FORECASTING PERFORMANCE EVALUATION OF GRACES IN GUANGDONG PROVINCE
-
摘要: 利用2016-2019年广东省国控站实况监测数据对华南区域大气成分数值模式系统(GRACEs)预报性能进行了综合评估。除空气质量指数AQI外，重点对PM2.5、O3及NO2进行了分析评估。(1) 模式预报性能存在年际差异，对各要素的预报值总体偏低。(2) 模式预报能较好地反映空气质量的空间分布，PM2.5中心在珠三角西北部，而O3-8 h高值区在珠三角核心区和粤东沿海，但模式对O3-8 h高值区存在显著预报低估现象。(3)模式可较好地模拟出PM2.5月变化的单峰型特征和O3-8 h月变化双峰型特征，但模式对AQI的秋冬季主峰值和春季次峰值的预报存在低估，分别与模式对O3-8 h、PM2.5的低估有关。(4) 模式能较好体现O3的午后峰值和NO2双峰值的日变化规律；模式对O3前体物NO2的预报偏差，有可能是导致随后几小时对O3浓度预报偏差的重要原因。(5) 日平均浓度预报效果检验显示模式可较好预测AQI和3种污染物的变化趋势，但对夏秋季高O3-8 h浓度预报显著偏低；模式对轻度污染及以上等级预报能力偏低，亟需提升模式对污染天气的预报能力。Abstract: Based on the observed data from national environment stations in Guangdong Province during 2016-2019, comprehensive evaluation is conducted on the forecasting performance of the Guangzhou Regional Atmospheric Composition and Environment Forecasting System (GRACEs). Apart from Air Quality Index (AQI), PM2.5, O3 and NO2 forecasting is also evaluated. The results are as follows: (1) There were inter-annual differences in the forecasting performance of GRACEs and the forecasted value of each parameter was universally low. (2) The model was skillful in forecasting the spatial distribution of air quality. The high concentration center of PM2.5 was in the northwest of the Pearl River Delta. The region of high O3-8 h was located in the core area of the Pearl River Delta and the eastern coast of Guangdong, though there was a significant underestimation in the high concentration region of O3-8 h. (3) In terms of monthly variation, GRACEs exhibited acceptable performance for the single-peak variation of PM2.5 and the bimodal variation of O3-8 h, but the main peak of AQI in autumn and the second peak in spring were underestimated, which were related to the underestimation of O3-8 h and PM2.5, respectively. (4) In terms of diurnal variation, GRACEs showed good capability in capturing the peak of O3 in the afternoon and the double peak of NO2. The forecast deviation of NO2, a precursor of O3, probably led to the forecast deviation of O3 concentration in the following hours. (5) The trend of the daily average concentration of AQI and 3 pollutants predicted by GRACEs was in accordance with that observed, but O3-8 h was significantly underestimated in summer and autumn. The capacity of GRACEs to forecast light pollution and above was low, so it is urgent to develop the model.
-
Key words:
- air quality /
- atmospheric composition and environment forecasting system /
- GRACEs /
- PM2.5 /
- O3 /
- forecast verification
-
图 3 2016-2019年广东省102个环保国控站AQI、PM2.5、O3-8 h、NO2的观测值(a1，b1，c1，d1)、预报值(a2，b2，c2，d2)及两者差值(预报减观测)(a3，b3，c3，d3)的分布图
表 1 2016-2019年模式和观测数据相同时段的小时值样本数
年份 2016 2017 2018 2019 24 h 7 642 7 706 7 757 6 828 48 h 7 488 7 715 7 768 6 823 72 h 7 158 7 722 7 797 6 803 
下载: 导出CSV
表 2 2016-2019年模式和观测数据相同时段的日值样本数
年份 2016 2017 2018 2019 24 h 333 328 337 298 48 h 325 328 335 298 72 h 310 328 338 297
下载: 导出CSV
表 3 2016-2019年模式和观测数据相同时段的月值样本数
年份 2016 2017 2018 2019 24 h 12 12 12 12 48 h 12 12 12 12 72 h 12 12 12 12
下载: 导出CSV
表 3 2016-2019年广东省GRACEs模式预报结果检验
要素 24 h预报 48 h预报 72 h预报 RMSE MB NMB RMSE MB NMB RMSE MB NMB AQI 19.0 -8.4 -8.7% 20.1 -7.1 -6.3% 21.5 -6.7 -5.0% PM2.5 13.2 -5.1 -13.4% 14.3 -5.5 -14.9% 15.8 -6.7 -17.4% O3-8 h 29.6 -13.6 -7.1% 30.6 -9.3 -0.4% 32.6 -5.7 5.4% NO2 9.9 -1.7 1.2% 11.0 -2.6 -1.7% 12.0 -3.5 -4.8%
