Air-ground cooperative greenhouse gas flux monitoring and source analysis method

Abstract

The invention discloses an air-ground cooperative greenhouse gas flux monitoring and source analysis method, which comprises the following steps of: arranging ground observation systems in the upwind direction and the downwind direction of a prevailing wind direction of a park to realize long-term continuous monitoring of greenhouse gas concentration and meteorological parameters; the method comprises the following steps: carrying out vertical section observation on emission plume under typical meteorological conditions by using an unmanned aerial vehicle, and estimating canopy scale flux based on a mass balance method; quantifying the emission flux of each sub-source in combination with ground long-term observation data and a backward Lagrange random diffusion model; and further taking the canopy scale flux estimated by the unmanned aerial vehicle as a constraint condition to correct an inversion result of the ground model so as to realize long-term accurate estimation of the park scale greenhouse gas emission flux and dynamic quantitative splitting of sub-source contribution. Based on the method, a flux monitoring result with time continuity, space consistency and source resolvability can be provided for a typical composite emission system, so that accurate emission reduction and scientific management are supported.

Description

Technical Field

[0001] This invention relates to the field of greenhouse gas emission monitoring and quantification technology, specifically to a method for air-ground coordinated greenhouse gas flux monitoring and source apportionment. Background Technology

[0002] Accurate monitoring and quantification of greenhouse gas emissions are a crucial technological foundation for addressing climate change and implementing emission reduction strategies. Complex emission systems at the industrial park scale, such as large-scale livestock farms, landfills, wastewater treatment plants, and chemical industrial parks, have become a key and challenging area for greenhouse gas emission monitoring due to their diverse emission sources, dispersed spatial distribution, and dynamic changes in emission intensity.

[0003] Taking dairy farms as an example, intestinal fermentation in cattle and manure accumulation fermentation are two key processes in methane emissions, accounting for a significant proportion of total anthropogenic methane emissions. Therefore, accurate and comprehensive monitoring and assessment of methane flux from emission sources at the farm scale, such as dairy farms, is of paramount importance for accurately calculating anthropogenic greenhouse gas emission inventories and developing scientific livestock emission reduction strategies.

[0004] Currently, greenhouse gas monitoring of such park-scale emission sources mainly relies on the following technical means: 1. Traditional inventory estimation methods mainly employ emission factor methods and statistical models to estimate greenhouse gas emissions. This involves multiplying activity level data (such as animal population and manure treatment volume) by the corresponding emission factor to arrive at the total amount. However, this method heavily relies on empirical parameters and struggles to reflect the spatiotemporal heterogeneity of emissions caused by differences in management practices, manure treatment processes, and fluctuating meteorological conditions in specific industrial parks. Furthermore, it lacks empirical data directly based on atmospheric observations, resulting in significant uncertainty in the estimation results.

[0005] 2. Relying on in-situ ground-based observation technologies, some schemes employ ground-based observation stations combined with atmospheric diffusion models to invert emission fluxes, or directly measure the intensity of individual emission sources using devices such as the breathing chamber method, head-mounted breathing mask method, and ventilation rate method. Ground-based observation stations are typically deployed in the near-surface area of ​​the park, but due to limitations in installation height and observation location, the concentration and diffusion information captured cannot fully characterize the vertical structure of the emission plume within the canopy height range, leading to systematic biases in the total flux of the park inverted by the model. Furthermore, microscopic measurement methods such as the breathing chamber method can only obtain emission data from a single livestock or a single livestock shed, failing to effectively cover non-point source emissions from manure fermentation tanks, oxidation ponds, etc., and making it even more difficult to achieve overall flux assessment and multi-source contribution decomposition at the park scale.

[0006] 3. Unmanned Aerial Vehicle (UAV) remote sensing technology: In recent years, observation schemes using UAVs equipped with high-precision greenhouse gas analyzers have attracted attention due to their high spatial resolution and flexibility. This technology obtains two-dimensional concentration and wind field distribution data by flying within the vertical cross section downwind of the emission source, and then uses the mass balance method to estimate the total emission flux. However, existing UAV observation schemes still have the following limitations in monitoring complex sources: (1) The selection of background concentration lacks standardized criteria and is easily affected by local environmental interference; (2) A single flight is difficult to capture the three-dimensional structure of the plume completely, and the observation scheme needs to be repeatedly adjusted based on pre-experiments; (3) In the case of multiple dispersed emission sub-sources in the park, it is difficult to accurately invert the dynamic contribution ratio of each sub-source based solely on instantaneous cross-sectional observation data. The results are more biased towards the total flux integration of the cross section and cannot achieve refined source apportionment.

[0007] In summary, existing technologies cannot simultaneously meet the triple requirements of time continuity, spatial representativeness, and source contribution resolvability for greenhouse gas monitoring at the industrial park scale. There is a lack of a systematic technical solution that combines air-ground collaboration, observation-model coupling, and long-term stable operation. As a result, accurate carbon accounting and targeted emission reduction management of complex emission systems such as aquaculture parks still face technical bottlenecks. Summary of the Invention

[0008] The present invention aims to at least partially solve one of the technical problems existing in the related art.

[0009] The purpose of this invention is to provide a ground-air coordinated method for monitoring and source apportionment of greenhouse gas fluxes. By constructing a ground-air coordinated three-dimensional monitoring system, and introducing the total flux at the canopy scale from unmanned aerial vehicles (UAVs), this method is used to correct the sub-source emission results derived from ground observations and backward Lagrange random diffusion models, thereby overcoming the limitations of near-ground models in terms of vertical representativeness. Furthermore, by establishing a dynamic update mechanism for the contribution ratio of sub-sources, long-term, accurate, and sustainable monitoring and source apportionment of emission fluxes in industrial parks can be achieved.

