ATM Data Tools

Tools for air traffic data analysis. The package currently consists of 3 functions for reading ADSB files. A feature to allow filtering of track data by radius from an airport is WIP for the read_adsb_byairport() function.

Installation

This package is available via PyPi:

pip install atmdatatools

Import

You can import the package by using

import atmdatatools as adt

Functions

read_adsb()

read_adsb(fname, datestr, downsample=None, floor=None, ceiling=None)

Reads raw ADSB files, assuming that each file only has 1 day worth of data. Raw ADSB files should have the columns ['073:071_073TimeforPos','131:Latitude','131:Longitude','140:GeometricHeight','170:TargetID']. The column 073:071_073TimeforPos contains the time elapsed in seconds in UTC and the timezone of the file is currently hardcoded for UTC+8. Hence, the first row of this column starts at approximately 57,600s representing 16:00 UTC of the previous day.

Returns a GeoDataFrame with the columns ['id','datetime','unix_timestamp','geometry']. Each row represents 1 ADSB position report.

Parameters

• fname: the filename of the adsb csv file
• datestr: date string in the format 'YYYYMMDD'
• [Optional] downsample = None: downsample the track data, e.g. 2 will take every 2nd point in the track data
• [Optional] floor = None: cuts off tracks below this altitude in feet, recommend 100 feet if you wish to exclude ground movements
• [Optional] ceiling = None: cuts off tracks above this altitude in feet

Returns

• GeoDataFrame: A GeoDataFrame with the columns ['id','datetime','unix_timestamp','geometry']. Each row represents 1 ADSB position report.
  • id contains the flight number (e.g. SIA92). Increments the id of a position if a time gap of more than 15 minutes is detected. e.g. MEDIC77 becomes MEDIC77_1.
  • Times are in UTC
  • geometry column has Shapely Point objects with lat, long, and altitude information

read_adsb_byflightid()

read_adsb_byflightid(fname, datestr, flightid, downsample=None, floor=None, ceiling=None)

Reads raw ADSB files, assuming that each file only has 1 day worth of data, returning a GeoDataFrame with flight tracks of the desired flightid. Each row represents 1 ADSB position report. Parameters allow for additional filtering.

See the read_adsb() function documentation for raw ADSB file format.

Parameters

• fname: the filename of the adsb csv file
• datestr: date string in the format 'YYYYMMDD'
• airport: Accepts ICAO airport codes, currently accepts WSSS, WSSL.
• [Optional] downsample = None: downsample the track data, e.g. 2 will take every 2nd point in the track data
• [Optional] floor = None: cuts off tracks below this altitude in feet, recommend 100 feet if you wish to exclude ground movements
• [Optional] ceiling = None: cuts off tracks above this altitude in feet

Returns

• GeoDataFrame: A GeoDataFrame with the columns ['id','datetime','unix_timestamp','geometry']. Each row represents 1 ADSB position report.
  • id contains the flight number (e.g. SIA92). Increments the id of a position if a time gap of more than 15 minutes is detected. e.g. MEDIC77 becomes MEDIC77_1.
  • Times are in UTC
  • geometry column has Shapely Point objects with lat, long, and altitude information

read_adsb_byairport()

read_adsb_byairport(fname, datestr, airport, arrdep=None, downsample=None, floor=None, ceiling=None)

Reads raw ADSB files, assuming that each file only has 1 day worth of data, returning a GeoDataFrame with flight tracks filtered by airport of interest. Each row represents 1 flightpath. Parameters allow for additional filtering.

See the read_adsb() function documentation for raw ADSB file format.

Parameters

• fname: the filename of the adsb csv file
• datestr: date string in the format 'YYYYMMDD'
• airport: Accepts ICAO airport codes, currently accepts WSSS, WSSL.
• [Optional] arrdep = None: filter for arrival ("arr") or departure ("dep") traffic or both (None)
• [Optional] downsample = None: downsample the track data, e.g. 2 will take every 2nd point in the track data
• [Optional] floor = None: cuts off tracks below this altitude in feet, recommend 100 feet if you wish to exclude ground movements
• [Optional] ceiling = None: cuts off tracks above this altitude in feet

Returns

• GeoDataFrame: A GeoDataFrame with the columns ['id','geometry']. Each row represents 1 flightpath.
  • id contains the flight number (e.g. SIA92). Increments the id of a position if a time gap of more than 15 minutes is detected. e.g. MEDIC77 becomes MEDIC77_1.
  • geometry column contains a Shapely LineString representing the flight's flightpath