Inversion method, system and equipment of black carbon emission and medium

Abstract

The invention discloses an inversion method, system, equipment and medium for black carbon emission, and relates to the technical field of carbon emission list inversion, and the method comprises the steps: carrying out the space-time decomposition and dimension reduction of a black carbon emission list through employing an empirical orthogonal function model, and obtaining a plurality of emission source regions representing a black carbon emission source on a global scale; in each emission source area, considering linear and nonlinear interaction among a plurality of influence factors, and establishing a nonlinear regression model between the interaction influence of the plurality of influence factors and the black carbon emission; and inputting the satellite remote sensing observation truth values of the plurality of influence factors and the reanalysis meteorological field data into a nonlinear regression model to obtain a black carbon emission list in the emission source area. According to the method, accurate black carbon emission can be obtained through inversion, and then accurate black carbon emission list data are obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission, and in particular to a method, system, equipment and medium for inverting black carbon emission. Background Art

[0002] Black carbon is a kind of fine particulate matter PM in the atmosphere. 2.5 Black carbon is one of the important components of carbon dioxide, which is produced by incomplete combustion of fossil fuels and biomass. The surface of black carbon can adsorb a large amount of carcinogens such as volatile organic compounds, which enter the human respiratory system and cause harm to human health. Black carbon aerosol can act as cloud condensation nuclei, affecting the size distribution of cloud droplets, the liquid water content in the cloud and the cloud amount, thereby affecting rainfall, leading to frequent droughts or rainstorms in different regions. In addition, black carbon aerosol has a strong absorption effect on the visible light band, produces positive radiation forcing, changes the radiation energy balance of the surface and thus affects climate change. Therefore, it is crucial to count black carbon pollution and formulate targeted measures to control black carbon pollution according to local emission characteristics, in order to estimate the climate effect of black carbon, assess residents' exposure health and formulate emission inventories. The black carbon emission inventory is a tool to record and report the black carbon emissions of a specific region or industry over a certain period of time.

[0003] In the existing technology, the black carbon emission inventory is mainly used to count the black carbon pollution values ​​emitted into the atmosphere by various black carbon pollution sources. Currently, the black carbon emission inventory mostly calculates the emissions based on the classified emission sources and emission factors.

[0004] However, there are too many types of emission sources, and it is impossible to conduct a comprehensive statistics of emission sources in various regions. In addition, the black carbon emitted from emission sources is affected by many factors. The above method does not take into account the various factors affecting the emission sources when obtaining the black carbon emission inventory, resulting in errors in the black carbon emissions in the black carbon emission inventory. Summary of the invention

[0005] The embodiments of the present invention provide a method, system, device and medium for inverting black carbon emissions, which can solve the problem of errors in black carbon emissions in the black carbon emission inventory obtained in the prior art.

[0006] An embodiment of the present invention provides an inversion method for black carbon emissions, comprising the following steps: obtaining a black carbon emission inventory including black carbon emissions, and obtaining multiple influencing factors that affect the black carbon emissions, wherein the multiple influencing factors include nitrogen dioxide, ultraviolet aerosol index and wind; performing spatiotemporal decomposition and dimensionality reduction on the black carbon emission inventory using an empirical orthogonal function model to obtain multiple emission source areas including black carbon emission sources; within each emission source area, considering linear and nonlinear interactions among multiple influencing factors, establishing a nonlinear regression model between the black carbon emissions under the influence of multiple influencing factors; inputting multiple influencing factors into the nonlinear regression model to obtain black carbon emissions, and completing the inversion of black carbon emissions.

[0007] Furthermore, the obtaining includes multiple emission source areas of black carbon emission sources, and the specific steps include: using an empirical orthogonal function model to obtain multiple eigenvalue contribution rates for the black carbon emission inventory, selecting the top three largest eigenvalue contribution rates from all the eigenvalue contribution rates, and marking the selected top three largest eigenvalue contribution rates from large to small as EOF1, EOF2 and EOF3 respectively; screening the values ​​greater than 0.5 and less than -0.5 in the eigenvector corresponding to the eigenvalue contribution rate EOF1; screening the values ​​greater than 0.8 and less than -0.8 in the eigenvectors corresponding to the eigenvalue contribution rates EOF2 and EOF3; obtaining the geographical location of the preliminary emission source area from the black carbon emission inventory based on the numerical difference of the filtered results; excluding the area with an area of ​​less than 100 pixels in the preliminary emission source area, and using an edge detection algorithm to extract the regional boundaries of the excluded preliminary emission source area, and dividing the six emission source areas.

[0008] Furthermore, the nonlinear regression model between the black carbon emissions under the interaction of multiple influencing factors is established, and the specific steps include: multiple influencing factors reflecting the black carbon emissions include nitrogen dioxide NO2, ultraviolet aerosol index UVAI and wind Wind; the linear interaction between the influencing factors and the black carbon emissions needs to be considered, and the nonlinear interaction between the influencing factors needs to be considered, and the nonlinear interaction includes the interaction between the ultraviolet aerosol index UVAI and wind Wind, and the interaction between nitrogen dioxide NO2 and wind Wind;

[0009] The established nonlinear regression model formula is:

[0010] BC=β1NO2+β2UVAI+β3Wind+β4NO2×Wind+β5UVAI×Wind+C

[0011] Among them, β1, β2, β3, β4 and β5 are the coefficients of nitrogen dioxide NO2, ultraviolet aerosol index UVAI, wind Wind, the interaction between wind Wind and ultraviolet aerosol index UVAI, and the interaction between wind Wind and nitrogen dioxide NO2, respectively, and C is a constant.

[0012] Furthermore, the black carbon emissions are obtained by inputting multiple influencing factors into a nonlinear regression model, and the specific steps include: obtaining multiple months of monitoring data of nitrogen dioxide NO2, ultraviolet aerosol index UVAI and wind Wind, and inputting the multiple months of monitoring data into a linear regression model to obtain the black carbon emissions of each month; analyzing the annual average, seasonal average and monthly average black carbon emissions based on the obtained black carbon emissions, and comparing the differences with the black carbon emission inventory to determine the accuracy of the inversion results.

