Urban heat-energy-carbon correlation analysis method and device based on intermediary effect and threshold effect and medium
By constructing mediation effect and threshold effect models, the impact of energy consumption on carbon emissions was quantified, and the problem of lack of mediation effect and nonlinear analysis in existing research was solved, and a comprehensive understanding of the thermal-energy-carbon system was achieved, providing a scientific basis for urban planning and climate governance.
Patent Information
- Application Number
- CN202510443094.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-10
- Publication Date
- 2025-07-18
AI Technical Summary
Existing research lacks the intermediary effect analysis on the indirect impact of the surface thermal environment on carbon emissions through the energy consumption chain, ignores the nonlinear threshold effect, and the temporal and spatial scale is limited to a single city or a specific land use type, and fails to fully understand the complex interactive relationship of the thermal-energy-carbon system.
Build a urban thermal-energy-carbon correlation analysis method based on mediation effect and threshold effect. Through panel benchmark regression model, mediation effect model and threshold effect model, the impact of energy consumption on carbon emissions is quantified, nonlinear characteristics in different economic development stages and policy environments are identified, and large-scale and long-term analysis is carried out.
The system reveals the mediating role of the surface thermal environment on carbon emissions, identifies the nonlinear threshold effect of the thermal-energy-carbon system, provides accurate carbon reduction policies and urban planning basis, and improves the scientificity and applicability of the research.
Smart Images
Figure CN120339023A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the cross - technical field of urban science, environmental science, energy science and climate science. More specifically, it relates to a method, device and medium for analyzing the urban heat - energy - carbon association based on mediation effect and threshold effect. Background Art
[0002] Under the background of global climate change, climate anomalies characterized by frequent high temperatures and extreme weather are intensifying the potential threats to the social - economic system and ecological - environmental security. This not only challenges the traditional industrial structure and energy consumption pattern, but also puts forward new requirements for regional sustainable development. Existing studies have emphasized that in addition to traditional fossil - fuel combustion, the high - temperature effect caused by changes in surface environmental characteristics will have a profound impact on regional energy demand, consumption structure and the resulting carbon emissions, thus constituting a new complex feedback mechanism under climate change. In this process, the relationship between the surface thermal environment and carbon emissions has received increasing attention. As a typical manifestation, the Urban Heat Island (UHI) has become a concentrated reflection of the surface warming phenomenon caused by rapid urbanization and land - use change. Oke (1982) was the first to reveal the cause of UHI from the perspective of urban radiation balance. Subsequently, a large number of scholars further showed through empirical research that the conversion of land - use types (such as vegetation reduction, increased building density, and intensified road - surface hardening) and changes in surface albedo not only disrupt the natural energy balance, but also exacerbate the evolution of surface heat retention and radiation characteristics, leading to local climate warming. At the same time, UHI and related thermal - environment anomalies will significantly increase the cooling load of buildings and public facilities during high - temperature summers, thus increasing energy consumption such as electricity and gas, and having an amplifying effect on the carbon - emission level at the urban and even regional scales. Li's empirical research based on major cities in the Yangtze River Delta of China shows that when the environmental temperature is higher than 25 degrees Celsius, for every 1°C increase in air temperature, the electricity demand of Shanghai households increases by 14.5%, and an air temperature of 32 degrees Celsius will lead to a 174% increase in daily electricity consumption. Santamouris (2015) further pointed out that UHI not only affects carbon emissions by increasing cooling demand and changing the urban wind field, but also indirectly shapes the regional carbon - emission pattern by enhancing the enrichment and transmission efficiency of pollutants in the urban space.
[0003] In recent years, the analysis has gradually expanded from the perspective of a single city or specific surface environment to the comprehensive analysis of the thermal - energy - carbon system at a cross - regional scale. For example, Santamouris (2015) pointed out that the UHI not only directly increases the cooling demand, but also indirectly shapes the regional carbon emission pattern by affecting the urban wind field and pollutant enrichment effect. At the same time, due to the popularity of remote sensing, geographic information system (GIS) and high - resolution statistical data, researchers can break through the limitations of traditional static analysis and examine the complex interaction mechanism of the thermal - energy - carbon system from the perspective of long - term dynamic evolution. Some studies use multi - city panel data and econometric models to quantitatively analyze the relationship between UHI and energy - related carbon emissions, and find that it is affected by factors such as economic scale, industrial structure, and energy structure. However, existing research mainly focuses on the direct impact of thermal environment changes on carbon emissions, rarely discusses how the thermal environment has a mediating effect on carbon emissions through the energy consumption chain, and also lacks the analysis of threshold effects under different economic development levels or industrial structures. In addition, most studies focus on a single city or specific land use type, while ignoring the impact of the interaction between different underlying surfaces at a larger regional scale on the thermal - energy - carbon dynamic process.
