Daily data showing UK flight numbers and rolling seven-day average, including flights to, from, and within the UK. These are official statistics in development. Source: EUROCONTROL.

As new technologies are developed to handle the complexities of the Next Generation Air Transportation System (NextGen), it is increasingly important to address both current and future safety concerns along with the operational, environmental, and efficiency issues within the National Airspace System (NAS). In recent years, the Federal Aviation Administration’s (FAA) safety offices have been researching ways to utilize the many safety databases maintained by the FAA, such as those involving flight recorders, radar tracks, weather, and many other high-volume sensors, in order to monitor this unique and complex system. Although a number of current technologies do monitor the frequency of known safety risks in the NAS, very few methods currently exist that are capable of analyzing large data repositories with the purpose of discovering new and previously unmonitored safety risks. While monitoring the frequency of known events in the NAS enables mitigation of already identified problems, a more proactive approach of finding unidentified issues still needs to be addressed. This is especially important in the proactive identification of new, emergent safety issues that may result from the planned introduction of advanced NextGen air traffic management technologies and procedures. Development of an automated tool that continuously evaluates the NAS to discover both events exhibiting flight characteristics indicative of safety-related concerns as well as operational anomalies will heighten the awareness of such situations in the aviation community and serve to increase the overall safety of the NAS. This paper discusses the extension of previous anomaly detection work to identify operationally significant flights within the highly complex airspace encompassing the New York area of operations, focusing on the major airports of Newark International (EWR), LaGuardia International (LGA), and John F. Kennedy International (JFK). In addition, flight traffic in the vicinity of Denver International (DEN) airport/airspace is also investigated to evaluate the impact on operations due to variances in seasonal weather and airport elevation. From our previous research, subject matter experts determined that some of the identified anomalies were significant, but could not reach conclusive findings without additional supportive data. To advance this research further, causal examination using domain experts is continued along with the integration of air traffic control (ATC) voice data to shed much needed insight into resolving which flight characteristic(s) may be impacting an aircraft's unusual profile. Once a flight characteristic is identified, it could be included in a list of potential safety precursors. This paper also describes a process that has been developed and implemented to automatically identify and produce daily reports on flights of interest from the previous day.

Motivation

The data in this dataset is derived and cleaned from the full OpenSky dataset to illustrate the development of air traffic during the COVID-19 pandemic. It spans all flights seen by the network's more than 2500 members since 1 January 2019. More data will be periodically included in the dataset until the end of the COVID-19 pandemic.

License

See LICENSE.txt

Disclaimer

The data provided in the files is provided as is. Despite our best efforts at filtering out potential issues, some information could be erroneous.

• Origin and destination airports are computed online based on the ADS-B trajectories on approach/takeoff: no crosschecking with external sources of data has been conducted.
  Fields origin or destination are empty when no airport could be found.
• Aircraft information come from the OpenSky aircraft database. Fields typecode and registration are empty when the aircraft is not present in the database.

Description of the dataset

One file per month is provided as a csv file with the following features:

• callsign: the identifier of the flight displayed on ATC screens (usually the first three letters are reserved for an airline: AFR for Air France, DLH for Lufthansa, etc.)
• number: the commercial number of the flight, when available (the matching with the callsign comes from public open API)
• icao24: the transponder unique identification number;
• registration: the aircraft tail number (when available);
• typecode: the aircraft model type (when available);
• origin: a four letter code for the origin airport of the flight (when available);
• destination: a four letter code for the destination airport of the flight (when available);
• firstseen: the UTC timestamp of the first message received by the OpenSky Network;
• lastseen: the UTC timestamp of the last message received by the OpenSky Network;
• day: the UTC day of the last message received by the OpenSky Network;
• latitude_1, longitude_1, altitude_1: the first detected position of the aircraft;
• latitude_2, longitude_2, altitude_2: the last detected position of the aircraft.

Examples

Possible visualisations and a more detailed description of the data are available at the following page:
<https://traffic-viz.github.io/scenarios/covid19.html>

Credit

If you use this dataset, please cite:

Martin Strohmeier, Xavier Olive, Jannis Lübbe, Matthias Schäfer, and Vincent Lenders
"Crowdsourced air traffic data from the OpenSky Network 2019–2020"
Earth System Science Data 13(2), 2021
https://doi.org/10.5194/essd-13-357-2021

The number of flights performed globally by the airline industry has increased steadily since the early 2000s and reached **** million in 2019. However, due to the coronavirus pandemic, the number of flights dropped to **** million in 2020. The flight volume increased again in the following years and was forecasted to reach ** million in 2025.