[0010] To achieve the above objectives, the present invention provides a method for air-ground coordinated greenhouse gas flux monitoring and source apportionment, comprising the following steps: S1. Deploy ground observation systems in the upwind and downwind directions of the prevailing wind direction in the target area to conduct long-term continuous synchronous observations of greenhouse gas concentrations and meteorological parameters, and obtain time-series observation data under different wind conditions. S2. Based on the time series observation data, combined with the backward Lagrange random diffusion model, the emission flux of each sub-source is estimated by inversion, and the emission flux of each sub-source is summed to obtain the total emission estimate of the target area based on ground observation, and the corresponding sub-source contribution structure is obtained. S3. Under typical meteorological conditions, use UAVs to perform vertical profile flight in a section downwind of the target area and perpendicular to the prevailing wind direction, and acquire observation data, including greenhouse gas concentration distribution data and wind field spatial distribution data covering the canopy height range of the target area. S4. Based on UAV observation data, construct a spatially continuous concentration field and wind field within the canopy height range of the target area, and use the mass balance method to calculate the total canopy-scale emission flux of the target area during the corresponding observation period. S5. Using the total emission flux at the canopy scale and the greenhouse gas concentration distribution data obtained in step S4 as constraints, the total emission estimate of the target area based on ground observations obtained in step S2 is corrected, and the contribution ratio of each sub-source is further quantified. S6. Based on the calibrated contribution ratio of each sub-source and combined with the time series observation data of the ground observation system, the long-term emission flux of the target area is estimated; and the sub-source contribution benchmark is updated by periodic UAV observations to achieve dynamic and accurate monitoring of emission flux.

[0011] A further preferred embodiment of the present invention is that, in step S1, the ground observation system is deployed as follows: Based on the prevailing wind direction of the target area, at least one ground observation station is set up in the upwind and downwind areas surrounding the target area. Each ground observation station is equipped with a greenhouse gas analyzer, a three-dimensional ultrasonic anemometer, and a temperature and humidity sensor, which are used to synchronously and continuously collect greenhouse gas concentration data, three-dimensional wind field data, and basic meteorological parameter data.

[0012] Preferably, in step S2, based on the time-series observation data and combined with the backward Lagrange random diffusion model, the emission flux of each sub-source is estimated by inversion, and the emission fluxes of each sub-source are summed to obtain the total emission estimate of the target area based on ground observations; specifically: The greenhouse gas concentration data collected by the ground observation system and the three-dimensional wind field data collected by the three-dimensional ultrasonic anemometer are averaged according to a preset time window; the data of the observation point located upwind of the target area are selected, and the lowest greenhouse gas concentration value in that time window is determined as the background concentration value of the current time window. Friction velocity calculated based on raw 10Hz three-dimensional ultrasonic wind speed data. Obukhov length Surface roughness and average wind direction The key turbulence parameters are given by the following formulas: ; ; ; in, The instantaneous disturbance component of horizontal wind speed in the mean wind direction. This represents the instantaneous disturbance component of the vertical wind speed; the disturbance component is obtained by subtracting the average wind speed within the corresponding time window from the instantaneous wind speed. Kármán's constant, It is the acceleration due to gravity. For temperature, For sensible heat flux, air density, For isobaric specific heat capacity, The installation height of the three-dimensional ultrasonic anemometer. For horizontal wind speed, For stability correction function; After discarding data that do not meet the similarity theory assumptions or have anomalous boundary conditions, for each selected effective observation time window, a backward Lagrange random diffusion model is used, combined with the observed concentrations within that time window. Background concentration Based on the calculated key turbulence parameters and the predefined spatial geographic extent of each emission sub-source, the inverse equation can be performed for each sub-source. Emission rate during this time window , is represented as: ; In the formula, The simulated concentration-to-source intensity ratio is determined by the source shape, atmospheric motion, and concentration observation location. During the simulation, the friction velocity... Obukhov length Surface roughness and average wind direction As input parameters to the model, they are used to characterize the atmospheric turbulence intensity, stability conditions, and wind field structure within the current time window, thereby determining the random motion trajectory and diffusion characteristics of Lagrange particles. The calculation formula is: ; in, The vertical velocity component of the Lagrange particles released by the model as they back towards the emission source region. The number of Lagrange particles released, set to 50,000 by default; By combining data from multiple effective observation time windows under different prevailing wind directions, the emission rate of each emission sub-source is calculated and retrieved within all relevant effective time windows. average This is used as the typical emission rate estimate for the N sub-sources; then, the typical emission rate estimates for all N sub-sources within the target area are summed to obtain the overall greenhouse gas emission rate estimate for the target area. The calculation formula is: .

[0013] Preferably, in step S3, the UAV is equipped with a greenhouse gas analyzer and a three-dimensional ultrasonic anemometer, and is configured as follows: The observation mission is conducted under meteorological conditions of moderate wind speed and stable wind direction, requiring that the wind direction at that moment aligns the downwind direction of all potential emission sources with the same plane. The flight path is set to perform zigzag maneuvers within a vertical profile downwind of the target area and perpendicular to the prevailing wind direction. The vertical flight altitude range covers from above ground obstacles to 120 meters, and the horizontal flight range covers the downwind area of ​​all potential emission sources. If the flight observation is interrupted, the battery is replaced and the operation resumes at the point of interruption to obtain continuous and encrypted observation data for a complete profile.

