A method for constructing a localized high-precision bvocs emission inventory

Abstract

This invention belongs to the interdisciplinary field of atmospheric environment simulation and ecological remote sensing, and relates to a method for constructing a localized high-precision BVOCs emission inventory. First, high-resolution land cover remote sensing data and ground observation data are collected to construct a high-precision vegetation distribution database for China, including urban green spaces, obtaining vegetation growth type, canopy type, and leaf area index. Next, detailed localized plant species composition in China is obtained. Then, emission factor data for each plant species are matched based on a localized BVOCs emission factor database. Finally, the vegetation growth type, canopy type, leaf area index, emission factor data, and meteorological data are input into the MEGANv3.2 natural gas and aerosol emission model to simulate and obtain a high-quality BVOCs emission inventory. This reduces the significant uncertainty in vegetation cover data caused by the use of single satellite remote sensing data and improves the accuracy of the BVOCs emission inventory.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of atmospheric environment simulation and ecological remote sensing, and relates to a localized high-precision BVOCs emission inventory construction method. Background Technology

[0002] Currently, numerous scholars have conducted research on China's BVOCs emission inventory and achieved certain results. Based on published national-level BVOCs emission inventory studies, the annual BVOCs emissions nationwide range from 10.00 to 58.89 Tg yr. -1 Significant differences exist in BVOCs emission estimates across studies, with discrepancies exceeding five times, indicating a degree of uncertainty in the inventory. Besides differences in simulation years and emission algorithms, these discrepancies are primarily due to variations in the vegetation cover data and emission factor data used, as well as their sources.

[0003] Regarding vegetation cover data, previous studies have mostly relied on single land cover remote sensing data for calculating BVOC emissions. These data typically have a spatial resolution of only a few hundred meters, making it difficult to identify detailed features and leading to significant uncertainty in the estimation results. For example, in some megacities (such as Beijing and Shanghai), BVOC emissions from urban green spaces exceed those from non-urban vegetation. Therefore, constructing a high-precision BVOC emission inventory for my country based on high-precision vegetation distribution data that can identify urban green spaces is of significant synergistic importance for refined urban BVOC emission reduction and precise control of atmospheric compound pollution. Furthermore, because BVOC emissions are highly species-specific, with significant differences in emission levels among different species, accurate and detailed plant species composition distribution is crucial. However, previous studies often categorized plant species based on vegetation function type (PFT), lacking detailed plant species composition and resulting in high uncertainty in the emission inventory.

[0004] Regarding emission factor data, most existing studies use the default global average emission factor library in BVOCs emission estimation models, which is largely based on foreign BVOCs emission observations and cannot accurately characterize the BVOCs emission levels of native Chinese plants. Some studies update the emission factors of plant species using native BVOCs emission observation data, but these are mostly based on static closed sampling techniques, resulting in high uncertainty in the observation results. Furthermore, some studies update emission factors based on PFT rather than plant species, ignoring species-specific emissions. Therefore, the use of emission factors from native Chinese plant species is of great significance for improving the accuracy of inventory.

[0005] The paper “Ma et al., Environ. Sci. Technol. 2022, 56, 175-184” discloses a method for constructing a regional urban green space BVOCs (U-BVOCs) emission inventory in China. It uses land cover data from FROM-GLC10 (10 m resolution) and MODIS MCD12Q1 (500 m resolution) to classify all vegetation types in China into 16 plant functional types (PFTs). It directly uses global average emission factor data based on these 16 PFTs provided by Guenther et al., without any correction based on localized field observations in China. This method provides the first estimate of the total U-BVOCs in China as 28.91 ± 0.89 Gg yr⁻¹, accounting for 0.1% of the total BVOCs emissions in China. However, it has the following limitations: First, for a specific region, the vegetation growth type distribution data only comes from a single land cover dataset (such as FROM-GLC10 or MODIS). The MCD12Q1 data collection method suffers from several drawbacks. First, it categorizes all plant species in China into PFT (Plant Functional Factor) levels and uses leaf area index (LAI) and other vegetation cover data based on PFT, without specifying the exact plant species, leading to some discrepancies with actual plant emissions. Second, the emission factor data used in the inventory is entirely based on PFT rather than plant species, failing to consider the species specificity of plant BVOCs emissions. Furthermore, the emission factor data directly uses global average emission factor data without correction from localized field observations in China, making it difficult to characterize the emission levels of native Chinese plant species.

