Small-scale greenhouse gas tracing method based on numerical simulation and data assimilation

Abstract

The invention discloses a small-scale greenhouse gas tracing method based on numerical simulation and data assimilation, and belongs to the technical field of greenhouse gas tracing. The method comprises the following steps: acquiring real-time meteorological monitoring data in a park, and performing cyclic assimilation; establishing a high-precision backward trajectory tracking system to obtain refined backward trajectory data of a monitoring station; establishing a high-precision greenhouse gas traceability calculation model to obtain greenhouse gas potential emission source distribution data; averaging the emission source distribution data to finally obtain potential emission source comprehensive probability distribution data; according to the invention, through a high-resolution numerical simulation and data assimilation technology, diffusion, transmission and chemical reaction processes of greenhouse gas in a small-scale range can be accurately simulated; meanwhile, the grid empirical weighting function of the traceability algorithm is optimized, the adverse effect of the low-track end point number on the final traceability result is effectively reduced, the robustness of the model is enhanced, and it is ensured that the traceability result is more accurate, reasonable and stable.

Description

Technical Field

[0001] The invention relates to a small-scale greenhouse gas source tracing method based on numerical simulation and data assimilation, and belongs to the technical field of greenhouse gas source tracing. Background Art

[0002] As the problem of climate change becomes increasingly serious, the emission of greenhouse gases, especially carbon dioxide (CO2) and methane (CH4), has become one of the major challenges for global environmental protection. These greenhouse gases mainly come from the burning of fossil fuels, agricultural activities, waste treatment and industrial production processes. Greenhouse gas emissions not only exacerbate global warming, but also have a profound impact on the climate system, ecological environment and human society.

[0003] The propagation of greenhouse gases in the atmosphere has significant spatial and temporal heterogeneity, especially at small scales. Gases such as CO2 and CH4 have a long lifespan in the atmosphere, but their emission sources are usually unevenly distributed. For example, cities, industrial parks, and agricultural areas are concentrated emission areas. These areas have dense greenhouse gas emissions, resulting in local high-concentration areas near the emission sources. In addition, the diffusion and propagation of greenhouse gases are also affected by many factors such as atmospheric conditions, topography, and wind speed, making their spatial distribution complex and diverse. Especially at small scales, this heterogeneity is more obvious, and the heterogeneity is not only reflected in the horizontal spatial distribution, but also in the vertical concentration gradient.

[0004] Due to the uneven spatial distribution of greenhouse gas emissions and the complex physical and chemical processes in the atmosphere, it is very challenging to trace the source of greenhouse gases on a small scale. On a small scale, the sources of greenhouse gas emissions may be diverse and dispersed, such as multiple factories, roads with high traffic flow, etc., resulting in the need to deal with multiple types of emission sources in the traceability process. Traditional greenhouse gas traceability methods usually rely on global or regional climate models and large-scale monitoring networks. These methods often have difficulty capturing detailed changes in greenhouse gas emissions on a small scale, resulting in inaccurate traceability results, which limits the timely response and effective prevention and control of greenhouse gas emissions. Summary of the invention

[0005] In order to improve the accuracy of greenhouse gas source tracing within a small scale, the present invention provides a small-scale greenhouse gas source tracing method based on numerical simulation and data assimilation. The technical solution is as follows:

[0006] The greenhouse gas source tracing method of the present invention comprises:

[0007] Step 1: Collect global meteorological reanalysis data, establish a meteorological data forecast system, and obtain real-time atmospheric data over the park;

[0008] Step 2: Acquire real-time meteorological monitoring data in the park, and perform cyclic assimilation on the real-time meteorological monitoring data to obtain high-precision real-time meteorological data near the ground in the park;

[0009] Step 3: Based on the location of the monitoring stations in the park and the high-precision historical meteorological data, use the HySPLIT software to establish a high-precision backward trajectory tracking system to obtain the refined backward trajectory data of the monitoring stations;

[0010] Step 4: Based on the greenhouse gas concentration data time series and the refined backward trajectory data, the PSCF, CWT, CFA, RTWC, and SQTBA algorithms are used to establish a high-precision greenhouse gas source tracing calculation model to obtain the distribution data of potential greenhouse gas emission sources;

[0011] Step 5: Use the Max-Min normalization algorithm to normalize the distribution data of the five potential greenhouse gas emission sources, and then average the data to finally obtain the comprehensive probability distribution data of the potential greenhouse gas emission sources;

[0012] The empirical weight function W of the grid (i, j) in each algorithm in step 4 i,j The value criteria are as follows:

[0013]

[0014] Where n i,j is the number of trajectory endpoints in the grid (i, j), n ave is the average number of trajectory endpoints of the grid, α is a scaling parameter that controls the sensitivity of the weight coefficient change. The larger the α value, the faster the weight coefficient changes. β is a balance parameter used to adjust the balance of the weight coefficient.

