Foehn reconstruction for station Ellbögen and Innsbruck

Supplementary material for research article "Stauffer R., Zeileis A., Mayr G.J. Long-term foehn reconstruction combining unsupervised and supervised learning. International Journal of Climatology, The Royal Meteorological Society." containing data for station Ellbögen and Innsbruck (Austria).`ERA5predictors.txt` contains (pseudo-)code for all direct and derived ERA5 variables used in the study. The two ZIP archives `foehnix_Ellboegen.zip` and `foehnix_Innsbruck.zip` contain one comma separated value (CSV) file each, with the high-resolution in-situ observations from the corresponding automated weather stations as used for foehn classification (unsupervised learning; foehnix). The two ZIP archives `reconstruction_Ellboegen.zip` and `reconstruction_Innsbruck.zip` contain two CSV files with the hourly reconstructed time series (supervised learning) from 1940 to 2022. All CSV files contain a header with meta information regarding the format as well as the content of the corresponding file. All data provided is available under the Creative Commons Attribution 4.0 license (see also CSV meta information).

本数据集为研究论文《Stauffer R.、Zeileis A.、Mayr G.J.：结合无监督学习与有监督学习的长期焚风重建》（发表于《国际气候学杂志》（International Journal of Climatology），由英国皇家气象学会（The Royal Meteorological Society）出版）的补充材料，包含奥地利埃尔伯根（Ellbögen）与因斯布鲁克（Innsbruck）两地自动气象站的相关数据。`ERA5predictors.txt`包含本研究中使用的全部直接与衍生ERA5变量的伪代码。两个ZIP压缩包`foehnix_Ellboegen.zip`与`foehnix_Innsbruck.zip`各内含一个逗号分隔值（CSV）文件，存储了用于焚风分类（无监督学习；foehnix）的对应站点高分辨率原位观测数据。其余两个ZIP压缩包`reconstruction_Ellboegen.zip`与`reconstruction_Innsbruck.zip`各包含两份CSV文件，涵盖1940年至2022年的逐小时重建时序数据（有监督学习）。所有CSV文件均带有表头，其中包含对应文件的格式与内容相关元信息。本数据集所有数据均采用知识共享署名4.0许可协议（Creative Commons Attribution 4.0）发布，详细信息可参见各CSV文件的元信息。