[0014] Preferably, in step S4, based on UAV observation data, a spatially continuous concentration field and wind field within the canopy height range of the target area are constructed, and the total canopy-scale emission flux of the target area during the corresponding observation period is calculated using the mass balance method; specifically: Flux estimation was performed using observational data from UAVs that captured the complete plume structure. Discrete greenhouse gas concentration data and three-dimensional wind speed data collected by the UAVs on an observation cross-section perpendicular to the prevailing wind direction were reconstructed into a continuous two-dimensional gridded concentration field and wind field using ordinary Kriging interpolation; the horizontal boundary of the observation cross-section... Covering the downwind extent of all potential emission sources, vertical boundary This refers to the maximum flight altitude from above ground obstacles to the preset maximum flight altitude. Based on the generated two-dimensional gridded concentration field and wind field, the mass balance formula is used to integrate the observed cross-section, which is expressed as: ; in, The net greenhouse gas flux through the observation section is expressed in moles per second (mol·s⁻¹). For grid points The dry mole fraction of greenhouse gases at a given point, expressed in micromoles per mole (μmol·mol⁻¹). For background concentration, the average concentration observed from the upwind direction on the ground during the UAV flight period was used; For grid points The wind speed component perpendicular to the direction of the observation section, in meters per second (m·s⁻¹). For grid points The air molar density at a given location, expressed in moles per cubic meter (mol·m⁻³), is calculated using air pressure data from nearby weather stations and temperature data from ground stations. The formula is as follows: ; in, The pressure is in Pa, obtained from data from nearby weather stations. The value of is the ideal gas constant, which is 8.314 (J·K⁻¹). The air temperature is derived from temperature data from ground stations. The calculated net greenhouse gas flux through the observation section Multiply by the molar mass of the target greenhouse gas That is, to obtain the total emission flux at the canopy scale of the target area in units of mass (e.g., grams per second, g·s⁻¹). , is represented as: .

[0015] Preferably, in step S5, the total emission flux at the canopy scale and the greenhouse gas concentration distribution data from the observation data obtained in step S4 are used as constraints to correct the total emission estimate of the target area obtained in step S2 based on ground observations, and to further quantify the contribution ratio of each sub-source; specifically: First, the inverse concentration field simulation is performed on the sub-source contribution structure obtained from the ground inversion model; When the simulation results and the greenhouse gas concentration distribution data from UAV observations are inconsistent in terms of emission height, diffusion pattern, or wind field conditions, it is determined that the contribution structure of the sub-source does not meet the cooperative constraint conditions, and the contribution ratio of each sub-source is corrected; when the two are consistent, the contribution structure of the sub-source is considered reasonable. When it is necessary to correct the contribution ratio of sub-sources, the total emission flux at the canopy scale of the target area for the corresponding observation period, obtained from UAV observation data, will be used. Compared with the total emission estimates of the target area obtained based on time series observation data Compare and calculate the correction factor. , is represented as: ; Subsequently, each sub-source obtained from the inversion will be... Emission rate during this time window Multiply by the correction factor The corrected sub-source emission rate was obtained. , is represented as: ; Then based on the corrected emission rates of each sub-source Calculate its contribution to the total flux.

[0016] As a preferred approach, when correcting the contribution ratio of sub-sources, it is also necessary to combine ground-based time-series observation data to assess the background rationality and representativeness of the observation period of the UAV observation data; specifically: The time series observation data is used to determine whether the concentration in the non-plume region observed by the UAV is within a reasonable background range. When the background concentration of the UAV observation results is abnormally high and it is determined that there may be other interference sources, the UAV observation is only used as reference information and is not used to correct the ground observation results, so as to ensure the accuracy and reliability of long-term emission flux estimation.

[0017] Beneficial effects: This invention obtains continuous spatial emission cross-sectional information within the canopy height range of the park through UAV observation, which makes up for the shortcomings of ground observation stations, which are limited by installation height and observation location and cannot directly characterize canopy-scale emission flux; This invention overcomes the problem that UAV observation can only obtain instantaneous cross-sectional information and cannot reflect the characteristics of emission time variation through long-term continuous observation under different wind conditions by ground observation system. This invention introduces the total emission flux at the canopy scale of the park obtained by UAV as a correction factor into the inversion process of the ground-based backward Lagrange stochastic diffusion model, and constructs a method for estimating greenhouse gas flux at the park scale and quantifying source contributions that simultaneously satisfies spatial consistency and temporal continuity. The method of this invention can achieve long-term accurate monitoring of greenhouse gas emission fluxes at the park scale and stable decomposition of their sub-source contributions without the need for additional complex observation equipment. It establishes a long-term estimation and periodic update mechanism based on contribution ratio, enabling the observation system to adapt to changes in the emission sources themselves. This provides a stable and reliable data foundation for accurate carbon management and emission reduction assessment in parks, and has good engineering practicality and promotion value. Attached Figure Description

[0018] Figure 1 This is a flowchart of the air-ground coordinated greenhouse gas flux monitoring and source apportionment method of the present invention.

[0019] Figure 2 Location map of the ground observation stations for this invention.

[0020] Figure 3 This is a flight path diagram of the UAV of the present invention.

[0021] Figure 4 This is a diagram showing the methane concentration distribution and interpolation observed by the UAV of this invention.

[0022] Figure 5 This refers to the trajectory of the particle released in the backward Lagrange random diffusion model of the ground station in this invention.

[0023] Figure 6 The fluxes and contribution percentages of each emission source were calculated using the backward Lagrange random diffusion model of this invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, embodiments of this invention, and should not be construed as limiting the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention. In the description of this invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0025] The following is combined with Figures 1-6 This invention describes the air-ground coordinated greenhouse gas flux monitoring and source apportionment method provided by the present invention.