[0013] Furthermore, the specific steps of obtaining the black carbon emissions inventory include: using the Kalman filtering method to obtain an optimized estimate of black carbon emissions based on historical black carbon emissions at a global scale, and obtaining a scaling factor for making black carbon emissions conform to actual conditions based on the difference between the optimized estimate of black carbon emissions and historical black carbon emissions; scaling the monthly black carbon emissions data of the Global Atmospheric Research Emissions Database EDGAR to obtain a black carbon emissions inventory.

[0014] An embodiment of the present invention provides a black carbon emission inversion system, comprising:

[0015] A data acquisition module is used to obtain a black carbon emission inventory including black carbon emissions, and obtain multiple influencing factors that affect black carbon emissions, wherein the multiple influencing factors include nitrogen dioxide, ultraviolet aerosol index and wind; a model construction module is used to use an empirical orthogonal function model to perform spatiotemporal decomposition and dimensionality reduction on the black carbon emission inventory, and obtain multiple emission source areas including black carbon emission sources; within each emission source area, the linear and nonlinear interactions between multiple influencing factors are considered, and a nonlinear regression model between the black carbon emissions under the interaction of multiple influencing factors is established; a black carbon emission inversion module is used to input multiple influencing factors into a nonlinear regression model to obtain black carbon emissions, and complete the inversion of black carbon emissions.

[0016] An embodiment of the present invention provides a computer device, comprising: a memory and a processor; the memory stores a computer program, and the processor implements the above-mentioned method for inverting black carbon emissions when executing the computer program.

[0017] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned method for inverting black carbon emissions is implemented.

[0018] The embodiments of the present invention provide a method, system, device and medium for inversion of black carbon emissions. Compared with the prior art, the beneficial effects thereof are as follows:

[0019] The factors affecting black carbon emissions include nitrogen dioxide, ultraviolet aerosol index and wind. Then, considering the interaction between the influencing factors, a nonlinear regression model between black carbon emissions under the interaction of multiple influencing factors is established. This nonlinear regression model not only takes into account the influence of nitrogen dioxide, ultraviolet aerosol index and wind on black carbon emissions, but also takes into account the impact of multiple influencing factors on black carbon emissions when there is interaction. Finally, after inputting multiple influencing factors into the nonlinear regression model, accurate black carbon emissions can be inverted. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A technical flow chart provided for an embodiment of the present invention;

[0021] Figure 2 The PDF analysis diagram of each factor in the linear regression model of Region A provided by the embodiment of the present invention, wherein (a) is the analysis result of coefficient α1 in each emission source region, (b) is the analysis result of coefficient α2 in each emission source region, and (c) is the analysis result of coefficient α3 in each emission source region;

[0022] Figure 3 The PDF analysis diagram of each factor in the nonlinear regression model of Region A provided by the embodiment of the present invention, wherein (a) is the analysis result of coefficient β1 in each emission source area, (b) is the analysis result of coefficient β2 in each emission source area, (c) is the analysis result of coefficient β3 in each emission source area, (d) is the analysis result of coefficient β4 in each emission source area, and (e) is the analysis result of coefficient β5 in each emission source area;

[0023] Figure 4 The inverted global black carbon monthly mean time series provided by the embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0025] See also Figure 1 , an embodiment of the present invention provides a method for inverting black carbon emissions, comprising the following steps:

[0026] Step 1: Obtain a black carbon emission inventory including black carbon emissions, and obtain multiple influencing factors that affect black carbon emissions, including nitrogen dioxide, ultraviolet aerosol index and wind.

[0027] The data used in the present invention specifically include:

[0028] 1. Ozone monitor data

[0029] The present invention uses the Level-3 daily global gridded (0.25×0.25 degree) nitrogen dioxide product OMNO2d (OMI / TOMS Nitrogen Dioxide (NO2) Total and Tropospheric Column L3 Daily Best Pixel Data) from 2005 to 2018 in the official website of NASA (National Aeronautics and Space Administration) Goddard Earth Science Data and Information Service Center. The OMNO2d data product is a three-level grid product in which high-quality pixel-level data is averaged into a 0.25×0.25 degree grid. The product includes total column NO2 and total tropospheric column NO2, which are suitable for all atmospheric conditions and sky conditions with less than 30% cloud cover. Based on this, this article retains the daily tropospheric data of NO2 in the product, converts it into monthly data, and saves it for backup year by year.

[0030] The ultraviolet aerosol index can be used to indicate ultraviolet absorbing aerosols. Black carbon (BC) is the main light-absorbing substance in aerosols. There should be a positive correlation between black carbon emissions and changes in the absorption intensity monitored by satellites. The higher the concentration of absorbing aerosols, the larger the value of the ultraviolet aerosol index UVAI (Ultraviolet Aerosol Index). This article uses the near-ultraviolet aerosol optical thickness and single scattering albedo V3 product OMAERUVG data under the Level-2 global grid (0.25×0.25°) from 2005 to 2018 on the official website of NASA Goddard Earth Science Data and Information Service Center. The Level-2G daily global grid product OMAERUVG is based on the pixel-level OMI Level-2 Aerosol product OMAERUV. The OMAERUVG data product is a special Level-2 grid product in which the pixel-level product is divided into a grid of 0.25×0.25°. Contains all scene data with observation times between 00:00:00 and 23:59:59.999 UTC. OMAERUVG files are stored in the 5th version of the EOS hierarchical data format (HDF-OSE5). Each file contains daily data for about 15 orbits mapped on a global 0.25×0.25° grid. Based on this, the UVAI data is preprocessed, the daily irregular longitude and latitude grid is converted into the same regular longitude and latitude grid as the nitrogen dioxide data, and the monthly average data is calculated and saved for later use.