[0004] Existing research on the relationship between urban thermal environment and carbon emissions has the following characteristics and limitations:
[0005] First, it focuses on direct impacts and lacks mediating effect analysis: Although existing research has shown that high - temperature weather will increase energy consumption and lead to an increase in carbon emissions, there is a lack of in - depth discussion on how the surface thermal environment indirectly affects carbon emissions through the energy consumption chain.
[0006] Second, it lacks non - linear characteristics and threshold effect analysis: Existing research mainly uses methods such as linear regression to analyze the impact of UHI on carbon emissions, while ignoring the possible threshold effects under different economic development levels, industrial structure adjustments, and energy supply conditions. For example, due to differences in policy intensity, energy structure, and urban functions in different regions, the thermal - energy - carbon correlation relationship may show non - linear changes, and research in this area is still relatively limited.
[0007] Third, there are limitations in spatio - temporal scales: Existing research mostly focuses on a single city or specific land use type, without fully considering the impact of cross - administrative units, urban - rural transition zones, and regional differences on the thermal - energy - carbon system, which limits the understanding of the overall systematic and long - term relationship. Summary of the Invention
[0008] To solve the above - mentioned technical problems, the present application provides a method, device, and medium for analyzing the urban thermal - energy - carbon correlation based on mediating effect and threshold effect, to systematically quantify how the surface thermal environment affects carbon emissions through energy consumption and reveal the threshold characteristics under different economic development stages and policy environments. The purpose of the present application is:
[0009] ①Construct the urban heat - energy - carbon correlation analysis framework: Integrate remote sensing data, socioeconomic statistical data, and energy consumption data to establish a systematic analysis framework covering the surface urban heat island intensity, energy consumption level, and carbon emissions.
[0010] ②Analyze the mediating role of the surface thermal environment: Through the mediating effect model, quantitatively examine how the urban thermal environment indirectly affects carbon emissions through the energy consumption chain, making up for the deficiency of existing research that only focuses on direct effects.
[0011] ③Identify the threshold characteristics of the heat - energy - carbon system: Use the threshold regression model to explore the non - linear threshold effect of the thermal environment on carbon emissions under different economic development levels, energy structures, and industrial layouts, and identify key nodes and mechanisms.
[0012] ④Expand the spatio - temporal scale analysis: Based on long - time - series data, comprehensively analyze the heat - energy - carbon system covering multiple cities to reveal long - term evolution trends and regional heterogeneity.
[0013] First, this application provides a method for analyzing the urban heat - energy - carbon correlation based on the mediating effect and threshold effect. The method includes:
[0014] Obtain data; wherein, the data includes the explained variable, explanatory variables, and control variables. The explained variable includes carbon emission data, the explanatory variable includes the area of the heat island area, and the control variable is an influencing factor related to the scale of carbon emissions.
[0015] Construct a panel benchmark regression model based on the obtained data to characterize the impact of the surface thermal environment on carbon emissions.
[0016] Construct a mediating effect model based on the panel benchmark regression model to test the mechanism of the surface thermal environment on carbon emissions.
[0017] Construct a threshold effect model based on the panel benchmark regression model to test whether the impact of the surface thermal environment on carbon emissions is non - linear.
[0018] Further, obtain the area of the heat island area through the following method:
[0019] Obtain surface temperature data;
[0020] After projecting and transforming the format of the surface temperature data, perform annual synthesis based on the maximum composite method.
[0021] Normalize the surface temperature data after annual synthesis, unify the temperature range to 0 - 1, and on the basis of the normalization process, take the area with a temperature range of 0 - 0.1 as the extremely strong cold island area, divide it into 10 levels at intervals of 0.1, and take the interval of 0.6 - 1 as the area of the heat island area.
[0022] Furthermore, the influencing factors related to the carbon emission scale include socioeconomic factors and urban landscape pattern factors; among them, the socioeconomic factors include the proportion of fiscal expenditure, the level of financial development, and the level of population urbanization. The level of financial development is measured by the ratio of the balance of deposits and loans of financial institutions at the end of the year to the gross regional product, and the level of population urbanization is measured by the ratio of the permanent population of towns to the total permanent population. The urban landscape pattern factors include the contagion index, patch density, and patch density.
[0023] Furthermore, the panel benchmark regression model is expressed as:
[0024] lny it = α0 + α1×lnx it + α2×Control it + u it + ε it (1)
[0025] In the formula: i represents the region, and t represents the time; lny it represents the carbon emissions of region i in period t; lnx it represents the surface thermal environment of region i in period t; Control it represents the control variable; u it represents the regional fixed effect; ε it is the random disturbance term.
