Barcelona Route Part 2

Kyleakin Lofts · January 13, 2018

Which route do our pigeons

really take from Barcelona?

Part 2.

In the first part we presented our project using geolocators. How we came to

these devices, why we use them and not GPS loggers and how the first tests

went.

The Barcelona weekend was of course D-day, now it had to happen. Jelle

Jellema had three pigeons basketed with geolocators and had the necessary

confidence in them not to let us down. We all know now that Barcelona 2013

was a very tough race which only the valiant ones came through. The straight

line distance from Barcelona to Nijverdal is 1260 km so under these

circumstances it certainly was no piece of cake. Of the first four arriving

pigeons there were three with geolocators. These are the pigeonâ€™s arrival times

and the provisional National results of the ZLU:

18.58hrs 08-2027103 332nd

National (113th

hens)

19.39hrs 09-1783744 423rd

National (138th

hens)

07.01hrs 08-2027077 847th National (253rd

hens)

From this we can safely conclude that the geolocators do not impede the

pigeons from being able to put up a top performance, a lot different to all the

GPS loggers that we tried earlier. The same weekend Ultsje Jellema sent two

pigeons to Cahors. Ideally both flights would overlap and we could establish the

route taken by the pigeons and we could use the data from Cahors as a

Page

2

reference. Cahors, the cock that Ultsje sent had a number of leg feathers clipped

so that maximum light could fall on the geolocator. Of the two pigeons he sent

one was lost, the other was listed as the 3rd Steggerda pigeon and 18th in the

Frisian Fondclub, 09-1653620 timed at 15.51 won a decent prize from 901

birds. After returning home, the geolocators were read and the data analysed.

The computer programme developed for this makes it a breeze. The data is then

presented as in the picture below:

Figure 1: Two beams with wave patterns.

The top bar is a shrunken view over a longer time period. The area blocked in

red is the part selected (11th and 12th July 2012 in this case), which is then

shown magnified below. The blue pattern shows the light intensity on the day.

The light intensity is shown on the vertical axis, graded on the left starting at 1

and rising to 100,000 lux. Horizontally, the times of the day are shown between

a green and red vertical line. The green and the red line shows that the light

increases (sunrise) and decreases (sunset) from and to around 10 lux (twilight).

The night between them is very clear; the light level is not far above the 1 lux.

From anywhere in the Northern Hemisphere it is known for each day when the

Page

3

sun rises and sets. This is used by fanciers to set the times for neutralization or

hours of darkness.

This data, and the amount of light that the geolocators have saved, makes it

possible to calculate where the pigeon was overnight. A very brief explanation

is under Figure 2.

Figure 2. Showing the earth with the sun to the right.

The bold circle around the earth is the boundary between day and night. To the

left of the line is the night on earth and to the right, where the sun shines, it is

day. The right globe shows the earth about 12 hours later. The sun is still on the

right, but the earth has turned almost 180Â°. Circle B now gives the boundary

between day and night. Now if we want to find the place where the sun rose on

the left and where 12 hours later it rose on the right, then it must be at a point on

both A and B. This position is where the two lines intersect, indicated by a red

asterisk. If we know the times of sunrise and sunset and the date we can

calculate the exact location where these lines intersect and thus the position

around midnight.

Four pigeons, five nights.

In the first part it was indicated that the geolocators have mainly been used in

research on migratory birds such as waders, swallows and terns. These species

move over a greater period of time and therefore there are many observations

available over that long period containing many nights. So we used a device that

Page

4

is not specifically designed for racing pigeons, which is a shortcoming. This did

not allow us to show anything with the data from Ruffec, because it was just a

one day race and the pigeons did not have to stay out overnight. For Barcelona

we knew definitely that the pigeons would be out overnight and also Cahors the

heavy flight conditions made for an overnight stay. This information we could

use to make our calculations, based on the assumption that the pigeons would

not move between sunset and sunrise (although this is not necessarily a fact). In

addition, there must be sufficient light received, otherwise it is only possible to

estimate the latitude as it is a lot harder with longitude. For us this was the most

interesting aspect as we wanted to know whether the pigeons went along the

west coast of France to fly home or along the eastern side, so we needed to take

the risk. Below we describe each pigeon and the results we now have:

F956, trip with two nights.

On Sunday morning at 7 am this pigeon was back home, so for the experiment

this was perfect because she was twice overnight and therefore can be

calculated with more certainty. The luminosity which has been recorded is

shown in Figure 3 but it is not of the intensity shown in the loops in Figure 1.

The moments of liberation and homecoming are marked with a red arrow,

which are clearly distinguishable. It is also striking that the light pattern on day

one is much more stable than the second day. Only in the afternoon of July 6

there appear more lines which could mean that the pigeon on the first day

travelled a shorter distance.

Figure 3: intensity of F956 during the period July 5 / July 7.

Page

5

F957, Second to the loft.

A smooth but hard trip for this hen that was home just after 7.30pm on

Saturday, she was a little over 40 minutes later than her loft mate who came

first. Even with her you can see the wave pattern and a clear peak at

homecoming and at liberation. On July 7, she is already in the loft, which means

the sunset is less well defined; we have manually marked this (solid red line at

6

th July evening instead of dotted line). The measurements in the afternoon

clearly increased between three and seven hours, producing a clearer footprint.

