Data Evaluation and the Advantage of the Double Automatic Weather Stations
-
摘要: 象站双套运行的结果和优缺点进行了深入分析和探讨。从缺测率上分析了双套站数据的完整性,认为双套站可以明显降低数据的缺测率,具有保证数据完整性的优点。从差值的统计数字特征以及一致率、超差率和粗差率等指标上分析了两套数据的差异性,表明双套站大部分要素 (如气温、气压、相对湿度等) 具有差值小、差值变化幅度小、数据之间有较高的一致率、较低的粗差率和超差率等特点,但个别要素 (草面温度和5 cm地温) 差异略大。运行中通过对两套数据的差值、差值的滑动平均值、差值日变化特征的监控可以及时发现数据异常、探测仪器的系统性误差等情况,从而说明双套站具有保证数据质量的潜在优势,且从一定程度解决了目前自动气象站单套运行的不足之处。Abstract: Anhui Provincial Meteorological Service starts pilot run of double automatic weather stations at national weather stations in 2010. The stations are deployed in Xiuning, Anqing, Kuche and other districts, and they have been put into test run one after another since October of 2010. A detailed assessment is carried out to evaluate the hourly data of temperature, air pressure, relative humidity, 2 min average wind speed, 5 cm ground temperature and grass temperature of the double automatic weather stations in Xiuning, Anqing and Qimen from January to August of 2011. At the same time, the advantages and disadvantages of the double automatic weather stations are discussed.The data integrity of the double automatic weather stations is analyzed by calculating missing rate. The missing rate of air pressure of Station A is 0.38%, and Station B is 0.15% in Xiuning, but rate of Station A and B both missing at the same time is only 0.07%. So it is considered that the double automatic weather stations can reduce missing rate, and ensuring the integrity of data. The difference of the double sets of data is discussed by the statistical features, consistent rate, rate of gross errorand out-of-tolerance rates. Through the detailed analysis, it shows that most elements (such as temperature, air pressure and relative humidity) of the double automatic weather stations have stable and smaller difference.For example, the average air pressure of station A is lower than station B in Anqing, with the difference of 0.01 hPa. In 5812 effective samples of difference data, there are 1659 samples with difference of 0.0 hPa, and for 99.2% of all the samples, differences are within ±0.1 hPa. The standard deviation of difference is 0.08 hPa.The difference has higher concordance rate, lower out-of-tolerance rates and lower gross error. For example, the consistency rate of temperature is 96.6% in Qimen, the out-of-tolerance rates of humidity is 0.0% in Xiuning, the gross error rate of air pressure is 0.0% in Anqing. However, some elements of the double automatic weather stations have lager difference, such as the grass surface temperature and ground temperature. For example, the gross error of grass surface temperature is 12.7% in Xiuning, and the out-of-tolerance rate of ground temperature is 29.7% in Anqing.The data evaluation shows that the abnormal data, instrument systematic error can be found through monitoring the difference and the moving average of difference data, and whether the instruments have significant changes in performance can be determined by the daily change characteristic difference data of the double data. Thus, the double automatic weather stations have the potential advantages of ensuring the quality of data, and resolve the deficiency of current automatic stations in some degree in operation.
-
表 1 双套站气压缺测率统计 (单位:%)
Table 1 The monthly missing rate of the pressure from the double automatic weather stations (unit:%)
月份 休宁站 安庆站 祁门站 A站 B站 A,B站同时 A站 B站 A,B站同时 A站 B站 A,B站同时 1 0 0 0 2.55 2.55 2.55 0 3.23 0 2 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0.14 0 0 1.53 0 7 2.82 0.67 0.54 0 0 0 0 0 0 8 0.13 0.54 0 0 0 0 0 0 0 表 2 双套站各要素缺测率统计 (单位:%)
