1-3- Taylor et al 2017 Flashcards
What is route efficiency?
straight line distance from start to end divided by the distance travelled by the bird
What are wingbeat frequency and amplitude a proxy for?
energy expenditure or work rate
Dorsal body amplitude can be an indirect measure of what?
amplitude of wing motions
Power requirement of horizontal steady flight is proportional to the square of what?
wingbeat amplitude
Power requirement of horizontal steady flight is proportional to the cube of what?
wingbeat frequency
Increasing wingbeat frequency and decreasing amplitude reduces ______ _____ and optimises _________-
Increasing wingbeat frequency and decreasing amplitude reduces wing drag and optimises aerodynamic efficiency
What happened a month before the experiment?
pigeons did at least 24 solo or flock releases from four release sites 1-3km from home
How many times were the pigeons released?
20 times each
When were the birds released
when the sun was visible and the speed was <7 m s -1
What were the maximum amount of releasea day?
2 a day, 3 hrs in between
WHat was the gap between pigeons being released
10 minute interval. 4 minute if out of site
What happens if birds meet up
flight excluded. this happened 18 times
___ additional tracks were removed due to GPD issues
4
__ more tracks excluded because birds landed
4
how were the trackers fitted on the birds?
velcro strips glued to trimmed feathers on their backs
how much did the loggers and fastenings weigh?
27g
what % of subjects’ mean body mass did the loggers and fastenings weigh?
7%
How did they get the pigeons used to the weight?
placed 27g clay weights to their backs 2 weeks before the experiment
how was wind speed per minute nad running mean of wind uploaded
using professional wireless weather station
GPS and accelerometer synchronised to accuracy of what?
0.2s
Weather data, GPD and accelerometer linked using what?
timestamps from the weather station
how was orthodromic (great-circular) distance calculcated
haversine formula
how was the bird’s final bearing from previous point calculated?
with the forward azimuth
Route efficiency was the ratio between what and what
Route efficiency was the ratio between total straight line (great-circular) distance between release and home divided by the sum of the direct (great circular) distances between each successive GPS point
What does wind support represent?
the length of the wind vector perpendicular to the birds’ direction of travel
Airspeed is what?
the speed derived from the GPD taking into account wind support and crosswind
what is static acceleration
gravity
how was static acceleration removed?
by subtracting a running mean over 15 wingbeat cycles
why was running mean calculated over wingbeat cycles rather than over specific time periods
because variation in wingbeat frequencies would have meant including varying quantities of partial wingbeat cycles in a time-based running mean
Why was dorsal acceleration used?
dorsal acceleration signal was used to detect each wingbeat using the upper reversal point in acceleration
what is amplitude of DB displacement
the amount the body is displaced per wingbeat
how was the amplitude of DB displacement calculated
by the double integration of dorsal accelerometer measurements
How did they make sure they compared a steady flight?
data trimmed 1000m from start to end site
what was the shortest straight-line distance of the entire steady flight for far site?
5.08km
what was the shortest straight-line distance of the entire steady flight for near site?
1.85km
why were wingbeat frequencies of3 Hz and below removed?
to remove effects of gliding, idling or circling
what is tortuosity
the running mean fo the change in birds’ bearing every second of data (5 GPS points), with changes in direction over 3 degrees removed because of circling and to keep active straight line powered flight
Homing pigeons (Columba livia) modulate wingbeat characteristics as a function of what?
route familiarity
What do piecewise linear mixed effect models look at
the effect of repeated releases on route efficiency, median peak-to-peak dorsal body acceleration per wingbeat, median dorsal body amplitude per wingbeat, median wingbeat frequency and median airspeed
what is breakpoint?
an abrupt change in the DV
how were breakpoints estiamted?
using one-dimensional optimisation
what were the fixed effects in model 1?
median peak-to-peak DB acceleration and median wingbeat frequency
what were the fixed effects in model 2?
median dorsal body amplitude per wingbeat (dependent on acceleration & wingbeat freq)(force exerted on body)
Was group significant in either model
no
Why was route efficiency transformed?
because it was negatively skewed
what was route efficiency in release 1?
0.46
what was route efficiency in release 5?
0.82
when was the breakpoint in route efficiency?
between releases 5 and 6
did route efficiney significantly increase prior to breakpoint?
yes
what were the differecnes in route efficiency from release 6 to 20
no signif diff
___ site birds had more efficient routes than ___
near site brids had mroe efficient routes than far
did group have an effect on wingbeat characteristics?
no
when was the breakpoint for median peak-to-peak DB acceleration and median DB amplitude?
between 5 and 6. increased signif prior to
where would you think there is a breakpoint in wingbeat frequency on visual inspection?
about release 6
when was the median wingbeat freq breakpoint
between release 9 and 10. no signif increase prior
what happened when the breakpoint of wingbeat frequency was moved to around 6
it account for less variability but the decreas prior was significant
how big was the confidence interval for the wingbeat freq
it was large
what happened after the objective breakpoint for wingbeat freq
it increased significantly
High wind support was associated with lower ___ ___ and increased _____ ____ _____ but no effect on ____ ____ ____
• High wind support was associated with lower wingbeat frequency and increased dorsal body amplitude but no effect on peak-to-peak dorsal body acceleration
hwo did median crosswind effect wingbeat characteristics
no signif effect
median airspeed ______in the first five releases but _____slightly after
• Median airspeed increased in the first five releases but decreased slightly after
what was the relationship between median airspeed and release number like?
