industrial parks, environmental data, historical land use efficiency data, and driving factor data to obtain a basic evaluation database;

[0027] Step 200: Calculate the net carbon emissions of the target park's urban industrial park based on the assessment basic database, and calculate the land use efficiency of the target park's urban industrial park using the ε-metric model based on global superefficiency to obtain the current park data;

[0028] Step 300: Define the dependent variable, independent variable, and control variable;

[0029] Step 400: Construct an alternating difference-in-differences evaluation model based on the basic evaluation database, the current park data, the dependent variable, the independent variable, and the control variable;

[0030] Step 500: Use the staggered double difference evaluation model to evaluate the land use efficiency of the park to be evaluated, and obtain the measurement results of the land use efficiency of the urban industrial park.

[0031] Step 600: Perform benchmark comparison, robustness testing, and dual-effect analysis on the staggered difference evaluation model to obtain the impact assessment results.

[0032] Specifically, data on urban industrial parks, environmental data, historical land use efficiency data, and driving factor data are collected to obtain a basic assessment database, including:

[0033] Collect data on the urban industrial parks; the urban industrial park data includes: data on science and technology parks and data on industrial parks;

[0034] The environmental data is obtained by collecting carbon emission data, nighttime light data, and net primary productivity data.

[0035] Collect the aforementioned historical land use efficiency data;

[0036] The driving factor data is obtained by collecting labor force data, land data, capital data, and GDP data; the labor force data refers to the number of employees in the primary to tertiary industries; the land data refers to the built-up area.

[0037] By integrating the data from the urban industrial parks, the environmental data, the historical land use efficiency data, and the driving factor data, a basic evaluation database is obtained.

[0038] Furthermore, based on the aforementioned assessment database, the net carbon emissions of the target industrial park are calculated, and the land use efficiency of the target industrial park is calculated using an ε-metric model based on global superefficiency, yielding current park data, including:

[0039] The carbon sink formula is calculated based on the aforementioned assessment database to obtain the carbon sink; the urban carbon sink formula is as follows: ;in, For carbon sequestration; Total net primary productivity;

[0040] The pixel brightness of the remote sensing images in the basic evaluation database is summed to obtain the total pixel brightness.

[0041] Based on the sum of pixel brightness, the PSO-BP algorithm is used to calculate a weighted average of the pixel brightness in the remote sensing image, resulting in a weighted average brightness value. The expression for the weighted average brightness value is as follows: ;in, The weighted average value of the brightness; This is the i-th weighted weight; The brightness of the i-th remote sensing image pixel; The number of pixels contained in the image;

[0042] The carbon emissions for the city are calculated using the brightness-weighted average value formula; the carbon emission formula is as follows: ;in, The carbon emissions of the city; The amount of activity or energy consumption recorded in the energy balance sheet; Let k be the emission factor. The total number of energy types; The mapping function between brightness and carbon emissions is established based on the PSO-BP algorithm;

[0043] The difference between the city's carbon emissions and the carbon sink is calculated to obtain the net carbon emissions of the city's industrial park; the expression for the net carbon emissions of the city's industrial park is: ;in, The net carbon emissions of the urban industrial park.

[0044] Specifically, the net carbon emissions of the target park's urban industrial park are calculated based on the aforementioned assessment database, and the land use efficiency of the target park's urban industrial park is calculated using a global super-efficiency-based ε-metric model to obtain current park data. This also includes:

[0045] Define the input variables for the ε measurement model; the input variables include: output variables and input variables;

[0046] A production possibility set is constructed based on the input variables; the expression for the production possibility set is:

[0047] ;in, For the set of production possibilities; Let be the input amount of the j-th decision-making unit in period t; Let be the expected output of the j-th decision-making unit in period t; Let be the undesirable output of the j-th decision unit in period t; The weights are linear combination weights; This represents the input-output vector of the current target decision unit; For the number of periods; The number of decision-making units;

[0048] Under the constraints of the production possibility set, the target decision unit is solved based on the ε-metric model of global superefficiency to obtain the superefficiency score; the expression of the ε-metric model is:

[0049] ;in,

[0050] ; The score for the super-efficiency; These are the weights of the input factor, the expected output factor, and the undesired output factor, respectively. For input-oriented radial contraction factor; Output-oriented radial expansion factor; To introduce slack variables; To produce slack variables; To avoid producing slack variables; These are the non-radial adjustment weights for controlling inputs, expected outputs, and undesired outputs, respectively; These represent the input, expected output, and unexpected output of the park to be evaluated.

