More than 10 million sets of data flow into the MeteoSwiss data warehouse every day. This measurement data originates from various sources and comes in various types and formats. The objective of the data integration is to consolidate and standardise the different measurement data. This allows for the data to be stored over the long term in a standardised form, meaning it is available for a variety of analyses.

Data sources

The data that feed into the data warehouse are obtained in a variety of ways. The sources include ground-level point measurements, atmospheric measurements in the form of three-dimensional profile data, weather observations and aviation weather reports. Data are also gathered using evenly distributed (or grid-based) measurements as well as in the form of radar data and interpolated data or on the basis of location-specific forecasts.

Data types

The data that feed into the data warehouse are obtained in a variety of ways. The sources include ground-level point measurements, atmospheric measurements in the form of three-dimensional profile data, weather observations and aviation weather reports. Data are also gathered using evenly distributed (or grid-based) measurements as well as in the form of radar data and interpolated data or on the basis of location-specific forecasts.

Data formats

The different kinds of data require different formats. Point data usually take the form of decoded ASCII bulletins, while grid data take the form of netCDF files. A large number of other formats can be imported in the data warehouse. For numerous formats (including e-mail, text, xls, csv and xml), there are adjustable conversion procedures that can be expanded as required. This provides the data infrastructure with a great deal of flexibility.