Context

The U.S. Department of Transportation's (DOT) Bureau of Transportation Statistics tracks the on-time performance of domestic flights operated by large air carriers. I came across this useful data from DOT's database at working and figured this would be a really helpful dataset: Summary information on the number of on-time, delayed, canceled, and diverted flight.

Content

The datasets contain daily airline information covering from flight information, carrier company, to taxing-in, taxing-out time, and generalized delay reason of exactly 10 years, from 2009 to 2019. The DOT's database is renewed from 2018, so there might be a minor change in the column names.

Acknowledgements

The flight delay and cancellation data were collected and managed by the DOT's Bureau of Transportation Statistics, only included data related to time-analysis on each flight. For any inspiration, please see tasks.

This dataset contains the records of all the flights in the Northern California TRACON. The data was provided by the aircraft noise abatement office (http://www.flyquietsfo.com/) of San Francisco International Airport. The data cover Jan-Mar 2006. It is organized by day and flight. Each record contains some information about the flight and a sequence of 3D position and estimated speed. This data contains thousands of trajectories that can be used for trajectory clustering. The data is used by the Aircraft Noise Abatement Office to analyze the trajectories of aircraft flying in and out SFO. The objective is to minimize the noise pollution due to aircraft in the San Francisco Bay Area The files have the extension "lt6" and are organized as follow, one file per day. line number & explaination 1 TRACK OPNUM (TRACK header word and operation number) 2 eventid (Corralation number) 3 trackstart date (in time since 1900, A8 version four year digit) 4 trackstart time HH:MM:SS 5 trackend time HH:MM:SS 6 airportid 7 ACID (FLIGHTNUM/TAILNUMBER) 8 owner name 9 aircrafttype 10 aircraft category 11 beacon 12 adflag 13 waypoint 14 other_port (dest/origin) 15 runwayname 16 min alt 17 max alt 18 min range 19 max range 20 Count of trackpoints (to follow) 21 x,y,z,v,t (all points is meters relative to MRP, velocity and time from start of track)

Our Flight Events data feed combines Spire Global satellite/terrestrial ADS-B flight event data with ch-aviation’s fleet, operator, and airport data providing an overview of all flights operated by airlines, business and general aviation players on a daily basis.

The value of our Flight Events data feed lies in its high-resolution integration of ADS-B flight tracking with ch-aviation’s comprehensive aircraft and operator data, delivering unmatched visibility into global aircraft movements. By identifying the aircraft type and registration for approximately 98% of all ADS-B-tracked flights, we offer an industry-leading solution for lessors, insurers, airports, OEMs, and analysts seeking precise, reliable, and actionable aviation intelligence.

• High-Resolution ADS-B Integration - Satellite and terrestrial ADS-B flight tracking combined with enriched aircraft and operator data for maximum accuracy and visibility • Comprehensive Aircraft Identification - Aircraft type and registration identified for approximately 98% of all ADS-B-tracked flights, using proprietary matching with ch-aviation data and supplementary publicly available authority data sources. • Global Flight Coverage - Tracks approximately 160,000–190,000 flights per day across commercial aviation, business jet, and general aviation sectors worldwide. • ACMI (Wet-Lease) and Cargo Customer Tracking - Detailed monitoring of ACMI operations, including identification of wet-lease activity between different operators as well as cargo customers identifying flights operated for integrators like DHL Express or FedEx as well as cargo customers such as Amazon. • Aircraft Utilisation Tracking - Tracking of flight hours and cycles at both the operator and individual tail number (aircraft) level • Matched Operator and Aircraft Data - Every flight is linked to comprehensive ch-aviation datasets, including aircraft ID, history, operator, variant, callsign, and airport details allowing customers to leverage the industry’s most comprehensive integration between ADS-B flight event and fleet/operator/airport data. • Fallback Data Enrichment - Where ch-aviation data is unavailable, civil aviation authority and ANSP sources are used to ensure continuity in aircraft identification and data accuracy. • Use Case-Driven Insights - Tailored for industry stakeholders like lessors, insurers, OEMs, airports, and analysts seeking operational, commercial, and technical flight data intelligence.

ch-aviation integrates its Commercial Aviation Aircraft Data and Business Jet Aircraft Data with Spire Global’s satellite-based ADS-B data that is fused by Spire with terrestrial feeds from AirNav and Wingbits.

