Data and code from: Spring migration strategies differ among three waterfowl species that winter in southern New England, USA
Data files
May 08, 2026 version files 22.87 KB
-
AnnualVariationData.RDS
3.61 KB
-
InterspecificInterindividualVariationData.RDS
2.35 KB
-
README.md
7 KB
-
SpringMigStrategyCode.R
9.91 KB
Abstract
Data files (InterspecificInterindividualVariationData.RDS and AnnualVariationData.RDS) and code (SpringMigStrategyCode.R) associated with Journal of Avian Biology manuscript "Spring migration strategies differ among three waterfowl species that winter in southern New England, USA". These methods were used to investigate interspecific and intraspecific variation in the spring migration strategy of 3 sympatric-wintering waterfowl. Specifically, we used GPS telemetry data to quantify interspecific, interindividual, and annual variation in spring migration strategies of American black ducks (Anas rubripes [n = 20]), Atlantic brant (Branta bernicla hrota [n = 20]), and greater scaup (Aythya marila [n = 9]) that share a common wintering area in southern New England, USA, but differed in distance traveled between wintering and breeding areas, breeding latitude, breeding strategy, and breeding range size. We characterized spring migration strategies from GPS telemetry data by quantifying ordinal spring migration initiation date, ordinal spring migration completion date, migration duration, number of stopovers, average stopover duration, proportion of migration time spent in stopover, and a stopover-to-travel ratio. In the script "SpringMigStrategyCode.R", we use one-way ANOVAs and TukeyHSD tests to quantify interspecific variation and coefficients of variation (CV) to quantify interindividual variation in each spring migration metric using the "InterspecificInterindividualVariationData.RDS" dataset. We then use normalized differences to quantify annual variation in each spring migration metric for American black ducks and Atlantic brant using the "AnnualVariationData.RDS" dataset. American black ducks demonstrated the most time-minimizing spring migration strategy, whereas Atlantic brant Branta and greater scaup demonstrated more energy-minimizing spring migration strategies. Atlantic brant and greater scaup also had less intraspecific variation in metrics of spring migration strategy than American black ducks, particularly those associated with stopover behavior, highlighting the importance of key stopover sites for long-distance migrating species. Our findings demonstrate that different species of sympatric-wintering waterfowl adopt distinct spring migration strategies along the continuum from time- to energy-minimization and have differing extents of intraspecific variation in migration strategy, which together have important conservation implications. Our analyses might serve as a template for considering similar questions in comparable study systems.
Dataset DOI: 10.5061/dryad.v9s4mw79t
Description of the data and file structure
We used GPS telemetry to quantify interspecific, interindividual, and annual variation in spring migration strategies of 3 species of waterfowl (American black ducks [Anas rubripes; n = 20], Atlantic brant [Branta bernicla hrota; n = 20], and greater scaup [Aythya marila; n = 9]) that share a common wintering area in southern New England, USA, but differed in distance traveled between wintering and breeding areas, breeding latitude, breeding strategy, and breeding range size. We deployed 25‐g GPS‐GSM transmitters on adult American black ducks and Atlantic brant (OrniTrack‐25 4G and OrniTrack‐BR25 4G, respectively; Ornitela, UAB, Vilnius, Lithuania) using backpack‐style harnesses and 30-g implantable GPS-GSM transmitters with percutaneous antennas on greater scaup (OrniTrack‐I30 4G; Ornitela, UAB, Vilnius, Lithuania).
To allow sufficient time for individuals to recover from capture and handling and mitigate associated behavioral impacts of tagging, we censored the first 4 days following deployment of external transmitters (Cox and Afton 1998, Mezebish Quinn et al. 2024) and 5 days following deployment of implant transmitters (Lamb et al. 2020). Location collection interval varied by transmitter model, geographic location across the annual cycle, and transmitter battery percentage, and ranged from 30 min – 48 hr, but was ≤ 24 hr for 99.91% of collection intervals across study species. GPS collection interval was regularized differently for different aspects of the analysis, as described in the publication associated with these datasets. GSM connection interval also varied by transmitter model and geographic location across the annual cycle and was programmed to promote transmitter battery longevity. GSM connection interval ranged from every 1 day during the non-breeding season to every 24 days during the breeding season, when the majority of individuals were located beyond the range of GSM network coverage. External transmitters were solar powered and could last multiple years, while implant transmitter batteries lasted < 1 year.