[0031] 2. EDGAR BC data

[0032] The present invention uses the EDGAR2002-2018 HTAP_v3 version of monthly black carbon emissions data (https: / / edgar.jrc.ec.europa.eu / dataset_htap_v3), with a resolution of 0.1×0.1°. The data includes emissions from international shipping, industrial waste gas, land transportation and other sectors, in units of ton / month. HTAP_v3 was compiled within the scope of the ECE Air Convention to improve scientific knowledge on the intercontinental transport of air pollution over the Northern Hemisphere. It consists of global monthly air pollutants (SO2, NO x , CO, NMVOCs, NH3, PM10, PM 2.5, BC and OC), covering the time series from 2000 to 2018 and emissions from all anthropogenic emission sectors except land use, land-use change and forestry. The emission grid map is collected from official reported data such as EMEP (CAMS-REG-v5.1) in Europe and the United States Environmental Protection Agency (US EPA). The EDGARv6.1 air pollutant emission grid map has covered all other countries in the world. In addition, compared with the previous HTAPv1 and HTAPv2.2, HTAP_v3 aims to expand the temporal coverage of air pollutant emissions, industry classification and geographical coverage of official data for analyzing development trends, studying the cross-border transport of air pollutants, and helping decision makers to address and reduce emissions from relevant departments. Based on this, the sum of emissions from each sector is calculated, and the resolution is converted to 0.25×0.25° using mathematical conversion principles.

[0033] 3. Reanalysis data (NCEP-NCAR Reanalysis 1)

[0034] The monthly average wind value data of NOAA Physical Science Laboratory (PSL) is selected. NOAA Physical Science Laboratory (PSL) is a laboratory under the National Oceanic and Atmospheric Administration (NOAA) of the United States, dedicated to researching and providing data and information related to climate, atmospheric science and geophysics. Among them, u-wind mean monthly 10m data refers to the east-west wind speed data measured at a height of 10 meters. These data record the average wind speed and wind direction information for each month. By analyzing these data, we can understand the changes in wind speed in different regions and time periods, so as to predict and study climate and weather.

[0035] Step 2: Use the empirical orthogonal function model to perform spatiotemporal decomposition and dimensionality reduction on the black carbon emission inventory to obtain multiple emission source areas including black carbon emission sources; in each emission source area, consider the linear and nonlinear interactions between multiple influencing factors, and establish a nonlinear regression model between black carbon emissions under the influence of multiple influencing factors.

[0036] Step 3: Input multiple influencing factors into the nonlinear regression model to obtain black carbon emissions and complete the inversion of black carbon emissions. In addition, the inverted black carbon emissions have the characteristics of long time series and gridding.

[0037] The specific analysis process of the above steps is as follows:

[0038] 1. Constructing a priori black carbon emission inventory

[0039] This paper is based on the location information of monthly black carbon emissions data of EDGAR HTAP_v3 version from 2002 to 2018, combined with the optimized black carbon emissions estimate of 17.8±5.6Tg / year obtained by Cohen (2014) using the Kalman filter method for the first time, and the scaling factor variation range is calculated to be 2.21-12.06. A new a priori emission black carbon emissions inventory is constructed, and this inventory is used as the basis for subsequent calculations.

[0040] 2. Identification and analysis of black carbon emission areas

[0041] The empirical orthogonal function analysis method (EOF) is also called principal component analysis (PCA) in mathematical statistics.

[0042] The BC data from 2005 to 2010 were analyzed by (empirical orthogonal function) EOF to identify and divide the emission source areas. According to the EOF analysis results, the contribution rates of the first five eigenvalues ​​are 0.5296, 0.1653, 0.1166, 0.0456 and 0.0232, respectively. The contribution rates of the last two are relatively small, so the first three are retained for result analysis. Note: EOF1 = 0.5296, EOF2 = 0.1653, EOF3 = 0.1166. The image uniformly removes the cluttered values ​​with EOF absolute values ​​less than 0.1.

[0043] In order to make the regional differences between high and low black carbon values ​​more intuitive, the values ​​of EOF1, EOF2, and EOF3 were further screened. The values ​​of EOF1 greater than 0.5 and less than -0.5 were screened, and the values ​​of EOF2 and EOF3 greater than 0.8 and less than -0.8 were screened.

[0044] Six emission source areas are delineated, among which: source area 1 is area A which is mainly composed of part of country A, source area 2 is area B which is mainly composed of country B, source area 3 is area C, source area 4 is area D which is mainly composed of countries D, E and F, source area 5 is area E which is mainly composed of country G, and source area 6 is area F which is mainly composed of country H.

[0045] Assume there are m related variables, and each variable has n samples, forming a matrix X m×n , by linearly transforming X, p variables can be linearly combined into a new variable:

[0046] Z p×n =A p×m X m×n

[0047] Among them, Z is the principal component of the original variable, and A is the linear transformation matrix. This process maximizes the concentration of most of the information of multiple variables on the principal components of a few independent variables.

[0048] Principal component analysis (PCA) in the climate variable field is to transform the variable X, which is composed of n observations at m spatial points, into m×n It can be regarded as a linear combination of p spatial eigenvectors and their corresponding time weight coefficients:

[0049] X m×n =V m×p T p×n

[0050] Among them, V is the space function matrix and T is the time function matrix. In this process, several typical eigenvectors concentrate the main information of the variable field. This means that we can use fewer space function modes and time function modes to approximate the original data and achieve the effect of dimensionality reduction.

