2016年秋季新乡市空气质量模式预报效果评估

Evaluation on Air Quality Forecasting Model in the Fall of 2016 in Xinxiang City,Henan Province

基于新乡市空气质量数值预报平台,采用相关系数(r)、标准化平均偏差(NMB)等统计指标,系统评估2016年秋季新乡市嵌套网格空气质量预报模式(NAQPMS)和通用多尺度空气质量模式(CMAQ)的预报效果,对比分析2套模式不同预报时效和不同水平分辨率的空气质量等级预报准确率。结果显示:2套模式均较好地表征了各主要污染物的浓度变化特征,2套模式的等级预报准确率高于60%,其中CMAQ对中度及重度的预报等级准确率达到70%。对比模式24、48、72 h3种预报时效效果,24 h预报时效的统计数据最优,说明24 h预报时效模拟结果可作为业务预报重要的支撑。

Based on air quality forecasting platform in Xinxiang city,two statistical indicators,correlation coefficient（ r） and Normal Mean Bias（ NMB）,were used to evaluate the simulation of air quality model NAQPMS and CMAQ in Autumn of 2016 in Xinxiang City. The accuracy of air quality level forecasts from two model were calculated as well. The results indicated that the models can performance well with 24 hour forecasts when comparing to observation. The accuracies of Air Quality Index（ AQI） from NAQPMS and CMAQ were both over 60%,and CMAQ performed better on moderately polluted and heavily polluted level,with the accuracy exceeding 70%. The performance of 24 hour forecasts were better than 48 and 72 hour forecasts. To improve the forecasting accuracy,it is suggesting that ensemble method for air quality forecasts could be applied for multi-models forecasting system.