[0006] In summary, my country has made some progress in the methodology of compiling BVOCs emission inventories, but more accurate localized basic data is still needed to further improve the accuracy of the inventories. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing technologies and to design a localized, high-precision BVOCs emission inventory construction method. This method is based on higher-precision vegetation cover data and localized emission factor data for different plant species, and utilizes a reliable emission model to improve the accuracy of the BVOCs emission inventory.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: A localized, high-precision BVOCs emission inventory construction method specifically includes the following steps: S1. Collect high-resolution land cover remote sensing data and ground observation data to construct a high-precision vegetation distribution database of China that includes urban green spaces, obtain vegetation growth types, and the spatial resolution of the database is 1km×1km. S2. Based on the high-precision vegetation distribution database of China constructed in step S1, obtain the canopy type and leaf area index; S3. Based on the "Vegetation Map of China (1:1000000)", obtain a detailed composition of native plant species in China; S4. Based on the localized BVOCs emission factor database in China, match the emission factor data of each plant species, select the dynamic observation results in the emission factor database, and for species whose emission factors cannot be matched in the localized emission factor database, use the global emission factor database in the MEGANv3.2 emission factor processing module to obtain emission factor data. S5. Use Weather Research and Forecasting (WRF) models to simulate meteorological data and obtain the required meteorological data; S6. Input vegetation growth type, canopy type, leaf area index, emission factor data and meteorological data into the natural gas and aerosol emission model MEGANv3.2 to simulate and obtain a high-quality BVOCs emission inventory.

[0009] As a further technical solution of the present invention, the high-resolution land cover remote sensing data and ground observation data mentioned in step S1 are derived from a basic land cover dataset. This basic land cover dataset includes the China Multi-Period Land Use Remote Sensing Monitoring Dataset (CNLUCC Land Use Data, https: / / www.resdc.cn / DOI / doi.aspx?DOIid=54), MODISMCD12Q1 land cover data (https: / / lpdaac.usgs.gov / products / mcd12q1v006 / ), the China Vegetation Map (1:1,000,000), and the European Space Agency WorldCover land cover data (https: / / viewer.esa-worldcover.org / ). Using ArcGIS software, land cover / use types in various land cover data were resampled into four vegetation growth types: trees, shrubs, herbs, and crops. Based on the four sets of reclassified vegetation growth type data, vegetation growth types in urban and non-urban areas were processed separately to construct a high-precision Chinese vegetation distribution database that can identify urban green spaces. The specific process is as follows: urban land, rural settlements, and other construction land in CNLUCC land use data are classified as urban areas. The vegetation in this area is redefined using European Space Agency WorldCover land cover data with a spatial resolution of 10m to identify the vegetation growth types in urban areas. For non-urban areas, CNLUCC land use data is used as the benchmark, and MODIS MCD12Q1 land cover data is introduced for comparative analysis. For raster with inconsistent vegetation growth types in CNLUCC and MODIS MCD12Q1 data, the vegetation growth types are further corrected and redefined using the China Vegetation Map (1:1000000).