This data is enriched with mapped callsigns, corrected hexcodes, regional partnership decoding, and identification of wet-leases and cargo customers, enabling detailed insight into each individual flight.

Where ch-aviation data is unavailable, public data from civil aviation authorities and ANSPs is used to ensure broad and reliable aircraft identification and coverage.

The data set is available historically going back to January 1, 2018.

The data set is updated daily.

The sample data shows flights on 2025-03-30, with Swiss, Alaska Airlines, Horizon Air, Jet Aviation Business Jets, and RVR Aviation as operators or wet lease customers.

Contact us to get access to ch-aviation's AWS S3 sample data bucket as well allowing you to build proof of concepts with all of our sample data.

The direct bucket URL for this data set is: https://eu-central-1.console.aws.amazon.com/s3/buckets/dataservices-standardised-samples?region=eu-central-1&bucketType=general&prefix=flights/&showversions=false

Full Technical Data Dictionary: https://about.ch-aviation.com/flights-2/

Passengers enplaned and deplaned at Canadian airports, annual.

Daily aircraft utilisation is available for all commercial aviation and business jet aircraft showing the number of flight hours and cycles every day (in UTC) time based on a combination of Spire Global satellite/terrestrial ADS-B data and ch-aviation fleet data.

The data set includes hours, cycles, average stage length as well as data quality indicators for each record.

The data set is updated daily.

The sample data shows aircraft flown on 2025-03-30 by Swiss, Alaska Airlines, Horizon Air, Jet Aviation Business Jets, and RVR Aviation, with utilization metrics

Contact us to get access to ch-aviation's AWS S3 sample data bucket as well allowing you to build proof of concepts with all of our sample data.

The direct bucket URL for this data set is: https://eu-central-1.console.aws.amazon.com/s3/buckets/dataservices-standardised-samples?region=eu-central-1&bucketType=general&prefix=aircraft_utilisation_daily/&showversions=false

Full Technical Data Dictionary: https://about.ch-aviation.com/aircraft-utilisation-daily-ads-b-based-2/

Context

The U.S. Department of Transportation's (DOT) Bureau of Transportation Statistics (BTS) tracks the on-time performance of domestic flights operated by large air carriers. Summary information on the number of on-time, delayed, canceled and diverted flights appears in DOT's monthly Air Travel Consumer Report, published about 30 days after the month's end, as well as in on-time data posted on this website.

Content

A flight is considered delayed if it is late by more than 15 minutes.

Factors considered: Year: The year of the flight (all records are from 2013) Month: The month of the flight DayofMonth: The day of the month on which the flight departed DayOfWeek: The day of the week on which the flight departed - from 1 (Monday) to 7 (Sunday) Carrier: The two-letter abbreviation for the airline. OriginAirportID: A unique numeric identifier for the departure aiport OriginAirportName: The full name of the departure airport OriginCity: The departure airport city OriginState: The departure airport state DestAirportID: A unique numeric identifier for the destination aiport DestAirportName: The full name of the destination airport DestCity: The destination airport city DestState: The destination airport state CRSDepTime: The scheduled departure time DepDelay: The number of minutes departure was delayed (flight that left ahead of schedule have a negative value) DelDelay15: A binary indicator that departure was delayed by more than 15 minutes (and therefore considered "late") CRSArrTime: The scheduled arrival time ArrDelay: The number of minutes arrival was delayed (flight that arrived ahead of schedule have a negative value) ArrDelay15: A binary indicator that arrival was delayed by more than 15 minutes (and therefore considered "late") Cancelled: A binary indicator that the flight was cancelled

Acknowledgements

The flight delay and cancellation data was collected and published by the DOT's Bureau of Transportation Statistics.

India All Scheduled Airlines: Domestic: Number of Flight data was reported at 102,319.000 Unit in Mar 2025. This records an increase from the previous number of 92,291.000 Unit for Feb 2025. India All Scheduled Airlines: Domestic: Number of Flight data is updated monthly, averaging 48,100.000 Unit from Apr 2001 (Median) to Mar 2025, with 288 observations. The data reached an all-time high of 102,319.000 Unit in Mar 2025 and a record low of 188.000 Unit in Apr 2020. India All Scheduled Airlines: Domestic: Number of Flight data remains active status in CEIC and is reported by Directorate General of Civil Aviation. The data is categorized under India Premium Database’s Transportation, Post and Telecom Sector – Table IN.TA019: Airline Statistics: All Scheduled Airlines.