[0026] Furthermore, the mediating effect model is expressed as:
[0027] medi it = β0 + β1×lnx it + β2×Control it + u it + ε it (2)
[0028] lny it = γ0 + γ1×lnx it + γ2×lnmedi it + γ3×Control it + u it + ε it (3)
[0030] In the formula: medi it represents the mediating variable, that is, the energy consumption; β0 and γ0 represent the constant terms; β1 represents the influence coefficient of the independent variable lnx it on the mediating variable medi it ; β2 represents the control variable Controlit The influence coefficient of the mediating variable medi it ; γ1 represents the influence coefficient of the independent variable lnx it on the dependent variable lny it ; γ2 represents the influence coefficient of the mediating variable medi it on the dependent variable lny it ; γ3 represents the influence coefficient of the control variable Control it on the dependent variable lny it .
[0031] Further, the threshold effect model is expressed as:
[0032] lny it = θ0 + θ1×lnx it (gdp ≤ q1) + θ2×lnx it (q1 < gdp < q2) + … + θ m ×
[0033] lnx it (q m-1 < gdp < q m ) + θ m+1 ×lnx it (gdp > q m ) + δ×Control it + u it + ε it (4)
[0035] In the formula: θ0 represents the constant term; θ1, θ2, …, θ m , θ m+1 represent the coefficients of the core explanatory variables at different threshold levels; q1, q2, …, q m represent the corresponding threshold values; gdp is the threshold variable; δ represents the influence coefficient of the control variable Control it on the dependent variable lny it .
[0036] Further, after constructing the threshold effect model, the method includes: performing a model robustness check to ensure the robustness of the threshold effect analysis results.
[0037] Further, the ways of performing the model robustness check include:
[0038] Replacing the core explanatory variables and using different surface thermal environment measurement indicators to replace the original indicators for regression analysis; the different surface thermal environment measurement indicators include the mean surface temperature and / or night light brightness
[0039] Perform two-sided winsorization on the sample data to reduce the impact of extreme values on the regression results;
[0040] Adjust the sample space and perform repeated regressions using different regional samples respectively.
[0041] In a second aspect, the present application provides an urban heat-energy-carbon association analysis device based on mediation effect and threshold effect. The device includes:
[0042] A data acquisition unit configured to acquire data; wherein the data includes an explained variable, an explanatory variable, and a control variable. The explained variable includes carbon emission data, the explanatory variable includes the area of the heat island area, and the control variable is an influencing factor related to the scale of carbon emissions;
[0043] A benchmark regression model construction unit configured to construct a panel benchmark regression model based on the acquired data to characterize the impact of the surface thermal environment on carbon emissions;
[0044] A mediation effect model construction unit configured to construct a mediation effect model based on the panel benchmark regression model to test the action mechanism of the surface thermal environment on carbon emissions;
[0045] A threshold effect model construction unit configured to construct a threshold effect model based on the panel benchmark regression model to test whether the impact of the surface thermal environment on carbon emissions is non-linear.
[0046] In a third aspect, the present application provides a readable storage medium storing one or more programs, which can be executed by one or more processors to implement the method described above.
[0047] The present application has at least the following beneficial effects:
[0048] 1) The present application introduces mediation effect analysis: Aiming at the limitation that existing research only focuses on the direct impact of the thermal environment on carbon emissions, a mediation effect model is used to quantify the conduction path of energy consumption in the heat-carbon relationship and reveal the regulatory role of the urban energy system under the background of climate change.
[0049] 2) The present application introduces threshold effect analysis: Based on the threshold regression method, explore the non-linear impact of different economic development stages, energy structure optimization, etc. on the heat-energy-carbon system relationship, and fill the gap in the existing research's insufficient attention to threshold characteristics.
[0050] 3) The present application conducts a comprehensive analysis on a large scale and long time series: Break through the limitations of single-city or short-term research, take the urban agglomerations in the Yangtze River Economic Belt as the research area, and combine long time series data of more than 20 years to reveal the dynamic evolution characteristics of the heat-energy-carbon system during the process of rapid urbanization and industrial transformation.
[0051] 4) The present application constructs an urban heat - energy - carbon index system by integrating multi - source data: integrating remote sensing data (land surface temperature, vegetation cover), urban statistical data (energy consumption, industrial structure), and carbon emission measurement indicators to construct a more complete analysis framework for the urban heat - energy - carbon system, improving the scientificity and applicability of the research.