下载: 导出CSV
-
[1] LIN C Q, LI Y, YUAN Z B, et al. Using satellite remote sensing data to estimate the high-resolution distribution of ground-level PM2.5[J]. Remote Sensing of Environment, 2015, 156: 117-128. [2] 张远航, 郑君瑜, 陈长虹, 等. 中国大气臭氧污染防治蓝皮书(2020年)[M]. 北京: 中国环境科学学会臭氧污染控制专业委员会, 2020: 15-16、20-35. [3] NAM K, ZHANG X, ZHONG M, et al. Health effects of Ozone and particulate matter pollution in China: A province-level CGE analysis[J]. Annals of Regional Science, 2019, 63(2): 269-293. [4] LIU H, LIU S, XUE B, et al. Ground-level ozone pollution and its health impacts in China[J]. Atmos Environ, 2018, 173: 223-230. [5] WILSON S R, MADRONICH S, LONGSTRETH J D, et al. Interactive effects of changing stratospheric ozone and climate on tropospheric composition and air quality, and the consequences for human and ecosystem health[J]. Photochemical and Photobiological Sciences, 2019, DOI: 10.1039/c8pp90064g. [6] 陈昊. 2019年广东21地市PM2.5首次全部达标[N]. 北京: 中国环境报, 2020: 4. [7] 吴乃庚, 邓玉娇, 向昆仑, 等. 2019年广东省生态气象监测公报[Z]. 广州: 广东省气象局, 2020: 8. [8] 邓雪娇, 李菲, 吴兑, 等. 广州地区典型清洁与污染过程的大气湍流与物质交换特征[J]. 中国环境科学, 2011, 31(9): 1 424-1 430. [9] 刘建, 吴兑, 范绍佳, 等. 前体物与气象因子对珠江三角洲臭氧污染的影响[J]. 中国环境科学, 2017, 37(3): 813-820. [10] 毛敏娟, 杜荣光, 齐冰. 浙江省大气扩散能力的时空分布特征[J]. 热带气象学报, 2019, 35(5): 644-651. [11] 李明华, 甘泉, 曹静, 等. 惠州市臭氧污染特征及其与气象条件关系研究[J]. 热带气象学报, 2019, 35(3): 324-331. [12] 齐艳杰, 于世杰, 杨健, 等. 河南省臭氧污染特征与气象因子影响分析[J]. 环境科学, 2020, 41(2): 587-599. [13] KALBARCZYK R, SOBOLEWSKI R, KALBARCZYK E. Biometeorological determinants of the tropospheric ozone concentration in the suburban conditions of Wroclaw, Poland[J]. Journal of Elementology, 2016, 21(3): 729-744. [14] LEE K J, KAHNG H, KIM S B, et al. Improving environmental sustainability by characterizing spatial and temporal concentrations of ozone [J]. Sustainability, 2018, 10(12): 4551. [15] 蒲义良, 吴斯敏, 叶朗明, 等. 江门市城区臭氧浓度变化特征及气象影响因素分析[J]. 热带气象学报, 2020, 36(5): 650-659. [16] WANG N, LYU X, DENG X, et al. Aggravating O3 pollution due to NOx emission control in eastern China[J]. Science of the Total Environment, 2019, 677: 732-744. [17] 严茹莎. 德州市夏季臭氧敏感性特征及减排方案[J]. 环境科学, 2020, 41(9): 3 961-3 968. [18] 李冬青, 吴其重, 徐旗, 等. 不同分辨率CMAQ模式系统对北京PM2.5预报效果研究[J]. 环境科学学报, 2020, 40(5): 1 587-1 593. [19] 许艳玲, 易爱华, 薛文博. 基于模型模拟的成都市PM2.5污染来源解析[J]. 环境科学, 2020, 41(1): 50-56. [20] 麦健华, 于玲玲, 邓涛, 等. 基于GRAPES-CMAQ的中山市空气质量预报系统预报效果评估[J]. 热带气象学报, 2018, 34(1): 78-86. [21] 王茜, 吴剑斌, 林燕芬. CMAQ模式及其修正技术在上海市PM2.5预报中的应用检验[J]. 环境科学学报, 2015, 35(6): 1 651-1 656. [22] 谢敏, 钟流举, 陈焕盛, 等. CMAQ模式及其修正预报在珠三角区域的应用检验[J]. 环境科学与技术, 2012, 35(2): 102-107. [23] 邓涛, 邓雪娇, 吴兑, 等. 珠三角灰霾数值预报模式与业务运行评估[J]. 气象科技进展, 2012, 2(6): 38-44. [24] 薛纪善, 庄世宇, 朱国富, 等. GRAPES新一代全球/区域变分同化系统研究[J]. 科学通报, 2008, 53(20): 2 408-2 417. [25] 陈德辉, 薛纪善, 沈学顺, 等. 我国自主研制的全球/区域一体化数值天气预报系统GRAPES的应用与展望[J]. 中国工程科学, 2012, 14 (9): 46-54. [26] 邓雪娇, 邓涛, 麦博儒, 等. 华南区域大气成分业务数值预报GRACEs模式系统[J]. 热带气象学报, 2016, 32(6): 900-907. [27] 中国气象局预报与网络司. 全国环境气象数值模式预报质量检验评估办法(试行)[Z]. 北京: 中国气象局, 2015: 1-6. [28] 宋烺, 邓涛, 吴兑, 等. 广州地区典型灰霾过程及不同天气类型下边界层高度研究[J]. 环境科学学报, 2019, 39(5): 13-23. [29] 黄俊, 廖碧婷, 吴兑, 等. 广州近地面臭氧浓度特征及气象影响分析[J]. 环境科学学报, 2018, 38(1): 23-31. [30] 姚青, 马志强, 林伟立, 等. 天津夏季边界层低层大气中PAN和O3的输送特征分析[J]. 环境科学, 2019, 40(1): 67-75. [31] 周永宏, 郑大伟. 相关系数检验表的新算法[J]. 中国科学院上海天文台年刊, 1997, 18: 18-23. -
点击查看大图
图(9) / 表(5)
计量
- 文章访问数: 235
- HTML全文浏览量: 23
- PDF下载量: 15
- 被引次数: 0
粤公网安备 4401069904700003号