Writing NetCDF Files: Best Practices

• Conventions
• Coordinate Systems
• Variable Grouping
• Variable Attributes
• Strings and Variables of type char
• Calendar Date/Time
• Unsigned Data
• Packed Data Values
• Missing Data Values
• Miscellaneous tips

Conventions

While netCDF is intended for "self-documenting data", it is often necessary for data writers and readers to agree upon attribute conventions and representations for discipline-specific data structures. These agreements are written up as human readable documents called netCDF conventions.

• Use an existing Convention if possible. See the list of registered conventions.
• The CF Conventions are recommended where applicable, especially for gridded (model) datasets.
• Document the convention you are using by adding the global attribute "Conventions" to each netCDF file, for example:

:Conventions = "CF-1.3";

• This document refers to conventions for the netCDF classic data model. For recommendations about conventions for the netCDF-4 enhanced data model, see Developing Conventions for NetCDF-4.

Coordinate Systems

A coordinate variable is a one-dimensional variable with the same name as a dimension, which names the coordinate values of the dimension. It must not have any missing data (for example, no _FillValue or missing_value attributes) and must be strictly monotonic (values increasing or decreasing). A two-dimensional variable of type char is a string-valued coordinate variable if it has the same name as its first dimension, e.g.: char time( time, time_len); all of its strings must be unique. A variable's coordinate system is the set of coordinate variables used by the variable. Coordinates that refer to physical space are called spatial coordinates, ones that refer to physical time are called time coordinates, ones that refer to either physical space or time are called spatio-temporal coordinates.

• Make coordinate variables for every dimension possible (except for string length dimensions).
• Give each coordinate variable at least unit and long_name attributes to document its meaning.
• Use an existing netCDF Convention for your coordinate variables, especially to identify spatio-temporal coordinates.
• Use shared dimensions to indicate that two variables use the same coordinates along that dimension. If two variables' dimensions are not related, create separate dimensions for them, even if they happen to have the same length.

Variable Grouping

You may structure the data in a netCDF file in different ways, for example putting related parameters into a single variable by adding an extra dimension. Standard visualization and analysis software may have trouble breaking that data out, however. On the other extreme, it is possible to create different variables e.g. for different vertical levels of the same parameter. However, standard visualization and analysis software may have trouble grouping that data back together. Here are some guidelines for deciding how to group your data into variables:

• All of the data in a variable must be of the same type and should have the same units of measurement.
• A variable's attributes should be applicable to all its data.
• If possible, all of the coordinate variables should be spatio-temporal, with no extra dimensions.
• Use 4D spatio-temporal coordinate systems in preference to 3D. Use 3D spatio-temporal coordinate systems in preference to 2D.
• Vector valued (e.g. wind) parameters are legitimate uses of extra dimensions. There are trade-offs between putting vectors in the same variables vs. putting each component of a vector in a different variable. Check that any visualization software you plan to use can deal with the structure you choose.
• Think in terms of complete coordinate systems (especially spatio-temporal), and organize your data into variables accordingly. Variables with the same coordinate system implicitly form a group.

Variable Attributes

• For each variable where it makes sense, add a units attribute, using the udunits conventions, if possible.
• For each variable where it makes sense, add a long_name attribute, which is a human-readable descriptive name for the variable. This could be used for labeling plots, for example.

Strings and Variables of type char

NetCDF-3 does not have a primitive String type, but does have arrays of type char, which are 8 bits in size. The main difference is that Strings are variable length arrays of chars, while char arrays are fixed length. Software written in C usually depends on Strings being zero terminated, while software in Fortran and Java do not. Both C (nc_get_vara_text) and Java (ArrayChar.getString) libraries have convenience routines that read char arrays and convert to Strings.

• Do not use char type variables for numeric data, use byte type variables instead.
• Consider using a global Attribute instead of a Variable to store a String applicable to the whole dataset.
• When you want to store arrays of Strings, use a multidimensional char array. All of the Strings will be the same length.
• There are 3 strategies for writing variable length Strings and zero-byte termination:
  1. Fortran convention: pad with blanks and never terminate with a zero byte.
  2. C convention: pad with zeros and always terminate with a zero byte.
  3. Java convention: You don't need to store a trailing zero byte, but pad trailing unused characters with zero bytes.
• When reading, trim zeros and blanks from the end of the char array and if in C, add a zero byte terminator.

Calendar Date/Time

Time as a fundamental unit means a time interval, measured in seconds. A Calendar date/time is a specific instance in real, physical time. Dates are specified as an interval from some reference time e.g. "days elapsed since Greenwich mean noon on 1 January 4713 BCE". The reference time implies a system of counting time called a calendar (e.g. Gregorian calendar) and a textual representation (e.g. ISO 8601).

There are two strategies for storing a date/time into a netCDF variable. One is to store it as a String using a standard encoding and Calendar. The other is to encode it as a numeric value and a unit that includes the reference time, e.g. "seconds since 1992-10-8 15:15:42.5 -6:00". The latter is more compact if you have more than one date, and makes it easier to compute intervals between two dates.

Unidata's udunits package provides a convenient way to implement the second strategy. It uses the ISO 8601 encoding and a hybrid Gregorian/Julian calendar. It does not allow you to use other Calendars or encodings for the reference time.

• If your data uses real, physical time that is well represented using the Gregorian/Julian calendar, encode it as an interval from a reference time, and add a units attribute which uses a udunits-compatible time unit. Readers can use the udunits package to manipulate or format the date values.
• If your data uses a different calendar, use an existing Convention such as the CF convention, or create a new Convention which clearly documents what the calendar and encoding is. Make it compatible with existing date manipulation packages if possible (e.g. java.text.SimpleDate).
• Add multiple sets of time encodings if necessary to allow different readers to work as well as possible.
  