[0015] Optionally, the step 1 includes:

[0016] Step 11: Collect global meteorological reanalysis data;

[0017] Step 12: Use the WRF-ARW numerical forecast model, take the global meteorological reanalysis data as initial data, and adopt multi-layer nested grids to establish an automatic forecast system for meteorological data covering the park;

[0018] Step 13: Organize and verify the meteorological data output by the meteorological data automatic forecasting system, and archive it into a historical meteorological database.

[0019] Optionally, the step 2 includes:

[0020] Step 21: Collect meteorological data in the park in real time through automated meteorological instruments and sensors;

[0021] Step 22: Perform quality control on the collected data, organize all observation data by site and time, and convert the data format into a standard format that can be read by the model;

[0022] Step 23: Based on the 3DVAR three-dimensional variational data assimilation model, a meteorological data assimilation system is established to cyclically assimilate the ground meteorological data of the monitoring stations in the park to obtain high-precision real-time meteorological data near the ground in the park.

[0023] Optionally, in step 4, greenhouse gas monitoring micro stations are used to collect greenhouse gas concentration data, and the greenhouse gas monitoring micro stations cover the entire park and are evenly distributed.

[0024] Optionally, the scaling parameter α=0.7, and the balancing parameter β=2.

[0025] The present invention provides a greenhouse gas source tracing system, comprising: a data acquisition module and a source tracing module;

[0026] The data acquisition module includes: a greenhouse gas monitoring microstation for collecting real-time meteorological data and real-time greenhouse gas concentration data;

[0027] The traceability module includes:

[0028] The meteorological data forecast module is configured to collect global meteorological reanalysis data and use the WRF-ARW numerical forecast model to obtain real-time atmospheric data over the park;

[0029] A meteorological data acquisition module is configured to acquire real-time meteorological data collected by the data acquisition module;

[0030] The backward trajectory tracking module is configured to obtain refined backward trajectory data of the monitoring stations based on the locations of the monitoring microstations in the park and high-precision historical meteorological data using the HySPLIT software;

[0031] The greenhouse gas source tracing module is configured to establish a high-precision greenhouse gas source tracing calculation model based on the greenhouse gas concentration data time series, combined with refined backward trajectory data, using PSCF, CWT, CFA, RTWC, and SQTBA algorithms to obtain the distribution data of potential greenhouse gas emission sources;

[0032] The emission source location module is configured to use the Max-Min normalization algorithm to normalize the distribution data of the potential emission sources of five greenhouse gases, and then average the data to finally obtain the comprehensive probability distribution data of the potential emission sources of greenhouse gases;

[0033] The empirical weight function W of the grid (i, j) in each algorithm in the greenhouse gas tracing module i,j The value criteria are as follows:

[0034]

[0035] Where n i,j is the number of trajectory endpoints in the grid (i, j), n ave is the average number of trajectory endpoints of the grid, α is a scaling parameter that controls the sensitivity of the weight coefficient change. The larger the α value, the faster the weight coefficient changes. β is a balance parameter used to adjust the balance of the weight coefficient.

[0036] Optionally, the greenhouse gas monitoring microstations cover the entire park and are evenly distributed.

[0037] Optionally, the scaling parameter α=0.7, and the balancing parameter β=2.

[0038] The present invention provides a greenhouse gas source tracing device, comprising a memory and a processor;

[0039] The memory is used to store computer programs;

[0040] The processor is used to implement the greenhouse gas source tracing method as described in any one of the above items when executing the computer program.

[0041] The present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the greenhouse gas source tracing method as described in any one of the above items is implemented.

[0042] The beneficial effects of the present invention are:

[0043] The present invention can accurately simulate the diffusion, transmission and chemical reaction processes of greenhouse gases in a small scale by combining high-resolution numerical simulation with data assimilation technology of real-time observation data. High-precision numerical simulation and real-time observation data assimilation system complement each other in small-scale greenhouse gas source tracing. High-precision numerical simulation provides detailed meteorological background and physical process description, while the data assimilation system ensures that the model can reflect the latest observation data in real time through real-time correction and data integration, so that the source tracing model can respond to actual changes more quickly and accurately, significantly improving the accuracy and real-time performance of the source tracing model; at the same time, through real-time adjustment and integration of multi-source data, the output uncertainty of the source tracing model is significantly reduced, especially when dealing with complex terrain and multi-source emission environments in small-scale areas. This advantage is particularly obvious, making small-scale greenhouse gas source tracing more accurate and efficient, and better able to cope with complex local environments and multi-source emission challenges.