[0051] The land use efficiency of the urban industrial park is obtained by performing interval matching on the super-efficiency score according to the preset efficiency interval distribution.

[0052] Furthermore, the dependent variable, independent variable, and control variable are defined, including:

[0053] The land use efficiency of the urban industrial park is set as the dependent variable;

[0054] The result of whether the output value of the target city industrial park reaches the preset annual total output value is set as the independent variable. If the output value of the target city industrial park reaches the annual total output value in the current year, the independent variable is set to 1 in the current year and subsequent years; otherwise, it is 0.

[0055] The natural logarithm of GDP per capita, the ratio of financial institution deposits to local GDP, the ratio of financial institution loans to local GDP, the ratio of added value of secondary and tertiary industries to local GDP, the ratio of the number of foreign-invested enterprises to the number of industrial enterprises, and the natural logarithm of the number of urban industrial parks are set as the control variables.

[0056] Specifically, the expression for the staggered difference-in-differences evaluation model is: ;in, For year t, the land use efficiency of urban industrial parks in city i. The independent variable is... The control variable; For the intercept term; The effect coefficient of setting an annual total output value target for the industrial park; For the control variable coefficient vector; This is a time-fixed effect; For urban fixed effects; This is the random error term.

[0057] Furthermore, the staggered difference-in-differences evaluation model is subjected to benchmark comparison, robustness testing, and dual-effect analysis to obtain the impact assessment results, including:

[0058] With and without the control variables, the impact of reaching the preset annual total output value at the 5% significance level on the land use efficiency of urban industrial parks in the staggered difference-in-differences assessment model was evaluated to obtain benchmark comparison results.

[0059] The parallel trend test, difference method test, alternating difference method test, and dummy dependent variable test were used to evaluate the significance of the prior coefficients, and the robustness test results were obtained by performing parallel trend test, difference estimation, heterogeneity treatment effect test, and placebo test on the alternating difference evaluation model.

[0060] The effects of the set values ​​of the independent variables on the reduction of carbon sources and the enhancement of carbon sinks were evaluated using regression analysis with and without the control variables, respectively, to obtain the results of the dual-effect analysis.

[0061] The impact assessment results are obtained by integrating the benchmark comparison results, the robustness test results, and the dual-effects analysis results.

[0062] Specifically, in this embodiment, this example is used to combine Figure 2 (a) to Figure 2 (c) Provide a method for measuring the land use efficiency and assessing the impact of an industrial park in a certain city. This method includes:

[0063] 1) Data Preparation: In this embodiment, urban data of the urban industrial park, environmental data related to net-zero carbon calculation, historical land use efficiency data, and other driving factor data are collected. The collected data are then preprocessed to construct a basic assessment database. The preprocessing includes the following:

[0064] In this embodiment, the urban industrial park data includes data on a specific urban industrial park city. To avoid potential bias, cities that established urban industrial parks before 2009 were excluded. In this embodiment, the environmental data related to net zero carbon measurement includes carbon emission data, nighttime light data, and net primary productivity data. In this embodiment, the carbon emission data comes from an energy statistics yearbook (2007-2019), the nighttime light data (DMSP / OLS and NPP / VIIRS) is provided by the Earth Observatory Group, and the net primary productivity data comes from the MODIS net primary productivity product (MOD17A3). The above data are processed by data type unification and classification analysis.

[0065] In this embodiment, the land use efficiency data comes from a city's statistical yearbook (2007-2019), and the above data is processed by unifying the types and classifying the analysis.

[0066] In this embodiment, other driving factor data include data such as labor force, land, capital, and gross domestic product (GDP). Labor force data is represented by the number of employees in the primary, secondary, and tertiary industries; land data is represented by the built-up area; and capital data is represented by capital stock. The data comes from a city statistical yearbook (2007-2019), a land parcel-level database on a land market website, and a research data service platform, etc. The perpetual inventory method is used to calculate the capital stock. The above data is then processed by unifying and classifying the data types.