Table 2 The missing rate of the elements from the double automatic weather stations (unit:%)
要素 休宁站 安庆站 祁门站 A站 B站 A,B站同时 A站 B站 A,B站同时 A站 B站 A,B站同时 气温 0.38 0.15 0.07 0.33 0.34 0.33 0 0.60 0 相对湿度 0.38 0.15 0.07 0.33 0.34 0.33 0 0.60 0 风速 0.38 0.15 0.07 0.33 0.34 0.33 0 0.60 0 地温 0.38 0.15 0.07 0.62 0.34 0.33 0 0.60 0 草温 0.38 0.15 0.07 0.62 0.34 0.33 0 0.60 0 表 3 双套站各要素差值数字特征描述
Table 3 The statistical features of the double automatic weather stations
站点 要素 样本数 平均值 标准差 偏度
系数峰度
系数是否服从
正态分布一致率/% 粗差
率/%休宁站 气压/hPa 5805 -0.03 0.05 -0.74 -1.40 否 87.6 0.00 气温/℃ 5805 0.16 14.02 76.10 579.00 否 99.8 2.43 相对湿度/% 5805 -1.07 1.52 1.97 77.80 否 99.8 0.16 风速/(m·s-1) 5805 -0.01 0.09 0.50 9.70 否 93.0 1.41 地温/℃ 5805 0.02 0.58 -2.70 12.40 否 95.9 5.08 草温/℃ 5805 0.08 0.81 5.35 55.60 否 96.5 12.68 安庆站 气压/hPa 5812 -0.02 0.08 0.40 -1.30 否 99.2 0.00 气温/℃ 5812 -0.04 0.07 0.00 3.70 是 96.5 1.60 相对湿度/% 5812 1.52 2.97 16.00 308.50 否 99.5 1.00 风速/(m·s-1) 5812 -0.03 0.23 2.70 39.90 否 95.5 8.10 地温/℃ 5795 0.05 0.36 1.00 2.60 否 94.3 3.30 草温/℃ 5795 0.35 1.04 2.00 5.30 否 91.5 16.80 祁门站 气压/hPa 5797 0.00 0.31 27.90 799.30 否 99.9 1.10 气温/℃ 5797 0.03 0.14 -3.01 21.49 否 96.6 2.80 相对湿度/% 5797 0.34 1.44 -13.1 323.30 否 95.4 5.60 风速/(m·s-1) 5797 -0.02 0.26 -0.47 17.66 否 94.2 13.60 地温/℃ 5797 0.01 0.25 0.88 12.56 否 95.8 4.20 草温/℃ 5797 -0.01 0.33 1.88 33.00 否 95.0 12.80 表 4 双套站各要素超差率统计 (单位:%)
Table 4 The out-of-tolerance rates of the elements from the double automatic weather stations (unit:%)
站点 气压 气温 相对湿度 平均风速 5 cm地温 草面温度 安庆站 0.0 0.1 0.5 0.7 20.1 29.7 祁门站 0.1 1.9 0.1 1.0 2.5 9.4 休宁站 0.0 0.3 0.0 0.0 11.1 20.1 -
[1] 王颖, 刘小宁, 鞠晓慧.自动观测与人工观测差异的初步分析.应用气象学报, 2007, 18(6):849-855. doi: 10.11898/1001-7313.200706128 [2] 胡玉峰.自动与人工观测数据的差异.应用气象学报, 2004, 15(6):719-726. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=20040689&flag=1 [3] 王颖, 刘小宁.自动站与人工观测气温的对比分析.应用气象学报, 2002, 13(6):741-748. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=20020697&flag=1 [4] 余君, 胡玉峰, 刘均.我国中部地区自动站与人工站气温的差异及原因分析.气象, 2007, 33(5):94-99. doi: 10.7519/j.issn.1000-0526.2007.05.014 [5] 陈豫英, 陈晓光, 张智, 等.宁夏自动站与人工观测气温的差异对比分析.应用气象学报, 2006, 17(6):118-124. http://www.cnki.com.cn/Article/CJFDTotal-YYQX2006S1016.htm [6] 任芝花, 冯明农, 张洪政, 等.自动与人工观测降雨量的差异及相关性.应用气象学报, 2007, 18(3):358-364. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=20070358&flag=1 [7] Sevruk B. Correction of Precipitation Measurements. WMO/TD, 1985, 104: 13-23. http://agris.fao.org/agris-search/search.do?recordID=XF2015025357 [8] 余君, 牟容.自动站与人工站相对湿度观测结果的差异及原因分析.气象, 2008, 34(12):96-102. doi: 10.7519/j.issn.1000-0526.2008.12.013 [9] Hubbard K G, Lin X, Robbins K. Preliminary Results from a Field Comparison of Relative Humidity Sensors. 12th Symposium on Meteorological Observations and Instrumentation, 2003. [10] 沈艳, 任芝花, 王颖, 等.我国自动与人工蒸发量观测资料的对比分析.应用气象学报, 2008, 19(4): 463-469. doi: 10.11898/1001-7313.20080410 [11] 杨晶. 浅析自动站维护与提供高质量数据//第26届中国气象学会年会第三届气象综合探测技术研讨会分会场论文集, 2009: 215-226. [12] 胡玉峰.自动气象站原理与测量方法.北京:气象出版社, 2004. [13] 李雁, 裴翀, 孟昭林, 等.全国自动站运行能力统计及其影响因素分析.仪器仪表用户, 2010, 17(2):4-8. http://www.cnki.com.cn/Article/CJFDTOTAL-DZYQ201002002.htm [14] 中国气象局.地面气象观测规范.北京:气象出版社, 2003. [15] CIMO.气象仪器和观测方法指南 (第6版).北京:气象出版社, 1992. [16] 黄嘉佑.气象统计分析与预报方法.北京:气象出版社, 1989:24-25. [17] 中华人民共和国国家计量技术规范.测量误差及数据处理.北京:中国计量出版社, 1991. [18] 邓天宏, 米鸿涛, 王安国, 等.自动气象站资料评估方法及应用.河南气象, 2005(3):44-46. http://www.cnki.com.cn/Article/CJFDTOTAL-HNQX200503035.htm [19] 中国气象局气象信息中心. 对比观测期间监测资料评估技术方法. 1999. [20] 中国气象局气象探测中心. 新型自动气象 (气候) 站评估报告. 2010. [21] 李黄.自动气象站实用手册.北京:气象出版社, 2007.