大气科学  2017, Vol. 41 Issue (4): 882-896 PDF

Impacts of Different Concentrations of Anthropogenic Pollutants on a Rainstorm
YANG Taojin, LIU Yudi, SUI Min
Institute of Meteorology and Oceanography, PLA University of Science and Technology, Nanjing 211101
Abstract: The WRF-Chem model coupled with anthropogenic pollution sources was applied to investigate the impact of anthropogenic pollution on microphysical process and precipitation of a rainstorm occurred on 13 August 2011. Three different simulations including one normal experiment (Experiment Norm) and two extreme experiments (Experiments High and Low) according to the emission intensity of anthropogenic pollutants were conducted. The results showed that the initiation of precipitation was the same in all the three cases. In Experiment Low, the precipitation area remained unchanged while the intensity weakened at the precipitation center but enhanced in the surrounding area, and the intensity of precipitation overall enhanced (weakened) at the initial period (later period). In Experiment High, both the precipitation area and the intensity decreased. Changes in rain water and graupel were the major reason for the change in precipitation. The anthropogenic pollutants could influence the thermodynamic processes of atmosphere through their impacts on the microphysical process, and these changed atmospheric dynamic processes in turn affected microphysics and the growth of precipitable particles. All the above processes contributed to the precipitation change in the ground. The mechanism can be summarized as follows. The atmospheric heating rate decreased due to the decrease in water vapor condensation and ice depositional growth when the emission intensity of anthropogenic pollutants enhanced, and this would cause the decrease in convective activity and suppressed the growth of rain water and graupel, resulting in the decrease in precipitable particles and precipitation reduced eventually.
Key words: Intensity of anthropogenic pollution      Microphysical process      Precipitation      WRF-Chem model
1 引言

Li et al.（2011）Niu and Li（2011）运用10年地面观测和全球的A-Train卫星资料发现，燃烧产生的烟雾、自然的沙尘和城市排放的污染气溶胶对云和降水有很大的影响，对其他卫星资料如AVHRR（Rosenfeld and Lensky, 1998Rosenfeld et al., 2001）、MODIS（Koren et al., 2005）和TRMM（Rosenfeld, 1999, 2000）的研究也得出相似的结论。机载测量和现场等观测结果（Andreae et al，2004江琪等，2014陈思宇等，2012杨磊等，2013）的分析也显示不同气溶胶浓度下，云和降水会表现出不同的特征。Gao et al（2012）运用WRF-Chem模式耦合现在及工业化前的气溶胶排放研究了气溶胶对云的影响，并对气溶胶的增加引起的云滴数、云水路径和有效半径的变化进行了分析。Mashayekhi and Sloan（2014）的模拟结果显示，在高度城市化的大西洋沿岸地区，人为排放引起的非对流性降水增加了30%。肖辉和银燕（2011）模拟了污染气溶胶对山西一次降水过程的影响，结果显示降水区域没有明显变化，中心强度变强，在不同的时间段内降水的类型不同，降水的强度也发生变化。其他的研究结果（Lee et al., 2008; Li et al., 2009）显示，气溶胶的增加有利于对流的增强和降水的增加。Fan et al.（2015）提出了“气溶胶引起的条件性不稳定”机制解释了黑碳等吸收性气溶胶引起的四川盆地的一次大暴雨过程。

2 试验方案 2.1 模式设置

 图 1 模拟区域设置 Figure 1 The model domains

2.2 排放源的选择

2.3 气溶胶的作用方式

 $\frac{{\partial {N_k}}}{{\partial t}} = - {\left({\boldsymbol{V} \cdot \nabla N} \right)_k} + {D_k} - {C_k} - {E_k} + {S_k},$ (1)

2.4 方案设计

 图 2 北京及周边地区三个试验SO2的排放强度（单位：mol km-2 h-1）分布：（a）High、（b）Norm和（c）Low试验；（d）北京站、（e）香河站和（f）兴隆站观测和模拟的PM2.5浓度随时间的变化，其中黑线代表观测值（Obs），红线、绿线和蓝线分别代表试验High（数值为试验High的值的0.1倍）、试验Norm和试验Low的模拟值 Figure 2 The emission intensity (units: mol km-2 h-1) of SO2 for Experiments (a) High, (b) Norm, and (c) Low, respectively; observed and simulated PM2.5 concentrations at (d) Beijing, (e) Xianghe, and (f) Xinglon stations, respectively. The black line represents observations, the red is for Experiment High (multiplied by 0.1), and the green (blue) is for Experiment Norm (Low)
3 结果分析

3.1 人为气溶胶排放强度对降水的影响

 图 3 2011年8月13日12:00到14日00:00试验（a）Norm、（b）Low、（c）High和（d）观测（Obs）的12小时累积降水量以及（e）区域平均的总降水量随时间的变化 Figure 3 12-h accumulative precipitation simulated from Experiment (a) Norm, (b) Low, (c) High, and (d) observations (Obs), (e) time series of area averaged precipitation from 1200 BJT (Beijing time) 13 to 0000 BJT 14 August, 2011

 图 4 2011年8月13日12:00到14日00:00区域A中（a）12小时累积降水量各量级所占的比率和（b）平均小时降水量随时间的变化。（a）横轴表示12小时累积降水量，纵轴表示区域A中大于该降水量的区域在A中的比率，红线、黑线和蓝线分别表示试验High、试验Norm和试验Low Figure 4 (a) Percentage of each 12-h accumulative precipitation level in area A and (b) the averaged rain rate from 1200 BJT 13 to 0000 BJT 14 August 2011. The horizontal axis in (a) represents 12-h accumulative precipitation level and the vertical axis is the percentage for each level, and the red line represents Experiment High, the black (blue) line is for Experiment Norm (Low)