[0010] As a further technical solution of the present invention, the canopy types mentioned in step S2 include coniferous trees, broad-leaved trees, tropical trees, shrubs, herbs, and crops. Based on the constructed high-precision Chinese vegetation distribution database, the data of the six canopy types are updated to correspond to the growth types. Among them, the grid canopy type coverage ratio of shrubs, herbs, and crops is consistent with the vegetation growth type data. The grid coverage ratio of tropical trees is determined according to the data from the MEGAN official website (http: / / bai.ess.uci.edu / megan). Trees in the vegetation growth types are divided into coniferous trees and broad-leaved trees. Using MODIS MCD12Q1 land cover data and the Chinese vegetation map (1:1000000), trees in the vegetation growth types are divided into coniferous trees and broad-leaved trees. That is, ArcGIS software is used to reclassify the two datasets. For MODIS Based on the MCD12Q1 land cover data, evergreen coniferous forest, deciduous coniferous forest, 50% mixed forest, and 70% woody savanna in this layer are reclassified as coniferous forest; evergreen broad-leaved forest, deciduous broad-leaved forest, 50% mixed forest, and 70% woody savanna are reclassified as broad-leaved forest. For the China Vegetation Map (1:1,000,000), coniferous forest and 50% mixed coniferous and broad-leaved forest in this layer are reclassified as coniferous forest, and broad-leaved forest and 50% mixed coniferous and broad-leaved forest are reclassified as broad-leaved forest. Based on this method, the consistency between the canopy type distribution data and the vegetation growth type distribution data in the emission inventory can be guaranteed, and the estimation error caused by the difference between vegetation input data can be reduced.

[0011] As a further technical solution of the present invention, the leaf area index (LAI) in step S2 is quantified using the index LAIv. Specifically, the original LAI data is updated based on the constructed high-precision Chinese vegetation distribution database. When the vegetation coverage rate (VCF) of a vegetation growth type is detected in a certain grid, it indicates that there is vegetation distribution in that grid. When the LAI value is zero, the LAI of that grid is updated. The empirical value of PFT-LAI (plant functional type-leaf area index empirical value) with a time resolution of 1 month is used for correction, and the monthly data is interpolated to an 8-day resolution. The weighted average value of LAI is calculated based on the coverage ratio of each PFT in the grid and used as the average LAI of the grid. The VCF data comes from the sum of the coverage rates of the four vegetation growth types in the high-precision Chinese vegetation distribution database. The two are divided to generate LAIv data with an 8-day resolution. The maximum value of LAIv is set to 6 to eliminate the special case of sparse vegetation in the grid but estimated as high value.

[0012] As a further technical solution of the present invention, the plant species mentioned in step S3 include 166 native tree species, 149 shrub species, 148 herbaceous species and 31 crop species native to my country.

[0013] As a further technical solution of the present invention, the Chinese localized BVOCs emission factor database described in step S4 is established by combining actual field observations and literature surveys. Specifically, it first establishes a method for classifying and determining plant emission intensity based on statistics, and then determines a refined emission intensity classification interval and a more accurate emission factor characteristic value, thereby establishing a detailed and reliable Chinese localized emission factor database by plant and BVOCs group. The database updates the isoprene, monoterpene and sesquiterpene emission factor data of 202 local plant species. The vegetation species that update the localized emission factors account for about 40% of the total number of vegetation species.

[0014] As a further technical solution of the present invention, the specific process of step S6 is as follows: S61. The preprocessing module of the natural gas and aerosol emission model MEGANv3.2 standardizes the basic geographic information of the study area. The land use characteristics and meteorological parameters of the study area are uniformly converted to the preset spatiotemporal resolution grid system through spatial interpolation, thereby obtaining the vegetation distribution dataset required by the model. S62, the emission factor processing module matches vegetation cover data (including vegetation growth type, canopy type and plant species composition) with emission factors of various plant species to establish emission factors and light-dependent factors based on grid cells. S63. The emission calculation module estimates emission flux through multi-source data coupling. It combines gridded emission factor and light-dependent factor data with vegetation leaf area index (LAIv) and canopy type data to parameterize the canopy environment and soil-vegetation interface processes respectively. Through dynamic correction algorithms, it accurately quantifies the emission of BVOCs.

