土壤湿度产品对2021年3月华北强沙尘暴模拟结果的影响  

Impact of soil moisture products on the simulation results of super sandstorms during March of 2021 in North China

作　　者：丁涵泳 康汉青[1,2] 吕晶晶 Ding Hanyong;Kang Hanqing;Lv Jingjing(China Meteorological Administration Key Laboratory of Aerosol-Cloud and Precipitation,Nanjing University of Information Science and Technology,Nanjing 210044,China;China Meteorological Administration Key Laboratory of Atmospheric Chemistry,Chinese Academy of Meteorological Sciences,Beijing 100081,China)

机构地区：[1]南京信息工程大学中国气象局气溶胶与云降水重点开放实验室,江苏南京210044 [2]中国气象科学研究院中国气象局大气化学重点开放实验室,北京100081

出　　处：《中国沙漠》2024年第2期172-184,共13页Journal of Desert Research

基　　金：中国气象局大气化学重点开放实验室开放课题(2023B01)。

摘　　要：针对2021年3月华北两次强沙尘暴过程,研究了10款土壤湿度产品在沙尘源地(戈壁沙漠地区)的空间分布,并选择了其中4款(ERA5、 GLDAS、NCEP/FNL和SMAP L3)驱动WRF-CMAQ对两次沙尘暴过程进行模拟。最后结合110个受沙尘影响城市的实际观测结果,对4款产品的模拟效果进行评估,以研究其对两次沙尘过程模拟准确性和实用性的影响。结果表明:(1)仅基于卫星反演的产品,普遍存在缺测区域,且空间分布的异质性较强,4款产品中仅SMAP L3表现较好;“开放环”模型产品和经过卫星数据同化的模型产品在空间分布上均一性较强,质量较高。(2)在沙尘源地,ERA5、ERA5-Land和SMAP L3中的土壤湿度较低,AMSR2-JAXA、GLDAS、GLEAM、NCEP/FNL和SMAP L4中的土壤湿度相对较高。(3)第一次沙尘过程中4款产品对PM_(10)浓度的预测都有一定的低估;第二次沙尘过程中ERA5和SMAP L3高估而NCEP/FNL和GLDAS低估了统计站点的PM_(10)浓度;模拟值的高低与其所基于的沙尘源地土壤湿度均值呈负相关。在两次沙尘暴过程的模拟结果中,SMAP L3和ERA5两款产品表现较好,GLDAS次之,而NCEP/FNL产品造成了对沙尘浓度的明显低估。In the study of two severe sandstorms in North China during March of 2021,spatial distribution and temporal variations of 10 soil moisture products were examined,specifically in the sandstorm sources(Gobi Desert region).Four of these products(ERA5,GLDAS,NCEP/FNL,and SMAP L3)were employed to drive the WRF-CMAQ model for simulating these sandstorm events.The simulation results of these four products were assessed through comparison with actual observations from 110 cities impacted by the sandstorms,aiming to determine their accuracy and practicality in modeling these events.The findings reveal that:(1)Products solely based on satellite observations generally exhibited missing areas and strong spatial heterogeneity,with only SMAP L3 showing comparatively better performance.Both the"Open-loop"model products and satellite data assimilation model products demonstrated stronger uniformity in spatial distribution,and were comparatively of higher quality.(2)In the sandstorm source areas,ERA5,ERA5-Land,and SMAP L3 indicated relatively lower soil moisture,whereas AMSR2-JAXA,GLDAS,GLEAM,NCEP/FNL,and SMAP L4 showed higher soil moisture levels.(3)All four products slightly underestimated PM_(10) concentrations during the first sandstorm event.In the second event,ERA5 and SMAP L3 overestimated,while NCEP/FNL and GLDAS underestimated PM_(10) concentrations at statistical sites.The accuracy of the simulated values negatively correlated with the average soil moisture in the sandstorm source areas.Among the simulation results for both sandstorm events,the SMAP L3 and ERA5 products demonstrated superior performance,followed by GLDAS.However,the NCEP/FNL product led to a significant underestimation of sand dust concentration.

关 键 词：土壤湿度 沙尘暴 数值模拟 

分 类 号：P421[天文地球—大气科学及气象学]