This data set provides high-frequency wind speed and direction data for the C-23 Sherpa aircraft during airborne campaigns over the Alaskan and Canadian Arctic as part of the Carbon in Arctic Reservoirs Vulnerability Experiment (CARVE). The data were collected in situ using the Aventech AIMMS-30 Airborne Wind Sensor onboard the aircraft and are presented at 1-second intervals throughout each flight. The Winds instrument was available for flights in year 2015 only. The measurements included in this data set are most useful when paired with the scientific data collected by other CARVE airborne instruments.

This data set provides high-frequency geolocation, time, height, pitch, roll, and heading information for the C-23 Sherpa aircraft during airborne campaigns over the Alaskan and Canadian Arctic as part of the Carbon in Arctic Reservoirs Vulnerability Experiment (CARVE). The data were collected in situ using the Digital Air Data System (DADS) onboard the aircraft and are presented at 1-second intervals throughout each flight. Flight campaigns took place from 2012 to 2015 between the months of March and November to enable investigation of both seasonal and inter-annual variability in atmospheric gas content. The measurements included in this data set are useful for matching aircraft position with the scientific data collected by other CARVE airborne instruments.

India All Scheduled Airlines: International: Number of Flight data was reported at 18,502.000 Unit in Mar 2025. This records an increase from the previous number of 16,668.000 Unit for Feb 2025. India All Scheduled Airlines: International: Number of Flight data is updated monthly, averaging 7,797.000 Unit from Apr 2001 (Median) to Mar 2025, with 283 observations. The data reached an all-time high of 18,574.000 Unit in Jan 2025 and a record low of 273.000 Unit in May 2020. India All Scheduled Airlines: International: Number of Flight data remains active status in CEIC and is reported by Directorate General of Civil Aviation. The data is categorized under India Premium Database’s Transportation, Post and Telecom Sector – Table IN.TA019: Airline Statistics: All Scheduled Airlines.

Last known aircraft location shows where an aircraft completed its last flight before midnight UTC yesterday based on Spire satellite ADS-B and ch-aviation fleet data. The aircraft may have been flying at the time or after that, so it is just an indication of where in the world aircraft where at the time.

The data set is updated daily.

The sample data shows the last known locations of aircraft operated by Alaska Airlines, Swiss, Horizon Air, Jet Aviation Business Jets, and RVR Aviation.

Contact us to get access to ch-aviation's AWS S3 sample data bucket as well allowing you to build proof of concepts with all of our sample data.

The direct bucket URL for this data set is: https://eu-central-1.console.aws.amazon.com/s3/buckets/dataservices-standardised-samples?region=eu-central-1&bucketType=general&prefix=last_aircraft_location/&showversions=false

Full Technical Data Dictionary: https://about.ch-aviation.com/last-aircraft-location/

The Spire-ch-aviation Mapping data files allow Spire Global flight IDs to be mapped to ch-aviation Aircraft History Data (aircraft_id and the date_from_full (the combined primary key of ch-aviation’s aircraft histories data)), while also providing regional partnership, cargo customer, wet-lease customer, flight type information for each flight. This allows Spire Global ADS-B customers to map each flight to ch-aviation’s Aircraft Data (Aircraft History, Aircraft, Operators) without having to build any of the logic on your side. We do all the mapping on your behalf.

The data set is updated daily.

The sample data shows flights on 2025-03-30 involving Swiss, Alaska Airlines, Horizon Air, Jet Aviation Business Jets, and RVR Aviation as operators or wet lease customers.

Contact us to get access to ch-aviation's AWS S3 sample data bucket as well allowing you to build proof of concepts with all of our sample data.

The direct bucket URL for this data set is: https://eu-central-1.console.aws.amazon.com/s3/buckets/dataservices-standardised-samples?region=eu-central-1&bucketType=general&prefix=spire_ch_flight_mapping/&showversions=false

Full Technical Data Dictionary: https://about.ch-aviation.com/spire-ch-aviation-mapping/

Abstract

The airport Nantes Atlantique (IATA-Code: NTE, ICAO-Code: LFRS) in west France is a mid-sized international airport with around 7Mil passengers and 60.000 commercial flights annually (as of 2024). This dataset is a SQL dump of 3D flight paths and meta data of all arrival and departure flights which were equipped with an ADS-B transponder for the period 2020-2023.