[0051] 3. Correlation analysis between black carbon emissions and different factors

[0052] Black carbon and nitrogen dioxide are usually produced simultaneously during the combustion process. Black carbon mainly comes from solid particulate matter produced by incomplete combustion of organic matter, while nitrogen dioxide is a gaseous pollutant generated in high-temperature combustion reactions. Black carbon particles can serve as carriers of nitrogen dioxide. Nitrogen dioxide adsorbed on the surface of black carbon particles can be converted into nitrates in the atmosphere and then fall to the ground, which has a negative impact on the environment and human health. In addition, black carbon BC has a strong absorption effect on solar radiation in multiple bands, and the Ultraviolet Aerosol Index (UVAI) can be used to indicate the absorption of absorbing aerosols in the ultraviolet band. It is a remote sensing indicator for identifying absorbing aerosols in the atmosphere. The higher the value of the Ultraviolet Aerosol Index, the higher the concentration of absorbing aerosols in the atmosphere. Therefore, by monitoring the value of the Ultraviolet Aerosol Index, we can indirectly understand the content of absorbing aerosols in the atmosphere, including black carbon particles. At the same time, some meteorological elements, such as wind, also play an important role in the transmission and diffusion of black carbon in the atmosphere. Therefore, by constructing the multivariate linear regression relationship between black carbon and nitrogen dioxide, ultraviolet aerosol index, and wind, we can better understand the emission of air pollution sources and pollutants. By analyzing the relationship between them, we can determine the main pollution sources, emissions, and pollution source control strategies; we can also evaluate the contribution and impact of different factors on pollutant concentrations, and provide a scientific basis for the formulation of pollutant control strategies. And by adding nonlinear terms of the interaction between nitrogen dioxide and wind, and the interaction between ultraviolet aerosol index and wind, we can further verify the accuracy of the estimation and the impact of each factor on black carbon emissions.

[0053] During the analysis, a linear regression model and a nonlinear regression model were established and the results were compared to evaluate which model obtained more accurate results.

[0054] First, a linear regression model and a nonlinear regression model were established to explore the correlation between nitrogen dioxide, ultraviolet aerosol index, wind and black carbon in different regions and the degree of influence on black carbon. The linear regression model 1 is as follows:

[0055] BC=α1NO2+α2UVAI+α3Wind+C

[0056] In the linear regression model, both NO2 and UVAI are positively correlated with BC in Region A, while Wind is negatively correlated with BC; in Region B, NO2 is positively correlated with BC, and NO2 has a significant impact on BC in Region B, while UVAI and Wind have weaker impacts on BC, but are also positively correlated; in Region C, NO2 is negatively correlated with BC in the central and eastern regions, while UVAI is positively correlated with BC as a whole, and the impact of Wind is mainly in the eastern region of Country C, where it is positively correlated with BC; in Region D, except for a negative correlation in some areas of Country F, NO2 is positively correlated as a whole, while UVAI and Wind have both positively correlated and negatively correlated parts, and it can be seen that Wind has a greater impact on BC emissions in this region; in Country G, both NO2 and Wind are positively correlated with BC, and NO2 has a greater correlation with the entire BC emissions in Country G, while UVAI is mainly negatively correlated; in Region F of Country H, Wind is significantly positively correlated, and NO2 and UVAI are both positively and negatively correlated with BC, among which NO2 is mainly positively correlated and UVAI is mainly negatively correlated.

[0057] The use of three influencing factors in the linear regression model and their related physical meanings are explained as follows: nitrogen dioxide and black carbon have the same emission properties; the ultraviolet aerosol index represents the absorptivity of aerosols in the ultraviolet band, and black carbon has a strong absorption effect in the ultraviolet band; meteorological field data is used as an influencing factor due to the influence of surrounding black carbon transmission.

[0058] The nonlinear regression model 2 is established as follows:

[0059] BC=β1NO2+β2UVAI+β3Wind+β4NO2×Wind+β5UVAI×Wind+C

[0060] On the basis of linear regression 1, two nonlinear terms, NO2×Wind and UVAI×Wind, were added to introduce the interaction between independent variables. By introducing these nonlinear terms, the mutual influence and nonlinear relationship between NO2, UVAI and Wind on BC emissions were explored, and the impact mechanism of NO2, UVAI and Wind on BC was understood more comprehensively.

[0061] In the nonlinear regression model, in addition to the above three influencing factors, there are two nonlinear influencing factors, and their relevant physical meanings are explained as follows: NO2×Wind represents the impact of NO2 emission transmission on black carbon emissions, which mainly represents the aging effect of black carbon in the air; UVAI×Wind represents the impact of surrounding black carbon emission transmission on local black carbon emissions, which mainly represents the power transmission mechanism of black carbon emissions.

[0062] According to the nonlinear regression model, it can be analyzed that: in Region A, NO2 is mainly positively correlated with BC in the north, and mainly negatively correlated with BC in the south, UVAI is positively correlated with BC, Wind is positively correlated with BC in the northeast region except for H City and B City, UVAI×Wind is positively correlated with BC in general, and NO2×Wind is positively correlated with BC in general except for some areas of HL City and JL City. In Region B, NO2 is positively correlated with BC, and in this region, NO2 has the strongest impact on BC, Wind is positively correlated with BC, NO2×Wind has alternating positive and negative correlations with BC, and UVAI and UVAI×Wind have weaker impacts on BC in this region. In region C, NO2 is positively correlated with BC in the west, and negatively correlated in other regions; UVAI is negatively correlated with BC in the north, and positively correlated in other regions; Wind is staggered with BC; UVAI×Wind is negatively correlated with BC in the north, and positively correlated in other regions; NO2×Wind is negatively correlated with BC in the middle and west, and positively correlated in other regions; in region D, NO2 is negatively correlated with BC in the north and south, and positively correlated in the middle; UVAI is mainly negatively correlated; Wind is negatively correlated in the north, and positively correlated in the south; UVAI×Wind and NO2×Wind are mainly positively correlated. In country G, except for the southwest corner of country G, NO2 is positively correlated with BC as a whole, UVAI is mainly negatively correlated in the east, and positively correlated in the west, Wind is positively correlated with BC as a whole except for some areas in the southeast corner, NO2×Wind is negatively correlated in the middle, and positively correlated in all regions, while UVAI×Wind is positively correlated with BC in the east, and negatively correlated in the west. In the Ji region of country H, NO2, Wind and UVAI×Wind are mainly positively correlated, UVAI is mainly negatively correlated, and NO2×Wind is a mixture of positive and negative correlations.

[0063] 4. Impact of source region data on black carbon emissions

[0064] The influence of each data in the source area on black carbon emissions is studied from two aspects: the size and range of the coefficients corresponding to each factor and the change of the coefficients on the changes in the black carbon emissions results. Therefore, the probability density function (PDF) analysis is used to understand the distribution characteristics of each coefficient in the two linear regression models, such as the coefficient range, peak value, variance, median, etc., and the sensitivity analysis is used to evaluate the degree of correlation between the factors and black carbon, and determine which factors have an important influence on the black carbon emissions results and which factors have a small influence on the black carbon emissions results.