[0044] The present invention combines five algorithms, namely, PSCF, CWT, CFA, RTWC and SQTBA, comprehensively considers different spatial and temporal scales, reduces the errors that may be caused by a single algorithm, optimizes the grid empirical weight function of the algorithm, introduces a standard Sigmoid function, makes the value of the weight coefficient a function with the number of trajectory endpoints and the average number of grid trajectory endpoints as independent variables, can provide delicate weight adjustment when the value changes slightly, and avoids the discontinuity problem caused by the sudden change of the weight coefficient; introduces adjustable parameters α and β, provides a flexible adjustment mechanism, so that the weight coefficient can be optimized according to the specific application scenario; the present invention optimizes the weight coefficient by taking the number of trajectory endpoints n as the number of trajectory endpoints n. i,j and weight coefficient W i,j The relationship between is designed as a nonlinear function. Compared with the traditional weight coefficient calculation method, it effectively reduces the adverse impact of the low number of trajectory endpoints on the final tracing results, enhances the robustness of the model, and ensures that the tracing results are more reasonable and stable. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0046] Figure 1 It is a flow chart of the small-scale greenhouse gas source tracing method based on numerical simulation and data assimilation of the present invention.

[0047] Figure 2 This is a layout diagram of greenhouse gas monitoring microstations in a park in the second embodiment of the present invention.

[0048] Figure 3 It is a comprehensive probability distribution diagram of potential greenhouse gas emission sources in Example 2 of the present invention. DETAILED DESCRIPTION

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0050] Embodiment 1:

[0051] This embodiment provides a greenhouse gas source tracing method, characterized in that the method includes:

[0052] Step 1: Collect global meteorological reanalysis data, establish a meteorological data forecast system, and obtain real-time atmospheric data over the park;

[0053] Step 2: Acquire real-time meteorological monitoring data in the park, and perform cyclic assimilation on the real-time meteorological monitoring data to obtain high-precision real-time meteorological data near the ground in the park;

[0054] Step 3: Based on the location of the monitoring stations in the park and the high-precision historical meteorological data, use the HySPLIT software to establish a high-precision backward trajectory tracking system to obtain the refined backward trajectory data of the monitoring stations;

[0055] Step 4: Based on the greenhouse gas concentration data time series and the refined backward trajectory data, the PSCF, CWT, CFA, RTWC, and SQTBA algorithms are used to establish a high-precision greenhouse gas source tracing calculation model to obtain the distribution data of potential greenhouse gas emission sources;

[0056] Step 5: Use the Max-Min normalization algorithm to normalize the distribution data of the five potential greenhouse gas emission sources, and then average the data to finally obtain the comprehensive probability distribution data of the potential greenhouse gas emission sources;

[0057] The empirical weight function W of the grid (i, j) in each algorithm in step 4 i,j The value criteria are as follows:

[0058]

[0059] Where n i,j is the number of trajectory endpoints in the grid (i, j), n ave is the average number of trajectory endpoints of the grid, α is a scaling parameter that controls the sensitivity of the weight coefficient change. The larger the α value, the faster the weight coefficient changes. β is a balance parameter used to adjust the balance of the weight coefficient.

[0060] Embodiment 2:

[0061] This example provides a small-scale greenhouse gas source tracing method based on numerical simulation and data assimilation. Figure 1 , the method comprising:

[0062] Step S1: Collect global meteorological reanalysis data, establish a meteorological data forecast system, and obtain real-time atmospheric data over the park. Specifically, it includes the following steps:

[0063] Step S1.1: Collect global meteorological reanalysis data. Global reanalysis meteorological data is obtained by integrating and reanalyzing a variety of observation data around the world. It integrates data from a variety of observation platforms, including meteorological satellites, ground meteorological stations, buoys, and aircraft soundings. It can more comprehensively reflect the meteorological conditions of the earth's atmosphere, oceans, and land, and provide detailed global meteorological field information.

