MeteoSwiss - Open Data > Understanding MeteoSwiss' Open Data products > E. Forecast Data

E. Forecast Data

Forecasting systems calculate future atmospheric conditions on the basis of measurement data and observations. MeteoSwiss uses these weather models to create weather forecasts and to enable it to issue weather warnings in the event of imminent hazards.

The following forecast data are available:

1. Short-term forecast data 🟡 documentation upcoming
2. Numerical weather forecasting model data 🟡 documentation upcoming
3. Local forecast data 🟡 documentation upcoming

---

1. Short-term forecast data

Nowcasting involves high spatial and temporal resolution forecasts of weather developments for the next few minutes and up to a maximum of six hours ahead. MeteoSwiss uses these short-term forecasts to, among other things, predict thunderstorms, hail and heavy rainfall.

As MeteoSwiss is planning to replace the current 'INCA' nowcasting software, the following datasets are available from the start of our open data provision:

• Precipitation (10min values): quantitative chain (based on CombiPrecip, RR)
• Wind, wind gust and wind direction (10min values)
• Relative sunshine duration (10min values)
• Total cloudiness (10min values)

The following datsets will be provided next:

• Snowfall (10min values): quantitative chain (based on CombiPrecip, RS)
• ...
• ...

1.1. Data granularity, update frequency, format and volume

Data granularity is every 10min. Update frequency for the period 0h- +6h is specified per dataset in the table below.

Data format is NetCDF.

Dataset	Update frequency	Example data file	Productive version file name	Estimated volume per file (MB)
Precipitation (10min values): quantitative chain (based on CombiPrecip, RR)	every 10min	RR_INCA_202106280700.nc	ogd-nowcasting_RR-INCA_(date and time code).nc	1.7
Wind, wind gust and wind direction (10min values)	every 10min	...	ogd-nowcasting_(product name)_(date and time code).nc	...
Relative sunshine duration (10min values)	10min	SU_INCA_202106280700.nc	ogd-nowcasting_SU-INCA_(date and time code).nc	6.4
Total cloudiness (10min values)	10min	SU_INCA_202106280700.nc	ogd-nowcasting_SU-INCA_(date and time code).nc	6.4
Snowfall (10min values): quantitative chain (based on CombiPrecip, RS)	every 10min	RS_INCA_202106280700.nc	ogd-nowcasting_RS-INCA_(date and time code).nc	0.4

1.2. Parameter metadata

Parameter metadata is part of each NetCDF-File. See example data files in the table above.

1.3. Coordinate system

The coordinate system is Swiss LV95 EPSG:2056.

1.4. Data visualisation

See e.g. MeteoSwiss' ....

2. Numerical Weather Forecasting Model Data

MeteoSwiss uses two models, ICON-CH1-EPS and ICON-CH2-EPS, to forecast atmospheric changes in Switzerland and its surroundings over a longer period than nowcasting, providing predictions for up to five days. Both models include ensemble data assimilation.

2.1 Model Specification

Attributes	ICON-CH1-EPS	ICON-CH2-EPS
Collection	ch.meteoschweiz.ogd-forecasting-icon-ch1	ch.meteoschweiz.ogd-forecasting-icon-ch2
Horizontal Grid Size	1 km	2.1 km
Ensemble Members	11	21
Forecast Period	33 h	120 h
Grid	Native icosahedral	Native icosahedral
Temporal Resolution	1 h	1 h
Model Run Interval	every 3 h	every 6 h
Format	GRIB edition 2	GRIB edition 2

2.2 Available Parameters

Users can find information about available parameters, including metadata, in the collection level assets of the above collections.

2.2.1 Parameter Metadata

The parameter metadata is part of each GRIB file.

2.3 Accessing Forecast Data

The user can access the forecast model data from the last 24 hours. Data older than 24 hours is no longer available. The data in each collection is described in the table above.

2.4 3D Grid Structure and Representation

The model data is structured on both a horizontal and vertical grid. While some parameters extend across the entire three-dimensional grid, others are only available at specific vertical levels. Parameters are classified as either single-level or multi-level:

• Single-level parameters contain data at a specific vertical level.
• Multi-level parameters extend across multiple vertical layers.

For example, vertical velocity is stored at multiple vertical levels, while the two-meter temperature is available only at a single vertical level.

2.4.1 Vertical Grid

The vertical grid is a height based coordinate system that follows the terrain. It is divided into multiple layers. The closer the layer is to the surface, the narrower the layers are, as shown in the image below. The so-called half levels align with horizontal grid points, while the full levels represent an averaged value over a vertical interval. There are 81 discrete half levels and 80 full levels in our data.

Illustration of ICON's vertical levels, Working with the ICON Model 2024, Figure 3.2

All parameters have their information on full levels, except for the vertical velocity W. W is stored on half levels and therefore called staggered. This means that the value is exact in this point and not an avarage value over a layer stored in one point like the full levels. For more detailed information on the vertical grid, read section 3.4 in Working with the ICON Model.

2.4.2 Horizontal Grid

The horizontal grid of ICON-CH1-EPS and ICON-CH2-EPS model is based on a native icosahedral grid inherited by the original ICON model grid (illustrated below).

Illustration of the grid construction, Working with the ICON Model, Figure 2.1

Since the provided data is given in the native grid, note that the grid points correspond to the center of the circumcircle of each triangle and not to the vertices. Therefore, the longitude and latitude are based in the middle of each triangle on the grid mentioned before. For more detailed information on the horizontal grid, read section 2.1 in Working with the ICON Model.

2.5 Data Visualisation

See jupyter-notebook examples.

3. Local forecast data

...

...

Data granularity, update frequency, format and volume

There are files of data granularity ..., ..., ..., ... and update frequency hourly (now), daily (recent) or yearly (historical) for each station.

Data format is CSV with an estimated volume of ... MB per file.

See example data files: ....

Parameter metadata

See example parameter metadata files of data granularity: ... and ....

Station metadata

See example station metadata file.

Data visualisation

See e.g. MeteoSwiss' ....