Defining the EddyPro dataset

The averaging dataset is comprised of all numeric data columns that were found in the raw file(s), including sensitive and non-sensitive variables. From this dataset, the EddyPro dataset (or EP dataset) is derived through the following operations:

If at least one column of the raw file is defined as a flag, EddyPro scans that column and filters individual data records according to the flag policy and value of the flag threshold. Recall that EddyPro supports up to two flag columns.

Variables of the averaging dataset are split into sensitive and non-sensitive variables. Sensitive variables are stored in the EP dataset, while non-sensitive variables are stored in what we call the user dataset. Starting from here, the two datasets undergo a different path of processing. In particular, the EPE dataset is used for the flux computation.