Data Formats and Data Viewers

Due to the huge amount of data, satellite information cannot be stored in formats like ASCII or excel-spreadsheets. Highly specialized binary data formats have been developed to store and handle array-oriented data sets, like satellite data. With the help of such standardized data formats large data sets can be assessed and analyzed with specialized software. However, commonly available software, e.g. software that is included in standard Windows distributions, is not able to open and read such binary data files.

Common formats for satellite data are hdf5 (Hierarchical Data Format) and NetCDF (network Common Data Form). The operational CM SAF products and CM SAF climate data records are distributed in NetCDF. The workshop will focus on the use of data in NetCDF-format. Current CM SAF data records are in NetCDF3, while the next releases of CM SAF data records will be in NetCDF4. Please check if you have installed all necessary libraries for NetCDF4. Otherwise, there is the option to choose between NetCDF3 and NetCDF4 during the ordering process (wui.cmsaf.eu).   

To convert a binary NetCDF file into readable ASCII format, the command-line tool ncdump can be used. ncdump is distributed with the NetCDF-library, which is available for Unix/Linux and Windows operating systems.

To visualize data files in NetCDF-format software is required. For quicklooks of the data easy-to-use software is available:
For Unix/Linux operating systems, ncview is commonly used; the java-based ncBrowse and Panoply is available for all operating systems.

To analyse and for more advanced visualization of NetCDF data, more complex software packages are required. During the workshop we will use the 'cmsaf' R-package for the analysis of CM SAF NetCDF data. However, most data analysis software packages are able to work with CM SAF NetCDF data, e.g., QGIS, IDL, MATLAB.


Joerg Trentmann / Steffen Kothe, DWD

February 2017

Last edited: Christine Träger-Chatterjee, October 2017

Última atualização: sexta-feira, 25 jan. 2019, 20:29