Data recording

The ADS-B transmissions were recorded with an ADS-B receiver (by Flightaware), processed with custom written Python scripts and stored in a PostGis (v2.5) enabled PostGreSQL (v11.2) database. Only flight positions below 10.000 feet AGL (Above Ground Level, as determined by the aircraft) were stored, to filter out flights passing the area at cruise altitude (typically 30.000+ feet). Landing and takeoff events were identified using spatial relations and the change of the transmitted “on-ground” aircraft status. Based on the velocity vector of the aircraft during touchdown or takeoff, the runway was identified: 21 (compass direction 210°) or 03 (compass direction 30°).

Schemata and tables

The “public” schema includes the processed ADS-B data, broken down into four tables:

Positions table

Individual aircraft position records contain a unique id, parent flight id, a PointZ geometry (EPSG:4326 latitude & longitude and height AGL in meter), UTC time stamp and the “on-ground” boolean flag.

Flights table

This table contains two important ADS-B information: The ICAO hex code of the ADS-B transponder of the aircraft and the call sign of the flight. As transponder hardware is transferred between aircraft only in very rare cases, the hex code can be used to identify individual physical aircraft in the dataset. The call sign, on the other hand, is assigned to the flight by the airline and can change on a daily basis. As its first 3-4 characters is the ICAO abbreviation of the airline name, the full name can be deduced by joining these characters with the “airlines” table from the “meta” schema.

Landings and Takeoffs tables

These records contain the identified landing and takeoff events as flight id, landing or takeoff UTC time and the runway number.

The flight id acts as foreign key allowing the join of all four tables.

Limitations

Only ADS-B equipped flights, and which transmit their position, are contained in this dataset. By November 2024 approximately 97% of all civil aircraft in Europe were equipped with this transponder type (https://www.eurocontrol.int/service/adsb-equipage). Yet, military aircraft and some special flights might choose not to transmit.

The altitudes AGL of some flight tracks might have a bias of up to 100 feet relative to the airport altitude, such that there is a jump in altitude upon a landing or takeoff event (all “on-ground=TRUE” positions are forced to an altitude of 0m AGL). This bias is in the received ADS-B data and stored without correction.

Some flights land and takeoff within a short time period without switching off their ADS-B transponder, or practice touch-and-go maneuvers. This leads to their approach and departure legs forming one continuous track. Yet, each landing and takeoff event is identified and stored with a unique id.

Landing and takeoff events of the rescue helicopter of the Nantes hospital (CHU Nantes) are stored under its flight id SAMU44, as it lands for refueling at the airport.

This dataset contains data related to Air Traffic Management hotspots. Hotspots are created in the European airspaces when capacity for some pieces of airspace are foreseen to be infringed due to weather, congestion, strikes, etc. This anonymised dataset records around 5900 hotspots happening at 22 major European airports. These hotspots are generated through a simulator called Mercury that is fed with real data (in particular, real capacity reduction that happened in Europe for over a year, schedules etc) and simulates a day of operation, randomising events like delays, cancellation etc. More details on mercury can be found here [1] and [2].

The data, anonymised in terms of airports and airlines, is a dictionary which is structured as follows:

- the top level key is the id of the airport, the value is list a of all regulations available for this airport.

- each item of the list is a dictionary, with keys:

-- 'slot_times': list of all slots available to flights for this hotspot/regulation, in minutes since midnight.

-- 'etas': list of initial estimated arrival times of flights involved in the regulation, in minutes since midnight.

-- 'flight_ids': list of flight ids (in the same order than etas)

-- 'cost_vectors': list of cost vectors. Each item is a list itself, of length equal to the slot_times list. Each element of that list is the estimated cost that the airline owning the flight would incur, were the flight be assigned to this slot, in terms of: maintenance, crew, rebooking fees, market value loss, and curfew infringement, in 2014 euros. This cost is computed within the Mercury model and is based on [3].

-- 'airlines_flights': dictionary whose keys are airline ids and values are lists of ids of flights owned by the airline.

[1] https://www.sciencedirect.com/science/article/abs/pii/S0968090X21003600

[2] G. Gurtner, L. Delgado, and D.Valput, “An agent-based model for air transportation to capture network effects in assessing delay management mechanisms”, Transportation Research Part C: emerging Technologies, 2021.

