flight-tracer 0.1.7

A package to fetch, process, store and plot aircraft trace data from ADS-B Exchange

FlightTracer: Tracking ADS-B Exchange flights

FlightTracer is a Python package for fetching, processing, storing and plotting flight trace data from ADS-B Exchange. It provides tools for managing flight data and offers flexible options for users. The project is new and under development.

---

Features

• Fetches flight trace data from ADS-B Exchange
• Identifies flight legs by detecting time gaps and handling Zulu day transitions
• Converts raw flight trace data into GeoDataFrames for easy analysis
• Saves processed flight data in CSV, GeoJSON or shapefile formats
• Visualizes flight paths with customizable basemaps and clear leg differentiation
• Supports AWS S3 uploads for cloud storage
• Offers customization options for file formats and filtering
• Now includes a command-line interface (CLI) for easy execution

---

Installation

FlightTracer is available on PyPI and can be installed using:

pip install flight-tracer

Alternatively, install the latest development version from GitHub:

pip install git+https://github.com/stiles/flight_tracer.git

Dependencies

FlightTracer requires the following Python packages:

• requests
• pandas
• geopandas
• boto3
• matplotlib
• contextily
• shapely
• click (for CLI support)

These dependencies will be installed automatically with pip.

---

Using FlightTracer via CLI

With the new CLI support, FlightTracer can be run directly from the command line.

CLI options overview

Option	Description
--icao	ICAO aircraft ID (required, can be multiple)
--start	Start date (YYYY-MM-DD)
--end	End date (YYYY-MM-DD)
--output	Directory for saving fetched data
--input	Path to input file for processing/exporting/uploading
--format	Output format: csv, geojson, shp
--filter-ground	Filter out ground-level points (default: True)
--plot	Generate a visualization of the flight trace
--bucket	AWS S3 bucket name for uploads
--aws-profile	AWS profile for authentication when uploading

These commands ensure seamless of FlightTracer's features via the command line.

Basic command: Fetch raw flight trace data

flight-tracer fetch --icao A11F59 --start 2025-02-07 --end 2025-02-08 --output data/

This fetches flight data for aircraft A11F59 for the given date range and saves it as a CSV file.

Processing fetched data

flight-tracer process --input data/raw_A11F59_2025-02-07_2025-02-08.csv --filter-ground

This processes the fetched flight trace data, filtering out ground-level points and saving the result in a structured format.

Exporting processed data

flight-tracer export --input data/processed_A11F59_2025-02-07_2025-02-08.geojson --format geojson

This exports the processed flight data in GeoJSON format (CSV and shapefile options are also available).

Plotting the flight trace

flight-tracer plot --input data/processed_A11F59_2025-02-07_2025-02-08.geojson --output visuals/flight_map_A11F59_2025-02-07_2025-02-08.png

This generates a visualization of the flight trace and saves it as an image file.

Uploading processed data to AWS S3

flight-tracer upload --input data/processed_A11F59_2025-02-07_2025-02-08.geojson --bucket my-bucket --aws-profile my-profile

This uploads the processed flight trace data to the specified AWS S3 bucket using the provided AWS profile.

---

Using FlightTracer in Python

FlightTracer can also be used as a Python library for more flexibility.

Basic example

from flight_tracer import FlightTracer
from datetime import date

# Initialize the FlightTracer with an aircraft ID
tracer = FlightTracer(aircraft_ids=["A11F59"])

# Define the date range for fetching trace data
start = date(2025, 2, 7)
end = date(2025, 2, 8)

# Fetch flight data
raw_df = tracer.get_traces(start, end)

# Process the raw data into a GeoDataFrame
if not raw_df.empty:
    gdf = tracer.process_flight_data(raw_df)
    print(gdf.head())

---

---

Using FlightTracer in Python

FlightTracer can also be used as a Python library for more flexibility.

