基于WRF-NAQPMS的一次沙尘天气预报评估 |
Evaluation of a Dust Weather Forecast Performance Based on WRF-NAQPMS Model |
投稿时间:2022-09-20 修订日期:2023-11-18 |
DOI:10.19316/j.issn.1002-6002.2024.02.24 |
中文关键词: 沙尘天气预报 WRF NAQPMS 预报评估 PM10 |
英文关键词:dust forecast WRF NAQPMS evaluation PM10 |
基金项目:国家自然科学基金青年基金项目(42205083);干旱气象科学研究基金项目(IAM202207);甘肃省青年科技基金计划项目(21JR7RA691,21JR7RA501) |
|
通讯作者:孔祥伟* 中国气象局兰州干旱气象研究所, 甘肃 兰州 730020;兰州中心气象台, 甘肃 兰州 730020 |
摘要点击次数: 816 |
全文下载次数: 626 |
中文摘要: |
为提高沙尘天气的预报准确率,利用ECWMF再分析资料和近地面PM10小时质量浓度监测数据,评估WRFNAQPMS模式对2021年3月15—21日甘肃强沙尘过程的预报能力。结果表明,WRF-NAQPMS能够在一定程度上模拟此次污染过程:WRF对“3· 15”天气系统的模拟与实况整体趋势较为一致,随着预报时效延长,气象模拟场移动偏快,导致沙尘预报场发展偏快、沙尘二次传输影响下游时间提前;近地面风向的局地偏差是导致甘肃中东部地区沙尘浓度出现预报误差的主要气象因素。NAQPMS模式对PM10小时质量浓度的模拟随着预报时效增加和离沙源地距离的增大,预报误差逐步增大:在河西地区,沙尘影响时段和起沙浓度的模拟值均接近监测值,其中嘉峪关、酒泉、张掖的PM10小时质量浓度模拟值与监测值相关系数r>0.8;中部地区城市的沙尘影响时段预报略有偏差,且模拟值低于监测值;受复杂下垫面和气象场预报误差影响,省内其他地区沙尘预报结果参考性较低。 |
英文摘要: |
In order to improve the prediction accuracy of dust weather,using the ECWMF reanalysis data and the near-surface PM 10 hourly mass concentration monitoring data, the forecasting ability of WRF-NAQPMS model for the heavy dust process in Gansu Province from March 15 to 21,2021 was evaluated. The results showed that:WRF-NAQPMS could simulate the pollution process to a certain extent,it's simulation of the "3·15" dust process was consistent with the actual weather. With the extension of forecast time limit, the predicted meteorological field moved faster than the actual meteorological field, resulting in the rapid development of the dust forecast field and the advance of the time of the dust secondary transmission affecting downstream. The local deviation of near surface wind direction was the main meteorological factor leading to the prediction deviation of dust concentration in the Middle East of Gansu Province. During dust period by NAQPMS model,the prediction error of PM10 hour concentration increased gradually with the increase of aging and distance from the sand source. In Hexi area,the simulated values of the dust impact period and the sand-driven concentration were close to the monitoring values,and the correlation coefficient r> 0. 8 between the simulated value and the monitoring value of PM10 hour mass concentration in Jiayuguan,Jiuquan,and Zhangye. For the cities in the central region,the forecast of dust impact period was slightly biased,and the simulation value was lower than the monitoring value. In other areas of the province,due to the influence of complex underlying surface and meteorological field forecast error,the reference value of dust forecast results was low. |
查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|