[0015] Compared with the prior art, the present invention has the following advantages: (1) The high-precision Chinese vegetation distribution database containing urban green space constructed by this invention integrates four sets of high-resolution land cover remote sensing data and ground observation data, which can reduce the large uncertainty of vegetation cover data caused by the use of a single satellite remote sensing data. (2) Based on the "Vegetation Map of China (1:1,000,000)," this invention obtains a detailed composition of native tree, shrub, herb, and crop species in China. A total of 166 native tree species, 149 shrub species, 148 herb species, and 31 crop species in my country have been updated. The "Vegetation Map of China (1:1,000,000)" is currently the most detailed reference data on the spatial distribution of vegetation in my country. It integrates field surveys, remote sensing data, climate, soil, and geological studies, and accurately records the spatial distribution data of more than 2,000 dominant plant populations and major cultivated crops at the species level. Since plant BVOC emissions are species-specific, a detailed and accurate composition of native plant species is of great significance for improving the accuracy of emission inventories. (3) This invention utilizes the emission factor data of various plant species to match the Chinese localized BVOCs emission factor library established by the team through comprehensive field observation and literature research. Combined with high-precision vegetation growth type and localized species composition data, it is input into the emission factor processing module of MEGANv3.2 to obtain the gridded standard canopy emission factor in the simulation domain, so that the BVOCs emission inventory can be made as accurate and localized as possible. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the workflow of the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the embodiments and accompanying drawings.

[0018] Example 1: As Figure 1 As shown, this embodiment provides a localized, high-precision BVOCs emission inventory construction method, which specifically includes the following steps: S1. Collect high-resolution land cover remote sensing data and ground observation data to construct a high-precision vegetation distribution database for China, including urban green spaces, and obtain vegetation growth types. The spatial resolution of the database is 1km × 1km. The high-resolution land cover remote sensing data and ground observation data are derived from basic land cover datasets, which include the China Multi-Period Land Use Remote Sensing Monitoring Dataset (CNLUCC Land Use Data, https: / / www.resdc.cn / DOI / doi.aspx?DOIid=54), MODIS MCD12Q1 land cover data (https: / / lpdaac.usgs.gov / products / mcd12q1v006 / ), China Vegetation Map (1:1000000), and European Space Agency WorldCover land cover data (https: / / viewer.esa-worldcover.org / worldcover / ). In this implementation, CNLUCC land use data and MODIS... The MCD12Q1 land cover data are global land cover data interpreted from Landsat remote sensing imagery and Terra and Aqua satellite remote sensing data, with spatial resolutions of 1000m and 500m, respectively. The China Vegetation Map (1:1,000,000) is a detailed vegetation distribution data of my country generated by combining remote sensing imagery and field surveys. The European Space Agency's WorldCover product provides a 2020 global land cover map based on Sentinel-1 and Sentinel-2 Earth observation satellite data, with a resolution of 1000m. The high-resolution features of this data can accurately identify the distribution of vegetation growth types in urban areas. Using ArcGIS software, the land cover / use types in each land cover data set were resampled into four vegetation growth types: trees, shrubs, herbs, and crops. Based on these four reclassified sets of vegetation growth type data, the vegetation growth types in urban and non-urban areas were processed separately to construct a high-precision Chinese vegetation distribution database capable of identifying urban green spaces. Specifically, urban land, rural settlements, and other construction land in the CNLUCC land use data were classified as urban areas. The vegetation in this region was redefined using the European Space Agency's WorldCover land cover data with a spatial resolution of 10m to identify the vegetation growth types in urban areas. For non-urban areas, the CNLUCC land use data was used as a benchmark, and MODIS MCD12Q1 land cover data was introduced for comparative analysis. For raster data where the vegetation growth types were inconsistent between the CNLUCC and MODIS MCD12Q1 data, further corrections were made using the China Vegetation Map (1:1000000) and the vegetation growth types were redefined.