[0064] Step S1.2: Use the WRF-ARW numerical forecast model, take the global meteorological reanalysis data as the initial data, and adopt multi-layer nested grids to establish an automatic forecast system for meteorological data covering the park.

[0065] The WRF (Weather Research and Forecasting Model) model is a numerical weather forecast model widely used in meteorology and climate research. It has flexible horizontal and vertical grid settings and can adapt to the simulation needs of different geographical regions and meteorological events. ARW (Advanced Research WRF) is the numerical forecast module of the WRF model. The ARW equation group uses a fully compressible non-static Euler method, the vertical coordinate system uses terrain-following coordinates, the vertical layer uses a sigma-distributed pressure layer, and the top layer of the model is set as a constant pressure surface; the horizontal grid uses the Arakawa-C grid; the time integration scheme uses the third-order Runge-Kutta scheme; and the spatial discretization uses the 2nd to 6th order schemes. The parameter configuration used in this model makes it particularly effective in numerical simulations of mesoscale and small-scale areas.

[0066] Step S1.3: Sort and verify the meteorological data output by the system and archive it in the historical meteorological database. The forecast system uses a four-layer bidirectional nested grid, and the innermost horizontal resolution reaches 100m×100m, covering the entire park and its surrounding areas, providing high-precision meteorological data over the park. The model automatically makes a forecast once a day, and each forecast obtains 24 hours of weather data for the day, and archives the historical data.

[0067] Step S2: Collect real-time observation data from ground monitoring stations within the park, establish a meteorological data assimilation system, cyclically assimilate meteorological observation data, and obtain high-precision real-time meteorological data near the ground within the park.

[0068] The ground monitoring stations in this embodiment are greenhouse gas monitoring micro-stations, which are evenly distributed in the park according to grid distribution. The distance between micro-stations is 1 km to ensure the spatial coverage of the monitoring points and the uniformity of the data. Figure 2 As shown in the figure, each microstation has the ability to monitor the concentration of greenhouse gases such as CO2 and CH4, and is equipped with sensors to measure meteorological parameters such as wind speed, wind direction, temperature, and humidity; the microstation collects data in real time and transmits the data to the central database through a wireless network for use by the subsequent data assimilation system.

[0069] Data collection and assimilation specifically include the following steps:

[0070] Step S2.1: Obtain real-time meteorological monitoring data from all monitoring microstations in the park, including wind speed, wind direction, temperature, relative humidity, air pressure, etc. The monitoring microstations in the park include standard air stations and air microstations. Both stations can measure near-ground meteorological data, including key meteorological parameters such as wind speed, wind direction, temperature, relative humidity, and air pressure; these data can be collected in real time through automated meteorological instruments and sensors, and transmitted to the central data management system through wireless or wired networks; ensure that the frequency of data collection is high enough to accurately capture changes in meteorological conditions in the park.

[0071] Step S2.2: Perform quality control on the observation data, organize all observation data by site and time, and convert the data format into a standard format that can be read by the model. The acquired observation data must undergo quality control to ensure the accuracy and reliability of the data: first, check the data integrity and remove missing values ​​and outliers; then, perform data cleaning, classify and organize all observation data by site and time; next, convert the data format into a standard format that can be read by the model and write it into the ob.ascii file. This step ensures that the data meets the requirements of the data assimilation model input, and the consistency and accuracy of the data are guaranteed.

[0072] Step S2.3: Based on the 3DVAR three-dimensional variational data assimilation model, a meteorological data assimilation system is established to cyclically assimilate the ground meteorological data of the monitoring stations in the park to obtain high-precision real-time meteorological data near the ground in the park. The 3DVAR (Three-Dimensional Variational) assimilation method can significantly and efficiently improve the model's forecasting capabilities.

[0073] The 3DVAR model uses observation data and background meteorological fields for data assimilation to optimize the initial conditions in the meteorological model. The specific operations include: first, input the cleaned observation data into the 3DVAR model, and calculate the deviation of the observation data by comparing the observation data with the model forecast data. Then, adjust the model initial field through the optimization algorithm to reduce the deviation, thereby improving the accuracy of the model forecast. Finally, the assimilation process is repeated to integrate the continuously updated observation data into the meteorological model to generate high-precision real-time meteorological data near the ground in the park.