Pre-print available here: https://westminsterresearch.westminster.ac.uk/item/v956w/an-agent-based-model-for-air-transportation-to-capture-network-effects-in-assessing-delay-management-mechanisms

[3] A. J. Cook and G. Tanner, “European airline delay cost reference values - updated and extended values (Version 4.1),” University of Westminster, London, 2015a

One of the main goals of the HALO-(AC)3 Arctic airborne campaign conducted in spring 2022 was a quasi-Lagrangian sampling of air masses. This means that the same air masses were to be sampled twice. Such a measurement strategy allows for concrete observations of air mass transformations, which can for example be used to benchmark weather models. After finishing the campaign, trajectory calculations were conducted for all flights of the HALO aircraft to check whether the flight strategy was a success. For this, the trajectory calculation tool Lagranto was used in conjunction with wind fields from the ERA5 reanalysis. Latter has an output resolution of around 30 km and one hour. The hourly data was bi-linearly interpolated to one minute resolution. Air masses were initialized temporally every one minute along the flight track of HALO, and vertically every 5 hPa between 250 hPa and 10 hPa above the ground level. Horizontally, air masses were started within a 30 km radius circle centered around the location of HALO. In each circle, 30 air masses were initialized evenly spaced approximately every 10 km, allowing for a better statistical analysis. Trajectories were then calculated in one-minute steps up to 10 hours forward in time. At typical HALO flight times of up to 10 hours, around 2.7 million trajectories were calculated per research flight. A quasi-Lagrangian match is registered if the same air mass is seen again below HALO at a later time of the same day and within the same 30 km radius. In order to allow air masses to evolve, a minimum temporal threshold of one hour was applied between matches. For every research flight (RF) of the HALO aircraft during HALO-(AC)3, quasi-Lagrangian matches on the same day are calculated using Lagranto and ERA5 input. For every RF, two files were uploaded: a quicklook png file, and the match data itself. The match data looks like this:Reference date 20220312_0800 / Time range 600 mintime lon lat p alt------------------------------------0.22 19.66 67.81 715 22918.40 21.20 67.94 776 17750.22 19.61 67.72 720 21588.40 21.15 67.96 786 1675....In the header, the reference date is given as YYYYMMDD_HHMM (UTC time). Trajectories had been calculated 10 hours forward (time range: 600 min). Then, a long list of matches is given. Let's look at the first pair of matches. An air mass is seen first at 08:22 UTC (08 + 0.22) below HALO. The lon and lat coordinates are then given, as well as the pressure level of 715 hPa. This corresponds to an altitude of approx. 2291 m above ground level, using the ERA5 geopotential height. This same air mass is seen more than 8 hours later, at 16:40 UTC (08 + 8.40). As it was moving along the prevailing winds, it is now seen at slightly different lon/lat coordinates and a different altitude (776 hPa, which corresponds to 1775 m above ground).In the quicklook, the radar reflecitivity is plotted in the lowest 10 km above ground. At the bottom of the figure, the ERA5 sea-ice concentration is shown, to highlight the surface type HALO was passing over. Finally, all quasi-Lagrangian matches are overlaid. For each orange dot (first match), there is a corresponding red dot (second match), meaning that the same air mass was observed again at a later stage and somewhere below HALO.

This dataset was established in the context of the Master Thesis of Arthur Garreau and Kilian Meier in 2020 and 2021. And was published with the resulting paper: Kilian Meier, Richard Hann, Jan Skaloud, and Arthur Garreau: "Wind Estimation with Multirotor UAVs", Atmosphere 2022, 13(4), 551; https://doi.org/10.3390/atmos13040551 The dataset contains flight logs from DJI Phantom 4 drones flown in Switzerland and Norway, as well as corresponding weather data: wind, temperature, pressure and humidity. There are a total of 75 flights spread over 13 days (56 flights in Switzerland and 19 flights in Norway). Flight locations are in Adventadlen at 78°12’10.0"N15°49’41.0"E and on the EPFL campus 46°31’17.0"N 6°34’02.5"E. Flight logs are labeled by location (EPFL and Svalbard) and date. Each folder represents a given flight day and is name respecting the following naming convention "LocationYYYYMMDD". Each folder contains two subfolders: "FLIGHT" and "WEATHER". They contain respectively, flight data and weather data. The "FLIGHT" folder contains the original raw DJI flight log (.DAT, proprietary) and the extracted human readable data as a CSV file. This data was extracted using DatCon. The "WEATHER" folder contains weather sensor data in human readable text form (.txt, .dat). The exact content depends on the available sensors. The "DATA_OVERVIEW.xlsx" file contains a table describing each flight (start time, end time, path to data, etc). For more details, please consult the paper written by Meier et al. https://doi.org/10.3390/atmos13040551. More information and related thesis to be found on https://folk.ntnu.no/richahan/. GitHub repository https://github.com/meierkilian/WEMUAV