Urban carbon emission accounting method based on night light remote sensing and improved STIRPAT

Abstract

The present invention provides a method for calculating urban carbon emissions based on nighttime light remote sensing and improved STIRPAT, comprising the following steps: processing nighttime light remote sensing data; then constructing a city development factor CDE (City Development Element) for provincial and municipal regions; using provincial statistical yearbook data to construct an improved STIRPAT; obtaining a model intercept term through a panel data regression model, and inverting carbon emissions at the municipal scale. The present invention proposes a city development factor CDE constructed based on nighttime light remote sensing data, which solves the problem of inconsistent caliber and missing data items of energy statistical data at the municipal scale, and can more accurately invert carbon emissions from provincial scale to municipal scale, thereby improving calculation accuracy.

Description

Technical Field

[0001] The invention relates to the technical field of urban research, and in particular to an urban carbon emission accounting method based on nighttime light remote sensing and improved STIRPAT. Background Art

[0002] Climate change has become a global challenge that transcends national boundaries, and how to calculate and control carbon emissions is also an issue that the international community attaches great importance to. Since the 1760s, the Industrial Revolution has gradually emerged from the United Kingdom, and the rapid development of industrial technology has affected the global production technology changes. With the burning of a large amount of oil and coal, carbon dioxide emissions began to rise year by year. In response to global climate change, as of June 29, 2016, 178 parties around the world jointly signed the Paris Agreement. The long-term goal is to limit the increase in global average temperature to within 2 degrees Celsius compared with the pre-industrial period, and strive to limit the temperature increase to within 1.5 degrees Celsius. The World Meteorological Organization released the "Global Climate Status in 2020" report, which mentioned that the global average temperature is about 1.2℃ higher than the pre-industrial level. In the first part of its Sixth Assessment Report, Climate Change 2021: The Physical Science Basis, the IPCC (Intergovernmental Panel on Climate Change) mentioned new estimates of global warming levels exceeding 1.5°C in the coming decades, and found that unless immediate measures are taken and greenhouse gas emissions are reduced on a large scale, limiting warming to close to 1.5°C or even 2°C will be out of reach.

[0003] In recent years, many experts and scholars at home and abroad have conducted a lot of research on the accounting and prediction of carbon emissions. Among them, the prediction of carbon emissions mainly involves carbon emission image factors. Commonly used research methods include STIRPAT model and factor decomposition method. Factor decomposition method can be mainly divided into three types: influencing factor decomposition system, regression analysis system, and energy statistics system. As an influencing factor decomposition method, the most representative one is the Logarithmic Mean Divisia Index model. The LMDI model decomposes carbon emissions as a target from the spatial and temporal dimensions to form multiple single influencing factors, which can effectively calculate the contribution of one of the factors to the change of carbon emissions over time. Tang et al. used the LMDI model to decompose historical emissions and predicted emissions, and effectively evaluated the impact of factors such as population and industrial structure adjustment on carbon emissions in Jiangsu Province. Zhang et al. analyzed the driving factors of carbon emissions in Henan Province through the STIRPAT model, and judged the best mode of balance between future economic development and carbon emissions in Henan Province based on scenario analysis. In energy statistics methods, from the perspective of carbon emission process classification, it can be roughly divided into two types, namely, carbon emission statistics based on the production end and carbon emission statistics based on the consumption end. Carbon emission statistics based on the production end mainly calculates the carbon emissions caused by domestically produced and exported products. This method takes into account the carbon emissions at the production point but does not take into account the carbon emissions caused by users and places of use. Carbon emission statistics based on the consumption end mainly evaluate the carbon emissions caused by terminal consumption activities. This method is mainly used for carbon emission calculations at the national and provincial scales.

[0004] Due to the large number of studies on carbon emission calculation at the national and provincial levels at home and abroad, in recent years, many researchers have gradually turned to carbon emission accounting at the city and county levels. The intensity of night lights can be used to indicate the level of energy consumption. The most commonly used methods for simulating social economy using night light data include linear simulation, exponential simulation, logarithmic simulation, quadratic polynomial simulation, etc. Yang et al. used the STIRPAT model to study the heterogeneity of county-level carbon emissions in Ningxia Hui Autonomous Region, China. Wang et al. used DMSP / OLS night light data to calculate carbon emissions in cities across China. Ghosh et al. used population data and night light remote sensing data to construct a global carbon emission kilometer grid map. Chen et al. used a particle swarm inverse optimization algorithm combined with night light data DMSP / OLS and NPP / VIIRS data to estimate carbon emissions at the county level in China from 1997 to 2017. Gu Yangyang et al. inverted the 1km carbon emission distribution map of Henan Province from 1995 to 2010, and revealed the spatial distribution pattern of carbon emissions in various prefecture-level cities. Ma Zhongyu and others constructed a spatiotemporal geographic weighted regression model using China's provincial nighttime light data from 2000 to 2013 and per capita carbon emissions and carbon emissions per unit area. At the same time, they simulated the carbon emissions of each province by combining population size and land area. Su Yongxian and others used DMSP / OLS nighttime light data to estimate China's total carbon emissions at the city level and analyzed the spatiotemporal evolution characteristics of total carbon emissions and carbon emission density at different scales.

[0005] Although current research uses night light remote sensing data or relevant energy statistical data in the calculation of carbon emissions, there are problems such as missing data and inconsistent statistical caliber in municipal energy statistical data, and there are few studies on downscaling carbon emissions from the provincial level to the municipal level. Summary of the invention

[0006] The present invention proposes a method for calculating urban carbon emissions based on nighttime light remote sensing and improved STIRPAT, so as to solve the technical problem that the existing technology cannot estimate the city-level carbon emissions.

[0007] In order to solve the above technical problems, the present invention discloses a method for calculating urban carbon emissions based on nighttime light remote sensing and improved STIRPAT, comprising:

[0008] S1: Acquire nighttime light remote sensing data, including DMSP / OLS data and NPP / VIIRS data, and preprocess the acquired nighttime light remote sensing data;

[0009] S2: Construct urban development factors based on pre-processed nighttime light remote sensing data;

[0010] S3: Construct an improved STIRPAT model based on provincial statistical yearbook data and urban development factors;

[0011] S4: Calculate provincial carbon emissions using the IPCC factor method based on provincial energy statistical yearbook data and form a provincial carbon emissions inventory;

[0012] S5: Based on the provincial carbon emission inventory data, the improved STIRPAT model is subjected to panel regression to calculate the carbon emissions at the municipal level.

[0013] In one embodiment, in step S1, preprocessing the acquired nighttime light remote sensing data includes:

[0014] Conduct continuity correction, saturation correction, intra-annual fusion and inter-annual correction on DMSP / OLS data, and denoise NPP / VIIRS data;

[0015] The DMSP / OLS data and NPP / VIIRS data of the same year were used for data fusion.

[0016] In one embodiment, step S2 includes:

[0017] S2.1: Perform mean processing on the night light brightness values ​​and calculate the average brightness value of the study area. The formula is:

[0018]

[0019] Where MDN is the average brightness of the study area, n is the number of pixels in the night light image contained in the study area, and DN is i Represents the brightness value corresponding to the i-th pixel in the study area;

[0020] S2.2: Use the nighttime light data of the study area to calculate the total nighttime light brightness SUML of the study area, the grid area SA of the study area, and the light intensity value DNQ of the study area. The formula is:

[0021]

[0022] Where SUML is the sum of the nighttime light brightness values ​​represented by each pixel grid in the study area, SA is the sum of the nighttime light grid areas in the study area, DNQ represents the light intensity value in the study area, and area i Represents the area of ​​the i-th pixel;

[0023] S2.3: Combine the average light brightness value MDN, the total light brightness value SUML, and the light intensity value DNQ to construct the urban development factor:

[0024]

[0025] Where a, b, and c are the coefficients of the average light brightness value MDN, the total light brightness value SUML, and the light intensity value DNQ, respectively, and CDE is the urban development factor.

[0026] In one embodiment, step S3 includes:

[0027] Based on the constructed urban development factor as a dimensionless variable CDE extended to the STIRPAT model, the improved night light STIRPAT inversion model is obtained:

[0028]

[0029] Among them, Tan represents carbon emissions, the driving factor People represents population, GDP represents regional gross domestic product, SeGDP represents the added value of the secondary industry, JianZhu represents the total output value of the construction industry, a, b, c, d, f, g, h, j are the intercept terms of the variables, and e is the error term.

[0030] In one embodiment, step S4 includes:

[0031] S4.1: Calculate provincial carbon emissions using the IPCC factor method:

[0032]

[0033] CE is the direct carbon emissions of the city caused by energy consumption; i is the industry type; j is the different energy types; E ij is the consumption of different types of energy in different industries; j Standard coal conversion factor for different energy sources; f j is the carbon emission coefficient of different energy sources; 44 / 12 is the conversion factor of carbon to carbon dioxide;

[0034] S4.2: Construct a carbon emissions inventory based on the calculated provincial carbon emissions.

[0035] In one embodiment, step S5 includes:

[0036] S5.1: Based on the calculated provincial carbon emission inventory data, provincial statistical data and provincial night light remote sensing data, the improved STIRPAT model coefficient values ​​are obtained by panel model regression;

[0037] S5.2: Obtain city-level nighttime light data and city-level statistical data, and use the improved STIRPAT model to calculate city-level carbon emissions using the model coefficient values ​​obtained through panel model regression.

[0038] The above one or more technical solutions in the embodiments of the present application have at least one or more of the following technical effects:

[0039] The present invention proposes a city carbon emission accounting method based on night light remote sensing and improved STIRPAT. The model is constructed by using provincial night light remote sensing images, and provincial and municipal urban development factors are constructed according to preprocessed night light remote sensing data, respectively. The problem of accuracy in calculating brightness values ​​of night lights in different regions can be solved. The constructed urban development factor CDE is used as an extension factor in the STIRPAT model to construct an improved STIRPAT model. The municipal carbon emissions are further calculated through provincial energy statistical data, thereby achieving accurate calculation of urban carbon emissions, overcoming problems such as data missing and statistical caliber, and being able to comprehensively, comprehensively and accurately invert carbon emissions at the municipal scale. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A schematic diagram of a nighttime light remote sensing data preprocessing process in an embodiment of the present invention;

[0042] Figure 2 A flow chart of a method for calculating urban carbon emissions based on nighttime light remote sensing and improved STIRPAT proposed in an embodiment of the present invention;

[0043] Figure 3 This is a schematic diagram of the city-level (urban) carbon emission inversion process according to an embodiment of the present invention;

[0044] Figure 4 This is a schematic diagram of the accuracy verification of carbon emissions at provincial and municipal levels according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] Through extensive research and practice, the inventors of this application discovered that the contradiction between the urgent needs of regional economic development and the historical tasks of carbon emissions and carbon neutrality is becoming increasingly severe. How to efficiently and accurately quantitatively estimate the carbon emissions of various cities and prefectures in Hubei Province is a key issue in balancing sustainable development among regions and giving full play to the important role of various cities and prefectures in the province's carbon emission targets.

[0046] Based on this, the present invention proposes an improved STIRPAT night light urban carbon emissions accounting method based on STIRPAT, constructs a city development factor (CDE) using NPP / VIIRS night light data, and constructs a panel model from provincial statistical data to the city level, overcoming the problem of missing data in city-level energy statistics and achieving accurate estimation of city-level carbon emissions. The method of the present invention has the characteristics of high data accuracy, wide coverage, long time span, strong objectivity, etc., and has important scientific significance for carbon emissions accounting at the city level.

[0047] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] The embodiment of the present invention provides a method for calculating urban carbon emissions based on nighttime light remote sensing and improved STIRPAT, including:

[0049] S1: Acquire nighttime light remote sensing data, including DMSP / OLS data and NPP / VIIRS data, and preprocess the acquired nighttime light remote sensing data;

[0050] S2: Construct urban development factors based on pre-processed nighttime light remote sensing data;

[0051] S3: Construct an improved STIRPAT model based on provincial statistical yearbook data and urban development factors;

[0052] S4: Calculate provincial carbon emissions using the IPCC factor method based on provincial energy statistical yearbook data and form a provincial carbon emissions inventory;

[0053] S5: Based on the provincial carbon emission inventory data, the improved STIRPAT model is subjected to panel regression to calculate the carbon emissions at the municipal level.

[0054] Specifically, the steps of the method of the present invention are: processing night light remote sensing data; then constructing city development factor CDE (City Development Element) for provincial and municipal regions; using provincial statistical yearbook data to construct an improved STIRPAT; using the IPCC factor method to calculate provincial carbon emissions and form a provincial carbon emissions inventory; obtaining the model intercept term through a panel data regression model to invert the carbon emissions at the municipal scale. The present invention proposes a city development factor CDE constructed based on night light remote sensing data, which solves the problem of inconsistent caliber and missing data items of energy statistical data at the municipal scale, and can more accurately invert provincial carbon emissions to municipal carbon emissions, thereby improving the calculation accuracy.

[0055] See also Figure 2 , which is a flow chart of a method for calculating urban carbon emissions based on nighttime light remote sensing and improved STIRPAT proposed in an embodiment of the present invention.

[0056] During the specific implementation process, provincial statistical yearbook data and provincial energy statistical yearbook data can be obtained in advance. The provincial statistical yearbook data include: gross regional product (GDP), population size (total population at the end of the year), total value of the secondary industry, and total output value of the construction industry.

[0057] The provincial energy statistical yearbook data includes: total energy consumption by region, coal consumption by region, coke consumption by region, crude oil consumption by region, fuel oil consumption by region, and gasoline consumption by region.

[0058] The nighttime light remote sensing data in step S1 include provincial and municipal data, both of which come from the same set of images (existing nighttime light remote sensing images, National Centers for Environmental Information website), and are preprocessed in the same way, with the difference being the regional scale. S2: Based on the preprocessed provincial nighttime light remote sensing data and municipal nighttime light remote sensing data, urban development factors are constructed respectively.

[0059] In one embodiment, in step S1, preprocessing the acquired nighttime light remote sensing data includes:

[0060] Conduct continuity correction, saturation correction, intra-annual fusion and inter-annual correction on DMSP / OLS data, and denoise NPP / VIIRS data;

[0061] The DMSP / OLS data and NPP / VIIRS data of the same year were used for data fusion.

[0062] Night light remote sensing data refers to the use of remote sensing technology to capture the distribution of lights on the earth at night. It can effectively reflect the spatial distribution of human activities and is therefore often used for remote sensing inversion of various socio-economic data. Currently, the commonly used night light remote sensing data mainly come from two satellites. One is DMSP (Defense Meteorological Satellite Program), which is a polar-orbiting satellite program of the U.S. Department of Defense. The spatial resolution of the sensor is 3000m, and the spatial resolution of the night light remote sensing products produced is usually 1000m. At the same time, DMPS night light remote sensing data is the longest time series of night light remote sensing data (from 1992-2013), which can provide long-term continuous night light remote sensing monitoring. The other is Suomi NPP, a new generation of earth observation satellite launched in 2011. The satellite carries a visible light / infrared radiation imager (Visible Infrared Imaging Radiometer Suit, VIIRS) that can obtain new night light remote sensing images (Day / Night Band, DNB band), and the spatial resolution is also increased to 750m (hereinafter referred to as NPP-DNB). The spatial resolution of the night light remote sensing products produced is usually 500m.

[0063] See also Figure 1 , is a schematic diagram of the preprocessing process of nighttime light remote sensing data in an embodiment of the present invention.

[0064] In the specific implementation process, the processing of DMSP / OLS (Defense Meteorological Satellite Program / Operational LineScan System) and NPP / VIIRS (National Polar-Orbiting Partnership / Visible Infrared Imaging Radiometer Suite) data includes:

[0065] (1) Due to the inherent problems of DMSP / OLS and NPP / VIIRS data, corresponding processing is required when applying long-term series. DMSP / OLS data require continuity correction, saturation correction, intra-annual fusion, and inter-annual correction.

[0066] (2) De-noising of NPP / VIIRS data.

[0067] (3) DMSP / OLS and NPP / VIIRS data fusion: In order to unify the DMSP / OLS and NPP / VIIRS data, the DMSP / OLS and NPP / VIIRS data of the same year are used to realize data fusion based on the power function.