[0067] 2) Dynamic Measurement of Land Use Efficiency: In this embodiment, based on data from the basic assessment database, the net carbon emissions (NCE) of the urban industrial park are first quantitatively calculated. Then, based on the calculated net carbon emissions and other data, the land use efficiency (ULUE) of the urban industrial park is quantitatively calculated. The "firstly quantitatively calculating the net carbon emissions (NCE) of the urban industrial park" includes assessing carbon sources and sinks through carbon budgeting methods, specifically including the following steps:

[0068] S1: Calculating carbon sequestration using the net primary productivity approach: based on For every gram of dry biomass produced, 1.62 grams of carbon dioxide are sequestered. This biomass accounts for approximately 45% of total net primary productivity (NPP). Therefore, the urban carbon sink is calculated using the following formula:

[0069]

[0070] Here, CS stands for Carbon Sinks, and NPP stands for Total Net Primary Productivity.

[0071] S2: Urban carbon source prediction and calculation using Particle Swarm Optimization-Optimized Backpropagation Neural Network (PSO-BP): First, establish the correlation between carbon emissions and nighttime light data, represented by the sum of pixel brightness (DN) values ​​in remote sensing images. Then, use the Particle Swarm Optimization-Optimized Backpropagation Neural Network (PSO-BP) algorithm to calculate a weighted average of the remote sensing image pixel brightness, obtaining a weighted average brightness value to reduce local extrema. Specifically, sum the pixel brightness values ​​of each region... Its normalized brightness value is used as an input variable, and the corresponding administrative region's carbon emission statistics (calculated based on the energy balance sheet) are used as input variables. and Using light intensity as the output variable, the PSO-BP network is trained to establish a nonlinear mapping relationship between light intensity and carbon emissions. After training, the predicted light intensity weights for each park are calculated using this model. And further calculate the brightness-weighted average value;

[0072] Secondly, carbon emissions are predicted using the weighted average of DN values ​​through the PSO-BP algorithm, specifically:

[0073]

[0074] in, The sum of the luminance (DN) of all pixels within the study area. The number of pixels contained in the image;

[0075] Secondly, the pixel brightness of the remote sensing image is calculated using a particle swarm optimization-optimized backpropagation neural network (PSO-BP) algorithm to obtain a weighted average brightness value, thereby reducing local extrema. Specifically, the pixel brightness of each region is summed. Its normalized brightness value is used as an input variable, and the corresponding administrative region's carbon emission statistics (calculated based on the energy balance sheet) are used as input variables. and Using light intensity as the output variable, the PSO-BP network is trained to establish a nonlinear mapping relationship between light intensity and carbon emissions. After training, the predicted light intensity weights for each park are calculated using this model. And further calculate the brightness-weighted average, specifically:

[0076]

[0077] in, The weighted average of the DN values. The weight can be pixel area, population density, land use category weight, or other prior importance weight;

[0078] Finally, in accordance with the guidelines of the Intergovernmental Panel on Climate Change (IPCC), carbon emissions were calculated using energy consumption data from the energy balance sheet and compared with the results calculated by the aforementioned algorithm, using a brightness-weighted average corrected for the PSO-BP model. As a representative indicator of nighttime light intensity, and combined with energy statistics, a coupled estimation model of brightness-energy consumption-carbon emissions is established, specifically as follows:

[0079]

[0080] in, For urban carbon emissions. For each type of energy, The activity / energy consumption recorded in the energy balance sheet. For the corresponding emission factors, The total number of energy types to be considered; The brightness-carbon emission mapping function established based on the PSO-BP algorithm is used to correct the spatial distribution bias of statistical data.

[0081] S3: Calculate the net carbon emissions (NCE) of urban industrial parks, specifically:

[0082]

[0083] NCE stands for Net Carbon Emissions, CE stands for Carbon Emissions, and CS stands for Carbon Sinks.

[0084] In this embodiment, the assessment results of the average net carbon emission development trend of an industrial park in a certain city from 2006 to 2020 are obtained through analysis (e.g., Figure 3 As shown in the figure, the specific content of the result will be elaborated in the subsequent K1.

[0085] The second part, "based on the calculated net carbon emission data and other data of urban industrial parks, quantitatively calculates the land use efficiency (ULUE) of urban industrial parks," includes assessing land use efficiency through an ε-measurement model based on global superefficiency, specifically including the following steps:

[0086] T1: Set the input variables for the ε-measure model based on global superefficiency, including output variables and input variables, specifically:

[0087] Output variables include: expected output variable Gross Domestic Product (GDP) and unexpected output variable Net Carbon Emissions (NCE).

