Comma Separated Value (CSV) Files

CSV, or comma separated value, files are simple ASCII formatted files that can be read in by many programs.  Many database, data-processing, and analysis tools support CSV files, including Microsoft Excel, OpenOffice.org Calc, Gnumeric, and many more.  The CSV file format is standardized by RFC 4180.

BlueSky and CSV

BlueSky uses CSV as the simplest standard for importing and exporting small sets of data.  These include:

• Fire Information data
• Fuel Loading data
• Total Consumption data
• Time Rate of Consumption data
• Speciated Emissions data
• Vertical Plume Profile data

BlueSky Native File Format

The BlueSky "native" file format for BSF3 is actually three CSV files.  These should be named, respectively, fire_events.csv, fire_locations.csv, and fire_hourly.csv (or, on output, fire_emissions.csv).  Of these, only fire_locations.csv is mandatory; the other files can be assumed to contain null records if they are not present.  Also, while these file names are preferred, it is possible to use other names; in particular, in the case when only one file is present.

The following table lists the field names for each of the three files.  Some of the fields are required to be present (Required); others are derived from data during the course of the run (Derived); still others may be present in some input files, but must be derived from other data if they are not provided (Optional/Derived).  Finally, some fields are completely Optional, and some are not used by BlueSky but are passed through to output files unchanged (Metadata).

The same file format is used by the BlueSky Framework for input of fire information and for output of its results.  Obviously, it is expected that the input files will be mostly empty, and the output files mostly full.  It should be possible to use an output from an existing run to re-run through BlueSky; in this case, the modules should detect that the outputs have already been filled in and do nothing (or almost nothing), and the end result should be an equivalent output file.

If an input file contains a column with the same name as a derived output column (e.g. total_heat), the input value is renamed to something non-conflicting (e.g. metadata_total_heat) and passed through as metadata.