[0065] The probability density function PDF analysis results of the relative coefficients α1, α2, and α3 obtained from the linear regression model (1) in each source area are as follows: Figure 2 .

[0066] In Region A, the coefficient β1 data corresponding to NO2 showed a positively skewed distribution, with the 10% quantile and 90% quantile being -0.0816 and 0.3687 respectively. The average value of the coefficient α1 was 0.1164, and the median was 0.0874. In the range of -0.2 to +0.2, it accounted for 56.11% of the overall coefficient distribution, and its peak value was in the range of 0.04-0.08, accounting for 8.82%.

[0067] The coefficient α2 corresponding to UVAI is relatively concentrated, with the 10% and 90% quantiles being -0.0053 and 0.2197 respectively, the average coefficient being 0.1046, and the median being 0.0958. It accounts for 68.56% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0.04-0.08, accounting for 15.86%.

[0068] The coefficient α3 corresponding to Wind is more discrete than that of NO2 and UVAI. The 10% and 90% quantiles are -0.1252 and 0.5175 respectively. The average coefficient is 0.1255 and the median is 0.0473. It accounts for 51.77% of the overall coefficient distribution in the range of -0.2-+0.2. Its peak value is in the range of -0.04-0, accounting for 10.75%.

[0069] Depend on Figure 2 It can be seen that the proportions of NO2 and Wind corresponding coefficients α1 and α3 in the range of 0.4-0.8 are much higher than the UVAI corresponding coefficient α2. α1 and α3 account for 6.23% and 10.79% respectively, while α2 accounts for only 0.08%. The proportions of NO2 and Wind coefficients in the high value area are higher than those of UVAI coefficient. Therefore, in area A, the impact of NO2 and UVAI on BC emissions is higher than that of UVAI.

[0070] In area B, the coefficient β1 data corresponding to NO2 is normally distributed, with the 10% quantile and 90% quantile being -0.0685 and 0.0834 respectively. The average value of the coefficient α1 is 0.1175, and the median is 0.1063. It accounts for 71.07% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0.08-0.12, accounting for 13.15%.

[0071] The coefficient α2 corresponding to UVAI is relatively concentrated, with the 10% and 90% quantiles being -0.0342 and 0.2847 respectively, the average coefficient being 0.0105, and the median being 0.0153. It accounts for 92.87% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0-0.04, accounting for 27.04%.

[0072] The coefficient α3 corresponding to Wind is also distributed very concentratedly, with the 10% quantile and 90% quantile being -0.0780 and 0.0621 respectively. The mean and median are very small, at 0.0021 and 0.0072 respectively. It accounts for 93.28% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0-0.04, accounting for as much as 30.48%.

[0073] Among them, the coefficient α1 corresponding to NO2 has a wider data distribution range, and the proportion of high values ​​is also higher than the corresponding coefficients of UVAU and Wind. Therefore, in area B, the impact of NO2 on BC emissions is the greatest.

[0074] In area C, the coefficient α1 data corresponding to NO2 showed a negatively skewed distribution, with the 10% quantile and 90% quantile being -0.3051 and 0.2437 respectively. The average value of the coefficient α1 was -0.0205, and the median was -0.0155. It accounted for 45.24% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value was in the range of 0-0.04, accounting for 5.89%.

[0075] The coefficient α2 data corresponding to UVAI showed a positively skewed distribution, with the 10% quantile and 90% quantile being -0.0529 and 0.3158 respectively. The average coefficient α2 was 0.1072 and the median was 0.0828. It accounted for 49.28% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value was in the range of 0-0.04, accounting for 9.26%.

[0076] The coefficient α3 data corresponding to Wind presents a bimodal distribution. The first peak appears in the range of 0-0.04, accounting for 11.93%, and the second peak appears in the range of 0.2-0.24, accounting for 5.51%. The 10% quantile and 90% quantile are -0.0395 and 0.2660 respectively. The average value of coefficient α3 is 0.0948, and the median is 0.0709, accounting for 51.78% of the overall coefficient distribution in the range of -0.2-+0.2.

[0077] Among them, NO2 coefficient has the widest distribution range and more values ​​than 0, which has a negative impact on BC emissions. UVAI and Wind both have positive effects on BC emissions.

[0078] The PDF results of coefficients β1, β2, β3, β4 and β5 obtained by nonlinear regression formula 2 in each source area are as follows: Figure 3 shown.

[0079] In Region A, the coefficient β1 data corresponding to NO2 showed a positively skewed distribution, with the 10% quantile and 90% quantile being -0.3420 and 0.5767 respectively. The data distribution range was large, with the average value of the coefficient β1 being 0.0933 and the median being 0.0693. In the range of -0.2 to +0.2, it accounted for 41.81% of the overall coefficient distribution, and its peak value was in the range of 0-0.04, accounting for 5.91%.

[0080] The coefficient β2 data corresponding to UVAI showed a positively skewed distribution, with the 10% quantile and 90% quantile being 0.0242 and 0.3721 respectively, the coefficient average being 0.1760, and the median being 0.1477. It accounted for 51.57% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value was in the range of 0.08-0.12, accounting for 11.49%.

[0081] The coefficient β3 data corresponding to Wind shows a positively skewed distribution, with the 10% quantile and 90% quantile being -0.3398 and 0.9568 respectively, the average coefficient being 0.1731, and the median being 0.0228. It accounts for 41.37% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0-0.04, accounting for 9.72%.

[0082] The coefficient β4 data corresponding to UVAI×Wind shows a negatively skewed distribution, with the 10% and 90% quantiles being -0.2592 and 0.0508 respectively, the coefficient mean being -0.0716, and the median being -0.0314. It accounts for 65.70% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak is in the range of -0.04-0, accounting for 17.94%.