[0074] Real-time data assimilation of observation data within the park is the key to achieving high-precision numerical simulation. Numerical models often produce certain errors due to the uncertainty of initial conditions and boundary conditions. Data assimilation technology integrates real-time observation data (such as temperature, wind speed, air pressure, etc.) into the numerical model, dynamically adjusts the model output, and promptly corrects the model error to make the simulation results more in line with the actual situation. Especially in complex local environments, through the data assimilation system, the model can continuously receive the latest observation data and update meteorological data in real time, thereby providing more real-time and accurate prediction and traceability results. This is critical for responding to sudden emission events or rapidly changing meteorological conditions. Small-scale traceability has extremely high accuracy requirements. The data assimilation system integrates data from high-density monitoring stations and combines these data with the output of numerical simulations, greatly enhancing the ability of the traceability model to capture greenhouse gas diffusion paths and concentration changes, and significantly improving the accuracy of traceability.

[0075] Step S3: Based on the location of the monitoring micro-station in the park and the high-precision historical meteorological data obtained in step S2, use HySPLIT software to establish a high-precision backward trajectory tracking system to obtain the refined backward trajectory data of the monitoring station. Specifically, it includes the following steps:

[0076] Step S3.1: Analyze the specific monitoring data of the greenhouse gas monitoring station, determine the specific backward trajectory analysis period, and obtain high-precision real-time meteorological data for the required area and time period based on the specific location information of the monitoring station. First, conduct an in-depth analysis of the specific monitoring data of the greenhouse gas monitoring station: these data include information such as the concentration and composition of greenhouse gases at different time points. Through careful study of these data, it can be determined which time periods are of great significance for backward trajectory analysis; for example, if there is a significant change or abnormal value in the greenhouse gas concentration in a certain time period, then this time period may be selected as a key analysis period. At the same time, based on the specific location information of the monitoring station, obtain high-precision real-time meteorological data for the required area and time period: according to the specific latitude and longitude range of the study area, combined with the near-ground and high-altitude meteorological data obtained in step S2, obtain high-precision meteorological data within the specific park area, and the time period corresponds to the previously determined backward trajectory analysis period; the meteorological data is uniformly distributed grid data, specifically including wind speed, wind direction, temperature, air pressure and other meteorological elements.

[0077] Step S3.2: Based on the above data, use the HySPLIT model to establish a high-precision backward trajectory tracking system, take the monitoring site location as the starting point of the trajectory, perform backward trajectory tracking, and obtain refined backward trajectory data of the greenhouse gas emission receptor site. The HySPLIT model is a computational model used for airflow trajectory simulation and air quality research. The model is based on the Lagrangian method and combines the physical and chemical mechanisms of atmospheric transport and diffusion processes. It considers the influence of factors such as wind field, topography, atmospheric boundary layer and meteorological conditions on airflow movement. It can simulate the diffusion, transportation and deposition process of greenhouse gases in the atmosphere. It is widely used in research and application in the fields of environmental protection, meteorology, climatology, etc. Taking the monitoring site location as the starting point of the trajectory, start the backward trajectory tracking. The model will perform backward integration of greenhouse gases at the receptor site based on meteorological information such as wind direction and wind speed, track the possible source path of the airflow, and obtain refined backward trajectory data of the greenhouse gas emission receptor site.

[0078] Step S4: Based on the greenhouse gas concentration data time series and the refined backward trajectory data obtained in step S3, use the PSCF, CWT, CFA, RTWC, and SQTBA algorithms to establish a high-precision greenhouse gas source tracing calculation model to obtain the distribution data of potential greenhouse gas emission sources. Specifically, the following steps are included:

[0079] Step S4.1: Based on the refined backward trajectory data calculated in step S4, obtain the time series of greenhouse gas concentration data of the receptor site for the corresponding period: According to the starting time of the backward trajectory, select the greenhouse gas concentration data of the receptor site in the form of a time series; perform quality control and format standardization on the data in advance to make it correspond one-to-one with the trajectory data, so as to facilitate subsequent calculation and processing.

[0080] Step S4.2: Use PSCF, CWT, CFA, RTWC, and SQTBA algorithms to establish a high-precision greenhouse gas source tracing calculation model and obtain the distribution data of potential greenhouse gas emission sources. The calculation formulas of each algorithm are as follows:

[0081] The calculation formula of the PSCF method is:

[0082]

[0083] Where n i,j is the number of all trajectory endpoints passing through the grid (i, j), m i,j is the number of endpoints of greenhouse gas emission trajectories passing through grid (i, j), W i,j is the empirical weight function of the grid (i, j).

[0084] The calculation formula of the CWT method is:

[0085]

[0086] Where L is the total number of trajectories, C l is the concentration value observed by trajectory l at the target point, t i,j,l is the time it takes for the trajectory l to pass through the grid (i, j), W i,j is the empirical weight function of the grid (i, j).

