Example Data

This data set is a collection of netCDF files conforming to the NCAS-GENERAL Data Standard and can be used to aid the development of software and also as a template against which an actual observationally derived data file can be compared for compliance purposes. For a given data product all variables that are defined are represented and the data is totally arbitrary: what the data should look like will be discussed later. These files represent the maximum variable content associated with the data product: an actual observationally derived file will always contain the common global attributes, dimensions and variables (see the Data Product Supporting Documentation) and all or a subset of those defined for a specific data product: they will never contain more.

For every instrument registered as part of NCAS Observations an exemplar netCDF file for all associated data products and file naming variations has been produced for a totally spurious land-based deployment on 25th December 2900. A subset of files has been produced for spurious air and sea-based deployments also on 25th December 2900.  For a given variable the data provided is totally arbitrary: what the data should look like will be discussed later.

There are two data sets and these correspond to the two sets of data products: v1.1 and v2.0. The number and names of data products available in the two versions are the same and for products that are not profile related there is no difference at all between v1.1 and v2.0. The difference comes from how altitude is dealt with.

In v1.1 the altitude variable is dependent on the time and index dimensions while in v2.0 it is dependent on the altitude dimension. This approach is used to allow for deployments where the altitude varies with time over the period of the file. For example for a land deployment where the altitude remains constant the v2.0 version of the data product would be used. In contrast for a land deployment (say on a vehicle going up and down hills) in which the altitude of each point in the profile varies with time v1.1 of the data, product would be used.

1-D data

Figure 1a. Shows an example of 1D data and is the full data that is in a file – no quality control (QC) flags applied. To provide an examples of quality control flag use the data flag is defined such that:

• flag_value = 0: This is never used so will not appear as a value in the file

• flag_value = 1: The data is of good quality

• flag_value = 2: The data value is <-0.85 and is suspect

• flag_value = 3: The data value is equal to 0 and is suspect

• flag_value = 4: The data value is >0.85 and is suspect.

The variable attributes valid_min and valid_max are defined as the minimum and maximum values of the data where the flag_value = 1. The most basic QC that can be applied to the data in the file is to set any data value that is outside the valid_min and valid_max range to NaN (not-a-number) and the result of this is shown in figure 1.b

2D data

The 2D data is built up by repeating the 1D data time series at each value of the second dependent variable. Figure 2a and 2b show this for the same variable as represented in v1.1 and v2.0. QC flagging is the same for the 1D data and figure 2c shows the result of the application of the most basic level of QC

3D data

The 3D data is built from repeating the 2D surface for every value of the third dependent variable.