Format specifications for netCDF files

Mandatory metadata

All netCDF-data must contain a minimum of mandatory metadata. All following regulations are mandatory if not stated otherwise.

All Metadata regulations that are listed in this procedure are taken from the Binding Regulations for Storing Data as netCDF Files from Helmholtz-Zentrum Hereon. The website contains additional optional regulations.

1.1 Files

• Names must not contain spaces, umlauts, "ß", special characters (exceptions: "_", "-", ".").
• Names must start with a letter and foreshadow the underlying data (e.g. pre for precipitation).
• Names must contain start (and end) date of underlying data (e.g. pre_20230101.nc, pre_202301.nc, pre_2023).

1.2 Variables

• Names must not contain spaces, umlauts, "ß", special characters (exception: "_").
• Names can be freely chosen but should follow international standards (e.g. Climate and Forecast (CF) Conventions, P09 database).
• Names must honor case-sensitiveness.
• Files must not contain multiple variables with names only differentiating in upper/lower letters.
• Must be described using corresponding attributes.

1.3 Dimensions

• A variable can have any number of dimensions.
• All dimensions must have different names.
• A restriction to four dimensions is recommended.
• Names must not contain spaces, umlauts, "ß", special characters (exception: "_").
• Names must honor case-sensitiveness.
• Files must not contain multiple dimensions with names only differentiating in upper/lower letters.
• Dimensions of a variable should be arranged in the relative order of date/time, height/depth, latitude, longitude (T,Z,Y,X). If this is not possible (e.g. for spectra), other reasonable sequences should be created.
• Additional dimensions must be to the left of (T,Z,Y,X).
• The time variable dimension should be "UNLIMITED" to ensure expansion of the time series without difficulties.

1.4 Attributes

• Names must not contain spaces, umlauts, "ß", special characters (exception: "_").
• Names must honor case-sensitiveness.
• Files must not contain multiple attributes with names only differentiating in upper/lower letters.
• Only the netCDF data types char, byte, float, real, double are allowed.
• units is mandatory for each variable.
• long_name is recommended.
• standard_name should be provided for each variable. If it does not make sense long_name is mandatory and should follow the format for standard_name.

Attribute	Description/Format	Specification/Example
units	"string", use SI unit if possible. This also applies to dimensionless units, then use given format: "1", "1e±2", "1e±3", "1e±6", … "1e±24"	- 'm' - '1'
standard_name	The standard_name stems from CF Conventions Standard Names Table. In the case of ancillary variables, it is mandatory to use the Standard Name Modifiers (e.g. status_flag)	sea_water_salinity
long_name	Any text. If no standard_name exists, then the long_name should be constructed like a standard name.	Salinity from Seabird CTD

1.5 Global Attributes

Attribute	Description	Example
Conventions	convention specification of underlying data (fixed: CF-1.8)	CF-1.8
institution	(fixed: Helmholtz Centre for Environmental Research GmbH, Germany)	Helmholtz Centre for Environmental Research GmbH, Germany
title	tile following the file naming conventions
source	essential data and process steps to retrace data production (data sources, used, methods, models (name and version))	cosmo_100614_4.14_clm2 int2lm_120824_1.20_clm1 remote sensing in-situ
creation_date	creation date of the file	2023-04-12T15:00:00Z
originator	creator of dataset	doctor scientist
contact	contact information (name, email)	doctor.scientist@ufz.de
crs	coordinate reference system	EPSG:4326