[0087] The calculation formula of the CFA method is:

[0088]

[0089] Where L is the total number of trajectories, C l is the concentration value observed by trajectory l at the target point, t i,j,l is the time it takes for the trajectory l to pass through the grid (i, j), W i,j is the empirical weight function of the grid (i, j).

[0090] The calculation method of the RTWC method is:

[0091]

[0092] In the formula, C l,t is the redistributed concentration value of trajectory l in period t, c l is the monitored concentration value of trajectory l, N l is the total time experienced by the trajectory l (the total number of endpoints of the trajectory). When the trajectory l passes through the grid (i, j) in the t period:

[0093]

[0094] CF i,j The calculation method is as follows:

[0095]

[0096] In the formula, is the CF value of the grid (i, j) at the kth (k ≥ 1) iteration, L is the total number of trajectories, τ i,j,l,t is the number of endpoints of the grid (i, j) that the trajectory l passes through in the time period t. In the first iteration, that is, when k = 1, take When CF between two successful iterations i,j When the average deviation is less than 0.5%, the iteration ends.

[0097]

[0098] The final RTWC calculation formula is:

[0099]

[0100] Where W i,j is the empirical weight function of the grid (i, j).

[0101] The calculation method of SQTBA method is:

[0102]

[0103] Where Q(x,y,t|x′,y′,t′) is the transmission probability density function of the trajectory, that is, the probability that a point (x′,y′) on the trajectory at time t′ appears at the position (x,y) at the corresponding time t, where x,y are the coordinates of the grid center, x′,y′ are the coordinates of the corresponding point on the trajectory, and σ x and σ y is the standard deviation of the trajectory x and y in both directions (assuming it grows over time):

[0104] σ x (t′)=σ y (t′) = at′

[0105] Where a is the diffusion speed, which is 5.4 km / h.

[0106]

[0107] The above formula is the backward diffusion potential distribution function of trajectory l, which is obtained by backward integration of Q(x,y,t|x′,y′,t′) along the trajectory, and τ is the total duration of trajectory l.

[0108] The final SQTBA calculation formula is as follows:

[0109]

[0110] In the formula, c l is the monitored concentration value of trajectory l, L is the total number of trajectories, W i,j is the empirical weight function of the grid (i, j).

[0111] In the above algorithm, the empirical weight function W of the grid (i, j) is i,j The value criteria are as follows:

[0112]

[0113] Where n i,j is the number of trajectory endpoints in the grid (i, j), n ave is the average number of track endpoints of the grid, α is a scaling parameter that controls the sensitivity of the weight coefficient change. The larger the α value, the faster the weight coefficient changes. β is a balance parameter used to adjust the balance of the weight coefficient. In small-scale tracing, α = 0.7 and β = 2 are usually taken.

[0114] In order to meet the needs of high-precision traceability results within a small scale, it is usually necessary to use a high-resolution horizontal grid; however, in this case, the number of grids may exceed the total number of trajectory endpoints, resulting in the average number of trajectory endpoints in the grid being less than 1, causing calculation errors. The traditional method of determining the weight coefficient of the traceability algorithm has problems in the case of small-scale traceability, which can easily lead to excessively large weight coefficients, thereby affecting the credibility of the traceability results. To solve this problem, the present invention improves the method for determining the value of the weight coefficient: specifically, by introducing the standard Sigmoid function, the value of the weight coefficient is determined in terms of n. i,j and n ave is a function of the independent variable, and its value range is always between 0 and 1; when n i,j Relatively small (much smaller than n ave When), the weight coefficient W i,j The value is close to 0; when n i,j Increase and gradually exceed n ave When the weight coefficient W i,j The value of gradually increases and approaches 1. Among them, α is the scaling parameter, which controls the sensitivity of the weight coefficient change. By adjusting α, W can be controlled i,j For example, a larger value of α makes W i,j In n i,j Close to n ave The weight coefficient grows rapidly when the value is α, while a smaller value of α makes the weight coefficient grow more slowly; β is a balance parameter used to adjust the balance of the weight coefficient. By adjusting β, the starting point of the weight coefficient growth can be controlled. It has been verified in practice that α=0.7 and β=2 are usually taken in small-scale traceability.