[0026] Example: This example provides a method for air-ground coordinated greenhouse gas flux monitoring and source apportionment, such as... Figure 1 As shown, its core lies in constructing a three-dimensional technical system that integrates "space-ground" collaboration and cross-verification between models and observations. The following section uses a large-scale dairy farm as a typical application scenario to elaborate on the specific implementation of this invention.

[0027] S1. Deploy ground-based observation systems in the upwind and downwind directions of the prevailing wind direction in the target area to conduct long-term continuous synchronous observations of greenhouse gas concentrations and meteorological parameters, and obtain time-series observation data under different wind direction conditions.

[0028] Select a typical large-scale dairy farm as the implementation site. This farm should contain multiple potential methane emission sources, such as: semi-open or closed cowsheds, solid manure storage areas after solid-liquid separation, liquid manure storage ponds (biogas slurry ponds, oxidation ponds), and water bodies such as lotus ponds and fishponds that may be affected by manure seepage. Before construction, a site survey and farm management data are required to determine the boundaries, area, and spatial location of each emission source.

[0029] Based on years of local meteorological data, the prevailing wind direction in the area was determined, such as... Figure 2As shown, two observation points were selected on the periphery of the park. These two points are approximately 50-70 meters from the boundary of the nearest emission source and must be located in open areas, avoiding excessive obstruction from tall buildings or trees, to ensure effective capture of emission plumes from the site. Each ground observation station is equipped with the same instrument suite, mounted on a stable support approximately 3 meters high to minimize the direct impact of ground turbulence and local obstacles. The two ground observation systems are constructed using a combination of a portable greenhouse gas analyzer, a three-dimensional ultrasonic anemometer, and temperature and humidity sensors. The portable greenhouse gas analyzer is based on off-axis integrating cavity output spectroscopy technology, with a sampling frequency of 1 Hz, and can simultaneously measure CO2, CH4, and H2O concentrations. The measurement accuracy for CH4 is 0.6 × 10⁻⁻⁻⁶. 9 mol·mol⁻¹ (10 seconds). The GillWindMaster Pro 3D ultrasonic anemometer, with a sampling frequency set to 10Hz, can measure 3D wind and ultrasonic virtual temperature. Its wind speed accuracy is ±1.5% m·s⁻¹, wind direction accuracy is 2°, and acoustic temperature accuracy is ±2℃. In addition, a Vaisala HMP155A temperature and humidity sensor is used to record meteorological parameters, with a temperature measurement accuracy of ±0.17℃ and a humidity accuracy of ±1%RH (0%~90%RH).

[0030] Three-dimensional wind observation instruments and temperature and humidity observation instruments are synchronized through data acquisition devices (such as CR1000) to achieve data timestamp alignment and local storage, and observation personnel regularly maintain the equipment and copy the data.

[0031] S2. Based on the time series observation data, combined with the backward Lagrange random diffusion model, the emission flux of each sub-source is estimated by inversion, and the emission flux of each sub-source is summed to obtain the total emission estimate of the target area based on ground observation and the corresponding sub-source contribution structure is obtained.

[0032] The greenhouse gas concentration data collected by the ground observation system and the three-dimensional wind field data collected by the three-dimensional ultrasonic anemometer are averaged according to a preset time window; the data of the observation point located upwind of the target area are selected, and the lowest greenhouse gas concentration value in that time window is determined as the background concentration value of the current time window. Friction velocity calculated based on raw 10Hz three-dimensional ultrasonic wind speed data. Obukhov length Surface roughness and average wind direction The key turbulence parameters are given by the following formulas: ; ; ; in, The instantaneous disturbance component of horizontal wind speed in the mean wind direction. This represents the instantaneous disturbance component of the vertical wind speed; the disturbance component is obtained by subtracting the average wind speed within the corresponding time window from the instantaneous wind speed. Kármán's constant, It is the acceleration due to gravity. For temperature, For sensible heat flux, air density, For isobaric specific heat capacity, The installation height of the three-dimensional ultrasonic anemometer. For horizontal wind speed, For stability correction function; After discarding data that do not meet the similarity theory assumptions or have anomalous boundary conditions, for each selected effective observation time window, a backward Lagrange random diffusion model is used, combined with the observed concentrations within that time window. Background concentration Based on the calculated key turbulence parameters and the predefined spatial geographic extent of each emission sub-source, the inverse equation can be performed for each sub-source. Emission rate during this time window , is represented as: ; In the formula, The simulated concentration-to-source intensity ratio is determined by the source shape, atmospheric motion, and concentration observation location. During the simulation, the friction velocity... Obukhov length Surface roughness and average wind direction As input parameters to the model, they are used to characterize the atmospheric turbulence intensity, stability conditions, and wind field structure within the current time window, thereby determining the random motion trajectory and diffusion characteristics of Lagrange particles. The calculation formula is: ; in, The vertical velocity component of the Lagrange particles released by the model as they back towards the emission source region. The number of Lagrange particles released, set to 50,000 by default; By combining data from multiple effective observation time windows under different prevailing wind directions, the emission rate of each emission sub-source is calculated and retrieved within all relevant effective time windows. average This is used as the typical emission rate estimate for the N sub-sources; then, the typical emission rate estimates for all N sub-sources within the target area are summed to obtain the overall greenhouse gas emission rate estimate for the target area. The calculation formula is: .

[0033] like Figure 5 As shown, Figure 5 a, Figure 5 b, Figure 5 c and Figure 5 The values ​​of d in the figure represent the Lagrange particle release trajectory distribution results for the four types of emission sources: cattle shed, manure storage area, lotus pond, and fish pond.

[0034] S3. Under typical meteorological conditions, use UAVs to perform vertical profile flights in a section downwind of the target area and perpendicular to the prevailing wind direction, and acquire observation data, including greenhouse gas concentration and wind field spatial distribution data covering the canopy height range of the target area.

[0035] The UAV observation system utilizes a high-payload, high-stability hexacopter platform (such as the DJI Matrice 600 Pro). Its vertical hovering accuracy must be better than ±0.5 meters, horizontal hovering accuracy better than ±1.5 meters, and maximum payload capacity greater than 6 kg to ensure it can carry the necessary sensors and complete precise flight paths. The UAV carries the lightweight, high-precision greenhouse gas monitoring instrument GLA133-GPC, with a sampling frequency of 1 Hz. The sensor extends to the lower side of the UAV fuselage via a 1.2-meter-long carbon fiber sampling tube to reduce the impact of rotor disturbance on the air intake. A lightweight carbon fiber pole is installed approximately 0.6 meters above the UAV, with a miniature three-dimensional ultrasonic anemometer (TriSonica™ Mini) fixed at its top. This device simultaneously measures three-dimensional wind speed, wind direction, and temperature at flight altitude, with wind speed accuracy better than ±0.2 m / s, wind direction accuracy better than ±3°, and a sampling frequency of 1 Hz.

[0036] During long-term ground-based observation, UAV observation missions should be triggered during periods of suitable meteorological conditions. The criteria for determining an ideal observation window include: clear or partly cloudy weather, stable prevailing wind direction, and moderate wind speed. Specific flight mission execution is as follows: Before each experiment, a preliminary flight plane was selected based on the on-site meteorological conditions and wind direction to obtain the overall distribution characteristics of the methane plume from the dairy farm. The wind direction needed to ensure that the downwind direction of all emission sources was in the same direction. The preliminary experiment consisted of one horizontal gradient and one vertical profile. Based on the relevant research experience in Reference 1 (Allen G, Pitt JR, Hollingsworth P, et al. Measuring landfill methane emissions using unmannedaerial systems: field trial and operational guidance[C]. 2015), the preliminary experiment was conducted at a distance of 100m to 120m downwind from the dairy farm. The flight width of the horizontal gradient should cover the downwind range of the entire dairy farm's methane emission sources as much as possible (>500m), with a height of approximately 20m to 30m. The flight height of the vertical profile ranged from 0m to 120m. The preliminary experiment data was processed on-site to quickly determine the approximate vertical thickness, horizontal width, and center position of the plume.

[0037] Based on the plume morphology determined in the preliminary experiments, a formal zigzag profile observation route will be planned, such as... Figure 3 As shown. The sampling is conducted perpendicular to the current stable prevailing wind direction. The vertical range begins at a safe height (e.g., 10-15 meters) above ground obstacles (such as trees or power lines) and extends to above the top of the plume as determined in the pre-experiment (typically 80-120 meters). The horizontal range ensures coverage of all high-concentration areas shown in the pre-experiment, typically extending approximately 100-150 meters beyond the source area, for a total width between 400-600 meters. A complete profile of dense sampling is achieved using two consecutive flights. The first flight begins at low altitude, flying a zigzag pattern according to a preset altitude sequence. After battery replacement, the second flight is immediately executed, continuing at the same altitude as the first flight. The UAV's horizontal flight speed is controlled at 2-4 m / s (depending on the flight range), and its vertical climb / descent speed is controlled at 0.5-2 m / s.

[0038] By conducting on-site analysis of the preliminary experimental observations, the flight altitude and width range for the formal experiment were determined to ensure that the observation path could fully capture the spatial morphology of the methane plume. The formal experiment was conducted alternately by two sets of batteries, and the data from the two sets were merged to form a complete observation cross-section.

[0039] The ground observation system operates continuously during the drone's flight.

[0040] S4. Based on UAV observation data, construct a spatially continuous concentration field and wind field within the canopy height range of the target area, and use the mass balance method to calculate the total canopy-scale emission flux of the target area during the corresponding observation period.

[0041] Flux estimation was performed using observational data from UAVs that captured the complete plume structure. Discrete greenhouse gas concentration data and three-dimensional wind speed data collected by the UAVs on an observation cross-section perpendicular to the prevailing wind direction were reconstructed into a continuous two-dimensional gridded concentration field and wind field using ordinary Kriging interpolation; the horizontal boundary of the observation cross-section... Covering the downwind extent of all potential emission sources, vertical boundary This refers to the maximum flight altitude from above ground obstacles to the preset maximum flight altitude. like Figure 4 The diagram shown is a schematic of the complete plume structure captured by the UAV in this embodiment. Figure 4 In the diagram, 'a' represents the distribution of discrete methane concentration observation data points from UAV observations. Figure 4 In the diagram, b represents the continuous two-dimensional methane concentration distribution field reconstructed using ordinary kriging interpolation based on the discrete data. This field is used to characterize the spatial distribution characteristics of the methane plume within the canopy height range of the target region during the observation period, providing grid data for subsequent flux calculations.

[0042] Based on the generated two-dimensional gridded concentration field and wind field, the mass balance formula is used to integrate the observed cross-section, which is expressed as: ; in, The net greenhouse gas flux through the observation section is expressed in moles per second (mol·s⁻¹). For grid points The dry mole fraction of greenhouse gases at a given point, expressed in micromoles per mole (μmol·mol⁻¹). For background concentration, the average concentration observed from the upwind direction on the ground during the UAV flight period was used; For grid points The wind speed component perpendicular to the direction of the observation section, in meters per second (m·s⁻¹). For grid points The air molar density at a given location, expressed in moles per cubic meter (mol·m⁻³), is calculated using air pressure data from nearby weather stations and temperature data from ground stations. The formula is as follows: ; in, The pressure is in Pa, obtained from data from nearby weather stations. The value of is the ideal gas constant, which is 8.314 (J·K⁻¹). The air temperature is derived from temperature data from ground stations. The calculated net greenhouse gas flux through the observation section Multiply by the molar mass of the target greenhouse gas That is, to obtain the total emission flux at the canopy scale of the target area in units of mass (e.g., grams per second, g·s⁻¹). , is represented as: .