Basic example

from flight_tracer import FlightTracer
from datetime import date

# Initialize the FlightTracer with an aircraft ID
tracer = FlightTracer(aircraft_ids=["A11F59"])

# Define the date range for fetching trace data
start = date(2025, 2, 7)
end = date(2025, 2, 8)

# Fetch flight data
raw_df = tracer.get_traces(start, end)

# Process the raw data into a GeoDataFrame
if not raw_df.empty:
    gdf = tracer.process_flight_data(raw_df)
    print(gdf.head())

Converting to a Specific Time Zone

By default, ADS-B times are in UTC. Users can convert point_time to their local time zone as needed:

import pytz

# Convert to US/Pacific Time
gdf["point_time_pacific"] = gdf["point_time"].dt.tz_localize("UTC").dt.tz_convert("US/Pacific")

# Convert to Eastern Time
gdf["point_time_eastern"] = gdf["point_time"].dt.tz_localize("UTC").dt.tz_convert("America/New_York")

To see all available time zones:

import pytz
print(pytz.all_timezones)

---

Customizing output

FlightTracer provides options to save data in different formats and configure the output directory:

Supported file formats

• CSV
• GeoJSON
• Esri shapefile

Example: Exporting data

# Save processed data locally
tracer.export_flight_data(gdf, base_path="data/flight_traces", export_format="geojson") # or "shp"

---

AWS S3 integration

Easily upload processed data to AWS S3 for cloud storage. Provide your AWS credentials or use an AWS profile:

Example: Uploading to S3

aws_creds = {
    "aws_access_key_id": "your-access-key",
    "aws_secret_access_key": "your-secret-key"
}

tracer.upload_to_s3(
    gdf,
    bucket_name="your-bucket",
    csv_object_name="flight_data.csv",
    geojson_object_name="flight_data.geojson"
)

---

Advanced features

Metadata mapping

Enrich your flight data with custom metadata using mapping options.

Example: Adding metadata

meta_df = pd.DataFrame({
    "flight": ["AAL124", "UAL1053"],
    "airline": ["American Airlines", "United Airlines"]
})
mapping_info = (meta_df, "flight", "airline", "airline")

# Pass mapping_info to process_flight_data
gdf = tracer.process_flight_data(raw_df, mapping_info=mapping_info)

Custom time thresholds for legs

Customize the time gap threshold for detecting new flight legs:

Example: Adjusting time gap threshold

tracer.set_time_gap_threshold(minutes=45)

---

Outputs

The example above would output two GeoJSON files: One with point features for each moment captured during the aircraft's flight and another with lines representing the overall route(s). Legs of the flights are differentiated in the flight_leg item. The script also outputs a CSV and a simple map plot.

"features": [
    {
        "type": "Feature",
        "properties": {
            "point_time": "2025-02-08T02:38:02.920",
            "flight_date_pst": "2025-02-07",
            "altitude": "30000",
            "ground_speed": 408.4,
            "heading": 253.2,
            "lat": 35.968307,
            "lon": -97.348509,
            "icao": "a11f59",
            "call_sign": "UAL333",
            "leg_id": 1,
            "flight_leg": "UAL333_leg1"
        },
        "geometry": {
            "type": "Point",
            "coordinates": [
                -97.348509,
                35.968307
            ]
        }
    }
]

Notes:

• Values such as altitude and ground speed are raw and uncorrected
• ADS-B datetimes are UTC, or Zulu, but FlightTracer users can concert that to a local time zone.

The plot has different colors for the various legs that day to help you identify them more clearly as you use the data for more advanced visualizations using QGIS or other tools.

---

Roadmap

• CLI option: Add a command-line interface for easier usage
• Improved metadata integration: Automatically enrich flight data with external sources (e.g., FAA, ICAO)
• Parallel processing: Optimize for large datasets
• Better visualizations: Add support for tools like Altair or Plotly
• Analysis tools: Better understand a flight's speed and altitude changes

---

Credits

Thanks to ADS-B Exchange for providing open flight data. Please consider supporting their service by subscribing or contributing data.

---

License

This project is licensed under the Creative Commons CC0 1.0 Universal Public Domain Dedication.