Set Analysis And Optimization Of Meteorological Parameterization Schemes For Heavy Air Pollution Forecast

In recent years,the air pollution situation in China is severe,and the smog phenomenon with fine particulate matter(PM_2.5)as the main pollutant often occurs,especially in Beijing-Tianjin-Hebei region,where heavy pollution weather is more likely to occur in winter.Once air pollution occurs,it will not only endanger public safety and health,but also affect social and economic development.Air pollution control is an arduous and long-term task,and the country attaches great importance to it,and has issued“Air Pollution Prevention Action Plan”,and clearly requires that all regions need to establish an early warning system for heavily polluted weather.Effectively conducting air pollution forecasting and early warning work can help relevant departments to formulate prevention and control measures in a timely manner,reduce the negative effects of air pollution,and provide reference for major activities and public travel and health.It has a certain warning effect.Although the current heavy pollution weather warning system is relatively complete,there are still some situations in which the peak pollutant concentration in the heavy pollution forecast process and the generation and consumption nodes are difficult to grasp accurately.One of the factors leading to this situation is that there is a certain uncertainty in the air quality forecast model.This uncertainty not only comes from errors in the model itself.At the same time,the meteorological field as the input data of the air quality model system also affects the prediction results of the air quality model and is an important source of uncertainty in the air quality model.This study mainly combined and optimized microphysical processes,long-wave radiation processes,short-wave radiation processes,land surface processes,boundary layer processes,near-surface processes,and cumulus convection parameterization processes in WRF model.A total of 50 sets of parameterized schemes were designed to perform meteorological field simulation on the two pollution processes in Beijing-Tianjin-Hebei region and one pollution process in Beijing region,and use the meteorological field-driven air quality model(NAQPMS)to conduct pollutant simulation studies to explore the applicability of different parameterization scheme combinations to different pollution processes in different regions,and provide a certain reference for the prediction and warning of heavy pollution in the future.The simulation analysis of the two pollution processes in Beijing-Tianjin-Hebei region shows that for the simulated meteorological elements,there is a large uncertainty in the simulated temperature and relative humidity in the parameterization scheme during the two pollution processes.For the wind speed,the effects of the simulation results are small and the uncertainty is small.For the simulation of pollutants,from the simulation results of the first pollution process(November 23-27,2018),it can be seen that during the entire pollution period,the uncertainty of the long-wave radiation process and the boundary layer process is large,followed by the short-wave radiation process,while the microphysics process and the cumulus convection process have less uncertainty.From the simulation results of the second pollution process(November 28-December 3,2018),it can be seen that different parameterization schemes have different effects on the simulation results,especially for the period 2-3 during the heavy pollution period.The influence of parameterization schemes on the simulation results varies greatly,and the long-wave radiation process and the boundary layer process schemes still have the greatest influence.The maximum difference between the maximum and minimum schemes is about 300(?)g/m³.Therefore,when predicting the pollution process,the choice of parameterization schemes has a certain impact on the results,and it needs to be determined according to the characteristics of different pollution processes in different regions.On this basis,the article mainly focuses on a simulation analysis of a pollution process in Beijing(December 16-21,2016).The results show that the temperature simulation is the most sensitive to the parameterization scheme of the long-wave process,the set dispersion reaches 2.4-7.4?,which is again the parameterization scheme of the short-wave process;the relative humidity simulation is also the most sensitive to the parameterization scheme of the long-wave process,and again it is the land surface process;the sensitivity of wind speed simulation to different process parameterization schemes is not much different.Through statistical comparison of simulation results and observations,we have selected the scheme combination with the smallest simulation error as follows:Lin microphysics scheme,RRTMG long-wave scheme,RRTMG short-wave scheme,Tiedtke cumulus convection scheme,Noah land surface scheme,MYNN 3rd boundary layer scheme and MYNN near-surface scheme,and compare the best scheme to the ensemble mean and baseline scheme.For the ensemble mean,the correlation coefficient between temperature simulation and observation is 0.69,which is higher than the baseline scheme.The simulated deviation and root mean square error are 25%and 11%lower than the baseline scheme;the ensemble mean relative humidity and wind speed simulation are less variable than the baseline scheme.Compared with the ensemble mean,the best scheme can simultaneously improve the temperature,relative humidity and wind speed simulation,so that the temperature simulation deviation and root mean square error decreased by35%and 17%compared with the baseline scheme,and the relative humidity simulation deviation and root mean square error decreased by 43%and 13%,and the wind speed simulation deviation and root mean square error decreased by 33%and 24%.The above results show that the sensitivity test and optimization of the parameterization scheme can significantly reduce the simulation error of meteorological elements during heavy pollution.The improvement of heavy pollution prediction needs to focus on the uncertainty of the parametric scheme simulation.It is also found that the MYNN3rd boundary layer scheme has good performance in the meteorological element simulation of this heavy pollution process,which can provide a reference for the improvement of future heavy pollution forecasting.The meteorological field simulated by all physical parameterization schemes was used to drive the NAQPMS air quality model to simulate and analyze the polluted process.The results show that the combined disturbance optimization scheme shows high applicability.In the period 2 of the simulation of this pollution process,the combined disturbance optimization scheme processed three meteorological conditions conducive to the generation and accumulation of pollutants:low temperature,low wind speed,and high relative humidity.Therefore,the combined disturbance optimization scheme has higher applicability to the pollution process in Beijing this time,and the simulation effect is the best.