[0088] Input variables include: labor force (represented by the number of employees in the primary, secondary, and tertiary industries), land (represented by the area of ​​the built-up area), and capital (represented by the capital stock).

[0089] T2: Run the ε-metric model based on global superefficiency to calculate the superefficiency score K. Specifically, it includes the following steps:

[0090] T2.1: Calculating the Land Use Efficiency (ULUE) of Urban Industrial Parks. The city is considered a decision unit (DMU), and a global technology production possibility set is constructed. Specifically, for DMUi, there are m inputs, denoted as xi=(xi1,xi2,...xim), producing s expected outputs, denoted as yi=(yi1,yi2,...yis), and producing p undesired outputs, denoted as bi=(bi1,bi2,...bim).

[0091] T2.2: Based on the inputs and outputs of T2.1 above, define the production possibility set reflecting land use efficiency, specifically as follows:

[0092]

[0093] in, For the set of production possibilities; Let be the input amount of the j-th decision-making unit in period t; Let be the expected output of the j-th decision-making unit in period t; Let be the undesirable output of the j-th decision unit in period t; The weights are linear combination weights; This represents the input-output vector for the current target decision-making unit (urban industrial park). For the number of periods; The number of decision-making units; this set The construction of the model enables the connection between the input variables and the ε-metric model. That is, the input, output, and undesirable output data mentioned above are all used as input parameters for solving the ε-metric model, forming the constraint boundary for subsequent solutions.

[0094] T2.3: Based on step T2.2, let... The optimal solution of the model is represented by γ, where γ is a weight variable. Under the constraints of the production possibility set, the solution for the target decision unit (DMU) is calculated to obtain the superefficiency score; the expression of the ε metric model is:

[0095]

[0096]

[0097] in, The score for the super-efficiency; These are the weights of the input factor, the expected output factor, and the undesired output factor, respectively. For input-oriented radial contraction factor; Output-oriented radial expansion factor; To introduce slack variables; To produce slack variables; To avoid producing slack variables; These are the non-radial adjustment weights for controlling inputs, expected outputs, and undesired outputs, respectively; These represent the input, expected output, and unexpected output of the park to be evaluated.

[0098] The parameter ε determines the balance between radial and non-radial relaxation, and its range is between 0 and 1. When ε=0, the ε metric model based on global superefficiency is equivalent to the fixed-size reporting (CCR) model. When ε=1, the ε metric model based on global superefficiency is equivalent to the relaxation-based metric model.

[0099] T3: The superefficiency score K obtained based on step T2 above. (Dimensionless numerical value), representing the land use efficiency (ULUE) of urban industrial parks. it ),in:

[0100] K When the value is greater than 1, the land use efficiency of the city's industrial park is higher than the global frontier average, and the higher the value, the higher the efficiency.

[0101] K When =1, it is exactly at the global forefront;

[0102] K When the value is less than 1, the efficiency is lower than the global frontier, indicating room for improvement.

[0103] In this embodiment, the evaluation results of the development trend of land use efficiency in an industrial park in a certain city from 2006 to 2020 are analyzed (e.g., Figure 4 As shown in the figure), and the specific content of this result will be elaborated in subsequent K2.

[0104] 3) Impact Assessment: In this embodiment, based on the data in the basic assessment database and the calculation results of urban net carbon emissions and urban industrial park land use efficiency, the dependent variable (explained variable), independent variable (core explanatory variable), and control variables for model analysis are set. An alternating difference-in-differences (DID) assessment model is constructed, and benchmark comparison, robustness testing, and dual-effect analysis of the model are completed. The model's explanatory results are output, completing the dynamic assessment of urban industrial park land use efficiency. Specifically, "setting the dependent variable (explained variable), independent variable (core explanatory variable), and control variables for model analysis" includes the following:

[0105] In this embodiment, the dependent variable (explained variable) of the model analysis is set as "Urban Industrial Park Land Use Efficiency (ULUE)". it "), with dimensionless numerical super-efficiency score K The specific calculation method is shown in steps T1 to T3 above. The input dependent variables include labor, land, and capital indicators, which are represented by the number of employees in the primary, secondary, and tertiary industries, the built-up area, and the capital stock, respectively. The output dependent variables include expected output and unexpected output, which are represented by GDP and net carbon emissions (NCE) of urban industrial parks, respectively.