[0083] The coefficient β5 data corresponding to NO2×Wind showed a positively skewed distribution, with the 10% quantile and 90% quantile being -0.4142 and 0.4574 respectively. The average coefficient was 0.0253 and the median was 0.0083. It accounted for 46.77% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value was in the range of -0.04-0, accounting for 10.32%.

[0084] The data ranges of coefficients β1, β3, and β5 corresponding to NO2, Wind, and NO2×Wind are relatively wide, while the coefficient β2 corresponding to UVAI is mostly greater than 0, and the coefficient β4 corresponding to UVAI×Wind is mostly less than 0. Among them, nO2 and UVAI have the greatest impact on BC emissions in Region A.

[0085] In area B, the coefficient β1 data corresponding to NO2 is approximately normally distributed, with the 10% quantile and 90% quantile being -0.0749 and 0.3384 respectively. The data distribution range is large, with the average value of the coefficient β1 being 0.1229 and the median being 0.1145, accounting for 63.34% of the overall coefficient distribution in the range of -0.2-+0.2, and its peak value being in the range of 0.12-0.16, accounting for 10.04%.

[0086] The coefficient β2 data corresponding to UVAI shows a negatively skewed distribution, with the 10% and 90% quantiles being -0.0833 and 0.0837 respectively, the average coefficient being 0.0043, and the median being 0.0115. The data is in the range of -0.2 to +0.2, accounting for 91.70% of the overall coefficient distribution. The data is relatively concentrated, with a peak value in the range of 0-0.04, accounting for 26.05%.

[0087] The coefficient β3 data corresponding to Wind shows a positively skewed distribution, with the 10% quantile and 90% quantile being -0.1642 and 0.1548 respectively. The average coefficient is -0.0046 and the median is -0.0025. It accounts for 80.93% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of -0.04-0, accounting for 18.08%.

[0088] The coefficient β4 data corresponding to UVAI×Wind are highly concentrated, with the 10% and 90% quantiles being -0.0174 and 0.0408 respectively, the average coefficient being 0.0079, and the median being 0.0026. It accounts for 93.52% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0-0.04, accounting for 45.63%.

[0089] The coefficient β5 data corresponding to NO2×Wind shows a negatively skewed distribution, with the 10% quantile and 90% quantile being -0.1232 and 0.1106 respectively. The average coefficient is -0.0020 and the median is 0.0021. It accounts for 86.52% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0-0.04, accounting for 24.09%.

[0090] The coefficients β2 and β4 data corresponding to UVAI and UVAI×Wind are highly concentrated in the range of -0.2 to +0.2, and the coefficient β5 corresponding to NO2×Wind is also relatively concentrated. Combined with the numerical values ​​of the coefficients, it can be seen that Wind×UVAI and NO2×Wind have the smallest impact on BC emissions in area B, and NO2 has the greatest impact on BC emissions.

[0091] In area C, the coefficient β1 data distribution range corresponding to NO2 is relatively large, with certain values ​​in each interval between -0.8 and +0.8. The 10% quantile and 90% quantile are -0.7858 and 0.5706 respectively. The average value of coefficient β1 is -0.0903, and the median is -0.0731. In the range of -0.2 to +0.2, it accounts for 22.47% of the overall coefficient distribution. Its peak value is in the range of 0-0.04, accounting for 2.54%.

[0092] The coefficient β2 data corresponding to UVAI is irregularly distributed, with the 10% quantile and 90% quantile being -0.2630 and 0.4143 respectively, the average coefficient being 0.0791, and the median being 0.0727. It accounts for 35.73% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of -0.04-0, accounting for 4.93%.

[0093] The coefficient β3 data corresponding to Wind presents a flat-top distribution, with the 10% quantile and 90% quantile being -0.5582 and 0.5414 respectively. The average coefficient is 0.0034 and the median is 0.0145. It accounts for 27.41% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0-0.04, accounting for 3.35%.

[0094] The coefficient β4 data corresponding to UVAI×Wind is approximately normally distributed, with the 10% quantile and 90% quantile being -0.1678 and 0.2521 respectively, the average coefficient being 0.0276, and the median being 0.0101. It accounts for 53.56% of the overall coefficient distribution in the range of -0.2 to +0.2, and its peak value is in the range of 0-0.04, accounting for 9.32%.

[0095] The data distribution trend of coefficient β5 corresponding to NO2×Wind is similar to that of coefficient β3. The 10% quantile and 90% quantile are -0.4589 and 0.6392 respectively. The average coefficient is 0.0681 and the median is 0.0382. It accounts for 28.44% of the overall coefficient distribution in the range of -0.2 to +0.2. Its peak value is in the interval of 0-0.04, accounting for 3.73%. The proportion in the interval of -0.04-0 is almost the same as that in the interval of 0-0.04, which is 3.72%.

[0096] In Region C, except for the coefficient β4 corresponding to UVAI×Wind, which is approximately normally distributed, the distribution of other coefficients has obvious high values, but the distribution is scattered, and each interval has a large proportion, which is consistent with the situation reflected by the coefficient images under the linear regression model 2. From the coefficient image, it can be seen that the coefficients in Region C are interlaced with positive and negative values, the situation is more complicated, and there is no unified high-value area. The PDF analysis results are in line with this law. Among them, the coefficient β1 corresponding to NO2 is mostly negative, so NO2 has a negative effect on BC emissions in Region C. On the positive side, UVAI and Wind have a greater impact on BC emissions.

[0097] An embodiment of the present invention provides a black carbon emission inversion system, comprising:

[0098] The data acquisition module is used to obtain a black carbon emission inventory including black carbon emissions, and obtain multiple influencing factors affecting black carbon emissions, wherein the multiple influencing factors include nitrogen dioxide, ultraviolet aerosol index and wind.

[0099] The model building module is used to perform spatiotemporal decomposition and dimensionality reduction on the black carbon emission inventory using an empirical orthogonal function model to obtain multiple emission source areas including black carbon emission sources; within each emission source area, the linear and nonlinear interactions between multiple influencing factors are considered, and a nonlinear regression model between the black carbon emissions under the interaction of multiple influencing factors is established.