3.2 对微物理过程的影响

 图 5 2011年8月13日10:00到18:00区域A中区域平均过饱和度在0.1%的CCN数浓度：（a）试验High；（b）试验Norm；（c）试验Low Figure 5 The number concentration of CCN (Cloud Condensation Nuclei) at a supersaturation rate of 0.1% averaged over area A from 1000 BJT to 1800 BJT 13 August 2011: (a) Experiment High; (b) Experiment Norm; (c) Experiment Low

 图 6 2011年8月13日10:00到18:00区域A中区域平均过饱和度在0.1%的（a1、b1、c1）云滴数、（a2、b2、c2）云水混合比和（a3、b3、c3）云水自动转换率：试验High（左列）；试验Norm（中间列）；试验Low（右列） Figure 6 (a1, b1, c1) The number concentration of cloud drop, (a2, b2, c2) cloud mixing ratio, and (a3, b3, c3) auto-conversion rate of cloud water at a supersaturation rate of 0.1% averaged over area A from 1000 to 1800 13 August, 2011. The left column is for Experiment High, and the middle (right)column is for Experiment Norm (Low)

 图 7 2011年8月13日10:00到18:00区域A平均的（a1、b1、c1）雨水和（a2、b2、c2）霰含量的时间—高度剖面：试验Norm（左列）；试验High与试验Norm的差值（中间列）；试验Low与试验Norm的差值（右列） Figure 7 Time–height cross sections of (a1, b1, c1) rain water content and (a2, b2, c2) graupel content averaged over area A from 1000 BJT to 1800 BJT 13 August, 2011.The left column is for Experiment Norm, the middle (right) column is the difference between Experiment High (Low) and Experiment Norm

 ${P_{{\rm{tot}}}} = \left\{ \begin{array}{l} {P_{{\rm{SAUT}}}} + {P_{{\rm{SACI}}}} + {P_{{\rm{SACW}}}} + {P_{{\rm{SFI}}}} + {P_{{\rm{SFW}}}} + {P_{{\rm{SSUB}}}}(1 - {\delta _1}) + \\ {\rm{ }}{P_{{\rm{SDEP}}}}({\delta _1}) + {P_{{\rm{RAUT}}}} + {P_{{\rm{RACW}}}}{P_{{\rm{RACI}}}} + {P_{{\rm{REVP}}}}(1 - {\delta _1}) + \\ {\rm{ }}{P_{{\rm{GACW}}}} + {P_{{\rm{GACI}}}} + {P_{{\rm{GSUB}}}}(1 - {\delta _1}){\rm{ }}T < 0℃\\ {P_{{\rm{RAUT}}}} + {P_{{\rm{RACW}}}} + {P_{{\rm{SACW}}}} + {P_{{\rm{GACW}}}} + {P_{{\rm{REVP}}}}{\rm{ }}T \ge 0℃ \end{array} \right.,$ (2)

 图 8 同图 7a1–c1，但为可降水粒子总的产生率Ptot Figure 8 Same as Fig. 7a1–c1, but for the production rate of precipitable particles Ptot

 图 9 同图 7，但为PGACW（第一行）、PRACW（中间行）和PREVP（第三行） Figure 9 Same as Fig. 7, but for PGACW (top panels), PRACW (middle panels), and PREVP (bottom panels)

3.3 对动力过程的影响

 图 10 同图 7a1–c1，但为微物理过程导致的潜热加热率 Figure 10 Same as Fig. 7a1–c1, but for latent heating rate caused by microphysical process

 图 11 （a、b）水汽凝结和（c、d）冰晶凝华增长的变化：（a、c）试验High与试验Norm的差值；（b、d）试验Low与试验Norm的差值 Figure 11 Changes in (a, b) water vapor condensation and (c, d) ice depositional growth: (a, c) The difference between Experiment High and Experiment Norm; (b, d) the difference between Experiment Low and Experiment Norm

 图 12 同图 7a1–c1，但为区域A平均的上升速度 Figure 12 Same as Fig. 7a1–c1, but for the vertical velocity averaged over region A

4 结论

（1）不同排放强度下降水开始的时间没有变化，低排放情况下降水区域未发生变化，高排放情况下降水区域明显减小；此外，暴雨前期试验Low的降水增强，后期减弱，降水中心的强度减弱，降水更多的分散到周围区域；当人为气溶胶排放增大到10倍时，不论是降水强度还是各个量级的降水区域都是减少的。

（2）排放强度的变化使CCN浓度的变化超过两倍，由此导致的高排放情况下云滴数增多，云滴向雨滴的自动转换过程受到抑制；低排放情况下云滴数减少，自动转换过程增强。

（3）对水成物粒子的分析发现，雨水和霰含量的变化是导致降水发生变化的原因。高排放情况下，雨水和霰的含量减少，降水减弱；低排放情况下，降水前期，雨水和霰含量增加，降水增强，后期雨水和霰减少，降水减弱。

（4）通过对可降水粒子总的产生率、雨水和霰的增长过程、微物理加热率及上升速度的分析，得出不同排放强度的污染气溶胶对此次降水的影响机理。当污染气溶胶的排放强度增强，微物理过程中水汽凝结和冰晶凝华过程的减弱导致的总的大气加热减弱，可用于发展对流的能量减少，上升运动减弱，对流强度的减弱抑制了了雨水和霰收集云水的增长，导致了可降水粒子含量的减少，降水减弱。当排放减弱时，初期大气加热增强，上升运动更加强盛，加快了雨水和霰的生成，降水增多，后期大气加热减弱导致降水减弱。不同强度的污染气溶胶通过影响微物理过程影响大气的热力和动力过程，大气动力过程的变化反过来又通过微物理过程影响降水粒子的形成，从而影响地面降水。