Limits...
Automated telemetry reveals age specific differences in flight duration and speed are driven by wind conditions in a migratory songbird.

Mitchell GW, Woodworth BK, Taylor PD, Norris DR - Mov Ecol (2015)

Bottom Line: We found that juveniles departed under wind conditions that were less supportive relative to adults and that this resulted in juveniles taking 1.4 times longer to complete the same flight trajectories as adults.We also found that groundspeeds were 1.7 times faster along the coast than over the ocean given more favourable tailwinds along the coast and because birds appeared to be climbing in altitude over the ocean, diverting some energy from horizontal to vertical movement.Our results provide the first evidence that adult songbirds have considerably more efficient migratory flights than juveniles, and that this efficiency is driven by the selection of more supportive tailwind conditions aloft.

View Article: PubMed Central - PubMed

Affiliation: Department of Integrative Biology, University of Guelph, Guelph, ON N1G 2W1 Canada ; Wildlife Research Division, National Wildlife Research Center, Environment Canada, Ottawa, ON K1H 0H3 Canada.

ABSTRACT

Background: Given that winds encountered on migration could theoretically double or half the energy expenditure of aerial migrants, there should be strong selection on behaviour in relation to wind conditions aloft. However, evidence suggests that juvenile songbirds are less choosy about wind conditions at departure relative to adults, potentially increasing energy expenditure during flight. To date, there has yet to be a direct comparison of flight efficiency between free-living adult and juvenile songbirds during migration in relation to wind conditions aloft, likely because of the challenges of following known aged individual songbirds during flight. We used an automated digital telemetry array to compare the flight efficiency of adult and juvenile Savannah sparrows (Passerculus sandwichensis) as they flew nearly 100 km during two successive stages of their fall migration; a departure flight from their breeding grounds out over the ocean and then a migratory flight along a coast. Using a multilevel path modelling framework, we evaluated the effects of age, flight stage, tailwind component, and crosswind component on flight duration and groundspeed.

Results: We found that juveniles departed under wind conditions that were less supportive relative to adults and that this resulted in juveniles taking 1.4 times longer to complete the same flight trajectories as adults. We did not find an effect of age on flight duration or groundspeed after controlling for wind conditions aloft, suggesting that both age groups were flying at similar airspeeds. We also found that groundspeeds were 1.7 times faster along the coast than over the ocean given more favourable tailwinds along the coast and because birds appeared to be climbing in altitude over the ocean, diverting some energy from horizontal to vertical movement.

Conclusions: Our results provide the first evidence that adult songbirds have considerably more efficient migratory flights than juveniles, and that this efficiency is driven by the selection of more supportive tailwind conditions aloft. We suggest that the tendency for juveniles to be less choosy about wind conditions at departure relative to adults could be adaptive if the benefits of having a more flexible departure schedule exceed the time and energy savings realized during flight with more supportive winds.

No MeSH data available.


Related in: MedlinePlus

Relationship between ∆AICc values for models relating tail and crosswind components and their interaction to flight times over the ocean (open circles connected by hatched line) and along the coast (open squares connected by solid line) at different altitudes (m) for Savannah sparrows. Wind data is from the NCEP/NOAA dataset and was accessed through the Environmental-Data Automated Track Annotation Service provided by Movebank (32 x 32 km spatial and 3 h temporal resolution; Dodge et al. [29]). All wind conditions were interpolated over space and time using inverse distance weighting to the estimated spatial and temporal midpoint of a bird’s track over the ocean and along the coast
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4537592&req=5

Fig2: Relationship between ∆AICc values for models relating tail and crosswind components and their interaction to flight times over the ocean (open circles connected by hatched line) and along the coast (open squares connected by solid line) at different altitudes (m) for Savannah sparrows. Wind data is from the NCEP/NOAA dataset and was accessed through the Environmental-Data Automated Track Annotation Service provided by Movebank (32 x 32 km spatial and 3 h temporal resolution; Dodge et al. [29]). All wind conditions were interpolated over space and time using inverse distance weighting to the estimated spatial and temporal midpoint of a bird’s track over the ocean and along the coast

Mentions: All statistical modelling was done in R 3.1.2 [40]. We visually assessed the fit of all models using residual plots. To determine the altitude at which winds were most strongly correlated with flight duration for each flight stage, we modelled flight duration as a function of tailwind component, crosswind component, and their interaction (e.g., [36]) for each of the altitudes described above. For the ocean stage, we included a random effect for ‘nest ID’ to account for potential correlations in flight duration among related individuals (ocean stage: n = 8 parent-offspring pairs; lme4 package). We did not include a random effect for ‘nest ID’ for the coastal stage models because we only tracked three parent offspring pairs. Prior to model fitting, we visually assessed the linearity of the relationship between flight duration and the wind components, and included a 2nd order term in the model when there was evidence for a curvilinear relationship. To determine the most parsimonious model for each altitude for both the coastal and ocean flight stages, we carried out an AICc model selection procedure (e.g., [41, 42]), where we compared AICc statistics for all possible model subsets (MuMIn package). In all cases the best fitting model was at least two ΔAICc units less than the model. After identifying the best model for each altitude, we then compared models among altitudes within each flight stage to determine the altitude for which model fit was best as evidenced by the lowest AICc value (Fig. 2 and Additional file 1: Tables S1 and S2). We then used the wind data from these altitudes (i.e., one altitude per flight stage) to parameterize our multilevel path model.Fig. 2


Automated telemetry reveals age specific differences in flight duration and speed are driven by wind conditions in a migratory songbird.