[0019] S2. Based on the high-precision vegetation distribution database of China constructed in step S1, obtain the canopy type and leaf area index; The canopy types include conifers, broad-leaved trees, tropical trees, shrubs, herbs, and crops. Canopy type data affects the determination of the vegetation canopy environment, thus influencing BVOCs emission simulation. Currently, most methods use default data from the MEGAN emission model or are based on single remote sensing monitoring data, which can lead to the canopy type distribution differing from the growth type distribution, resulting in uncertainty in emission estimation. Therefore, this embodiment updates the data of six canopy types based on a constructed high-precision Chinese vegetation distribution database to correspond them with growth types. Specifically, the grid canopy type coverage ratios for shrubs, herbs, and crops are consistent with the vegetation growth type data. The grid coverage ratio for tropical trees is determined based on data from the MEGAN website (http: / / bai.ess.uci.edu / megan). Trees in the vegetation growth types are divided into conifers and broad-leaved trees, and MODIS is used. The MCD12Q1 land cover data and the China Vegetation Map (1:1,000,000) classify trees in vegetation growth types into coniferous and broad-leaved trees. This involves reclassifying the two datasets using ArcGIS software. For the MCD12Q1 land cover data, evergreen coniferous forest, deciduous coniferous forest, 50% mixed forest, and 70% woody savanna are reclassified as coniferous forests; evergreen broad-leaved forest, deciduous broad-leaved forest, 50% mixed forest, and 70% woody savanna are reclassified as broad-leaved forests. For the China Vegetation Map (1:1,000,000), coniferous forest and 50% mixed coniferous-broadleaved forest are reclassified as coniferous forests, and broad-leaved forest and 50% mixed coniferous-broadleaved forest are reclassified as broad-leaved forests. This method ensures consistency between canopy type distribution data and vegetation growth type distribution data in the emission inventory, reducing estimation errors caused by differences in vegetation input data. The Leaf Area Index (LAI) is the ratio of the total leaf area of ​​plants per unit land area to the land area, and is an important indicator for measuring plant growth status. This embodiment uses the LAIv index to quantify the impact of LAI. The raw LAI data used is from the global 0.1° reanalysis data developed by Yuan Hua's team at Sun Yat-sen University, updated every 8 days. The data is then updated based on a constructed high-precision Chinese vegetation distribution database. The update method is based on the empirical values ​​of plant functional types and leaf area index (PFT-LAI) from Zhang et al. When the vegetation cover rate (VCF) in a vegetation growth type is detected in a certain grid, it indicates that vegetation distribution exists in that grid. When the LAI value is zero, the LAI of the grid is updated. The empirical value of PFT-LAI (plant functional type-leaf area index empirical value) with a time resolution of 1 month is used for correction, and the monthly data is interpolated to an 8-day resolution. The weighted average of LAI is calculated in the grid based on the coverage ratio of each PFT, and used as the average LAI of the grid. The VCF data comes from the sum of the grid coverage of four vegetation growth types in the China High-Precision Vegetation Distribution Database. The two are divided to generate LAIv data with an 8-day resolution. The maximum value of LAIv is set to 6 to eliminate the special case of sparse vegetation in the grid but estimated as high value.

[0020] S3. Based on the "Vegetation Map of China (1:1,000,000)," this study obtains a detailed composition of native plant species in China. The emission characteristics of different plant species vary significantly. To improve the accuracy of emission inventories, obtaining accurate native plant species composition data is crucial. Most existing research uses global average emission factor libraries for simulation, and these emission factors are mostly based on PFTs rather than plant species, greatly neglecting the specific emissions between vegetation species. Therefore, obtaining a detailed and accurate composition of native vegetation species in my country is of great significance for improving the accuracy of simulation results. The "Vegetation Map of China (1:1,000,000)" is currently the most detailed reference for the spatial distribution of vegetation in my country, integrating field surveys, remote sensing data, climate, soil, and geological research. It accurately records the spatial distribution data of over 2,000 dominant plant populations and major cultivated crops at the species level. This embodiment obtains a detailed composition of native trees, shrubs, herbs, and crops in China based on the "Vegetation Map of China (1:1,000,000)." A total of 166 native tree species, 149 shrub species, 148 herbaceous species, and 31 crop species were updated.