Files and variables
File: InterspecificInterindividualVariationData.RDS
Description: Data for quantifying interspecific and interindividual variation in spring migration strategy.
Variables (columns) in dataset:
- individual: individual identifier (transmitter ID number)
- species: abdu = American Black Duck, atbr = Atlantic Brant, grsc = Greater Scaup
- sex: F = female, M = male
- age: ASY = after second year (adult), SY = second year (juvenile)
- mig.dist.km: sum of distances for all 24 hr movements from initiation to completion of spring migration to a suspected breeding site (km)
- breed.lat: latitude of individuals identified settled breeding location (decimal degrees)
- initiation.odate: ordinal spring migration initiation date
- completion.odate: ordinal spring migration completion date
- mig.duration.days: spring migration duration; difference in days between ordinal spring migration completion and initiation dates
- numb.stopover: the total number of unique stopover events identified
- avg.stopdur: average duration (days) of unique stopover events
- prop.stop: proportion of migration time (days) spent in stopover; total stopover duration divided by migration duration
- stop.travel.rat: stopover to travel ratio (days/km); total number of days stopped over divided total migration distance
File: AnnualVariationData.RDS
Description: Data for quantifying annual variation in spring migration strategy.
Variables (columns) in dataset:
- individual: individual identifier (transmitter ID number)
- sex: F = female, M = male
- age: ASY = after second year (adult), SY = second year (juvenile)
- species: abdu = American Black Duck, atbr = Atlantic Brant, grsc = Greater Scaup
- mig.year: one = individual's first spring migration, two = individual's second spring migration
- initiation.odate: ordinal spring migration initiation date
- completion.odate: ordinal spring migration completion date
- mig.duration.days: spring migration duration; difference in days between ordinal spring migration completion and initiation dates
- numb.stopover: the total number of unique stopover events identified
- avg.stopdur: average duration (days) of unique stopover events
- prop.stop: proportion of migration time (days) spent in stopover; total stopover duration divided by migration duration
- stop.travel.rat: stopover to travel ratio (days/km); total number of days stopped over divided total migration distance
File: SpringMigStrategyCode.R
Description: Code used to implement analyses described in the associated Journal of Avian Biology publication. The analysis involved the use of GPS telemetry data to quantify interspecific, interindividual, and annual variation in spring migration strategies of 3 species of waterfowl (American black ducks [Anas rubripes; n = 20], Atlantic brant [Branta bernicla hrota; n = 20], and greater scaup [Aythya marila; n = 9]) that share a common wintering area in southern New England, USA, but differed in distance traveled between wintering and breeding areas, breeding latitude, breeding strategy, and breeding range size. Lines 13-46 can be used to read in the data for quantifying interspecific and interindividual variation in spring migration metrics and describe variables in the "InterspecificInterindividualVariationData.RDS" data file. Lines 32-46 can be used to read in the data for quantifying annual variation in spring migration metrics and describe variables in the "AnnualVariationData.RDS" data file. Lines 51-154 can be used to quantify interspecific variation via ANOVA and TukeyHSD tests, as well as interindividual variation in spring migration metrics via the coefficient of variation (CV). Lines 157-205 can be used to quantify annual variation in spring migration metrics as a normalized difference. Spring migration metrics evaluated in all analyses include: ordinal spring migration initiation date, ordinal spring migration completion date, spring migration duration (days), number of stopovers, average stopover duration (days), proportion of migration in stopover, and stopover to travel ratio (days/km).
Code/software
All data manipulation and analyses in R version 4.2.3.
Access information
GPS telemetry data relevant to this study are available on Movebank (Movebank ID: 1442384859) and will be shared by the corresponding author upon reasonable request.
The data and code provided here are also publicly available on GitHub: