Sulfur dioxide emission source optimization method based on mode initial field assimilation

Abstract

The invention discloses a sulfur dioxide emission source optimization method based on mode initial field assimilation, and belongs to the field of atmospheric pollution numerical simulation, and the method comprises the following steps: 1, obtaining sulfur dioxide ground observation data, meteorological observation data, meteorological reanalysis grid point data and an emission source counted by abottom-to-top method; 2, assimilating the sulfur dioxide ground observation data based on an atmospheric chemical mode and a three-dimensional variational (3Dvar) assimilation method to obtain an optimal sulfur dioxide concentration field and a sulfur dioxide prediction error; and 3, converting the sulfur dioxide prediction error into a sulfur dioxide emission source error so as to obtain an optimal sulfur dioxide emission source, and verifying the optimal sulfur dioxide emission source.

Description

technical field [0001] The invention relates to the technical field of air quality model forecasting, relates to a method for optimizing sulfur dioxide emission sources based on model initial field assimilation, and belongs to the field of numerical simulation of air pollution. Background technique [0002] Sulfur dioxide is one of the main air pollutants, and it is also an important content of the air quality model forecast. However, due to the influence of factors such as the initial field of the model, the description of dynamic processes such as sedimentation and diffusion, the description of aerosol chemical reaction processes, and the accuracy of sulfur dioxide emission sources, etc., There are still large errors in sulfur dioxide forecasting. [0003] Among the above factors, the uncertainty of emission sources is considered to be the main factor affecting the accuracy of SO2 forecast. Improving the accuracy of emission sources, especially the accuracy of total emiss...

AI Technical Summary

Problems solved by technology

[0002] Sulfur dioxide is one of the main air pollutants, and it is also an important content of the air quality model forecast. However, due to the influence of factors such as the initial field of the model, the description of dynamic processes such as sedimentation and diffusion, the description of aerosol chemical reaction processes, and the accuracy of sulfur dioxide emission sources, etc., Sulfur dioxide forecast still has large errors

[0005] There are two disadvantages in using traditional methods to make emission sources. One is that the production process requires a lot of manpower and material resources. With the rapid economic and social development, the actual emission sources change rapidly, but it is difficult to update them quickly and effectively.

Second, there are great uncertainties in the process of three-dimensional gridding and time division of emission sources. Especially in terms of time division, the emission sources counted by the "bottom-up" method only includes the total daily emissions and does not Including hourly change information, it is necessary to assume the hourly factor of the emission source (the proportion of the hourly emission to the daily emission) to achieve time distinction, thus introducing more errors to the emission source

[0006] At present, there are two methods for improving emission sources through data assimilation, one is the ensemble Kalman filter method, but this method requires a large number of ensemble members, and each member is a large-scale numerical forecast simulation, so this method is used for The inversion of emission sources has very high requirements on computing resources and computer performance, and it is difficult to meet the needs of large-scale and long-term simulations

The second is the four-dimensional variational method (4Dvar). By establishing an adjoint model (inverse modeling), the optimization and adjustment of emission sources can be realized. However, 4Dvar also requires a large amount of computing resources, and the adjoint model is complicated to write, and maintenance and upgrades are extremely difficult. Difficult, there will be many restrictions in application

Images