地面PM2.5和气溶胶激光雷达联合变分同化对天津冬季一次重污染过程数值模拟改进研究

杨旭; 樊文雁; 蔡子颖; 朱玉强; 唐颖潇; 董琪如

PDF(2047 KB)

中国环境科学 ›› 2023, Vol. 43 ›› Issue (10) : 5088-5097.

大气污染与控制

地面PM_2.5和气溶胶激光雷达联合变分同化对天津冬季一次重污染过程数值模拟改进研究

杨旭^1,2, 樊文雁^1,2, 蔡子颖^1,2, 朱玉强^1,2, 唐颖潇^1,2, 董琪如³

作者信息 +

Improvement on numerical simulation by variational assimilation of surface PM_2.5 and aerosol lidar during a heavy pollution episode in winter in Tianjin

YANG Xu^1,2, FAN Wen-yan^1,2, CAI Zi-ying^1,2, ZHU Yu-qiang^1,2, TANG Ying-xiao^1,2, DONG Qi-ru³

Author information +

文章历史 +

摘要

针对输送型重污染过程，大气化学模式垂直模拟偏差较大，地面观测同化对污染垂直初始场改进不明显，高空输送模拟偏低的问题，以2023年1月5~6日天津地区一次重污染过程为例，通过PM_2.5地面观测和气溶胶激光雷达垂直观测联合应用，发展相应的气溶胶变分同化技术，改进大气化学模式污染垂直模拟效果，服务重污染天气预报预警和成因解析.研究表明：此次重污染过程由于大气化学模式垂直模拟能力不足，导致S1发展阶段地面PM_2.5增长速率和S2维持阶段的峰值浓度明显低估，垂直分布上污染层高度模拟偏低200~300m，白天输送阶段高空PM_2.5浓度高于地面、夜间稳定边界层以上污染高值特征均未能有效反映；通过同化地面PM_2.5观测有效实现PM_2.5空间分布模拟优化，过程期间上游城市石家庄和衡水的模拟误差分别下降35.3μg/m³和77.3μg/m³，天津地面PM_2.5质量浓度模拟误差下降39.7μg/m³，但地面同化对垂直分布模拟改进有限，尤其是对于输送型污染过程中PM_2.5垂直分布的复杂变化，基本无法准确反映；通过增加气溶胶激光雷达PM_2.5垂直观测同化，模式PM_2.5垂直模拟效果进一步提升，450m左右均方根误差减小幅度最大，约19.8μg/m³.在同化时刻及临近2~3h正效果明显，后续时效中S2阶段表现更优，夜间稳定边界层以上污染高值得到有效模拟，污染层高度模拟差异减小至50~100m，低于未同化CTRL试验（200~300m）和地面同化DA_SFC试验（100~200m）；S1阶段随着时效延长垂直模拟偏差明显增加，但同化的正作用依旧存在，污染层高度模拟误差比DA_SFC试验降低幅度不低于50m.

Abstract

For the transport-type heavy pollution episodes, the vertical simulation deviation of atmospheric chemical model was large, the ground observation assimilation could not significantly improve the vertical initial field of pollution, and the upper air transport was underestimated. A heavy pollution episode was taken as an example in Tianjin, which occurred from January 5 to 6, 2023, the aerosol variational assimilation technology was developed for the joint application of surface PM_2.5 and aerosol lidar vertical observations, to improve the vertical pollution simulation of atmospheric chemical model and serve forecast, early warning and mechanism analysis of heavy pollution episodes. The results showed that during the heavy pollution episode, growth rate and peak concentration of ground PM_2.5 was significantly underestimated, simulated pollution layer height was about 200~300m lower than observed value and characteristics such as higher PM_2.5 concentration in upper air than that on the ground in transportation phase of daytime and high pollution values above the stable boundary layer at night were not effectively reflected; Assimilation of surface PM_2.5 observations effectively optimized PM_2.5 spatial distribution, average simulation error decreased by 35.3μg/m³ and 77.3μg/m³ respectively in Shijiazhuang and Hengshui, the upstream cities and decreased by 39.7μg/m³ in Tianjin. However, its improvement on vertical distribution simulation was limited, especially for the complex change of PM_2.5 vertical distribution in transport-type pollution episode, which was not accurately reflected. By adding assimilation of vertical PM_2.5 observations obtained from aerosol lidar, vertical PM_2.5 simulation was further improved, the maximum reduction of PM_2.5 mass concentration RMSE was 19.8μg/m³ at about 500m. There was obvious positive effect at the assimilated time and within 2~3h, as for subsequent time, simulated pollution layer height was basically same with observation in S2 phase, the difference between them reduced to 50~100m, compared with 200~300m in CTRL experiment without assimilation and 100~200m in DA_SFC experiment with surface PM_2.5 assimilation. High value of pollution above stable boundary layer was effectively simulated and the simulation error of PM_2.5 mass concentration in the heavy polluted layer was significantly reduced; Assimilation effect in the daytime on January 5 was not good as that of S2 phase, accurate simulation of vertical PM_2.5 could be achieved in a short time, but the deviation of vertical simulation had been significantly enlarged again with simulation time increasing. In spite of this, positive effect still existed compared with DA_SFC and the simulation error of pollution layer height was reduced by at least 50m.