[0043] S5. Using the total emission flux at the canopy scale and the greenhouse gas concentration distribution data obtained in step S4 as constraints, the total emission estimate of the target area obtained in step S2 based on ground observation is corrected, and the contribution ratio of each sub-source is further quantified.

[0044] Ground-based observation inversion models typically invert emission fluxes from multiple emission sources based on near-surface concentration observations and statistical wind field conditions. However, due to limitations imposed by installation height and spatial sampling location, the representativeness of ground-based observation results in the vertical direction is uncertain; that is, the vertical diffusion patterns of emissions from different sources may have multiple equivalent solutions.

[0045] Unmanned aerial vehicle (UAV) observations, through vertical profile flights downwind of the target area, can acquire information on the distribution of greenhouse gas concentrations and wind fields at different altitudes, thus directly reflecting the vertical structural characteristics of emission plumes within the canopy height range. Vertical concentration distribution characteristics include the overall pattern of concentration changes with altitude, the altitude range where the plume is mainly concentrated, and the width of concentration enhancement distribution in the vertical direction; wind field structure information includes the characteristics of wind speed and direction changes with altitude, used to characterize atmospheric diffusion conditions at different altitude levels.

[0046] In this method, the vertical concentration distribution characteristics and wind field structure information are not directly used to calculate emission fluxes, but are introduced into the ground observation inversion model as constraints on the rationality of the ground inversion results. Specifically, when the emission flux allocation or sub-source contribution structure obtained from the ground inversion model is physically inconsistent with the vertical concentration distribution pattern and wind field structure observed by the UAV, the inversion result is determined not to meet the cooperative constraint conditions, and is corrected, downweighted, or eliminated; when the two are consistent in emission height range, diffusion pattern, and wind field conditions, the inversion result is considered to meet the cooperative constraint requirements.

[0047] By introducing the above constraint mechanism, the vertical concentration distribution characteristics and wind field structure information obtained by UAV observation can effectively limit the feasible solution space of the ground inversion model, avoid the emission structure ambiguity caused by relying solely on near-ground observation data, and thus improve the physical consistency and reliability of the ground inversion results at the vertical scale.

[0048] In layman's terms, this involves quantifying the proportion of each source using ground data. After quantification, the concentration field can be simulated in reverse. This simulated concentration field is then compared with the vertical concentration field observed by the UAV, allowing for fine-tuning of the proportion of each source until the concentration fields from the two methods are similar. The specific method is as follows: First, the inverse concentration field simulation is performed on the sub-source contribution structure obtained from the ground inversion model; When the simulation results and the greenhouse gas concentration distribution data from UAV observations are inconsistent in terms of emission height, diffusion pattern, or wind field conditions, it is determined that the contribution structure of the sub-source does not meet the cooperative constraint conditions, and the contribution ratio of each sub-source is corrected; when the two are consistent, the contribution structure of the sub-source is considered reasonable. When it is necessary to correct the contribution ratio of sub-sources, the total emission flux at the canopy scale of the target area for the corresponding observation period, obtained from UAV observation data, will be used. Compared with the total emission estimates of the target area obtained based on time series observation data Compare and calculate the correction factor. , is represented as: ; Subsequently, each sub-source obtained from the inversion will be... Emission rate during this time window Multiply by the correction factor The corrected sub-source emission rate was obtained. , is represented as: ; Then based on the corrected emission rates of each sub-source Calculate its contribution to the total flux.

[0049] However, it's important to note that the concentration field observed by the drone itself will include non-plume regions, which will also provide a background concentration value. If the observed background concentration value is higher than expected and inconsistent with long-term upwind data from the ground station, the drone observation results may be influenced by other non-concern emission sources. Therefore, the drone results in this instance can only be used as a reference and cannot be used to correct subsequent ground station results. Another drone observation and correction should be conducted at a later time.

[0050] When correcting the contribution ratio of sub-sources, it is also necessary to combine ground time-series observation data to judge the background rationality and representativeness of the observation period of the UAV observation data; specifically: The time series observation data is used to determine whether the concentration in the non-plume region observed by the UAV is within a reasonable background range. When the background concentration of the UAV observation results is abnormally high and it is determined that there may be other interference sources, the UAV observation is only used as reference information and is not used to correct the ground observation results, so as to ensure the accuracy and reliability of long-term emission flux estimation.

[0051] like Figure 6 The figure shows the estimated emission fluxes of each emission source in the target area of ​​this embodiment. It can be seen that the emission rate of manure from the cattle farm is 27.54 g·s⁻¹, accounting for approximately 47% of the total emission; the emission rate from the cattle shed is 14.92 g·s⁻¹, accounting for 23.2%; the emission rate from the lotus pond is 13.40 g·s⁻¹, accounting for 25.9%; and the emission rate from the fishpond is 1.83 g·s⁻¹, accounting for 25.9%.

[0052] S6. Based on the calibrated contribution ratio of each sub-source and combined with the time series observation data of the ground observation system, the long-term emission flux of the target area is estimated; and the sub-source contribution benchmark is updated by periodic UAV observations to achieve dynamic and accurate monitoring of emission flux.