[0106] In this embodiment, the independent variable (core explanatory variable) of the model analysis is set as "whether the output value of the urban industrial park reaches the preset annual total output value (UPDZ)". it If the city's industrial park reaches the preset annual total output value in the current year, then the value of this variable is set to 1 in the current year and subsequent years; otherwise, it is set to 0.

[0107] In this embodiment, the control variables for model analysis are set as economic development (Pgdp), financial development (Fin), industrial structure (Indus), foreign direct investment (Fdi), and the number of urban industrial parks (Num), which are respectively represented by the natural logarithm of GDP per capita, the ratio of deposits and loans of financial institutions to local GDP, the ratio of added value of secondary and tertiary industries to local GDP, the ratio of the number of foreign-invested enterprises to the number of industrial enterprises, and the natural logarithm of the number of urban industrial parks.

[0108] Specifically, "constructing an alternating difference-in-differences (DID) evaluation model, completing benchmark comparisons, robustness tests, and dual-effect analysis of the model, outputting the model's interpretation results, and completing the dynamic evaluation of land use efficiency in urban industrial parks" includes the following steps:

[0109] U1: Construction of the staggered DID model: Based on the dependent variable "Urban Industrial Park Land Use Efficiency (ULUE)" it The independent variable is "whether the output value of the urban industrial park has reached the preset annual total output value (UPDZ)". it Based on the five types of control variables affecting ULUE, as well as city fixed effects and time fixed effects, an alternating DID assessment model is constructed, specifically as follows:

[0110]

[0111] in, For year t, the land use efficiency of urban industrial parks in city i. The independent variable is... The control variable; For the intercept term; The effect coefficient of setting an annual total output value target for the industrial park; For the control variable coefficient vector; This is a time-fixed effect; For urban fixed effects; This is the random error term.

[0112] U2: Evaluation of the staggered DID model: The staggered DID model constructed in step U1 above is subjected to benchmark model comparison analysis, robustness testing, and dual-effect analysis. The model interpretation results are then output, specifically including the following steps:

[0113] U2.1: Benchmark Model Comparative Analysis: With and without control variables included, assess the impact of achieving the preset annual total output value on the land use efficiency of urban industrial parks at the 5% significance level, and demonstrate the rationality of the variable settings in the evaluation model. Specifically:

[0114]

[0115] in, The dependent variable is the land use efficiency index of urban industrial parks. (Including control variables) and (Excluding control variables) is an estimate of the average treatment effect; This is a dummy variable, representing whether the output value of the urban industrial park in year t reaches the preset annual total output value. As a control variable affecting the land use efficiency of urban industrial parks It is the urban fixed effect. For time-fixed effects, It is a random error term;

[0116] U2.2: Robustness testing: The robustness of the model is comprehensively tested by: conducting parallel trend tests through prior coefficient significance assessment; testing the comprehensive difference estimates of the model through the comprehensive difference method; testing the heterogeneity treatment effect of the model through the alternating difference method; and completing the placebo trial of the model through dummy dependent variable testing.

[0117] In this embodiment, the robustness test specifically includes:

[0118] The "parallel trend test" specifically assesses whether the land use efficiency of urban industrial parks follows the assumption of the same trend before reaching the preset annual total output value by testing the significance of the prior coefficients. If the prior coefficients are not significant, the parallel trend test is passed.

[0119]

[0120] in, as a unit The first processing time, indicating the function During the relative event period Take 1 if the event is not statistically significant (usually at the 5% level), otherwise take 0. The assumption of parallel trend is acceptable when all the event-preceding coefficients are not statistically significant (usually at the 5% level).

[0121] In this embodiment, the specific results of the parallel trend test and dynamic effect analysis are obtained, such as... Figure 5 As shown.

[0122] "Comprehensive difference estimation" specifically involves using individual and time weights to match the trends of urban industrial parks before reaching the preset annual total output value between the experimental and control urban groups, balancing the periods before and after to improve comparability. Specifically, it uses a weighted least squares solution in the form of WLS to balance the trends of urban industrial parks before reaching the preset annual total output value.

