设为首页
加入收藏
Decision Tree Forecasting Models of Sea Fog for the Coast of Guangdong Province
-
摘要: 利用汕头、珠海和湛江地面观测站2000—2008年1—5月的海雾历史观测资料和NCEP/NCAR FNL再分析资料，采用分类与回归树 (CART) 方法对海雾及其生成前24 h的海洋气象条件进行分类分析，建立了海雾决策树预报模型；并根据现有的海雾理论认识，对海雾预报规则包含的物理意义进行讨论。10次交叉检验的结果表明：采用CART方法建立的海雾决策树预报模型有较好的预报性能，对广东沿岸海雾的预报准确率总体上可达到73%以上。根据决策树预报模型建立的海雾预报判别流程，可在业务工作中直接用于有雾/无雾判别。海雾预报判别流程同时也具有明确的物理意义，能够较好地反映水汽与海表冷却条件对平流冷却雾形成的重要性，CART方法可作为海雾业务预报的有效建模工具。Abstract: Sea fog is a phenomenon of water vapor condensation or sublimation in marine atmospheric boundary layer and is also one of the main disastrous weathers on the coast of Guangdong Province in spring. However, there is no suitable method for operational sea fog forecasting in Guangdong due to the complexity of physical processes involved in the formation of sea fog. Therefore, historical sea fog reports from Shantou, Zhuhai and Zhanjiang surface meteorological observation and NCEP/NCAR FNL reanalysis for the period of 2000—2008 are analyzed to explore the feasibility of sea fog forecasting with a 24-hour lead time. The relationship between marine atmospheric conditions and sea fog events is examined by Classification and Regression Trees (CART), employing the NCEP/NCAR reanalysis data 24 hours before the sea fog events. Then, the decision tree models for sea fog forecasting are developed based on results of classification analysis. Finally, the physical significance of the forecasting rules is discussed based on existing theoretical knowledge on sea fog.The validation results by 10 cross-validation show that the forecasting accuracy of sea fog decision tree models developed by CART can reach 83.7%, 73.7% and 82.4% respectively for Shantou, Zhuhai and Zhanjiang on the coast of Guangdong Province. It can be interpreted or understood easily due to the clear logical relationship. The decision-making procedure can be developed and used directly to make fog/no-fog identification in operational sea fog forecasting with clear physical meanings. It also reflects the importance of the water vapor and the cooling effect of cold sea surface in the formation of advective cooling fog well. Simple calculation processes and relatively high classification accuracy make the CART an effective tool to develop sea fog forecasting model.
-
图 1 地面观测站与预报参考点地理位置图
Fig. 1 The location of the surface observation stations and forecasting reference points
图 2 汕头、珠海和湛江的决策树节点数与相对成本关系
Fig. 2 The relationship between relative costs and the number of nodes for the decision trees of Shantou, Zhuhai and Zhanjiang
图 5 湛江海雾判别流程图
Fig. 5 The decision-making procedure of sea fog forecasting for Zhanjiang
表 1 用于CART方法的预报变量
Table 1 The predictor variables used in CART analysis
序号 预报变量名 海洋气象要素 1 Tsea 海表温度, 单位：℃ 2 T2 m 2 m高度气温, 单位：℃ 3 D2 m 2 m高度露点温度, 单位：℃ 4 α10 m 10 m高度风向, 单位：(°) 5 V10 m 10 m高度风速, 单位：m/s 6 T1000 1000 hPa高度气温, 单位：℃ 7 D1000 1000 hPa高度露点温度, 单位：℃ 8 α1000 1000 hPa高度风向, 单位：(°) 9 V1000 1000 hPa高度风速, 单位：m/s 10 T850 850 hPa高度气温, 单位：℃ 11 D850 850 hPa高度露点温度, 单位：℃ 12 α850 850 hPa高度风向, 单位：(°) 13 V850 850 hPa高度风速, 单位：m/s 14 T2 m-Tsea 气温—海表温度差, 单位：℃ 15 D2 m-Tsea 露点—海表温度差, 单位：℃ 
下载: 导出CSV
表 2 决策树的预报准确率检验结果
Table 2 The testing results for the classification/forecasting accuracy of sea fog decision trees
站点 类别 样本数 训练误分率/% 验证误分率/% 训练成功率/% 验证成功率/% 汕头 0 2267 31.8 32.2 68.2 67.8 1 110 13.7 17.3 86.3 83.7 珠海 0 4480 18.9 21.4 81.1 78.6 1 133 18.8 26.3 81.2 73.7 湛江 0 5031 26.8 28.5 74.2 71.5 1 412 13.4 17.6 86.6 82.4
下载: 导出CSV
-
