新洲国家气象观测站于2017年9月开始启用DSG5型降水现象仪。从日常实际观测结果来看,新洲降水现象自动观测还是存在判识结果不准确的问题。新洲降水现象综合判识一直使用的是统一下发的质控算法,未进行过本地化的优化处理。而对于不同...新洲国家气象观测站于2017年9月开始启用DSG5型降水现象仪。从日常实际观测结果来看,新洲降水现象自动观测还是存在判识结果不准确的问题。新洲降水现象综合判识一直使用的是统一下发的质控算法,未进行过本地化的优化处理。而对于不同地区、不同气候条件,其质控算法适应性会有差别,为进一步提高新洲降水现象自动识别准确率,本文利用新洲国家气象观测站降水现象平行观测数据进行研究,基于地面气象观测业务人员在开展降水现象人工观测时,同时会关注当前的相对湿度、空气温度、风向、风速以及降水量,而且观测人员会根据自身积累的经验,结合上述气象要素来综合判断具体的天气现象类型。本文将人工观测综合感知的方法与仪器可以实现的自动观测要素相结合,通过分析台站人工观测的天气现象结果与仪器自动观测的各类气象要素对比数据,提出了雨、毛毛雨、雪、雨夹雪四种现象的优化算法模型。The Xinzhou National Meteorological Observation Station began using the DSG5-type precipitation phenomenon instrument in September 2017. From the perspective of daily actual observation results, there are still issues with the accuracy of automated identification of precipitation phenomena in Xinzhou. The comprehensive identification of precipitation phenomena in Xinzhou has always relied on a standardized quality control algorithm that has not undergone localized optimization. However, the adaptability of quality control algorithms can vary across different regions and climate conditions. To further improve the accuracy of automatic recognition of precipitation phenomena in Xinzhou, this paper utilizes parallel observation data from the Xinzhou National Meteorological Observation Station. It is based on the fact that meteorological observation personnel, when conducting manual observations of precipitation phenomena, also pay attention to current relative humidity, air temperature, wind direction, wind speed, and precipitation amount. Moreover, observers use their accumulated experience in conjunction with the afore mentioned meteorological elements to comprehensively judge specific weather phenomenon types. This paper combines the method of comprehensive perception from manual observations with the automatic observation elements that can be achieved by the instruments. By analyzing the comparison data between the weather phenomenon results from manual observations and the various meteorological elements from automated observations, it proposes optimized algorithm models for four types of phenomena: rain, drizzle, snow, and sleet.展开更多
文摘新洲国家气象观测站于2017年9月开始启用DSG5型降水现象仪。从日常实际观测结果来看,新洲降水现象自动观测还是存在判识结果不准确的问题。新洲降水现象综合判识一直使用的是统一下发的质控算法,未进行过本地化的优化处理。而对于不同地区、不同气候条件,其质控算法适应性会有差别,为进一步提高新洲降水现象自动识别准确率,本文利用新洲国家气象观测站降水现象平行观测数据进行研究,基于地面气象观测业务人员在开展降水现象人工观测时,同时会关注当前的相对湿度、空气温度、风向、风速以及降水量,而且观测人员会根据自身积累的经验,结合上述气象要素来综合判断具体的天气现象类型。本文将人工观测综合感知的方法与仪器可以实现的自动观测要素相结合,通过分析台站人工观测的天气现象结果与仪器自动观测的各类气象要素对比数据,提出了雨、毛毛雨、雪、雨夹雪四种现象的优化算法模型。The Xinzhou National Meteorological Observation Station began using the DSG5-type precipitation phenomenon instrument in September 2017. From the perspective of daily actual observation results, there are still issues with the accuracy of automated identification of precipitation phenomena in Xinzhou. The comprehensive identification of precipitation phenomena in Xinzhou has always relied on a standardized quality control algorithm that has not undergone localized optimization. However, the adaptability of quality control algorithms can vary across different regions and climate conditions. To further improve the accuracy of automatic recognition of precipitation phenomena in Xinzhou, this paper utilizes parallel observation data from the Xinzhou National Meteorological Observation Station. It is based on the fact that meteorological observation personnel, when conducting manual observations of precipitation phenomena, also pay attention to current relative humidity, air temperature, wind direction, wind speed, and precipitation amount. Moreover, observers use their accumulated experience in conjunction with the afore mentioned meteorological elements to comprehensively judge specific weather phenomenon types. This paper combines the method of comprehensive perception from manual observations with the automatic observation elements that can be achieved by the instruments. By analyzing the comparison data between the weather phenomenon results from manual observations and the various meteorological elements from automated observations, it proposes optimized algorithm models for four types of phenomena: rain, drizzle, snow, and sleet.