[0123]

[0124] in, Design the matrix (including constant terms, individual / time virtual terms, treatment terms, and control variables). The vector of the explained variables (corresponding to the panel) ), It is a diagonal weight matrix (the weights are generated based on individual and time weights);

[0125]

[0126] in, For the observations of the processing unit (or merged processing unit) at time t, Candidate control unit In time The observed values, The control unit weights (composite control weights) are solved by minimizing the problem described above.

[0127] The "Heterogeneous Treatment Effect Test" specifically involves dividing a sample of urban industrial parks into subgroups, estimating their effect on achieving a preset annual total output value, and then summing these effects using a specific strategy to calculate the average treatment effect (ATT) over the period.

[0128]

[0129]

[0130] in, The definition of group ATT (group g), The total ATT is a weighted composite.

[0131] Specific weighting strategies can effectively reduce potential biases, namely:

[0132] (I) Simple weighted ATT, using equal weights, specifically:

[0133]

[0134] in, The number of groups with equal weights.

[0135] (II) Dynamic ATT, which weights each group based on multiple processing at different time points, specifically:

[0136]

[0137] in, Let g be the number of observable post-processing periods in the g-th group, and normalize the sum to 1.

[0138] (III) Year ATT, weighted according to year, specifically:

[0139]

[0140] in, The year is used as the weight, and the group effect of different years is weighted accordingly (for example, emphasizing recent years or key years that reach the preset value of annual total output). Let g be the number of observations in year y.

[0141] (IV) Group ATT, weighted according to the time after the first treatment, specifically:

[0142]

[0143] in, For group The length of the observation period after processing.

[0144] A "placebo trial" specifically involves generating spurious treatment variables, including incorporating the actual first-treatment time. A "time placebo" is constructed by moving forward or backward several years; a group of urban industrial parks, unaffected by actual changes, are randomly selected spatially as a "spatial placebo," and tested in both unrestricted and restricted mixing placebo tests to ensure that changes in land use efficiency in urban industrial parks are due to achieving a predetermined annual total output value rather than other factors. Specifically:

[0145]

[0146] In this embodiment, the specific analytical results of the placebo trial are obtained, such as... Figure 6 As shown;

[0147] Based on the above robustness test details, this embodiment obtains the robustness test evaluation results, and the specific details of these results will be elaborated in subsequent section K3.

[0148] U2.3: Dual-Effect Analysis: Regression analysis is used to assess the impact of urban industrial parks reaching their pre-set annual gross output value on carbon source reduction and carbon sink enhancement, both with and without control variables. This completes the dual-effect analysis of the model's net carbon emission target, further explaining the potential of urban industrial parks reaching their pre-set annual gross output value to promote carbon sink enhancement while reducing emissions. Specifically, the case with control variables is as follows:

[0149]

[0150] in, This is a dummy variable, representing whether the output value of the urban industrial park in year t reaches the preset annual total output value. As a control variable affecting the land use efficiency of urban industrial parks It is the urban fixed effect. For time-fixed effects, It is a random error term.

[0151] The case without control variables is as follows:

[0152]

[0153] The meaning of the variables is the same as in the above-mentioned "cases containing control variables".

[0154] By obtaining and The estimated value is then used to calculate the net effect of carbon source attenuation and carbon sink enhancement. The variance and standard error of the net effect are then calculated using the Delta rule. Specifically:

[0155]

[0156] In this embodiment, the evaluation results of the dual-effects analysis are obtained, and the specific content of the results will be elaborated in subsequent K4.

[0157] In summary, this embodiment yields the following conclusions:

[0158] K1: Since 2011, the average net carbon emissions growth of urban industrial parks that have reached the preset industrial added value has slowed significantly compared to those that have not.

[0159] K2: Since 2013, the average land use efficiency of urban industrial parks that have achieved the preset industrial added value has significantly improved compared with those that have not.