Unsigned Data

NetCDF-3 does not have unsigned integer primitive types.

• To be completely safe with unknown readers, widen the data type, or use floating point.
• You can use the corresponding signed types to store unsigned data only if all client programs know how to interpret this correctly.
• A new proposed convention is to create a variable attribute _Unsigned = "true" to indicate that integer data should be treated as unsigned.

Packed Data Values

Packed data is stored in a netCDF file by limiting precision and using a smaller data type than the original data, for example, packing double-precision (64-bit) values into short (16-bit) integers. The C-based netCDF libraries do not do the packing and unpacking. (The netCDF Java library will do automatic unpacking when the VariableEnhanced Interface is used. For details see EnhancedScaleMissing).

• Each variable with packed data has two attributes called scale_factor and add_offset, so that the packed data may be read and unpacked using the formula:

  unpacked_data_value = packed_data_value * scale_factor + add_offset

• The type of the stored variable is the packed data type, typically byte, short or int.
• The type of the scale_factor and add_offset attributes should be the type that you want the unpacked data to be, typically float or double.
• To avoid introducing a bias into the unpacked values due to truncation when packing, the data provider should round to the nearest integer rather than just truncating towards zero before writing the data:

  packed_data_value = nint((unpacked_data_value - add_offset) / scale_factor)

Depending on whether the packed data values are intended to be interpreted by the reader as signed or unsigned integers, there are alternative ways for the data provider to compute the scale_factor and add_offset attributes. In either case, the formulas above apply for unpacking and packing the data.

A conventional way to indicate whether a byte, short, or int variable is meant to be interpreted as unsigned, even for the netCDF-3 classic model that has no external unsigned integer type, is by providing the special variable attribute _Unsigned with value "true". However, most existing data for which packed values are intended to be interpreted as unsigned are stored without this attribute, so readers must be aware of packing assumptions in this case. In the enhanced netCDF-4 data model, packed integers may be declared to be of the appropriate unsigned type.

Let n be the number of bits in the packed type, and assume dataMin and dataMax are the minimum and maximum values that will be used for a variable to be packed.

• If the packed values are intended to be interpreted as signed integers (the default assumption for classic model data), you may use:

  scale_factor =(dataMax - dataMin) / (2ⁿ - 1)

  add_offset = dataMin + 2^{n - 1} * scale_factor

• If the packed values are intended to be interpreted as unsigned (for example, when read in the C interface using the nc_get_var_ubyte() function), use:

  scale_factor =(dataMax - dataMin) / (2ⁿ - 1)

  add_offset = dataMin

• In either the signed or unsigned case, an alternate formula may be used for the add_offset and scale_factor packing parameters that reserves a packed value for a special value, such as an indicator of missing data. For example, to reserve the minimum packed value (-2^{n - 1}) for use as a special value in the case of signed packed values:

  scale_factor =(dataMax - dataMin) / (2ⁿ - 2)

  add_offset = (dataMax + dataMin) / 2

• If the packed values are unsigned, then the analogous formula that reserves 0 as the packed form of a special value would be:

  scale_factor =(dataMax - dataMin) / (2ⁿ - 2)

  add_offset = dataMin - scale_factor

• Example, packing 32-bit floats into 16-bit shorts:

      variables:
        short data( z, y, x);
  	    data:scale_offset = 34.02f;
  	    data:add_offset = 1.54f;

• The units attribute applies to unpacked values.

Missing Data Values

Missing data is a general name for data values that are invalid, never written, or missing. The netCDF library itself does not handle these values in any special way, except that the value of a _FillValue attribute, if any, is used in pre-filling unwritten data. (The Java-netCDF library will assist in recognizing these values when reading, see class VariableStandardized).

• Default fill values for each type are available in the C-based interfaces, and are defined in the appropriate header files. For example, in the C interface, NC_FILL_FLOAT and NC_FILL_DOUBLE are numbers near 9.9692e+36 that are returned when you try to read values that were never written. Writing, reading, and testing for equality with these default fill values works portably on the platforms on which netCDF has been tested.
• The _FillValue attribute should have the same data type as the variable it describes. If the variable is packed using scale_factor and add_offset attributes, the _FillValue attribute should have the data type of the packed data.
• Another way of indicating missing values for real type data is to store an IEEE NaN floating point value. The advantage of this is that any computation using a NaN results in a NaN. Client software must know to look for NaNs, however, and detection of NaNs is tricky, since any comparison with a NaN is required to return false.
  • In Java, you can use Double.NaN and Float.NaN constants.
  • In many C compilers, you can generate a NaN value using double nan = 0.0 / 0.0;
• Alternatively or in addition, set the valid_range attribute for each variable that uses missing values, and make sure all valid data is within that range, and all missing or invalid data is outside of that range. Again, the client software must recognize and make use of this information. Example:

      variables:
        float data( z, y, x);
  	    data:valid_range = -999.0f, 999.0f;
      

  If the variable is packed using scale_factor and add_offset attributes, the valid_range attribute should have the data type of the packed data.

Miscellaneous tips

• To define a file whose structure is known in advance, write a CDL file and create the netCDF file using ncgen. Then write the data into the netCDF file using your program. This is typically much easier than programming all of the create calls yourself.
• For the netCDF classic or 64-bit-offset formats, it's possible to reserve extra space in the file when it is created so that you can later add additional attributes or non-record variables without copying all the data. (This is not necessary for netCDF-4 files, because metadata can be added efficiently to such files.) See the C man-page reference documentation (or the Fortran reference documentation) for nc__create and nc__enddef (nf__create and nf__enddef for Fortran) for more details on reserving extra space in the header.