[0052] Therefore, the technical solution disclosed in the present application helps to more comprehensively understand the complex interaction relationship between the urban heat environment and carbon emissions, provides a scientific basis for formulating accurate carbon reduction policies and optimizing urban spatial planning, and also provides a reference for global climate governance. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 shows the flow of a method for analyzing the urban heat - energy - carbon association based on mediating effect and threshold effect according to an embodiment of the present application Figure 1 ;
[0054] Figure 2 shows the flow of a method for analyzing the urban heat - energy - carbon association based on mediating effect and threshold effect according to an embodiment of the present application Figure 2 ;
[0055] Figure 3 shows the structural diagram of a device for analyzing the urban heat - energy - carbon association based on mediating effect and threshold effect according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0056] To enable those skilled in the art to better understand the technical solution of the present application, the present application will be described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings and specific examples, but shall not be construed as a limitation to the present application. For the various steps described herein, if there is no necessity for a sequential relationship between them, the order in which they are described as examples herein shall not be regarded as a limitation. Those skilled in the art should know that they can be adjusted in order as long as the logic between them is not destroyed and the entire process cannot be realized.
[0057] An embodiment of the present application provides a method for analyzing the urban heat - energy - carbon association based on mediating effect and threshold effect. As Figure 1 shown, the method for analyzing the urban heat - energy - carbon association based on mediating effect and threshold effect includes the following steps S10 to S40.
[0058] S10: Obtain data; wherein, the data includes an explained variable, an explanatory variable, and a control variable. The explained variable includes carbon emission data, the explanatory variable includes the area of the heat island area, and the control variable is an influencing factor related to the scale of carbon emissions.
[0059] In an exemplary embodiment, the annual carbon emission data of each city in the Yangtze River Economic Belt from 2000 to 2022 is used as the explained variable. The data is sourced from the global high-resolution carbon emission dataset released by ODIAC (Open-source Data Inventory for Anthropogenic CO). The ODIAC data is a 1-kilometer spatial resolution land grid, and the emission estimation is carried out by integrating power plant profile information (including emission intensity and geographical location) and satellite-observed nightlight data (Oda et al., 2018). This dataset has been widely recognized by the international research community for its reliability, diverse sources, and strict verification process, and has strong comparability. In terms of data processing, we first performed annual synthesis and spatial cropping operations on the original data, and through outlier removal and data smoothing, ensured the consistency and accuracy of the time-series data. Finally, the long-time series carbon emission data of each city in the Yangtze River Economic Belt was extracted and generated.
[0060] The core explanatory variable is the annual area of the heat island from 2000 to 2022. The land surface temperature data uses the MOD11A2 data provided by NASA from 2000 to 2022. After projection and format conversion, the annual synthesis is carried out based on the maximum composite method MVC. At the same time, in order to eliminate the problem that it is difficult to directly compare the land surface temperatures of different time phases, the annual synthesized land surface temperature data is normalized, and the temperature range is unified to between 0 and 1. Based on the normalized data, with 0 - 0.1 as the extremely strong cold island area, it is divided into 10 levels at intervals of 0.1, and the interval of 0.6 - 1 is taken as the heat island area.
[0061] To further avoid other spurious regression problems such as variable omission and endogeneity, social and economic factors and urban landscape pattern factors closely related to the carbon emission scale are used as control variables for investigation. Social and economic factors mainly include 3 categories: 1) The proportion of fiscal expenditure: measured by the ratio of local general budgetary expenditure to regional GDP; 2) The level of financial development: measured by the ratio of the balance of deposits and loans of financial institutions at the end of the year to regional GDP; 3) The level of population urbanization: measured by the ratio of the permanent urban population to the total permanent population. Urban landscape pattern factors are all calculated in fragstats based on the CLCD dataset from 2000 to 2022, and the indicators cover the contagion index CONTAG, patch density PD, and patch fractal dimension PAFRAC.
[0062] S20: Based on the obtained data, construct a panel benchmark regression model to characterize the impact of the land surface thermal environment on carbon emissions.
[0063] In an exemplary embodiment, to explore the impact of the surface thermal environment on carbon emissions, the following panel benchmark regression model is constructed:
[0064] lny it =ɑ0 + ɑ1×lnx it + ɑ2×Control it + u it + ε it (1)
[0065] Where: i represents the region, and t represents the time; lny it represents the carbon emissions in the i region at time t; lnx it represents the surface thermal environment in the i region at time t. To eliminate the influence of heteroscedasticity, in this embodiment, logarithms are taken for y and x; Control it represents a series of control variables, including the degree of government intervention, the degree of financial development, the level of urbanization, the sprawl index, the patch density, and the perimeter-area fractal dimension; u it represents the regional fixed effect; ε it is the random disturbance term.