[0100] The black carbon emission inversion module is used to input multiple influencing factors into the nonlinear regression model to obtain black carbon emissions and complete the inversion of black carbon emissions.

[0101] An embodiment of the present invention provides a computer device, comprising: a memory and a processor; the memory stores a computer program, and the processor implements the steps of a method for inverting black carbon emissions when executing the computer program.

[0102] An embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of a method for inverting black carbon emissions are implemented.

[0103] A specific embodiment is as follows:

[0104] Analysis of spatiotemporal distribution and transport characteristics of black carbon inversion emission inventory.

[0105] Based on the linear regression model, the black carbon emission inventory from 2011 to 2018 was inverted, and the spatiotemporal characteristics of the black carbon inversion emission inventory were analyzed. Two high-value emission source areas were selected: Area A and Area B. The global and these two high-value emission source areas black carbon inversion emission inventories were calculated from the three aspects of annual mean, seasonal mean, and monthly mean, and their spatiotemporal similarities and differences were analyzed and compared. Taking advantage of the similarity and homology of black carbon and carbon monoxide (CO), the MOPITT CO data and black carbon data were used to calculate the overlapping areas, and the spatiotemporal transmission of black carbon and CO was compared to verify the regional accuracy of the black carbon emission inventory. At the same time, the transmission of CO in different atmospheric layers is used to indicate the transmission of black carbon emissions. The main conclusions are as follows:

[0106] After calculating the black carbon emission inversion inventory for the world and regions A and B, it was found that:

[0107] 1. Global situation. Global black carbon emissions generally experienced a process of first growth, then decline, and then rise between 2011 and 2018. Specifically, there was a slight increase between 2011 and 2013, a decrease between 2014 and 2015, and a slow increase between 2016 and 2018. High emission areas are mainly concentrated in Region A in the east of Country A, Region B in Country B, Region C in Country Z, Country X, and Country V, Country N, Country G, Country H, Country F, etc. Although the minimum black carbon emissions in each year vary in months, they are usually concentrated between June and August. This may be because most of the major emission source areas in the northern hemisphere are in the summer rainfall season or wind direction change season, resulting in a reduction in emission sources, especially reduced winter heating emissions. In contrast, global black carbon emissions reach their maximum from November to February, which is the winter in the northern hemisphere, and increased combustion heating leads to an increase in black carbon emissions. In addition, this period is also a period of frequent celebrations in many countries, so black carbon emissions will appear in a short-term cluster.

[0108] 2. Northern, eastern and central regions of Country A. From 2011 to 2013, black carbon emissions in the northern, eastern and central regions of Country A gradually increased, and then showed a downward trend from 2013 to 2015. Emissions in 2016 were basically the same as in 2015, and from 2016 to 2018, black carbon emissions in the region rose slowly. Overall, the trend of the annual average time series in the region is similar to the global annual average time series, and the trend of black carbon emissions is also relatively consistent. Black carbon emissions in the northern, eastern and central regions of Country A show obvious seasonal changes. The high-value emission area is smaller in summer and the largest in winter. Its seasonal average emission time series is highly consistent with the global seasonal average emission time series trend. The monthly average emissions show obvious changes. From February to August, the scope of the high-value emission area gradually shrinks, especially in many eastern provinces and the eastern region. The changes are more intuitive. In addition, the black carbon emissions in Region A continued to decline from January to June of each year, reaching the lowest value in June, then fluctuated slightly from June to September, and continued to increase from September to December.

[0109] 3. Region B. The high-value emission areas of Country B in Region B are mainly concentrated near the Gangetic Plain in the north. This region is one of the most densely populated areas in the world and one of the important agricultural, political, economic and cultural centers of Country B. From 2015 to 2017, black carbon emissions in the region increased significantly. According to the annual average black carbon emission data and the corresponding time series of Region B, it was found that black carbon emissions increased slightly from 2011 to 2012, slowly decreased from 2012 to 2015, and rose sharply and then fell from 2015 to 2017. The seasonal average time series of black carbon emissions in Region B is also highly consistent with the trend of the global seasonal average time series. Emissions decrease from spring to summer and begin to rise after summer. The monthly average data of black carbon emissions in Region B are more complex. The black carbon emission values ​​are the highest in December and March. The range of the high-value area of ​​black carbon emissions in Region B is greatly reduced in May. Emissions increase from June to July and decrease again from August to September.

[0110] The main research contents of this paper include the correlation analysis between black carbon and nitrogen dioxide, ultraviolet aerosol index and wind, the analysis of the impact of various factors on black carbon, and the analysis of the temporal and spatial distribution of the black carbon inversion emission inventory. The research in this paper helps to establish an accurate black carbon emission inventory, better understand the source, distribution and impact of black carbon, provide a scientific basis for reducing black carbon emissions, and facilitate the subsequent global black carbon emission reduction work. The main conclusions are as follows:

[0111] 1. The scaling factor between the constructed prior emission inventory and the EDGAR emission inventory varies from 2.21 to 12.06. High-value emission areas include Region A, Region B, and Region C, and the high-value years are 2010 to 2013. Black carbon emissions reached an inflection point in 2010, increasing year by year from 2005 to 2010, and decreasing year by year from 2011 to 2018.

[0112] 2. There are differences in the correlation between each factor and black carbon, the distribution of coefficients, and the dominant influencing factors in different regions. In Region A and Region B, nitrogen dioxide has the greatest impact on black carbon emissions, and nitrogen dioxide and the ultraviolet aerosol index are positively correlated with black carbon; in Region C, nitrogen dioxide is still the main influencing factor, but the correlation is more complicated. Nitrogen dioxide is negatively correlated with black carbon in some areas, and the ultraviolet aerosol index is positively correlated with black carbon overall. In Region C, but in Region D, the impact of nitrogen dioxide on black carbon emissions is higher than that of wind and ultraviolet aerosol index. In Region F of Country H, wind has the greatest impact on black carbon emissions. Globally, nitrogen dioxide has the greatest impact on black carbon emissions, followed by the ultraviolet aerosol index, and wind has the least impact. There are significant differences in the impact of meteorological factors and pollutants on black carbon emissions in different regions, showing obvious regional characteristics.

