LSTW: Large-Scale Traffic and Weather Events Dataset

Published:

Description

LSTW is a large-scale, country-wide dataset for transportation and traffic research, which contains traffic and weather event data for the United States. In terms of traffic, we have several types of events including accident, congestion, construction, etc. In terms of weather events, we have several types including rain, snow, storm, cold weather event, etc. This dataset is continuously being collected from August 2016, and today it contains about 29.5 million traffic and weather events. Please read below descriptions for further details on this dataset.

Frequency distribution of traffic events in Bay Area from August 2016 to June 2019 using Kepler.gl

Acknowledgment

Please cite the following paper if you use this dataset:

This dataset is being distributed only for Research purposes, under Creative Commons Attribution-Noncommercial-ShareAlike license (CC BY-NC-SA 4.0). By clicking on the download buttons below, you are agreeing to use this data only for non-commercial, research, or academic applications. You may cite the above paper if you use this dataset.

Download (Version 3, December 2019)

In this version, we provide data which is collected from August 2016 to the end of December 2019 for the Contiguous United States. In this set, we have about 25.1 million traffic and 4.4 million weather events.

  • Download Traffic data from here.
  • Download Weather data from here.

The next version will be available by December 2020.

Traffic Events

Traffic event is a spatiotemporal entity, where such an entity is associated with location and time. Following is the description of available traffic event types:

  • Accident: Refers to traffic accident which can involve one or more vehicles.
  • Broken-Vehicle: Refers to the situation when there is one (or more) disabled vehicle(s) in a road.
  • Congestion: Refers to the situation when the speed of traffic is lower than the expected speed.
  • Construction: Refers to an on-going construction, maintenance, or re-paring project in a road.
  • Event: Refers to the situations such as sport event, concert, and demonstration.
  • Lane-Blocked: Refers to the cases when we have blocked lane(s) due to traffic or weather condition.
  • Flow-Incident: Refers to all other types of traffic events. Examples are broken traffic light and animal in the road.

The traffic data is provided in terms of a CSV file with the following attributes:

#AttributeDescriptionNullable
1EventIdThis is the identifier of a recordNo
2TypeThe type of an event; examples are accident and congestion.No
3SeverityThe severity of an event, wherever applicable. For a traffic event, severity is a number between 1 and 4, where 1 indicates the least impact on traffic (i.e., short delay as a result of the event) and 4 indicates a significant impact on traffic (i.e., long delay).No
4TMCEach traffic event has a Traffic Message Channel (TMC) code which provides a more detailed description on type of the event.No
5DescriptionThe natural language description of an event.No
6StartTime (UTC)The start time of an event in UTC time zone.No
7EndTime (UTC)The end time of an event in UTC time zone.No
8TimeZoneThe US-based timezone based on the location of an event (eastern, central, mountain, and pacific).No
9LocationLatThe latitude in GPS coordinate.Yes
10LocationLngThe longitude in GPS coordinate.Yes
11Distance (mi)The length of the road extent affected by the event.Yes
12AirportCodeThe closest airport station to the location of a traffic event.Yes
13NumberThe street number in address record.Yes
14StreetThe street name in address record.Yes
15SideThe relative side of a street (R/L) in address record.Yes
16CityThe city in address record.Yes
17CountyThe county in address record.Yes
18StateThe state in address record.Yes
19ZipCodeThe zipcode in address record.Yes

Weather Events

Weather event is a spatiotemporal entity, where such an entity is associated with location and time. Following is the description of available weather event types:

  • Severe-Cold: The case of having extremely low temperature, with temperature below -23.7 degrees of Celsius.
  • Fog: The case where there is low visibility condition as a result of fog or haze.
  • Hail: The case of having solid precipitation including ice pellets and hail.
  • Rain: The case of having rain, ranging from light to heavy.
  • Snow: The case of having snow, ranging from light to heavy.
  • Storm: The extremely windy condition, where the wind speed is at least 60 km/h.
  • Other Precipitation: Any other type of precipitation which cannot be assigned to previously described event types.

Visit our paper to learn how we determine type and severity of weather events.

The weather data is provided in terms of a CSV file with the following attributes:

#AttributeDescriptionNullable
1EventIdThis is the identifier of a recordNo
2TypeThe type of an event; examples are rain and snow.No
3SeverityThe severity of an event, wherever applicable.Yes
4StartTime (UTC)The start time of an event in UTC time zone.No
5EndTime (UTC)The end time of an event in UTC time zone.No
6TimeZoneThe US-based timezone based on the location of an event (eastern, central, mountain, and pacific).No
7LocationLatThe latitude in GPS coordinate.Yes
8LocationLngThe longitude in GPS coordinate.Yes
9AirportCodeThe airport station that a weather event is reported from.Yes
10CityThe city in address record.Yes
11CountyThe county in address record.Yes
12StateThe state in address record.Yes
13ZipCodeThe zipcode in address record.Yes

Coverage

The data coverage is country-wide. It currently contains data for the Contiguous United States. Following diagram shows the current frequency distribution of traffic events across 50 difference states in US.

Note: For the following states, we have no traffic data from August 2016 to August 2017: AL, AR, AZ, CO, ID, KS, KY, LA, ME, MN, MS, MT, NC, ND, NH, NM, NV, OK, OR, SD, TN, UT, VT, WI, and WY.

traffic_frequency

The next diagram shows the current frequency distribution of weather events for the same set of states.

weather_frequency

Following diagram shows the current distribution of different traffic event types:

traffic_types

Similarly, current distribution of weather event types is shown in below diagram:

weather_types

Applications of Dataset

This dataset can be used for plenty of purposes such as traffic analysis and prediction, impact prediction, accident prediction, routing engine optimization, travel time estimation, and many other research applications.



Page Visits