Data from: American robin behavioral responses to wildfire smoke in Oregon, USA
Data files
Jun 22, 2026 version files 78.06 KB
-
README.md
4.17 KB
-
robin_smoke_database_dryad.csv
73.90 KB
Abstract
Wildfire smoke is an increasing disturbance across much of the globe that has received little attention as a selective force in animal behavior. Animals may have evolved behavioral strategies to cope with smoke exposure, but what those strategies are and how they affect health outcomes are poorly described. We hypothesized that American robins (Turdus migratorius), a highly mobile, partial migrant songbird living in a fire-prone habitat, would leave impacted areas to avoid smoke from nearby wildfires during the post-breeding season. However, we found that they made shorter-distance movements as wildfire smoke intensified. Robins did not alter orientation behavior relative to the direction of the wind under light smoke, but they were more likely to orient into the wind when heavy smoke was present, which tended to reduce smoke exposure. Our results suggest that American robins shift their behavior during wildfire smoke events, potentially to reduce exposure to toxic air. While they do not immediately relocate, they may use different strategies as smoke becomes more intense – supporting our hypothesis that animal responses to wildfire smoke are complex and dose-dependent. Data were collected from GPS based satellite tracking of American robin songbirds in Oregon, USA during the wildfire season in 2023 - 2024 to determine if movement behavior changed under variable air quality conditions. Air quality was estimated using historical maps generated by AirNow.gov, which uses the NowCast algorithm and incorporates data from ground-based sensors and the NOAA-GOES-R satellites. The data were used in a variety of model structures to determine if distance moved, direction traveled or direction traveled relative to prevailing wind between successive reported robin locations changed under heavy smoke conditions. The dataset contains capture data (age, sex, tag ID, date, year) and tracking location data (longitude, latitude, date, year, wind speed, wind direction, AQI at origin, AQI at destination) and calculated data (movement direction, movement interval, distance moved, change in AQI and angle difference between robin movement direction and prevailing wind direction).
Dataset DOI: 10.5061/dryad.vhhmgqp8z
Description of the data and file structure
Files and variables
File: robin_smoke_database_dryad.csv
Description: Database of American robin locations. Each row represents a location reported by a GPS-satellite tag and the calculated movement data (e.g., distance, direction, etc) using the prior reported location, alongside air quality estimates, wind speed and wind direction for the day of the reported location.
Variables
- ID: The identification number of the satellite tag deployed on each individual robin.
- sex: Birds were identified using secondary sexual characteristics and size measurements as male (M), female (F) or unknown sex (U)
- age: Age of bird was identified based on plumage characteristics. All birds were adults (A).
- date: Date (month, day, year) for each location of the robin reported by the tag.
- day_of_year: Ordinal date where January 1 = 1 and December 31 = 365.
- doy_sc: day of year centered to a mean of zero and scaled to a standard deviation of 1.
- year: Year of tag deployment, which also determined how frequently locations were reported (daily in 2023 or once every 7 days in 2024).
- interval: Number of days that has passed since the prior reported location.
- long: Longitude of the reported location in decimal degrees.
- lat: Latitude of the reported location in decimal degrees.
- lat_sc: Latitude in decimal degrees centered to a mean of zero and scaled to a standard deviation of 1.
- robin_direction: Direction of the robin's movement that resulted in the current location, expressed as an angle from 0 - 360 degrees, where 0 is north, 90 is east, 180 is south and 270 is west.
- distance_km: Distance (km) of the movement that resulted in the current reported location.
- distance_interval: Distance (km) divided by the Interval (days) to give a distance moved per day.
- AQI_origin: Air quality category at the origin point of the movement, where 0 = PM2.5 < 9 μg/m3, 1 = PM2.5 between 9 μg/m3 to 35.59 μg/m3, 2 = PM2.5 between 35.6 μg/m3 and 55.4 μg/m3, and 3 = PM2.5 > 55.5 μg/m3.
- AQI_origin_cat: Air quality category at the origin point of the movement as described above, but with categories 2 and 3 collapsed into category 2 (i.e., PM2.5 > 35.6 μg/m3).
- AQI_destination: Air quality category at the destination point of the movement, where 0 = PM2.5 < 9 μg/m3, 1 = PM2.5 between 9 μg/m3 to 35.59 μg/m3, 2 = PM2.5 between 35.6 μg/m3 and 55.4 μg/m3, and 3 = PM2.5 > 55.5 μg/m3.
- AQI_differential: The AQI at destination minus the AQI at origin to calculate the change in AQI category as a result of the movement.
- robin_ew: East-West vector component of movement after trigonometric transformation in km.
- robin_ns: North-south vector component of movement after trigonometric transformation in km.
- wind_speed: Average wind speed (m/s) for the date of the reported location extracted from the ERA5 reanalysis from the European Centre for Medium-Range Weather Forecasts.
- wind_sc: Wind speed centered to a mean of zero and scaled to a standard deviation of 1.
- wind_direction: Direction from which wind originates on a 0 - 360 degree scale where 0 is north on the date of the reported location, extracted from the ERA5 reanalysis from the European Centre for Medium-Range Weather Forecasts.
- relative2wind: Smallest difference in degrees (0 - 180) between the direction of the robin movement and the direction of the wind.
Code/software
The data are provided in a CSV file that can be viewed in most spreadsheet or text applications.
Access information
Data were derived from the following sources:
- Wind data were derived from https://climate.copernicus.eu/climate-reanalysis (Creative Commons Attribution 4.0 International (CC BY 4.0)
- AQI data were derived from the AirNow.gov historical air quality maps, which are U.S. government works and are free for use in the public domain.
American robins were tagged with Lotek GPS-satellite tags in the Willamette Valley and Central Cascades of Oregon. Subsequent locations reported either daily (2023) or weekly (2024) were used to determine how behavior changed under variable air quality conditions throughout the study period.