[0160] K3: Parallel trend test indicates that the improvement in land use efficiency resulting from the achievement of the preset industrial added value in urban industrial parks will continue and intensify over time, representing a lasting economic and environmental benefit, satisfying the parallel trend hypothesis; comprehensive difference estimation shows that the regression coefficients of the independent variables remain positive and significant regardless of whether control variables are included; heterogeneous treatment effect test shows that only 7.3% of the subgroups have potential estimation bias, and the estimation results are largely unaffected by bias, while all four types of ATT indicate that the land use efficiency of urban industrial parks that achieve the preset industrial added value is significantly improved, consistent with the baseline results; placebo test shows that the change in land use efficiency of urban industrial parks is due to their achievement of the preset annual total output value rather than other factors.

[0161] K4: With and without control variables, urban industrial parks that achieve the preset annual total output value significantly reduce urban carbon emissions and significantly enhance carbon sinks, demonstrating a significant dual effect in reducing carbon emissions and enhancing carbon sinks.

[0162] The beneficial effects of this invention are as follows:

[0163] This invention integrates multiple modeling techniques to quantitatively calculate the current status of net carbon emissions in urban industrial parks. Based on this result, it further quantitatively calculates the current status of land use efficiency in urban industrial parks and extracts the influencing mechanisms of land use efficiency, innovatively developing a relatively scientific method for assessing urban industrial park land use efficiency. This method combines the net primary productivity method with the PSO-BP algorithm to comprehensively quantify the carbon emission status of urban industrial parks from both carbon source and carbon sink dimensions. It utilizes an ε-metric model based on global superefficiency to handle complex multivariate input-output systems, scientifically measuring the input-output efficiency of urban industrial park land use. It employs the staggered difference method to control the influence of difficult-to-observe sample heterogeneity and common trend factors, thereby more accurately estimating the actual impact of a preset annual total output value on urban industrial park land use efficiency. The key point of this invention is the innovative method for measuring urban net carbon emissions and land use efficiency in urban industrial parks. Based on existing assessments, it introduces factors such as the background of net-zero carbon construction and dynamic land use exit mechanisms, which is more conducive to accurately identifying potential inefficiencies in urban industrial parks from the perspective of land use efficiency. This helps to achieve a positive interaction between economic development and environmental protection, and provides scientific guidance for promoting the development of urban industrial parks towards green, low-carbon, and efficient directions and the goal of carbon neutrality.

[0164] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0165] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for measuring and evaluating the land use efficiency of an urban industrial park, characterized in that, The method comprises the following steps: Collecting city industrial park data, environmental data, historical land use efficiency data, and driving factor data to obtain an evaluation basic database; According to the evaluation basic database, the net carbon emissions of the target park are calculated, and the land use efficiency of the city industrial park is calculated based on the global super-efficiency ε measurement model to obtain the current park data; Setting dependent variables, independent variables, and control variables; According to the evaluation basic database and the current park data, the interlaced double difference evaluation model is constructed using the dependent variables, independent variables, and control variables; Using the interlaced double difference evaluation model to evaluate the land use efficiency of the to-be-evaluated park to obtain the city industrial park land use efficiency measurement result; Benchmark comparison, robustness test, and double effect analysis are performed on the interlaced double difference evaluation model to obtain the impact evaluation result. 2.The method of claim 1, wherein, Collecting city industrial park data, environmental data, historical land use efficiency data, and driving factor data to obtain an evaluation basic database, comprising: Collecting the city industrial park data; the city industrial park data includes: science and technology park data, industrial park data; Collecting carbon emission data, night light data, and net primary productivity data to obtain the environmental data; Collecting the historical land use efficiency data; Collecting labor data, land data, capital data, and gross domestic product to obtain the driving factor data; the labor data is the number of employees in the first to third industries; the land data is the built-up area; Integrating the city industrial park data, the environmental data, the historical land use efficiency data, and the driving factor data to obtain the evaluation basic database. 3.The method of claim 1, wherein, According to the evaluation basic database, the net carbon emissions of the target park are calculated, and the land use efficiency of the city industrial park is calculated based on the global super-efficiency ε measurement model to obtain the current park data, comprising: According to the evaluation basic database, a city carbon sink formula is calculated to obtain the carbon sink; the city carbon sink formula is: ; wherein, is the carbon sink; is the total net primary productivity; Summation calculation is performed on the brightness of the remote sensing image pixels in the evaluation basic database to obtain the total brightness of the pixels; According to the pixel brightness sum, the PSO-BP algorithm is used for weighted average calculation of the remote sensing image pixel brightness, so as to obtain a brightness weighted average value; an expression of the brightness weighted average value is: ; wherein, is the brightness weighted average value; is the i-th weighted weight; is the i-th remote sensing image pixel brightness; is the number of pixels contained in the image. According to the luminance weighted average value, a carbon emission formula is calculated to obtain the urban carbon emission; the carbon emission formula is: ; wherein, is the urban carbon emission; is the activity or energy consumption recorded in the energy balance table; is the kth emission factor; is the total number of energy types; is a mapping function of luminance and carbon emission established according to the PSO-BP algorithm. The urban carbon emission and the carbon sink are subtracted to obtain the net carbon emission of the urban industrial park; and an expression of the net carbon emission of the urban industrial park is: ; wherein, is the net carbon emission of the urban industrial park.