[0115] Step S5: Use the Max-Min normalization algorithm to normalize the distribution data of the five potential greenhouse gas emission sources obtained in step S4, and then average the data to finally obtain the comprehensive probability distribution data of the potential greenhouse gas emission sources. Specifically, the following steps are included:

[0116] Step S5.1: Perform Max-Min standardization on the obtained data of the five potential greenhouse gas emission sources, and linearly normalize the data to a specified interval. The Max-Min standardization formula is:

[0117]

[0118] Where, X norm is the standardized data, X is the data to be processed, and X max and X min are the maximum and minimum values ​​of the data set respectively;

[0119] Step S5.2: average the normalized data of the five potential greenhouse gas emission sources to obtain a comprehensive probability distribution of the potential greenhouse gas emission sources;

[0120] Step S5.3: Visualize the obtained comprehensive probability distribution data, see Figure 3 The spatial distribution characteristics of potential greenhouse gas emission sources are displayed in the form of a scattered pixel distribution map. The depth of color represents different probability levels of potential emission sources, which helps users quickly understand the distribution of emission sources and provides intuitive and accurate distribution information of potential greenhouse gas emission sources for environmental protection departments and decision makers.

[0121] In this embodiment, the high-precision numerical simulation and real-time observation data assimilation system complement each other in small-scale greenhouse gas tracing. High-precision numerical simulation provides a detailed description of the meteorological background and physical process, while the data assimilation system ensures that the model can reflect the latest observation data in real time through real-time correction and data integration, so that the tracing model can respond to actual changes more quickly and accurately, significantly improving the accuracy and real-time performance of the tracing model. At the same time, through real-time adjustment and integration of multi-source data, the output uncertainty of the tracing model is significantly reduced, especially when dealing with complex terrain and multi-source emission environments in small-scale areas. This advantage is particularly obvious, making small-scale greenhouse gas tracing more accurate and efficient, and better able to cope with complex local environments and multi-source emission challenges.

[0122] Traditional weight coefficient calculation methods may use simple linear relationships or fixed thresholds. This embodiment introduces the Sigmoid function to smoothly process weight changes. The Sigmoid function has the characteristics of an S-shaped curve, which can provide delicate weight adjustments when the value changes are small, avoiding the discontinuity problem caused by sudden changes in weight coefficients. At the same time, the Sigmoid function expands the weight change from linear to nonlinear, so that the weight changes with the number of trajectory endpoints n. i,j The change is more sensitive, especially near the mean value n ave When , the weight changes more naturally and smoothly.

[0123] At the same time, this embodiment naturally ensures the weight coefficient W through the definition of the Sigmoid function i,j The value range of is between 0 and 1, and no additional correction or constraint conditions are required. This design avoids the problem that in traditional methods, additional conditions may need to be introduced to ensure that the weight coefficient is within a reasonable range, making the calculation process simpler and ensuring the physical meaning of the result (that is, the weight coefficient will not exceed the reasonable physical range).

[0124] In addition, this embodiment provides a flexible adjustment mechanism by introducing adjustable parameters α and β, so that the weight coefficient can be optimized according to the specific application scenario. The α parameter controls the sensitivity of the weight coefficient change, allowing the speed of weight change to be adjusted in different environments; the β parameter provides an initial adjustment to the weight distribution, so that the method can adapt to different average trajectory endpoint number distributions. This dual-parameter control mechanism provides a high degree of flexibility for weight calculation, which is a major improvement over the traditional fixed formula.

[0125] In small-scale tracing, the average number of trajectory endpoints n ave The traditional linear method may cause some grid weights to be too high, thus affecting the accuracy of the traceability results. i,j and weight coefficient W i,j The relationship between is designed as a nonlinear function, which effectively reduces the adverse impact of the low number of trajectory endpoints on the final tracing results, enhances the robustness of the model, and ensures that the tracing results are more reasonable and stable.

[0126] Some steps in the embodiments of the present invention may be implemented using software, and the corresponding software program may be stored in a readable storage medium, such as a CD or a hard disk.

