U. S. Flight Track Database: Data Types

Data Type Description

The data file naming convention is tsmftYYYYMMDD.EXT where tsmft indicates processed flight track data, YYYY is the four digit year of the data, MM is the two digit month of the year of the data, DD is either the two digit day of the year of the data or the letters mo indicating consolidated monthly data, and the extension, EXT, is any of the following:

(no extension) - Daily individual sorted flight tracks. These data files consist of raw downloaded data that have been quality checked and sorted by a procedure delineated here.
seg - Daily individual flight track segments within each hypercube. These data files consist of a series of header lines, each of which is followed by a series of data lines. Each data line contains the four values needed to define an endpoint in the hypercube. Connecting the endpoints in order creates a series of segments defining the flight path of the flight identified in the header line. Each endpoint is either a reported aircraft position that falls within a hypercube or an interpolated position that falls on the boundary separating adjacent hypercubes.
hed - Header file consisting of just the header lines from the segment file described immediately above.
cel - Daily summary statistics for each hypercube.
cel - Monthly statistics for each cell (where vertical layers are 1 km thick).
ce2 - Monthly statistics for each cell (where vertical layers are 2 km thick).
ce4 - Monthly statistics for each cell (where vertical layers are 4 km thick).
ce8 - Monthly statistics for each cell (where vertical layers are 8 km thick).

Additionally there is a file YYYYMMhrs.xls or YYYYMMehr.xls that is a Microsoft Excel spreadsheet containing a consolidated monthly summary of hourly totals of both number of segments and cumulative segment length. There is also a file tmoYYYYMM.val that is a monthly validation file of the following format.

What's all this stuff about hypercubes?

The hypercube terminology is a heuristic concept that is intended to help in understanding how the analysis region is subdivided. The analysis region for a day is 8 km thick (from 7 km to 15 km MSL), 75 degrees wide (from 135 degrees West to 60 degrees West), 30 degrees high (from 20 degrees North to 50 degrees North), and 24 hours in duration (from 0 UTC to 24 UTC). Although it is clearly understood that this region is curved, and that the methods used for subdividing the region into smaller regions account for the curvature, it is easier to think of everything all nice and flat and orthogonal, and hypercube is easier than hyperwhateverelseyoucanthinkoftocallit. The 'hyper' derives from the fact that it has four dimensions while your standard ordinary every-day run-of-the-mill cube has but three. You can then think of the large analysis region divided into smaller regions 1 km by 1 degree by 1 degree by 1 hour that are the small hypercubes referenced in the Data Type Description section above. At one point I described them as 'cells', which accounts for the (.cel) extension. You may think of them as tesseracts if that makes you more comfortable.

If you have questions about this site, you may send email to Don Garber at donald.p.garber@nasa.gov.

This page was last modified on 27 July 2004.