导出引用

杨旭, 樊文雁, 蔡子颖, 朱玉强, 唐颖潇, 董琪如. 地面PM_2.5和气溶胶激光雷达联合变分同化对天津冬季一次重污染过程数值模拟改进研究[J]. 中国环境科学. 2023, 43(10): 5088-5097

YANG Xu, FAN Wen-yan, CAI Zi-ying, ZHU Yu-qiang, TANG Ying-xiao, DONG Qi-ru. Improvement on numerical simulation by variational assimilation of surface PM_2.5 and aerosol lidar during a heavy pollution episode in winter in Tianjin[J]. China Environmental Science. 2023, 43(10): 5088-5097

中图分类号： X513

参考文献

[1] 韩素芹,冯银厂,边海,等.天津大气污染物日变化特征的WRF-Chem数值模拟[J]. 中国环境科学, 2008,28(9):828-832. Han S Q, Feng Y C, Bian H, et al. Numerical simulation of diurnal variation of major pollutants with WRF-Chem model in Tianjin[J]. China Environmental Science, 2008,28(9):828-832.
[2] 李珊珊,程念亮,徐峻,等.2014年京津冀地区PM_2.5浓度时空分布及来源模拟[J]. 中国环境科学, 2015,35(10):2908-2916. Li S S, Cheng N L, Xu J, et al. Spatial and temporal distributions and source simulation of PM_2.5 in Beijing-Tianjin-Hebei region in 2014[J]. China Environmental Science, 2015,35(10):2908-2916.
[3] Zhang L, Gong S, Zhao T, et al. Development of WRF/CUACE v1.0model and its preliminary application in simulating air quality in China[J]. Geoscientific Model Development, 2021,14(2):703-718.
[4] 崔应杰,王自发,朱江,等.空气质量数值模式预报中资料同化的初步研究[J]. 气候与环境研究, 2006,11(5):616-626. Cui Y J, Wang Z F, Zhu J, et al. A preliminary study on data assimilation for numerical air quality model prediction[J]. Climatic and Environmental Research (in Chinese), 2006,11(5):616-626.
[5] Tombette M, Mallet V, Sportisse B. PM₁₀ data assimilation over Europe with the optimal interpolation method[J]. Atmospheric Chemistry and Physics, 2009,9(1):57-70.
[6] Lin C, Wang Z, Zhu J. An Ensemble Kalman Filter for severe dust storm data assimilation over China[J]. Atmospheric Chemistry and Physics, 2008,8(11):2975-2983.
[7] Pagowski M, Grell G A. Experiments with the assimilation of fine aerosols using an ensemble Kalman filter[J]. J. Geophys. Res., 2012,117:D21302.
[8] Peng Z, Liu Z Q, Chen D, et al. Improving PM_2.5forecast over China by the joint adjustment of initial conditions and source emissions with an ensemble Kalman filter[J]. Atmospheric Chemistry and Physics, 2017, 17:4837-4855.
[9] Jiang Z Q, Liu Z Q, Wang T J, et al. Probing into the impact of 3DVAR assimilation of surface PM₁₀ observations over China using process analysis[J]. Journal of Geophysical Research:Atmospheres, 2013,118(12):6738-6749.
[10] Feng S Z, Jiang F, Jiang Z Q, et al. Impact of 3DVAR assimilation of surface PM_2.5 observations on PM_2.5 forecasts over China during wintertime[J]. Atmospheric Environment, 2018,187:34-49.
[11] 靳璐滨,臧增亮,潘晓滨,等.PM_2.5和PM_2.5~10资料同化及在南京青奥会期间的应用试验[J]. 中国环境科学, 2016,36(2):331-341. Jin L B, Zang Z L, Pan X B, et al. Data assimilation and application experiments of PM_2.5 and PM_2.5~10 during Nanjing Youth Olympic Games[J]. China Environmental Science, 2016,36(2):331-341.
[12] 陈杰,李正强,常文渊,等.基于改进GSI系统的气溶胶变分同化对WRF-Chem PM_2.5分析和预报的影响评估[J]. 大气与环境光学学报, 2020,15(5):321-333. Chen J, Li Z Q, Chang W Y, et al. Impact evaluation of aerosol variational assimilation based on improved GSI system on WRF-Chem PM_2.5 analysis and forecast[J]. J. Atmospheric and Environmental Optics, 2020,15(5):321-333.
[13] 蔡子颖,唐邈,肖致美,等.基于源反演和气溶胶同化方法天津空气质量模式预报能力改进[J]. 环境科学, 2022,43(5):2415-2426. Cai Z Y, Tang M, Xiao Z M, et al. Improvement of environmental model prediction based on inversion and aerosol assimilation[J]. Environmental Science, 2022,43(5):2415-2426.
[14] Zang Z L, Li Z J, Pan X B, et al. Aerosol data assimilation and forecasting experiments using aircraft and surface observations during CalNex[J]. Tellus B:Chemical and Physical Meteorology, 2016,68(1):29812.