[0127] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A greenhouse gas source tracing method, characterized in that: The method comprises: Step 1: Collect global meteorological reanalysis data, establish a meteorological data forecast system, and obtain real-time atmospheric data over the park; Step 2: Acquire real-time meteorological monitoring data in the park, and perform cyclic assimilation on the real-time meteorological monitoring data to obtain high-precision real-time meteorological data near the ground in the park; Step 3: Based on the location of the monitoring stations in the park and the high-precision historical meteorological data, use the HySPLIT software to establish a high-precision backward trajectory tracking system to obtain the refined backward trajectory data of the monitoring stations; Step 4: Based on the greenhouse gas concentration data time series and the refined backward trajectory data, the PSCF, CWT, CFA, RTWC, and SQTBA algorithms are used to establish a high-precision greenhouse gas source tracing calculation model to obtain the distribution data of potential greenhouse gas emission sources; Step 5: Use the Max-Min normalization algorithm to normalize the distribution data of the five potential greenhouse gas emission sources, and then average the data to finally obtain the comprehensive probability distribution data of the potential greenhouse gas emission sources; The empirical weight function W of the grid (i, j) in each algorithm in step 4 i,j The value criteria are as follows: Where n i,j is the number of trajectory endpoints in the grid (i, j), n ave is the average number of trajectory endpoints of the grid, α is a scaling parameter that controls the sensitivity of the weight coefficient change. The larger the α value, the faster the weight coefficient changes. β is a balance parameter used to adjust the balance of the weight coefficient.

2. The greenhouse gas source tracing method according to claim 1, characterized in that: The step 1 comprises: Step 11: Collect global meteorological reanalysis data; Step 12: Use the WRF-ARW numerical forecast model, take the global meteorological reanalysis data as initial data, and adopt multi-layer nested grids to establish an automatic forecast system for meteorological data covering the park; Step 13: Organize and verify the meteorological data output by the meteorological data automatic forecasting system, and archive it into a historical meteorological database.

3. The greenhouse gas source tracing method according to claim 1, characterized in that: The step 2 comprises: Step 21: Collect meteorological data in the park in real time through automated meteorological instruments and sensors; Step 22: Perform quality control on the collected data, organize all observation data by site and time, and convert the data format into a standard format that can be read by the model; Step 23: Based on the 3DVAR three-dimensional variational data assimilation model, a meteorological data assimilation system is established to cyclically assimilate the ground meteorological data of the monitoring stations in the park to obtain high-precision real-time meteorological data near the ground in the park.

4. The greenhouse gas source tracing method according to claim 1, characterized in that: In step 4, greenhouse gas monitoring microstations are used to collect greenhouse gas concentration data. The greenhouse gas monitoring microstations cover the entire park and are evenly distributed.

5. The greenhouse gas source tracing method according to claim 1, characterized in that: The scaling parameter α=0.7 and the balancing parameter β=2.

6. A greenhouse gas source tracing system, characterized in that: The system comprises: a data acquisition module and a traceability module; The data acquisition module includes: a greenhouse gas monitoring microstation for collecting real-time meteorological data and real-time greenhouse gas concentration data; The traceability module includes: The meteorological data forecast module is configured to collect global meteorological reanalysis data and use the WRF-ARW numerical forecast model to obtain real-time atmospheric data over the park; A meteorological data acquisition module is configured to acquire real-time meteorological data collected by the data acquisition module; The backward trajectory tracking module is configured to obtain refined backward trajectory data of the monitoring stations based on the locations of the monitoring microstations in the park and high-precision historical meteorological data using the HySPLIT software; The greenhouse gas source tracing module is configured to establish a high-precision greenhouse gas source tracing calculation model based on the greenhouse gas concentration data time series, combined with refined backward trajectory data, using PSCF, CWT, CFA, RTWC, and SQTBA algorithms to obtain the distribution data of potential greenhouse gas emission sources; The emission source location module is configured to use the Max-Min normalization algorithm to normalize the distribution data of the potential emission sources of five greenhouse gases, and then average the data to finally obtain the comprehensive probability distribution data of the potential emission sources of greenhouse gases; The empirical weight function W of the grid (i, j) in each algorithm in the greenhouse gas tracing module i,j The value criteria are as follows: Where n i,j is the number of trajectory endpoints in the grid (i, j), n ave is the average number of trajectory endpoints of the grid, α is a scaling parameter that controls the sensitivity of the weight coefficient change. The larger the α value, the faster the weight coefficient changes. β is a balance parameter used to adjust the balance of the weight coefficient.

7. The greenhouse gas source tracing system according to claim 6, characterized in that: The greenhouse gas monitoring microstations cover the entire park and are evenly distributed.

8. The greenhouse gas source tracing system according to claim 6, characterized in that: The scaling parameter α=0.7 and the balancing parameter β=2.

9. A greenhouse gas source tracing device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the greenhouse gas source tracing method according to any one of claims 1 to 5 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the greenhouse gas source tracing method according to any one of claims 1 to 5 is implemented.

Images