[0113] 3. The interannual and seasonal variations of the black carbon inversion emission inventory are obvious. Globally, from 2011 to 2018, black carbon emissions generally showed a trend of first increasing, then decreasing, and then increasing. Black carbon emissions reach their minimum from June to August and their maximum from November to February, which is related to the summer rainy season or increased combustion heating in the northern hemisphere winter. In the northern, eastern and central regions of country A, black carbon emissions gradually increased from 2011 to 2013, then decreased from 2013 to 2015, and slowly increased from 2016 to 2018. Black carbon emissions have obvious seasonal variations, with smaller black carbon emissions in summer and the largest in winter. In region B, the seasonal variation of black carbon emissions is consistent with the global trend, decreasing from spring to summer and increasing after summer. In addition, black carbon emissions in region B are mainly concentrated near the Gangetic Plain in the north of country B, and increased significantly from 2015 to 2017. Figure 4 Shown are the research results of the present invention.

[0114] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for inverting black carbon emissions, characterized in that: The following steps are involved: Obtaining a black carbon emission inventory including black carbon emissions, and obtaining a plurality of influencing factors affecting the black carbon emissions, wherein the plurality of influencing factors include nitrogen dioxide, ultraviolet aerosol index, and wind; The black carbon emission inventory is decomposed and reduced in time and space using an empirical orthogonal function model to obtain multiple emission source areas including black carbon emission sources; within each emission source area, the linear and nonlinear interactions between multiple influencing factors are considered, and a nonlinear regression model between the black carbon emissions under the interaction of multiple influencing factors is established; Multiple influencing factors are input into the nonlinear regression model to obtain black carbon emissions and complete the inversion of black carbon emissions.

2. The method for inverting black carbon emissions according to claim 1, characterized in that: The obtaining of multiple emission source areas including black carbon emission sources specifically comprises the following steps: The empirical orthogonal function model is used to obtain multiple eigenvalue contribution rates for the black carbon emission inventory, and the top three largest eigenvalue contribution rates are selected from all the eigenvalue contribution rates. The top three largest eigenvalue contribution rates are marked as EOF1, EOF2 and EOF3 from large to small. The values ​​in the eigenvector corresponding to the eigenvalue contribution rate EOF1 that are greater than 0.5 and less than -0.5 are screened; the values ​​in the eigenvector corresponding to the eigenvalue contribution rates EOF2 and EOF3 that are greater than 0.8 and less than -0.8 are screened; According to the numerical differences of the screened results, the geographical locations of the preliminary emission source areas are obtained from the black carbon emission inventory; The areas with preliminary emission source areas less than 100 pixels were excluded, and the edge detection algorithm was used to extract the regional boundaries of the excluded preliminary emission source areas, dividing them into six emission source areas.

3. The method for inverting black carbon emissions according to claim 1, characterized in that: The specific steps of establishing a nonlinear regression model between the black carbon emissions under the interaction of multiple influencing factors include: Multiple influencing factors that reflect black carbon emissions include nitrogen dioxide (NO2), ultraviolet aerosol index (UVAI) and wind (Wind). The linear interaction between the influencing factors and black carbon emissions needs to be considered, and the nonlinear interaction between the influencing factors needs to be considered, including the interaction between the ultraviolet aerosol index (UVAI) and wind (Wind), and the interaction between nitrogen dioxide (NO2) and wind (Wind). The established nonlinear regression model formula is: BC=β1NO2+β2UVAI+β3Wind+β4NO2×Wind+β5UVAI×Wind+C Among them, β1, β2, β3, β4 and β5 are the coefficients of nitrogen dioxide NO2, ultraviolet aerosol index UVAI, wind Wind, the interaction between wind Wind and ultraviolet aerosol index UVAI, and the interaction between wind Wind and nitrogen dioxide NO2, respectively, and C is a constant.

4. The method for inverting black carbon emissions according to claim 1, characterized in that: The step of inputting multiple influencing factors into a nonlinear regression model to obtain black carbon emissions specifically includes: Obtain multiple months of monitoring data on nitrogen dioxide (NO2), ultraviolet aerosol index (UVAI) and wind (Wind), and input the multiple months of monitoring data into a linear regression model to obtain the black carbon emissions for each month; Based on the obtained black carbon emissions, the annual average, seasonal average and monthly average black carbon emissions are analyzed, and the differences are compared with the black carbon emission inventory to determine the accuracy of the inversion results.

5. The method for inverting black carbon emissions according to claim 1, characterized in that: The specific steps of obtaining the black carbon emission inventory include: Based on the historical black carbon emissions at the global scale, the Kalman filter method is used to obtain the optimized estimate of black carbon emissions, and based on the difference between the optimized estimate of black carbon emissions and the historical black carbon emissions, a scaling factor is obtained to make the black carbon emissions conform to the actual situation; The monthly black carbon emission data of EDGAR, the global emission database for atmospheric research, was scaled to obtain the black carbon emission inventory.

6. A black carbon emission inversion system, characterized in that: include: A data acquisition module, used to acquire a black carbon emission inventory including black carbon emissions, and acquire multiple influencing factors affecting black carbon emissions, wherein the multiple influencing factors include nitrogen dioxide, ultraviolet aerosol index and wind; A model building module is used to perform spatiotemporal decomposition and dimensionality reduction on the black carbon emission inventory using an empirical orthogonal function model to obtain multiple emission source areas including black carbon emission sources; In each emission source area, the linear and nonlinear interactions among multiple influencing factors are considered, and a nonlinear regression model between the interaction of multiple influencing factors and black carbon emissions is established; The black carbon emission inversion module is used to input multiple influencing factors into the nonlinear regression model to obtain black carbon emissions and complete the inversion of black carbon emissions.

7. A computer device comprising: Memory and processor; The memory stores a computer program, wherein the processor implements a black carbon emission inversion method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a method for inverting black carbon emissions as claimed in any one of claims 1 to 5 is implemented.

Images