[0053] Based on the calculated contribution ratio of the sub-sources and long-term continuous data from the ground-based observation system, the flux of the corresponding sub-source under the current wind direction can be calculated, thereby estimating the total emission flux at the canopy scale of the park. To address potential changes in emission source activities within the park (such as changes in livestock inventory, emptying or covering of manure pits, etc.), drone observations are repeated at regular intervals (such as every quarter or half a year) to recalibrate the contribution ratio of the sub-sources, thereby achieving dynamic, accurate, and sustainable monitoring of greenhouse gas emissions across the entire park.

[0054] In summary, this invention provides a method and system for estimating greenhouse gas fluxes and decomposing source contributions in a park based on a combination of ground station and UAV observations. The method constructs an air-ground collaborative observation system consisting of fixed ground observations and UAV observations, simultaneously collecting long-term continuous ground concentration and turbulence data, as well as high spatial resolution vertical profile data, achieving complementarity between temporal continuity and spatial resolution. Based on UAV observation data, a continuous observation cross-section is constructed by mapping the concentration and wind field information of discrete points to a two-dimensional plane perpendicular to the wind direction and performing spatial interpolation. The greenhouse gas emissions of the entire park are obtained by integrating the concentration enhancement value and vertical wind speed component of each grid on the cross-section. The emission rates of each sub-source are inverted based on ground observation data and a backward Lagrange random diffusion model. The total emissions at the canopy scale estimated by UAVs are introduced to correct the ground model inversion results, accurately quantifying the contribution ratio of each sub-source. Finally, based on the corrected contribution ratio relationship and combined with long-term continuous observation data from the ground system, long-term accurate estimation of park emissions is achieved. The sub-source contribution ratio is updated periodically through UAV observations, forming a dynamic and sustainable monitoring system. This invention not only achieves accurate and rapid estimation of total greenhouse gas emissions at the park scale and reliable decomposition of sub-source contributions, but also constructs a complete technical closed loop of "observation-correction-calculation-update", providing a systematic and long-term operational solution for accurate carbon accounting, inventory verification and targeted emission reduction management at the typical park scale such as farms, landfills and chemical industrial parks.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for air-ground coordinated greenhouse gas flux monitoring and source apportionment, characterized in that, Includes the following steps: S1. Deploy ground observation systems in the upwind and downwind directions of the prevailing wind direction in the target area to conduct long-term continuous synchronous observations of greenhouse gas concentrations and meteorological parameters, and obtain time-series observation data under different wind conditions. S2. Based on the time series observation data, combined with the backward Lagrange random diffusion model, the emission flux of each sub-source is estimated by inversion, and the emission flux of each sub-source is summed to obtain the total emission estimate of the target area based on ground observation, and the corresponding sub-source contribution structure is obtained. S3. Under typical meteorological conditions, use UAVs to perform vertical profile flight in a section downwind of the target area and perpendicular to the prevailing wind direction, and acquire observation data, including greenhouse gas concentration distribution data and wind field spatial distribution data covering the canopy height range of the target area. S4. Based on UAV observation data, construct a spatially continuous concentration field and wind field within the canopy height range of the target area, and use the mass balance method to calculate the total canopy-scale emission flux of the target area during the corresponding observation period. S5. Using the total emission flux at the canopy scale and the greenhouse gas concentration distribution data obtained in step S4 as constraints, the total emission estimate of the target area based on ground observations obtained in step S2 is corrected, and the contribution ratio of each sub-source is further quantified. S6. Based on the calibrated contribution ratio of each sub-source and combined with the time series observation data of the ground observation system, the long-term emission flux of the target area is estimated; and the sub-source contribution benchmark is updated by periodic UAV observations to achieve dynamic and accurate monitoring of emission flux.

2. The air-ground coordinated greenhouse gas flux monitoring and source apportionment method according to claim 1, characterized in that, In step S1, the ground observation system is deployed as follows: Based on the prevailing wind direction of the target area, at least one ground observation station is set up in the upwind and downwind areas surrounding the target area. Each ground observation station is equipped with a greenhouse gas analyzer, a three-dimensional ultrasonic anemometer, and a temperature and humidity sensor, which are used to synchronously and continuously collect greenhouse gas concentration data, three-dimensional wind field data, and basic meteorological parameter data.