Mitchell GW, Woodworth BK, Taylor PD, Norris DR - Mov Ecol (2015)

Relationship between ∆AICc values for models relating tail and crosswind components and their interaction to flight times over the ocean (open circles connected by hatched line) and along the coast (open squares connected by solid line) at different altitudes (m) for Savannah sparrows. Wind data is from the NCEP/NOAA dataset and was accessed through the Environmental-Data Automated Track Annotation Service provided by Movebank (32 x 32 km spatial and 3 h temporal resolution; Dodge et al. [29]). All wind conditions were interpolated over space and time using inverse distance weighting to the estimated spatial and temporal midpoint of a bird’s track over the ocean and along the coast
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4537592&req=5

Fig2: Relationship between ∆AICc values for models relating tail and crosswind components and their interaction to flight times over the ocean (open circles connected by hatched line) and along the coast (open squares connected by solid line) at different altitudes (m) for Savannah sparrows. Wind data is from the NCEP/NOAA dataset and was accessed through the Environmental-Data Automated Track Annotation Service provided by Movebank (32 x 32 km spatial and 3 h temporal resolution; Dodge et al. [29]). All wind conditions were interpolated over space and time using inverse distance weighting to the estimated spatial and temporal midpoint of a bird’s track over the ocean and along the coast
Mentions: All statistical modelling was done in R 3.1.2 [40]. We visually assessed the fit of all models using residual plots. To determine the altitude at which winds were most strongly correlated with flight duration for each flight stage, we modelled flight duration as a function of tailwind component, crosswind component, and their interaction (e.g., [36]) for each of the altitudes described above. For the ocean stage, we included a random effect for ‘nest ID’ to account for potential correlations in flight duration among related individuals (ocean stage: n = 8 parent-offspring pairs; lme4 package). We did not include a random effect for ‘nest ID’ for the coastal stage models because we only tracked three parent offspring pairs. Prior to model fitting, we visually assessed the linearity of the relationship between flight duration and the wind components, and included a 2nd order term in the model when there was evidence for a curvilinear relationship. To determine the most parsimonious model for each altitude for both the coastal and ocean flight stages, we carried out an AICc model selection procedure (e.g., [41, 42]), where we compared AICc statistics for all possible model subsets (MuMIn package). In all cases the best fitting model was at least two ΔAICc units less than the model. After identifying the best model for each altitude, we then compared models among altitudes within each flight stage to determine the altitude for which model fit was best as evidenced by the lowest AICc value (Fig. 2 and Additional file 1: Tables S1 and S2). We then used the wind data from these altitudes (i.e., one altitude per flight stage) to parameterize our multilevel path model.Fig. 2

Bottom Line: We found that juveniles departed under wind conditions that were less supportive relative to adults and that this resulted in juveniles taking 1.4 times longer to complete the same flight trajectories as adults.We also found that groundspeeds were 1.7 times faster along the coast than over the ocean given more favourable tailwinds along the coast and because birds appeared to be climbing in altitude over the ocean, diverting some energy from horizontal to vertical movement.Our results provide the first evidence that adult songbirds have considerably more efficient migratory flights than juveniles, and that this efficiency is driven by the selection of more supportive tailwind conditions aloft.

View Article: PubMed Central - PubMed

Affiliation: Department of Integrative Biology, University of Guelph, Guelph, ON N1G 2W1 Canada ; Wildlife Research Division, National Wildlife Research Center, Environment Canada, Ottawa, ON K1H 0H3 Canada.

ABSTRACT

Background: Given that winds encountered on migration could theoretically double or half the energy expenditure of aerial migrants, there should be strong selection on behaviour in relation to wind conditions aloft. However, evidence suggests that juvenile songbirds are less choosy about wind conditions at departure relative to adults, potentially increasing energy expenditure during flight. To date, there has yet to be a direct comparison of flight efficiency between free-living adult and juvenile songbirds during migration in relation to wind conditions aloft, likely because of the challenges of following known aged individual songbirds during flight. We used an automated digital telemetry array to compare the flight efficiency of adult and juvenile Savannah sparrows (Passerculus sandwichensis) as they flew nearly 100 km during two successive stages of their fall migration; a departure flight from their breeding grounds out over the ocean and then a migratory flight along a coast. Using a multilevel path modelling framework, we evaluated the effects of age, flight stage, tailwind component, and crosswind component on flight duration and groundspeed.

Results: We found that juveniles departed under wind conditions that were less supportive relative to adults and that this resulted in juveniles taking 1.4 times longer to complete the same flight trajectories as adults. We did not find an effect of age on flight duration or groundspeed after controlling for wind conditions aloft, suggesting that both age groups were flying at similar airspeeds. We also found that groundspeeds were 1.7 times faster along the coast than over the ocean given more favourable tailwinds along the coast and because birds appeared to be climbing in altitude over the ocean, diverting some energy from horizontal to vertical movement.

Conclusions: Our results provide the first evidence that adult songbirds have considerably more efficient migratory flights than juveniles, and that this efficiency is driven by the selection of more supportive tailwind conditions aloft. We suggest that the tendency for juveniles to be less choosy about wind conditions at departure relative to adults could be adaptive if the benefits of having a more flexible departure schedule exceed the time and energy savings realized during flight with more supportive winds.

No MeSH data available.


Related in: MedlinePlus