...
首页> 外文期刊>Advances in Geosciences >Investigation of trends in synoptic patterns over Europe with artificial neural networks
【24h】

Investigation of trends in synoptic patterns over Europe with artificial neural networks

机译:利用人工神经网络调查欧洲天气模式的趋势

获取原文
           

摘要

The present study is a comprehensive application of amethodology developed for the classification of synoptic situations usingartificial neural networks. In this respect, the 500 hPa geopotential heightpatterns at 12:00 UTC (Universal Time Coordinated) determined from thereanalysis data (ERA-40 dataset) of the European Centre for Medium rangeWeather Forecasts (ECMWF) over Europe were used. The dataset covers a periodof 45 years (1957–2002) and the neural network methodology applied is theSOM architecture (Self Organizing Maps). The classification of the synopticscale systems was conducted by considering 9, 18, 27 and 36 synopticpatterns. The statistical analysis of the frequency distribution of theclassification results for the 36 clusters over the entire 44-year periodrevealed significant tendencies in the frequency distribution of certainclusters, thus substantiating a possible climatic change. In the following,the database was split into two periods, the "reference" period thatincludes the first 30 years and the "test" period comprising the remaining14 years.
机译:本研究是使用人工神经网络对天气状况进行分类的方法学的综合应用。在这方面,使用了欧洲中距离天气预报中心(ECMWF)的分析数据(ERA-40数据集)确定的12:00 UTC(全球协调时间)处的500 hPa高度势模式。该数据集涵盖了45年(1957–2002年),神经网络方法是SOM架构(自组织图)。对天气尺度系统的分类是通过考虑9、18、27和36个同视光箱进行的。在整个44年的时间里对36个集群的分类结果的频率分布进行统计分析,揭示了某些集群的频率分布的显着趋势,从而证实了可能的气候变化。在下文中,数据库被分为两个时期,包括前30年的“参考”时期和包括其余14年的“测试”时期。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号