[1] 王彬华.海雾.北京:海洋出版社, 1983. [2] 刘小宁, 张洪政, 李庆祥, 等.我国大雾的气候特征及变化初步解释.应用气象学报, 2005, 16(2): 220-230. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=20050227&flag=1 [3] 张苏平, 鲍献文.近十年中国海雾研究进展.中国海洋大学学报, 2008, 38: 359-366. http://www.cnki.com.cn/Article/CJFDTOTAL-QDHY200803004.htm [4] 张春桂, 蔡义勇, 张加春. MODIS遥感数据在我国台湾海峡海雾监测中的应用.应用气象学报, 2009, 20(1):8-16. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=20090102&flag=1 [5] Lewis J M, Koracin D, Redmond K T. Sea fog research in the United Kingdom and United States, a historical essay including outlook. Bull Amer Meteor Soc, 2004, 75:395-408. http://connection.ebscohost.com/c/articles/12650863/sea-fog-research-united-kingdom-united-states-historical-essay-including-outlook [6] Edson J, Crawford T, Crescenti J, et al. The coupled boundary layers and air-sea transfer experiment in low winds. Bull Amer Meteor Soc, 2007, 88:341-356. doi: 10.1175/BAMS-88-3-341 [7] Taylor W. Mesoscale Case Study: Project Haar. Met Office Met. O.11 Tech, Note 239, 1987. [8] 傅刚, 张涛, 周发琇, 等.一次黄海海雾的三维数值模拟研究.青岛海洋大学学报 (自然科学版), 2002, 32:859-867. http://www.cnki.com.cn/Article/CJFDTOTAL-QDHY200206001.htm [9] Gao Shanhong, Lin Hang, Shen Biao, et al. A heavy sea fog event over the Yellow Sea in March 2005: Analysis and numerical modeling. Adv Atmos Sci, 2007, 24:65-81. doi: 10.1007/s00376-007-0065-2 [10] 周发琇.海雾:第四讲海雾预报.海洋预报, 1986, 6: 71-78. http://www.cnki.com.cn/Article/CJFDTOTAL-HYYB198904011.htm [11] 胡基福, 郭可彩, 鄢利农.应用模式输出统计作海雾出现判别预报.青岛海洋大学学报, 1996, 4:439-445. http://www.cnki.com.cn/Article/CJFDTOTAL-QDHY604.007.htm [12] 王厚广, 曲维政.青岛地区的海雾预报.海洋预报, 1997, 14(3):53-57. http://www.cnki.com.cn/Article/CJFDTOTAL-HYYB703.007.htm [13] 于润玲, 王亚男, 李永平.黄海和东海海雾短期客观预报业务系统.大气科学研究与应用, 2007(2): 28-36. http://www.cnki.com.cn/Article/CJFDTOTAL-DQTY200702006.htm [14] 刘科峰, 张韧, 江海英, 等.模糊逻辑仿真建模及其在青岛海雾分析中的应用.海洋湖沼通报, 2004(4):17-25. http://www.cnki.com.cn/Article/CJFDTOTAL-HYFB200404004.htm [15] 朱云清, 于世永.青岛近海海雾的数学预报方法.海洋预报, 1991, 8 (3):68-71. doi: 10.11737/j.issn.1003-0239.1991.03.012 [16] 李法然, 周之栩, 陈卫锋, 等.湖州市大雾天气的成因分析及预报研究.应用气象学报, 2004, 16(6): 794-803. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=200506103&flag=1 [17] 梅珏.人工神经元网络在辐射雾预报中的应用.应用气象学报, 1999, 10(4): 511-512. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=199904106&flag=1 [18] Steinberg D, Colla P. CART: Tree-structured Nonparametric Data Analysis. San Diego, CA, Salford Systems, 1995. [19] Lewis D M. Forecasting Advective Sea Fog with the Use of Classification and Regression Tree Analyses for Kunsan Air Base. AFIT/GM/ENP/04-08, Physics Graduate School of Engineering and Management, Air Force Institute of Technology, Air University, 2004. [20] Wantuch F. Visibility and fog forecasting based on decision tree method. IDOJARAS, 2003, 105:29-38. https://www.researchgate.net/publication/228420283_Visibility_and_fog_forecasting_based_on_decision_tree_method [21] 梁军, 张胜军, 隋洪起, 等.大连地区大雾特征.应用气象学报, 2009, 20(1):28-35. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=20090104&flag=1 [22] Breiman L, Friedman J H, Olshen R A, et al. Classification Regression Trees. Wadsworth, Belmont, CA, 1984. [23] 周发琇, 刘龙太.长江口及济州岛邻近海域综合调查报告 (第七节海雾).山东海洋学院学报, 1986, 16(1):115-131. http://kns.cnki.net/KCMS/detail/detail.aspx?filename=qdhy1989s1016&dbname=CJFD&dbcode=CJFQ [24] Zhang Suping, Xie Shangping, Liu Qinyu, et al. Seasonal variations of Huanghai sea fog: Observations and mechanisms. J Climate, 2009, 22: 6758-6772. doi: 10.1175/2009JCLI2806.1 [25] Huang Jian, Wang Xin, Zhou Wen, et al. The characteristics of sea fog with different airflow over the Huanghai Sea in boreal spring. Acta Oceanologica Sinica, 2010, 29(4):3-12. doi: 10.1007/s13131-010-0045-8 [26] 黄健, 王斌, 周发琇, 等.华南沿海暖海雾过程中的湍流热量交换特征.大气科学, 2010, 34(4):715-725. http://www.cnki.com.cn/Article/CJFDTOTAL-DQXK201004006.htm [27] 张朝锋.粤东海区海雾的气候特征分析.广东气象, 2002, 2: 2-21. doi: 10.3969/j.issn.1007-6190.2002.z1.001 [28] 林钢, 卢山, 薛登智.华南沿岸雾的气候特征及影响系统.广东省气象台研究技术报告, 2000. http://cdmd.cnki.com.cn/Article/CDMD-10730-1014297778.htm -
图(5) / 表(2)
计量
- 摘要浏览量: 3580
- HTML全文浏览量: 937
- PDF下载量: 2080
- 被引次数: 0