[0068] The above-mentioned nighttime light remote sensing data includes provincial and municipal nighttime light remote sensing data, which is obtained through the above-mentioned nighttime light remote sensing images. In one embodiment, step S2 includes:

[0069] S2.1: Perform mean processing on the night light brightness values ​​and calculate the average brightness value of the study area. The formula is:

[0070]

[0071] Where MDN is the average brightness of the study area, n is the number of pixels in the night light image contained in the study area, and DN is i Represents the brightness value corresponding to the i-th pixel in the study area;

[0072] S2.2: Use the nighttime light data of the study area to calculate the total nighttime light brightness SUML of the study area, the grid area SA of the study area, and the light intensity value DNQ of the study area. The formula is:

[0073]

[0074] Where SUML is the sum of the nighttime light brightness values ​​represented by each pixel grid in the study area, SA is the sum of the nighttime light grid areas in the study area, DNQ represents the light intensity value in the study area, and area i Represents the area of ​​the i-th pixel;

[0075] S2.3: Combine the average light brightness value MDN, the total light brightness value SUML, and the light intensity value NDQ to construct the urban development factor:

[0076]

[0077] Where a, b, and c are the coefficients of the average light brightness value MDN, the total light brightness value SUML, and the light intensity value DNQ, respectively, and CDE is the urban development factor.

[0078] Specifically, there are two types of areas in the city, urban and rural areas. The two areas are quite different in terms of development level. When the DN values ​​are at a low level, the carbon emissions will also have large differences. Therefore, the DN values ​​obtained using night light remote sensing data are averaged to construct an urban development factor to correct the difference in carbon emissions between urban and rural areas at the same DN value.

[0079] The present invention constructs the urban development factor through steps S2.1 to S2.3, and according to the different research areas, the urban development factors are respectively for the provincial and municipal levels. If the provincial urban development factor is to be constructed, the research area is the province, and the construction is based on the average brightness value, total light brightness, and light intensity value of the area. If the municipal urban development factor is to be constructed, the research area is the city, and the construction is based on the average brightness value, total light brightness, and light intensity value of the municipal area.

[0080] In one embodiment, step S3 includes:

[0081] Based on the constructed urban development factor as a dimensionless variable CDE extended to the STIRPAT model, the improved night light STIRPAT inversion model is obtained:

[0082]

[0083] Among them, Tan represents carbon emissions, the driving factor People represents population, GDP represents regional gross domestic product, SeGDP represents the added value of the secondary industry, JianZhu represents the total output value of the construction industry, a, b, c, d, f, g, h, j are the intercept terms of the variables, and e is the error term.

[0084] The formula of the scalable stochastic environmental impact assessment model (STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology)) is as follows:

[0085] LnTan=LnCDE+dLnPeople+fLnGDP+g(LnGDP) 2 +hLnSeGDP+jLnJianZhu+Lne

[0086] Based on the existing STIRPAT model, the present invention constructs an improved STIRPAT model shown in formula (6) based on constructed urban development factors and provincial statistical yearbook data.

[0087] In one embodiment, step S4 includes:

[0088] S4.1: Calculate provincial carbon emissions using the IPCC factor method:

[0089]

[0090] CE is the direct urban carbon emissions generated by energy consumption, that is, provincial carbon emissions; i is the industry type; j is the different energy types; Eij is the consumption of different types of energy in different industries; εj is the standard coal conversion coefficient of different energy sources; fj is the carbon emission coefficient of different energy sources; 44 / 12 is the conversion coefficient of carbon to carbon dioxide;

[0091] S4.2: Construct a carbon emissions inventory based on the calculated provincial carbon emissions.

[0092] Specifically, provincial carbon emissions are calculated using the IPCC (Intergovernmental Panel on Climate Change) factor method.

[0093] After calculating the provincial carbon emissions, the carbon emission inventory can be obtained, among which the standard coal conversion coefficient and carbon emission coefficient are shown in Table 1.

[0094] Table 1 Standard coal conversion factor and carbon emission factor

[0095]

[0096]

[0097] In one embodiment, step S5 includes:

[0098] S5.1: Based on the calculated provincial carbon emission inventory data, provincial statistical data and provincial night light remote sensing data, the improved STIRPAT model coefficient values ​​are obtained by panel model regression;

[0099] S5.2: Obtain city-level nighttime light data and city-level statistical data, and use the improved STIRPAT model to calculate city-level carbon emissions using the model coefficient values ​​obtained through panel model regression.

[0100] Specifically, panel data can be called time series cross-sectional data or mixed data. It is two-dimensional data with time and space cross-sections. The panel data model can make the variables reflect the changing rules and characteristics in time and space. At the same time, the construction of the panel data model can control the heterogeneity and endogeneity of individuals, increase the degree of freedom and thus improve the effectiveness of parameter estimation. Perform the Huasman test on the panel model, and select the random effect model or the fixed effect model according to the test results.