[0066] S30: Based on the panel benchmark regression model, a mediation effect model is constructed to test the mechanism of the impact of the surface thermal environment on carbon emissions.
[0067] In an exemplary embodiment, to test the mechanism of the impact of the surface thermal environment on carbon emissions, the stepwise regression method is used for verification. Therefore, based on equation (1), a mediation effect model is constructed:
[0068] medi it =β0 + β1×lnx it + β2×Control it + u it + ε it (2)
[0069] lny it =γ0 + γ1×lnx it + γ2×lnmedi it + γ3×Control it + u it + ε it (3)
[0071] Where: medi it represents the mediating variable, that is, the energy consumption; β0 and γ0 represent the constant terms; β1 represents the impact coefficient of the independent variable lnx it on the mediating variable medi it ; β2 represents the control variable Control itThe influence coefficient of the mediating variable medi it ; γ1 represents the influence coefficient of the independent variable lnx it on the dependent variable lny it ; γ2 represents the influence coefficient of the mediating variable medi it on the dependent variable lny it ; γ3 represents the influence coefficient of the control variable Control it on the dependent variable lny it . The remaining variables are the same as those in Equation (1). In this embodiment, the mechanism of action of energy consumption will be tested using Formulas (2) and (3).
[0072] S40: Construct a threshold effect model based on the panel benchmark regression model to test whether the influence of the surface thermal environment on carbon emissions is non-linear.
[0073] In an exemplary embodiment, to test whether the influence of the surface thermal environment on carbon emissions is non-linear, a threshold effect model is constructed based on Equation (1):
[0074] lny it = θ0 + θ1 × lnx it (gdp ≤ q1) + θ2 × lnx it (q1 < gdp < q2) + … + θ m ×
[0075] lnx it (q m-1 < gdp < q m ) + θ m+1 × lnx it (gdp > q m ) + δ × Control it + u it + ε it (4)
[0077] In the formula: θ0 represents the constant term; θ1, θ2, …, θ m , θ m+1 represent the coefficients of the core explanatory variables at different threshold levels; q1, q2, …, q m represent the corresponding threshold values; gdp is the threshold variable, that is, the level of economic development; δ represents the influence coefficient of the control variable Control it on the dependent variable lny it .
[0078] In an exemplary embodiment, as Figure 2 shown, after constructing the threshold effect model in step S40, the method includes step S50 to perform a model robustness check to ensure the robustness of the threshold effect analysis results.
[0079] In an exemplary embodiment, to ensure the robustness of the threshold effect analysis results, further tests need to be carried out from the following aspects: First, adopt the method of replacing the core explanatory variables, and use different surface thermal environment measurement indicators (such as the average surface temperature, night light brightness, etc.) to replace the original indicators for regression analysis; Second, perform two-sided winsorization on the sample data to reduce the impact of extreme values on the regression results; Third, adjust the sample space, and use different regional samples (such as only selecting cities in the eastern or western part of the Yangtze River Economic Belt) for repeated regression; Finally, re-estimate after excluding municipalities directly under the Central Government to rule out the interference of policy and economic characteristic heterogeneity that these cities may bring.
[0080] In summary, the embodiment of the present application provides an analytical method for urban heat-energy-carbon association based on mediation effect and threshold effect. Aiming at the problems in existing research such as the lack of mediation mechanism analysis, ignoring the non-linear threshold effect, spatial scale limitation, and insufficient time series investigation, a systematic analysis framework is constructed, and the scientificity and practicality of the research are improved through a series of technical means. Specifically, the technical solution of the present application has achieved beneficial effects in the following aspects:
[0081] First, aiming at the limitation of existing research that only focuses on direct effects, the mediation role of the surface thermal environment is systematically revealed. The present application uses a mediation effect model, introduces energy consumption as a key mediating variable, and quantifies how the surface thermal environment affects carbon emissions through the energy consumption chain through methods such as stepwise regression and Sobel test. Thus, on the one hand, the present application not only verifies the direct effect of the surface thermal environment on carbon emissions, but also further reveals its indirect effect through energy consumption, making up for the deficiency of existing research that only focuses on direct effects and making the understanding of the impact mechanism of urban carbon emissions more comprehensive. At the same time, by quantitatively analyzing the conduction effect of energy consumption between the thermal environment and carbon emissions, key energy consumption patterns can be identified, providing targeted policy basis for optimizing the energy structure and improving energy utilization efficiency.