[0021] S4. Emission factors are important parameters for characterizing the levels of different compounds released by plants and are also a significant source of uncertainty in emission inventories. Currently, most existing emission inventories use emission factor libraries based on earlier emission factor findings, standard limits, or global average emission factor libraries. However, these emission factors cannot accurately describe the emission levels of native plant species in my country. To improve the accuracy of emission inventories, this embodiment applies a comprehensive approach, combining field observations and literature reviews to establish a Chinese native BVOCs emission factor library, matching emission factor data for each plant species. The team selected relatively accurate dynamic observation results from this library and updated the isoprene, monoterpenes, and sesquiterpenes emission factor data for 202 native plant species. The vegetation species whose emission factors were updated represent approximately 40% of the total vegetation species. (See Han H, Jia Y, Shi R, Nie C, Kajii Y, Wu Y, Li L. 2026. A localized plantspecies-specific BVOC emission rate library of China established using a developed statistical approach based on field measurements. Atmos. Chem. Phys.,) 26(2): 1587-1604; For species whose emission factors could not be matched in the localized emission factor library, emission factor data of these plant species were obtained from the global emission factor library in the MEGANv3.2 emission factor processing module. Combined with high-precision vegetation growth type and localized species composition data, the data were input into the MEGANv3.2 emission factor processing module to obtain the gridded standard canopy emission factors in the simulation domain. S5. Use Weather Research and Forecasting (WRF) models to simulate meteorological data and obtain the required meteorological data; S6. Input vegetation growth type, canopy type, leaf area index, emission factor data, and meteorological data into the natural gas and aerosol emission model MEGANv3.2. The natural gas and aerosol emission model MEGANv3.2 adopts the publicly available model from http: / / bai.ess.uci.edu / megan to simulate and obtain a high-quality BVOCs emission inventory, specifically: S61. The preprocessing module of the natural gas and aerosol emission model MEGANv3.2 standardizes the basic geographic information of the study area. The land use characteristics and meteorological parameters of the study area are uniformly converted to the preset spatiotemporal resolution grid system through spatial interpolation, thereby obtaining the vegetation distribution dataset required by the model. S62, the emission factor processing module matches vegetation cover data (including vegetation growth type, canopy type and plant species composition) with emission factors of various plant species to establish emission factors and light-dependent factors based on grid cells. S63. The emission calculation module estimates emission flux through multi-source data coupling. It combines gridded emission factor and light-dependent factor data with vegetation leaf area index (LAIv) and canopy type data to parameterize the canopy environment and soil-vegetation interface processes respectively. Through dynamic correction algorithms, it accurately quantifies the emission of BVOCs.

[0022] Example 2: Based on the technical solution of Example 1, this example explores the spatiotemporal distribution characteristics of BVOCs emissions in my country in 2020, and obtains the following experimental results: (1) In 2020, China's total BVOC emissions were 27.58 Tg, of which urban areas accounted for 4.60 Tg and non-urban areas accounted for 22.98 Tg. Isoprene, monoterpenes, sesquiterpenes and other VOCs contributed 22.24%, 26.08%, 4.82% and 46.86% of the total emissions, respectively. In terms of individual compounds, isoprene, butane (which belongs to other VOCs category) and 2-methylbutene contributed the most to the total BVOC emissions, contributing 22.25%, 13.13% and 7.92% of the total emissions, respectively.

[0023] (2) my country's BVOCs emissions exhibit significant seasonal variations, mainly concentrated in summer, when emissions reach their annual peak, accounting for 62.96% of the total annual emissions. This is closely related to the suitable environmental conditions in summer (such as higher temperatures and strong solar radiation) and the vigorous growth of vegetation. In contrast, winter emissions are the lowest, accounting for only 3.42% of the total annual emissions, mainly due to poor temperature and light conditions, leading to withered vegetation. Emissions in spring and autumn are relatively close, accounting for 17.3% and 16.29% of the total annual emissions, respectively. In terms of monthly emissions, BVOCs monthly emissions show a unimodal distribution, reaching the annual maximum in August and dropping to the minimum level in December, with a difference of about 40 times. Among the various components of BVOCs, isoprene shows the most significant seasonal variation, with its contribution to total BVOCs emissions showing a high summer and low winter characteristic. Unlike isoprene, the emission contribution of monoterpenes and other VOCs is characterized by high levels in winter and low levels in summer. This is mainly because monoterpenes are primarily derived from evergreen conifers, whose leaves remain active throughout the year, allowing compounds such as monoterpenes to be continuously emitted throughout the year.