3. The air-ground coordinated greenhouse gas flux monitoring and source apportionment method according to claim 2, characterized in that, In step S2, based on the time-series observation data and combined with the backward Lagrange random diffusion model, the emission flux of each sub-source is estimated by inversion, and the fluxes of each sub-source are summed to obtain the total emission estimate of the target area based on ground observations; specifically: The greenhouse gas concentration data collected by the ground observation system and the three-dimensional wind field data collected by the three-dimensional ultrasonic anemometer are averaged according to a preset time window; the data of the observation point located upwind of the target area are selected, and the lowest greenhouse gas concentration value in that time window is determined as the background concentration value of the current time window. Friction velocity calculated based on raw 10Hz three-dimensional ultrasonic wind speed data. Obukhov length Surface roughness and average wind direction The key turbulence parameters are given by the following formulas: ； ； ； in, The instantaneous disturbance component of horizontal wind speed in the mean wind direction. This represents the instantaneous disturbance component of the vertical wind speed; the disturbance component is obtained by subtracting the average wind speed within the corresponding time window from the instantaneous wind speed. Kármán's constant, It is the acceleration due to gravity. For temperature, For sensible heat flux, air density, For isobaric specific heat capacity, The installation height of the three-dimensional ultrasonic anemometer. For horizontal wind speed, For stability correction function; After discarding data that do not meet the similarity theory assumptions or have anomalous boundary conditions, for each selected effective observation time window, a backward Lagrange random diffusion model is used, combined with the observed concentrations within that time window. Background concentration Based on the calculated key turbulence parameters and the predefined spatial geographic extent of each emission sub-source, the source can be inverted. Emission rate during this time window , is represented as: ； In the formula, The simulated concentration-to-source intensity ratio is determined by the source shape, atmospheric motion, and concentration observation location. During the simulation, the friction velocity... Obukhov length Surface roughness and average wind direction As input parameters to the model, they are used to characterize the atmospheric turbulence intensity, stability conditions, and wind field structure within the current time window, thereby determining the random motion trajectory and diffusion characteristics of Lagrange particles. The calculation formula is: ； in, The vertical velocity component of the Lagrange particles released by the model as they back towards the emission source region. The number of Lagrange particles released, set to 50,000 by default; By combining data from multiple effective observation time windows under different prevailing wind directions, the emission rate of each emission sub-source is calculated and retrieved within all relevant effective time windows. average This is used as the typical emission rate estimate for the N sub-sources; then, the typical emission rate estimates for all N sub-sources within the target area are summed to obtain the overall greenhouse gas emission rate estimate for the target area. The calculation formula is: 。 4. The air-ground coordinated greenhouse gas flux monitoring and source apportionment method according to claim 1, characterized in that, In step S3, the drone is equipped with a greenhouse gas analyzer and a three-dimensional ultrasonic anemometer, and is configured as follows: The observation mission is conducted under meteorological conditions of moderate wind speed and stable wind direction, requiring that the wind direction at that moment aligns the downwind direction of all potential emission sources with the same plane. The flight path is set to perform zigzag maneuvers within a vertical profile downwind of the target area and perpendicular to the prevailing wind direction. The vertical flight altitude range covers from above ground obstacles to 120 meters, and the horizontal flight range covers the downwind area of ​​all potential emission sources. If the flight observation is interrupted, the battery is replaced and the operation resumes at the point of interruption to obtain continuous and encrypted observation data for a complete profile.

5. The air-ground coordinated greenhouse gas flux monitoring and source apportionment method according to claim 4, characterized in that, In step S4, based on UAV observation data, a spatially continuous concentration field and wind field within the canopy height range of the target area are constructed, and the total canopy-scale emission flux of the target area during the corresponding observation period is calculated using the mass balance method; specifically: Flux estimation was performed using observational data from UAVs that captured the complete plume structure. Discrete greenhouse gas concentration data and three-dimensional wind speed data collected by the UAVs on an observation cross-section perpendicular to the prevailing wind direction were reconstructed into a continuous two-dimensional gridded concentration field and wind field using ordinary Kriging interpolation; the horizontal boundary of the observation cross-section... Covering the downwind extent of all potential emission sources, vertical boundary This refers to the maximum flight altitude from above ground obstacles to the preset maximum flight altitude. Based on the generated two-dimensional gridded concentration field and wind field, the mass balance formula is used to integrate the observed cross-section, which is expressed as: ； in, This represents the net greenhouse gas flux through the observation section; For grid points Dry mole fraction of greenhouse gases at the location; For background concentration, the average concentration observed from the upwind direction on the ground during the UAV flight period was used; For grid points The wind speed component perpendicular to the direction of the observation section; For grid points The air molar density at that location was calculated using air pressure data from nearby weather stations and temperature data observed at ground stations. The calculation formula is as follows: ； in, The air pressure was obtained from data from a nearby weather station. It is the ideal gas constant; The air temperature is derived from temperature data from ground stations. The calculated net greenhouse gas flux through the observation section Multiply by the molar mass of the target greenhouse gas Obtain the total emission flux at the canopy scale in the target region. , is represented as: 。 6. The air-ground coordinated greenhouse gas flux monitoring and source apportionment method according to claim 4, characterized in that, In step S5, the total emission flux at the canopy scale and the greenhouse gas concentration distribution data obtained in step S4 are used as constraints to correct the total emission estimate of the target area based on ground observations obtained in step S2, and the contribution ratio of each sub-source is further quantified; specifically: First, the inverse concentration field simulation is performed on the sub-source contribution structure obtained from the ground inversion model; When the simulation results and the greenhouse gas concentration distribution data from UAV observations are inconsistent in terms of emission height, diffusion pattern, or wind field conditions, it is determined that the contribution structure of the sub-source does not meet the cooperative constraint conditions, and the contribution ratio of each sub-source is corrected; when the two are consistent, the contribution structure of the sub-source is considered reasonable. When it is necessary to correct the contribution ratio of sub-sources, the total emission flux at the canopy scale of the target area for the corresponding observation period, obtained from UAV observation data, will be used. Compared with the total emission estimates of the target area obtained based on time series observation data Compare and calculate the correction factor. , is represented as: ； Subsequently, each sub-source obtained from the inversion will be... Emission rate during this time window Multiply by the correction factor The corrected sub-source emission rate was obtained. , is represented as: ； Then based on the corrected emission rates of each sub-source Calculate its contribution to the total flux.

7. The air-ground coordinated greenhouse gas flux monitoring and source apportionment method according to claim 6, characterized in that, When correcting the contribution ratio of sub-sources, it is also necessary to combine ground time-series observation data to judge the background rationality and representativeness of the observation period of the UAV observation data; specifically: The time series observation data is used to determine whether the concentration in the non-plume region observed by the UAV is within a reasonable background range. When the background concentration of the UAV observation results is abnormally high and it is determined that there may be other interference sources, the UAV observation is only used as reference information and is not used to correct the ground observation results, so as to ensure the accuracy and reliability of long-term emission flux estimation.

Images