[0101] During the specific implementation process, city-level night light data can be obtained based on the aforementioned night light remote sensing images, and city-level statistical data can be obtained through the statistical yearbooks of each city.

[0102] It should be noted that the data used in the improved STIRPAT model used in S5.2 are city-level nighttime light data and city-level statistical data. The data used in the improved STIRPAT model in S5.1 are provincial carbon emission inventory data, provincial statistical data, and provincial nighttime light remote sensing data.

[0103] See also Figure 3 , which is a schematic diagram of the city-level (urban) carbon emission inversion process according to an embodiment of the present invention.

[0104] Although the prior art uses nighttime light remote sensing data to estimate carbon emissions, it does so by extracting urban construction land and building energy carbon emissions inversion models for different dominant industry types. The present invention proposes an urban development factor, which is obtained from nighttime light data. Adding this factor to the improved model can improve the accuracy of carbon emissions accounting.

[0105] The main body of the present invention is to improve the STIRPAT model and propose a new concept of urban development factor based on night light data to solve the current problems of missing energy statistical data and inconsistent statistical caliber at the city level.

[0106] Compared with the existing solutions, it has the following advantages or beneficial technical effects:

[0107] 1. In the calculation of carbon emissions at the municipal level, there is a significant improvement in accuracy, and the calculation accuracy can reach above 0.9.

[0108] 2. Solved the problems of missing data and inconsistent statistical caliber.

[0109] In general, the method provided by the present invention utilizes nighttime light remote sensing data, the STIRPAT model, and the Environmental Kuznets Curve (EKC) to downscale and invert carbon emissions from the provincial scale to the municipal scale, overcoming the problems of missing energy statistical yearbooks and inconsistent calibers at the municipal scale, and providing a high-precision estimate of carbon emissions at the municipal scale, with the characteristics of wide range, long time series, and high precision.

[0110] The method proposed in the present invention is described in detail below through specific examples.

[0111] The study area takes 17 cities and prefectures in Hubei Province as an example, and uses the NPP-VIIRS from 2012 to 2018 as the research data to construct the urban development factor CDE. The provincial and municipal statistical yearbook data are used to build an improved STIRPAT model, and the panel data is used for regression to obtain the intercept term of the model factor. Finally, the municipal carbon emission data released by CEADs is used as the verification data set.

[0112] 1. Night light remote sensing data processing

[0113] 2. Calculate the urban development factor CDE

[0114] The urban development factors of cities in Hubei Province from 2012 to 2018 and the urban development factors of 30 provinces in China from 2012 to 2018 are calculated based on the region.

[0115] 3. Constructing an improved STIRPAT model based on provincial statistical yearbook data and urban development factors

[0116] 4. Use provincial carbon emissions to perform panel data regression to obtain model coefficient values.

[0117] By constructing a provincial carbon emission panel data model to invert city-level carbon emissions, the constructed provincial carbon emission panel model is subjected to a Huasman test. According to the calculated provincial panel data Huasman test result P value greater than 0.05, the original hypothesis is accepted and the random effect model is adopted.

[0118]

[0119] Where i = 1, 2, ..., N represents individual members, and t = 1, 2, ..., T represents the time span. it represents the disturbance term, a i represents the constant term, β i is the coefficient value, Y it represents the carbon emissions of the ith city in year T, x it It represents the influencing factors of carbon emissions in the ith city in the Tth year, and k represents the number of influencing factors.

[0120] LLC and ADF tests were performed, and all variables passed the test (Table 2), indicating that there was no pseudo regression in the panel data model. Therefore, the intercept terms of each factor in the improved model were obtained through panel data regression of the random effects model (Table 3).

[0121] Table 2 LLC and ADF test

[0122]

[0123]

[0124] Table 3 Intercept and error of each factor in panel data regression (Coefficient represents intercept, Std.Error represents standard deviation, T-Statistic represents T statistic, Prob represents significance)

[0125]

[0126] 5. Carbon emissions of 17 cities and prefectures in Hubei Province (2012-2018) were inverted through the improved model. Table 4

[0127] Table 4 Carbon emissions by city and prefecture in Hubei Province from 2012 to 2018

[0128]

[0129] 6. Accuracy Verification

[0130] See also Figure 4 , which is a schematic diagram of the verification of the accuracy of carbon emissions at the provincial and municipal levels according to an embodiment of the present invention.