[0024] (3) In 2020, the spatial distribution of BVOCs emissions in China showed a gradient change from southeast to northwest. Among them, high emission areas were mainly distributed in the subtropical region south of the Yangtze River. The North China Plain and Changbai Mountain area also contributed a certain amount of BVOCs emissions due to the widespread distribution of crops and trees. The Northwest region's dominant plant type is herbaceous plants with low emission potential, and the arid climate limited BVOCs emissions. High isoprene emission areas were mainly distributed in Changbai Mountain, southern Qinling Mountains, South China, southern Yunnan, and Hainan Province. The spatial pattern of monoterpene emissions was similar to that of isoprene, but the emissions in southern China and the North China Plain were generally higher than those of isoprene. The emission level of sesquiterpenes was generally low, and the emission intensity of other VOCs was higher than that of isoprene. The main reason is that other VOCs contain a wide variety of compounds and account for a large proportion of the total BVOCs emissions. BVOC emissions in urban areas show a gradient decreasing from southeast to northwest. High emission areas are consistent with densely populated urban areas, mainly concentrated in the Pearl River Delta, Yangtze River Delta and the middle reaches of the Yangtze River urban agglomeration.

[0025] The model structures and algorithms not described in detail in this article are all general techniques in this field.

[0026] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A localized, high-precision BVOCs emission inventory construction method, characterized in that, Specifically, the following steps are included: S1. Collect high-resolution land cover remote sensing data and ground observation data to construct a high-precision vegetation distribution database of China that includes urban green spaces, obtain vegetation growth types, and the spatial resolution of the database is 1km×1km. S2. Based on the high-precision vegetation distribution database of China constructed in step S1, obtain the canopy type and leaf area index; S3. Based on the "Vegetation Map of China (1:1000000)", obtain a detailed composition of native plant species in China; S4. Based on the localized BVOCs emission factor database in China, match the emission factor data of each plant species, select the dynamic observation results in the emission factor database, and for species whose emission factors cannot be matched in the localized emission factor database, use the global emission factor database in the MEGANv3.2 emission factor processing module to obtain emission factor data. S5. Use weather research and forecasting models to simulate meteorological data and obtain the required meteorological data; S6. Input vegetation growth type, canopy type, leaf area index, emission factor data and meteorological data into the natural gas and aerosol emission model MEGANv3.2 to simulate and obtain a high-quality BVOCs emission inventory.

2. The method for constructing a localized high-precision BVOCs emission inventory according to claim 1, characterized in that, The high-resolution land cover remote sensing data and ground observation data mentioned in step S1 are derived from the basic land cover dataset, which includes the MCD12Q1 land cover data of the China multi-period land use remote sensing monitoring dataset, the "China Vegetation Map (1:1,000,000)" and the European Space Agency WorldCover land cover data. Using ArcGIS software, the land cover / use types in each land cover dataset are resampled into four vegetation growth types: trees, shrubs, herbs, and crops. Based on the four sets of reclassified vegetation growth type data, the vegetation growth types of urban and non-urban areas are processed separately to construct a high-precision Chinese vegetation distribution database capable of identifying urban green spaces. Specifically, urban land, rural settlements, and other construction land in the CNLUCC land use data are classified as urban areas. The vegetation in this area is redefined using the European Space Agency WorldCover land cover data with a spatial resolution of 10m to identify the vegetation growth types in urban areas. For non-urban areas, the CNLUCC land use data is used as a benchmark, and MODIS data is introduced. A comparative analysis of MCD12Q1 land cover data was conducted. For raster data where vegetation growth types were inconsistent between CNLUCC and MODIS MCD12Q1 data, the vegetation growth types were further corrected and redefined using the "Vegetation Map of China (1:1000000)".