[0082] Second, break through the linear assumption and identify the threshold characteristics at different stages of economic development. This application uses the Threshold Regression Model (TRM) and, based on the idea of piecewise regression, identifies the non-linear effects of factors such as economic development level, energy consumption type, and industrial structure on the heat-energy-carbon relationship, and then depicts the threshold effect. This application finds that at different stages of economic development, the impact path of the surface thermal environment on carbon emissions may change. For example, in more developed regions, the optimization of the energy structure may weaken the carbon emission effect of the thermal environment, while in less developed regions, where energy consumption depends on traditional fossil fuels, the carbon emission effect caused by the thermal environment is more significant. This finding makes up for the deficiency of existing research in paying insufficient attention to non-linear characteristics. At the same time, due to the different industrial structures and energy consumption patterns in different regions, this application identifies the key nodes in each region through the threshold regression model, which helps to formulate phased and regional low-carbon development strategies and improve the accuracy and effectiveness of policy intervention.
[0083] Third, break through the limitations of single-city and short-time-series research and construct a large-scale and long-time-series analysis framework. This application uses long-time-series remote sensing data (2000 - 2022), socioeconomic data, and panel econometric models to conduct cross-regional and multi-period dynamic analysis. Based on long-time-series data, this application systematically analyzes the long-term effects of factors such as urbanization process, climate change, and policy evolution on the heat-energy-carbon system, making the research conclusions more stable and applicable, and making up for the problem of short time span in existing research, which is difficult to reveal long-term trends. At the same time, compared with single-city research, this application's cross-regional analysis based on multiple cities in the Yangtze River Economic Belt covers urban agglomerations with different industrial structures, energy types, and economic development levels, making the research conclusions more extrapolable and providing experience for urban planning and carbon emission reduction in other regions.
[0084] Fourth, improve the accuracy of the heat-energy-carbon system research and construct a multi-source data fusion measurement system. This application uses remote sensing data, GIS information, and energy statistical data to construct a comprehensive measurement system for the surface thermal environment and carbon emissions, and introduces multi-dimensional variables such as the Surface Urban Heat Island Ratio Index (SUHI), energy consumption data, and industrial structure indicators to improve the explanatory power of the model. Compared with the traditional method that only relies on statistical data, this application inverses the urban thermal environment through remote sensing data and conducts cross-validation in combination with socioeconomic statistical data, making the carbon emission calculation more scientific. At the same time, by fusing multi-source data, the heat-energy-carbon analysis framework constructed in this application can be applied to the research of different urban agglomerations, industrial structures, and energy consumption patterns, making it highly applicable in various scenarios.
[0085] The embodiment of this application also provides an urban heat-energy-carbon association analysis device based on the mediation effect and threshold effect, as Figure 3 shown. This device includes:
[0086] A data acquisition unit 301, configured to acquire data; wherein, the data includes an explained variable, an explanatory variable, and a control variable, the explained variable includes carbon emission data, the explanatory variable includes the area of the heat island area, and the control variable is an influencing factor related to the scale of carbon emissions;
[0087] A benchmark regression model construction unit 302, configured to construct a panel benchmark regression model based on the acquired data, for characterizing the impact of the surface thermal environment on carbon emissions;
[0088] A mediation effect model construction unit 303, configured to construct a mediation effect model based on the panel benchmark regression model, for testing the action mechanism of the surface thermal environment on carbon emissions;
[0089] A threshold effect model construction unit 304, configured to construct a threshold effect model based on the panel benchmark regression model, for testing whether the impact of the surface thermal environment on carbon emissions is non-linear.
[0090] In some embodiments, the data acquisition unit is further configured to acquire the area of the heat island area in the following manner:
[0091] Acquire surface temperature data;
[0092] After performing projection and format transformation on the surface temperature data, perform annual synthesis based on the maximum synthesis method;
[0093] Normalize the surface temperature data after annual synthesis, unify the temperature range to between 0 and 1, and on the basis of the normalization process, take the range of 0 - 0.1 as the extremely strong cold island area, divide it into 10 levels at intervals of 0.1, and take the interval of 0.6 - 1 as the area of the heat island area.
[0094] In some embodiments, the influencing factors related to the scale of carbon emissions include social and economic factors and urban landscape pattern factors; wherein, the social and economic factors include the proportion of fiscal expenditure, the level of financial development, and the level of population urbanization. The level of financial development is measured by the ratio of the balance of deposits and loans of financial institutions at the end of the year to the regional GDP, and the level of population urbanization is measured by the ratio of the permanent population of towns to the total permanent population. The urban landscape pattern factors include the contagion index, patch density, and patch density.
[0095] In some embodiments, the panel benchmark regression model is expressed as:
[0096] lny it = α0 + α1×lnx it + α2×Control it + u it + εit (1)
[0097] Where: i represents the region, and t represents time; lny it represents the carbon emissions in region i at time t; lnx it represents the surface thermal environment in region i at time t; Control it represents the control variable; u it represents the regional fixed effect; ε it is the random disturbance term.
[0098] In some embodiments, the mediation effect model is expressed as:
[0099] medi it =β0 + β1×lnx it +β2×Control it +u it +ε it (2)
[0100] lny it =γ0 + γ1×lnx it +γ2×lnmedi it +γ3×Control it +u it +ε it (3)
[0102] Where: medi it represents the mediating variable, i.e., the energy consumption; β0 and γ0 represent the constant terms; β1 represents the influence coefficient of the independent variable lnx it on the mediating variable medi it ; β2 represents the influence coefficient of the control variable Control it on the mediating variable medi it ; γ1 represents the influence coefficient of the independent variable lnx it on the dependent variable lny it ; γ2 represents the influence coefficient of the mediating variable medi it on the dependent variable lny it ; γ3 represents the influence coefficient of the control variable Control it on the dependent variable lny it ;
[0103] In some embodiments, the threshold effect model is expressed as:
[0104] lny it =θ0 + θ1×lnx it (gdp ≤ q1)+θ2×lnx it(q1 < gdp < q2)+...+θ m ×lnx it (q m-1 <gdp<q m )+θ m+1 ×lnx it (gdp>q m )+δ×Control it +u it +ε it (4)
[0106] In the formula: θ0 represents the constant term; θ1, θ2, …, θ m , θ m+1 represent the coefficients of the core explanatory variables at different threshold levels; q1, q2, …, q m represent the corresponding threshold values; gdp is the threshold variable; δ represents the influence coefficient of the control variable Control it on the dependent variable lny it .
[0107] In some embodiments, the device further includes a model robustness verification module, which is configured to perform model robustness verification to ensure the robustness of the threshold effect analysis results.
[0108] In some embodiments, the model robustness verification module is further configured to replace the core explanatory variables, use different surface thermal environment measurement indicators to replace the original indicators for regression analysis; the different surface thermal environment measurement indicators include the mean surface temperature and / or night light brightness to perform two-sided winsorization on the sample data to reduce the influence of extreme values on the regression results; adjust the sample space and perform repeated regression using different regional samples respectively.
[0109] It should be noted that the structures of the various urban heat - energy - carbon correlation analysis devices based on mediation effect and threshold effect described in this embodiment belong to the same technical concept as the previously described urban heat - energy - carbon correlation analysis method based on mediation effect and threshold effect, and achieve the same beneficial effects through the same principle, which will not be elaborated here.
[0110] The embodiment of the present application also provides a readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the method described in any of the above embodiments.
[0111] The above description is intended to be illustrative and not restrictive. For example, the above examples (or one or more aspects thereof) may be used in combination with each other. For instance, those of ordinary skill in the art may use other embodiments when reading the above description. Additionally, in the above detailed description, various features may be grouped together to simplify the present application. This should not be construed as an intention that a feature of an unclaimed application is necessary for any claim. On the contrary, the subject matter of the present application may be less than all of the features of a particular application embodiment. Thus, the following claims are hereby incorporated into the detailed description by way of example or embodiment, where each claim stands on its own as a separate embodiment, and it is contemplated that these embodiments may be combined with each other in various combinations or permutations. The scope of the present application should be determined with reference to the appended claims and the full scope of equivalents to which those claims are entitled.
Claims
1. A method for analyzing the urban heat - energy - carbon correlation based on mediation effect and threshold effect, characterized in that, The method includes: Obtaining data; wherein, the data includes an explained variable, an explanatory variable, and a control variable, the explained variable includes carbon emission data, the explanatory variable includes the area of the heat island area, and the control variable is an influencing factor related to the scale of carbon emissions; Constructing a panel benchmark regression model based on the obtained data to characterize the impact of the surface thermal environment on carbon emissions; Constructing a mediating effect model based on the panel benchmark regression model to test the mechanism of the surface thermal environment on carbon emissions; Constructing a threshold effect model based on the panel benchmark regression model to test whether the impact of the surface thermal environment on carbon emissions is non-linear.
2. The method for analyzing the urban heat-energy-carbon correlation based on the mediating effect and the threshold effect according to claim 1, characterized in that The area of the heat island area is obtained in the following manner: Obtaining surface temperature data; After performing projection and format transformation on the surface temperature data, annual synthesis is performed based on the maximum synthesis method; The surface temperature data after annual synthesis is normalized, and the temperature range is unified to between 0 and 1. On the basis of the normalization process, taking 0 - 0.1 as the extremely strong cold island area, with an interval of 0.1, 10-level division is performed, and the interval of 0.6 - 1 is taken as the area of the heat island area.