low
• Higher airspeeds were associated with higher ___ ____ ___ ___ and lower ___ ___ (when wingbeat & DB accelerations are fixed effects)
• Higher airspeeds were associated with higher peak-to-peak dorsal body accelerations and lower wingbeat frequencies (when wingbeat & DB accelerations are fixed effects)
• In model with DB amplitude as a fixed effect, DB amplitude was _____associated with airspeed
• In model with DB amplitude as a fixed effect, DB amplitude was positively associated with airspeed
• Route efficiency as a fixed effect between releases 1 and 6. ____ ___ ___ ___, ___ ___ ___ ___ ___ and ___ ___ positively related to route efficiency
• Route efficiency as a fixed effect between releases 1 and 6. Median peak-to-peak DB acceleration, median DB amplitude per wingbeat and median airspeed positively related to route efficiency
• Median peak-to-peak acceleration and release number were only influenced by route efficiency with no significant effect of what?
wind support or crosswind
what was associated with lower route efficiencies in releases 1 to 6
DB amplitude, wingbeat frequencies and airspeed
• Although there was a significant difference between groups in route efficiency in the piecewise LME model, there was no significant difference between groups when relating ____ ____ and ____to route efficiency
• Although there was a significant difference between groups in route efficiency in the piecewise LME model, there was no significant difference between groups when relating wingbeat characteristics and airspeed to route efficiency
• Analysing only data with flap frequencies > 3 Hz and tortuosity < 3 deg resulted in a shift in the breakpoint in wingbeat frequency from release what to what?
9.6 to 6.2
what were the effects of removing circling and glidign
decreasing median wingbeat frequency and median peak-to-peak DB acceleration but increasing median DB amplitude and median airspeed
o Confidence interval in the breakpoint for wingbeat frequency remains approximately ________on visual inspection
o Confidence interval in the breakpoint for wingbeat frequency remains approximately in the same and that on visual inspection
• During the first 6 releases, what increased?
efficiency, DB acceleration, DB amplitude, median airspeed all increased
after stabilisation, what increased?
DB acceletation and DB amplitude decreased whereas median wingbeat frequency increased
did decreasing DB amplitude and increasing wingbeat frequency result in a higher airspeed?
no
was there a significant difference between groups in wingbeat characteristics and speed
no
___ ___ ___ ___ and _______ were particularly low during the first few releases
• Median peak-to-peak DB acceleration and airspeeds were particularly low during the first few releases
higher airspeeds were associated with higher ___ ____ and lower ____ ____
• Established that higher airspeeds were associated with higher peak-to-peak accelerations and lower wingbeat frequencies (peak-to peak DB had larger effect)
what had the overall greatest effect on airspeed
• DB displacement amplitude (dependent on DB acceleration and wingbeat frequency)
High peak-to-peak acceleration and low wingbeat frequency (aka long in time and high in force) leads to _____displacement and _____airspeed
• High peak-to-peak acceleration and low wingbeat frequency (aka long in time and high in force) leads to greater displacement and higher airspeed
what is an alternative explanation for changes in wingbeat characteristics and airspeed
changes in flight behaviour like circling and gliding
why is the alternative explanation of changes in flight behaviour like circling and gliding not that impressive
Removing them didn’t make that big of a difference though apart from moving breakpoints
This suggests that changes in wingbeat characteristics occurs during straight-line powered flight
why might flying at a slower speed may be beneficial
as brids can gain local ambient info
DB acceleration and amplitude and efficiency all increased and had a breakpoint in a similar time frame. what does this indicate?
that DB movements change as a function of navigational knowledge
o Higher wingbeat frequencies were associated with lower ____ _____ in first 6 releases
o Higher wingbeat frequencies were associated with lower route efficiencies in first 6 releases
what do the large CIs and late breakpoint in wingbeat freq imply?
that birds continue to learn routes home even after route efficinency plateaus
how did removing circling and gliding alter wingbeat frequency breakpoint? it went from
it went from 9.5 to 6.2
why could fitness increase not explain the improved efficiency?
brids flew lots beforehand and had clay on their backs
Changes in wingbeat characteristics shortly after route efficiency breakpoint could be related to what?
acquisition of navigational knowledge
after stabilising, what happened?
wingbeat freq increased whereas DB amplitude and acceleration decreased
median wingbeat freq increased by what percent from release 6 to 18
5.8%
median wingbeat freq increased by what to what from release 6 to 18
from 4.9 to 5.2
DB acceleration was not affected by wind. What does this mean
o Suggests that in winds under 7 m s -1, birds compensate for the wind by modulating wingbeat freq
The breakpoint change in wingbeat frequency from ___to ___after controlling for circling could reflect birds utilising circling and gliding to compensate was the wind
• The breakpoint change in wingbeat frequency from9.6 to 6.2 after controlling for circling could reflect brids utilising circling and gliding to compensate was the wind
• Higher route efficiency was associated with higher ___ ____ and lower _____
• Higher route efficiency was associated with higher wind support and lower crosswind
what was the near distance from home
3.85km
what was the far distance from home
7.06km
what is the amplitude of DB displacement?
the amount the body is displaced per wingbeat. this is calculated by the double integration of dorsal accelerometer measures
how were butterworth cut off frequencies chosen
they were determined by visualising the data using fast Fourier transforms. drift data had cutoff of 1Hz whereas integration for displacement had 2.5Hz
what was added as a random effect on the intercept?
individual
what was pooled for most analyses?
the two groups