3. The method for constructing a localized high-precision BVOCs emission inventory according to claim 2, characterized in that, Step S2 describes canopy types including conifers, broadleaf trees, tropical trees, shrubs, herbs, and crops. Based on the constructed high-precision Chinese vegetation distribution database, the data for the six canopy types are updated to correspond to the growth types. Specifically, the grid canopy type coverage ratios for shrubs, herbs, and crops are consistent with the vegetation growth type data. The grid coverage ratio for tropical trees is determined based on data from the MEGAN website. Trees within the vegetation growth types are divided into conifers and broadleaf trees. Using MODIS MCD12Q1 land cover data and the "China Vegetation Map (1:1000000)", trees within the vegetation growth types are further divided into conifers and broadleaf trees. This involves reclassifying the two datasets using ArcGIS software. For MODIS... Based on the MCD12Q1 land cover data, evergreen coniferous forest, deciduous coniferous forest, 50% mixed forest, and 70% woody savanna in this layer are reclassified as coniferous forest; evergreen broad-leaved forest, deciduous broad-leaved forest, 50% mixed forest, and 70% woody savanna are reclassified as broad-leaved forest. For the "Vegetation Map of China (1:1,000,000)," coniferous forest and 50% mixed coniferous and broad-leaved forest in this layer are reclassified as coniferous forest, and broad-leaved forest and 50% mixed coniferous and broad-leaved forest are reclassified as broad-leaved forest. Based on this method, the consistency between the canopy type distribution data and the vegetation growth type distribution data in the emission inventory can be guaranteed, and the estimation error caused by the difference between vegetation input data can be reduced.

4. The method for constructing a localized high-precision BVOCs emission inventory according to claim 3, characterized in that, Step S2 describes the Leaf Area Index (LAI) being quantified using the index LAIv. Specifically, the raw LAI data is updated based on a constructed high-precision Chinese vegetation distribution database. When the vegetation cover rate (VCF) of a vegetation growth type is detected in a grid, it indicates that vegetation distribution exists in that grid. When the LAI value is zero, the LAI of that grid is updated. The empirical value of PFT-LAI with a time resolution of 1 month is used for correction, and the monthly data is interpolated to an 8-day resolution. The weighted average of LAI is calculated based on the coverage ratio of each PFT in the grid and used as the average LAI of that grid. The VCF data comes from the sum of the coverage rates of the four vegetation growth types in the high-precision Chinese vegetation distribution database. The two are divided to generate LAIv data with an 8-day resolution. The maximum value of LAIv is set to 6 to eliminate the special case where the vegetation in the grid is sparse but the estimated value is high.

5. The method for constructing a localized high-precision BVOCs emission inventory according to claim 4, characterized in that, The plant species mentioned in step S3 include 166 native tree species, 149 shrub species, 148 herbaceous species, and 31 crop species native to my country.

6. The method for constructing a localized high-precision BVOCs emission inventory according to claim 5, characterized in that, The Chinese native BVOCs emission factor database described in step S4 was established by combining actual field observations and literature surveys. Specifically, it first established a method for classifying and determining plant emission intensity based on statistics, and then determined a refined emission intensity classification interval and a more accurate emission factor characteristic value. This resulted in the establishment of a detailed and reliable Chinese native emission factor database by plant and BVOCs group. The database updated the isoprene, monoterpene, and sesquiterpene emission factor data of 202 native plant species. The vegetation species updated with native emission factors accounted for approximately 40% of the total number of vegetation species.

7. The method for constructing a localized high-precision BVOCs emission inventory according to claim 5, characterized in that, The specific process of step S6 is as follows: S61. The preprocessing module of the natural gas and aerosol emission model MEGANv3.2 standardizes the basic geographic information of the study area. The land use characteristics and meteorological parameters of the study area are uniformly converted to the preset spatiotemporal resolution grid system through spatial interpolation, thereby obtaining the vegetation distribution dataset required by the model. S62, the emission factor processing module matches vegetation cover data with emission factors of various plant species to establish emission factors and light-dependent factors based on grid cells. The vegetation cover data includes vegetation growth type, canopy type and plant species composition. S63. The emission calculation module estimates emission flux through multi-source data coupling. It combines gridded emission factor and light-dependent factor data with vegetation leaf area index (LAIv) and canopy type data to parameterize the canopy environment and soil-vegetation interface processes respectively. Through dynamic correction algorithms, it accurately quantifies the emission of BVOCs.

Images