[15] 张艳品,陈静,钤伟妙,等.石家庄冬季典型污染过程气溶胶激光雷达观测[J]. 中国环境科学, 2020,40(10):4205-4215. Zhang Y P, Chen J, Qian W M, et al. Aerosol lidar observation of typical pollution process in Shijiazhuang in Winter[J]. China Environmental Science, 2020,40(10):4205-4215.
[16] Yumimoto K, Uno I, Sugimoto N, et al. Adjoint inversion modeling of Asian dust emission using lidar observations[J]. Atmospheric Chemistry And Physics, 2008,8(11):2869-2884.
[17] Wang Y, Sartelet K N, Bocquet M, et al. Assimilation of ground versus lidar observations for PM₁₀ forecasting[J]. Atmospheric Chemistry and Physics, 2013,13:269-283.
[18] Wang Y, Sartelet K N, Bocquet M, et al. Assimilation of lidar signals:Application to aerosol forecasting in the western Mediterranean basin[J]. Atmospheric Chemistry and Physics, 2014b,14:12031-12053.
[19] 郑海涛.地面及激光雷达观测数据同化对PM_2.5预报的改进[D]. 合肥:中国科学技术大学, 2018. Zheng H T. Improvement of PM_2.5 Forecast by data assimilation of ground and Lidar observation[D]. Hefei:University of Science and Technology of China, 2018.
[20] Xiang Y, Zhang T S, Ma C Q, et al. Lidar vertical observation network and data assimilation reveal key processes driving the 3-D dynamic evolution of PM_2.5 concentrations over the North China Plain[J]. Atmospheric Chemistry and Physics, 21,7023-7037,2021.
[21] Cheng X H, Liu Y L, Xu X D, et al. Lidar data assimilation method based on CRTM and WRF-Chem models and its application in PM_2.5 forecasts in Beijing[J]. Science of the Total Environment, 2019, 682:541-552.
[22] Liang Y F, Zang Z L, Liu D, et al. Development of a three-dimensional variational assimilation system for lidar profile data based on a size-resolved aerosol model in WRF-Chem model v3.9.1and its application in PM_2.5 forecasts across China[J]. Geoscientific Model Development, 13,6285-6301,2020.
[23] 蔡子颖,杨旭,韩素芹,等.基于天气背景天津大气污染输送特征分析[J]. 环境科学, 2020,41(11):4855-4863. Cai Z Y, Yang X, Han S Q, et al. Transport characteristics of air pollution in Tianjin based on weather background[J]. Environmental Science, 2020,41(11):4855-4863.
[24] 王晓琦,郎建垒,程水源,等.京津冀及周边地区PM_2.5传输规律研究[J]. 中国环境科学, 2016,36(11):3211-3217. Wang X Q, Lang J L, Cheng S Y, et al. Study on transportation of PM_2.5 in Beijng-Tianjin-Hebei (BTH) and its surrounding area[J]. China Environmental Science, 2016,36(11):3211-3217.
[25] 杨旭,蔡子颖,韩素芹,等.基于无人机探空和数值模拟天津一次重污染过程分析[J]. 环境科学, 2021,42(1):9-18. Yang X, Cai Z Y, Han S Q, et al. Heavy pollution episode in Tianjin based on UAV meteorological sounding and numerical model[J]. Environmental Science, 2021,42(1):9-18.
[26] Li M., Liu H, Geng G, et al. Anthropogenic emission inventories in China:a review[J]. National Science Review, 2017,4:834-866.
[27] Li Z, Zang Z, Li Q, et al. A three-dimensional variational data assimilation system for multiple aerosol species with WRF/Chem and an application to PM_2.5 prediction[J]. Atmospheric Chemistry and Physics, 2013,13(8):4265-4278.
[28] 杨旭,唐颖潇,蔡子颖,等.基于气溶胶三维变分同化天津PM_2.5数值预报研究[J]. 中国环境科学, 2021,41(12):5476-5484. Yang X, Tang Y X, Cai Z Y, et al. Impact of aerosol data assimilation with 3-DVAR method on PM_2.5 forecast over Tianjin[J]. China Environmental Science, 2021,41(12):5476-5484.
[29] Lv L H, Liu W Q, Zhang T S, et al. Observations of particle extinction, PM_2.5 mass concentration profile and flux in north China based on mobile lidar technique[J]. Atmospheric Environment, 2017,164:360- 369.
[30] Tao Z M, Wang Z Z, Yang S J, et al. Profiling the PM_2.5 mass concentration vertical distribution in the boundary layer[J]. Atmospheric Measurement Techniques, 2016,9:1369-1376.