[0131] The provincial carbon emission data and the municipal carbon emission inventory data released by CEADs were used to verify the accuracy. The provincial scale accuracy was 0.7976, the municipal scale accuracy in Hubei Province was 0.9641, and the R 2 Greater than 0.9, indicating that the improved model has a good degree of fit and high calculation accuracy.

[0132] It should be understood that the above description of the preferred embodiment is relatively detailed and cannot be regarded as limiting the scope of patent protection of the present invention. Under the enlightenment of the present invention, ordinary technicians in this field can also make substitutions or modifications without departing from the scope of protection of the claims of the present invention, which all fall within the scope of protection of the present invention. The scope of protection requested for the present invention shall be based on the attached claims.

Claims

1. The urban carbon emission accounting method based on night light remote sensing and improved STIRPAT is characterized by: include: S1: Acquire nighttime light remote sensing data, including DMSP / OLS data and NPP / VIIRS data, and preprocess the acquired nighttime light remote sensing data; S2: Construct urban development factors based on pre-processed nighttime light remote sensing data; S3: Construct an improved STIRPAT model based on provincial statistical yearbook data and urban development factors; S4: Calculate provincial carbon emissions using the IPCC factor method based on provincial energy statistical yearbook data and form a provincial carbon emissions inventory; S5: Based on the provincial carbon emission inventory data, the improved STIRPAT model is subjected to panel regression to calculate the carbon emissions at the municipal level; Wherein, step S2 comprises: S2.1: Perform mean processing on the night light brightness values ​​and calculate the average brightness value of the study area. The formula is: Where MDN is the average brightness of the study area, n is the number of pixels in the night light image contained in the study area, and DN is i Represents the brightness value corresponding to the i-th pixel in the study area; S2.2: Use the nighttime light data of the study area to calculate the total nighttime light brightness SUML of the study area, the grid area SA of the study area, and the light intensity value DNQ of the study area. The formula is: Where SUML is the sum of the nighttime light brightness values ​​represented by each pixel grid in the study area, SA is the sum of the nighttime light grid areas in the study area, DNQ represents the light intensity value in the study area, and area i Represents the area of ​​the i-th pixel; S2.3: Combine the average light brightness value MDN, the total light brightness value SUML, and the light intensity value DNQ to construct the urban development factor: In the formula, a, b, c are the coefficients of the average light brightness value MDN, the total light brightness value SUML, and the light intensity value DNQ respectively, and CDE is the urban development factor; Step S3 includes: Based on the constructed urban development factor as a dimensionless variable CDE extended to the STIRPAT model, the improved STIRPAT model is obtained: Among them, Tan represents carbon emissions, the driving factor People represents population, GDP represents regional gross domestic product, SeGDP represents the added value of the secondary industry, JianZhu represents the total output value of the construction industry, a, b, c, d, f, g, h, j are the intercept terms of the variables, and e is the error term.

2. The urban carbon emission accounting method according to claim 1, characterized in that: In step S1, preprocessing the acquired nighttime light remote sensing data includes: Conduct continuity correction, saturation correction, intra-annual fusion and inter-annual correction on DMSP / OLS data, and denoise NPP / VIIRS data; The DMSP / OLS data and NPP / VIIRS data of the same year were used for data fusion.

3. The urban carbon emission accounting method according to claim 1, characterized in that: Step S4 includes: S4.1: Calculate provincial carbon emissions using the IPCC factor method: CE is the direct carbon emissions of the city caused by energy consumption; i is the industry type; j is the different energy types; E ij is the consumption of different types of energy in different industries; j Standard coal conversion factor for different energy sources; f j is the carbon emission coefficient of different energy sources; 44 / 12 is the conversion factor of carbon to carbon dioxide; S4.2: Construct a carbon emissions inventory based on the calculated provincial carbon emissions.

4. The urban carbon emission accounting method according to claim 1, characterized in that: Step S5 includes: S5.1: Based on the calculated provincial carbon emission inventory data, provincial statistical data and provincial night light remote sensing data, the improved STIRPAT model coefficient values ​​are obtained by panel model regression; S5.2: Obtain city-level nighttime light data and city-level statistical data, and use the improved STIRPAT model to calculate city-level carbon emissions using the model coefficient values ​​obtained through panel model regression.

Images