3. The method for analyzing the urban heat-energy-carbon correlation based on the mediating effect and the threshold effect according to claim 1, wherein The influencing factors related to the scale of carbon emissions include socio-economic factors and urban landscape pattern factors; wherein, the socio-economic factors include the proportion of fiscal expenditure, the level of financial development, and the level of population urbanization. The level of financial development is measured by the ratio of the balance of deposits and loans of financial institutions at the end of the year to the gross regional product, and the level of population urbanization is measured by the ratio of the permanent population in towns to the total permanent population. The urban landscape pattern factors include the sprawl index, patch density, and patch density.
4. The method for analyzing the urban heat-energy-carbon correlation based on the mediating effect and the threshold effect according to claim 1, characterized in that, The panel benchmark regression model is expressed as: lny it = α0 + α1×lnx it + α2×Control it + u it + ε it (1) Where: i represents the region, and t represents the time; lny it represents the carbon emissions in region i during period t; lnx it represents the surface thermal environment in region i during period t; Control it represents the control variable; u it represents the regional fixed effect; ε it is the random disturbance term.
5. The method for analyzing the urban heat-energy-carbon correlation based on the mediating effect and the threshold effect according to claim 4, wherein The mediating effect model is expressed as: medi it = β0 + β1×lnx it + β2×Control it + u it + ε it (2) lny it = γ0 + γ1 × lnx it + γ2 × lnmedi it + γ3 × Control it + u it + ε it (3) Where: medi it represents the mediating variable, i.e., energy consumption; β0, γ0 represent the constant terms; β1 represents the independent variable lnx it The influence coefficient on the mediating variable medi it ; β2 represents the control variable Contral it The influence coefficient on the mediating variable medi it ; γ1 represents the independent variable lnx it The influence coefficient on the dependent variable lny it ; γ2 represents the mediating variable medi it The influence coefficient on the dependent variable lny it ; γ3 represents the control variable Control it The influence coefficient on the dependent variable lny it ; and γ3 represents the influence coefficient on the dependent variable lny.
6. The method for analyzing the urban heat - energy - carbon correlation based on the mediating effect and the threshold effect according to claim 4, wherein The threshold effect model is expressed as: lny it = θ0 + θ1 × lnx it (gdp ≤ q1) + θ2 × lnx it (q1 < gdp < q2) + … + θ m × lnx it (q m-1 < gdp < q m ) + θ m+1 × lnx it (gdp > q m ) + δ × Control it + u it + ε it (4) Where: θ0 represents the constant term; θ1, θ2, …, θ m , θ m+1 represent the coefficients of the core explanatory variables at different threshold levels; q1, q2, …, q m represent the corresponding threshold values; gdp is the threshold variable; δ represents the impact coefficient of the control variable Control it on the dependent variable lny it .
7. The method for analyzing the urban heat-energy-carbon association based on the mediating effect and the threshold effect according to claim 1, characterized in that After constructing the threshold effect model, the method includes: performing model robustness verification to ensure the robustness of the threshold effect analysis results.
8. The method for analyzing the urban heat-energy-carbon association based on the mediating effect and the threshold effect according to claim 7, wherein The methods for performing model robustness verification include: Replacing the core explanatory variable, and using different surface thermal environment measurement indicators to replace the original indicators for regression analysis; the different surface thermal environment measurement indicators include the mean surface temperature and / or night light brightness Performing two-sided winsorization on the sample data to reduce the impact of extreme values on the regression results; Adjusting the sample space, and performing repeated regression using different regional samples respectively.
9. An urban heat - energy - carbon correlation analysis device based on mediating effect and threshold effect, characterized in that, The device includes: A data acquisition unit configured to acquire data; wherein, the data includes an explained variable, an explanatory variable, and a control variable, the explained variable includes carbon emission data, the explanatory variable includes the area of the heat island area, and the control variable is an influencing factor related to the scale of carbon emissions; A benchmark regression model construction unit configured to construct a panel benchmark regression model based on the obtained data to characterize the impact of the surface thermal environment on carbon emissions; A mediating effect model construction unit configured to construct a mediating effect model based on the panel benchmark regression model to test the mechanism of the surface thermal environment on carbon emissions; The threshold effect model construction unit is configured to construct a threshold effect model based on the panel benchmark regression model, and is used to test whether the impact of the surface thermal environment on carbon emissions is non-linear.
10. A non-transitory computer-readable storage medium storing instructions, which, when executed by a processor, execute the method according to any one of claims 1 to 8.