4. The method according to claim 1, wherein, According to the evaluation basic database, the net carbon emissions of the target park are calculated, and the land use efficiency of the city industrial park is calculated based on the global super-efficiency ε measurement model to obtain the current park data, further comprising: Setting input variables of the ε measurement model; the input variables include output-type variables and input-type variables; Building a production possibility set according to the input variables; the expression of the production possibility set is: ; wherein, is the production possibility set; is the input quantity of the jth decision unit in the tth period; is the expected output quantity of the jth decision unit in the tth period; is the unexpected output quantity of the jth decision unit in the tth period; is the linear combination weight; is the input-output vector of the current target decision unit; is the time period number; is the decision unit number; Solving the target decision unit based on the global super-efficiency ε measurement model under the constraint of the production possibility set to obtain a super-efficiency score; the expression of the ε measurement model is: ; wherein, ; is the super-efficiency score; are the weights of the input factors, the desirable output factors and the undesirable output factors, respectively; is the input-oriented radial contraction factor; is the output-oriented radial expansion factor; is the input slack variable; is the desirable output slack variable; is the undesirable output slack variable; are the non-radial adjustment weights of the control inputs, the desirable outputs and the undesirable outputs, respectively; are the input, desirable output and undesirable output values of the park to be evaluated, respectively; According to the preset efficiency interval distribution, the super-efficiency score is matched in an interval to obtain the land use efficiency of the city industrial park.

5. The urban industrial park land use efficiency measurement and influence evaluation method according to claim 1, characterized in that, Setting dependent variables, independent variables, and control variables, comprising: Setting the land use efficiency of the city industrial park as the dependent variable; Setting the result of whether the target city industrial park output value reaches the preset annual total output value as the independent variable, if the target city industrial park output value reaches the annual total output value in the current year, setting the independent variable in the current year and the subsequent years as 1, otherwise 0; Setting the natural logarithm of per capita GDP, the ratio of financial institution deposits to local GDP, the ratio of financial institution loans to local GDP, the ratio of secondary and tertiary industry added value to local GDP, the ratio of foreign-funded enterprise number to industrial enterprise number, and the natural logarithm of city industrial park number as the control variables.

6. The urban industrial park land use efficiency measurement and influence evaluation method according to claim 1, characterized in that, The expression of the staggered double difference evaluation model is: ; wherein, is the land use efficiency of the urban industry park of city i in t years; is the independent variable; is the control variable; is the intercept term; is the effect coefficient of the preset annual total output value target of the industry park; is the control variable coefficient vector; is the time fixed effect; is the city fixed effect; is the random error term.

7. The urban industrial park land use efficiency measurement and influence evaluation method according to claim 1, characterized in that, Performing benchmark comparison, robustness test and double effect analysis on the staggered double difference evaluation model to obtain the influence evaluation results, including: Respectively under the conditions of including and not including the control variables, evaluating the influence level of reaching the preset annual total output value at the 5% significant level on the land use efficiency of the city industrial park of the staggered double difference evaluation model to obtain the benchmark comparison results; Using pre-coefficient significance evaluation, difference method test, staggered difference method test, and virtual dependent variable test to perform parallel trend test, difference estimation, heterogeneity processing effect test, and placebo test on the staggered double difference evaluation model to obtain the robustness test results; Respectively under the conditions of including and not including the control variables, using regression analysis to evaluate the influence of the set value of the independent variable on carbon source weakening and carbon sink enhancement to obtain the double effect analysis results; Integrating the benchmark comparison results, the robustness test results, and the double effect analysis results to obtain the influence evaluation results.

Images