Data Warehouse

All datasets have a monthly granularity. In case you are interested in longer time series, just subsets of offered parameters or a joint dataset from different stations you may use the Data Warehouse.

  • Choose all files you are interested in. This can be done using the link-tables offered in the lower part of Opens internal link in current windowData retrieval via PANGAEA or by any query using Opens external link in new windowhttps://www.pangaea.de. As result you will get a list of files.
  • Before you can retrieve all files at once you must be logged in at the very top right corner of this page.
  • After logging in, you will see the green button "Data Warehouse" just below the area you performed the login (see screenshot).
  • A click on this button leads you into the Data Warehouse. Choose the parameter you are looking for. 
  • Arithmetic averages can be obtained as "Method" if the parameter DATE/TIME was selected. Warning: Arithmetic averages can lead to undesired results when the dataset contains gaps! Please prefer more sofistic averaging procedures as published in doi:10.5194/amt-4-339-2011.
  • Start your query by clicking on the button "Start Data Warehouse Query".
  • A window will pop up to download the output-file on your local computer.
  • You may open this file in any text-editor, or visualization software like PanPlot.

This example may show how to use the Data Warehouse. Output files from a Data Warehouse query may contain quality flags which are described in PangaWiki.

Missing Values: The output from the Data Warehouse are tab-delimeted ascii-files. Missing values are left blank. In case you prefer other missing values unix user may may use: 

gsed -e "s/\t\t/\t-9999\t/g".

Windows user may take the Opens external link in new windowBSRN Toolbox. Download one or many file(s) from the Data Warehouse and drag these files onto the window of the Toolbox. Start "Tools->Convert files to unformated format". The new filename is "zz_Filename.txt".

⇒ Please try short time series first, the size of the output file can quickly get very huge, and it might take hours to download it!!!

It might be advisable in these cases to split up the download process or use daily/monthly averages (see above).