Figure 4: Intensity of F957, morning flight July 5

th to evening July 6

th

.

F958, first to the loft.

This was the first hen to arrive with a geolocator at 18:58 and is the 3

rd pigeon

in NIC Borne. It has a much more stable light pattern than F957, both on the

first and on the second day. Striking on her first flying day is the brightness

reduced at night, just before 2100 pm, she would then be roosting (almost 12

hour flight). On the second day there is a dip in brightness around 1400 hours

(1600 hours our time), perhaps a drink break? We cannot retrieve the intensity

measurements or the temperature.

Page

6

Figure 5: Intensity of F958, the first home on July 6.

The F960 at Cahors, a different picture on the same day.

With this cock of Ultsje we first trimmed the leg feathers so that they would

keep the geolocator free from the shade as much as possible. The fruit of this is

shown by Figure 6. While the three hens at Barcelona show intensity values

between 1,000 and 10,000 lux, this is very different for F960. The light levels

are truncated at 100,000 but during the afternoon of July 5 they were probably

still higher. It is also a much less erratic pattern, although of course we are

dealing with a more constant airspeed and altitude. The liberation at 14:00 (12h

UTC) is clearly visible, as well as the return the next day at 15:15 (13:15

standard time).

Figure 6: Light intensity measured during Cahors F960 on 5

th and 6

th July.

Page

7

For F960 there is more data available because Ultsje Jellema has been in the

habit of weighing the pigeons. Regularly the weights of the pigeons are

recorded both at basketing and returning so the condition gradient becomes

visible. Pigeons being prepared for basketing put on some weight and pigeons

generally lose weight before they arrive at home. The amount of weight loss

also says something about the deviation of the flight line after all the more

distance a pigeon travels the more fat they will use to get home. These

measurements have now been available for several years and have been already

developed in a graph which is shown in figure 7. As the pigeons have to travel

more miles, the percentage weight loss is greater equally if a pigeon is flying a

little it does not need to consume too much energy. In particular, the steepness

of the line can be used to say 1) whether a pigeon is economical in its energy

use or not, and 2) how far a pigeon has been flying if you know what the weight

loss in % has been.

Figure 7: weight loss versus distance,

the result of three years of weight measurements U. Jellema

When F960 was basketed this cock weighed 480 grams and on return a weight

of 400 grams, which means that 17% of its weight at basketing has been

consumed during the flight. Of course this is not an exact science, because

while staying in the basket one pigeon might eat and drink more than another,

causing weight in the digestive tract. With this weight and the early prize (direct

Page

8

flight) we can deduce from the figures that this pigeon must have flown

between 900 and 1100 km, roughly. This seems to be fairly accurate.

Probable overnight place.

Previously this technique required all kinds of complicated calculations to

adjust and determine the location nowadays this happens in the background

within the program. The key question is, where were the pigeons themselves

during their stay? That is, given the low accuracy of this technique it cannot be

said with certainty. As indicated above, the latitude especially is pretty accurate

and clearly 5Â° east was not the latitude for all three pigeons. 5Â° east is important

because almost the entire Rhone Valley is at that latitude. In the table below are

the probable coordinates displayed for the 4 pigeons. The latitude is a lot more

secure than the longitude, which may differ just one or two points, meaning that

the pigeon may have been found 50 km to the north or south. Furthermore the

distance â€œas the crow fliesâ€ is shown in the last column. This shows that the

pigeons fly more mileage than we account for, for all four pigeons deviate from

a straight line, F960 is still the least on the first flying day. That is also the

closest to an average rate of speed of a pigeon with moderate headwind.

Table 1: calculated on eight sites and perpendicular distance from liberation to night

landing place.

Of course, these coordinates can also be plotted on a map, and we have done

that in the final image.

GEOLOCATOR NR. FLIGHT LATITUDE LONGITUDE DISTANCE TO NIGHT LANDING

PLACE

F956 BARCELONA 51 3,7 1090 km (2nd Night)

F957 BARCELONA 47 1,7 633 km in 11 hours (57 km/h)

F958 BARCELONA 49 1,2 835 km in 13 hours (64 km/h)

F960 CAHORS 48 3,7 480 km in 7 hours (70 km/h)

Page

9

Figure 8: Markers of the Barcelona birds night spots (A, B, C), the blue marker shows

the night spot of the Cahors cock. At the top are the two lofts of Jellema Sr. and Jr. and

under the planes the liberation points Barcelona and Cahors

It is striking that the F960 from Cahors has picked the easternmost night spot.

This pigeon is on the same latitude as the F956 and we know that she was at

home on July 7th at 7 pm and may therefore have been sitting just two hours

from the loft. This means that they are already close to the Belgian border.

Where this pigeon has been the first night is to less well known and we

Page10

therefore prefer to omit it until we have better information. The fact is that a lot

of pigeons came home much later than Jelleâ€™s three pigeons, since all three have

earned a neat position on the result. On Monday they were still timing in

Belgium and also in (Northern) Netherlands with less than 25% home. These

pigeons have spent the night further south than the pigeons in this experiment

and it is obvious that they remain "stuck" in the Rhone Valley or that they have

gone around the Pyrenees for their first night. The map above shows that all

pigeons have taken the west coast of France route to go home, something that

no one had expected in advance. Depending on the location of the home loft, it

can be better for the pigeons to take the east side but with the Mistral present it

may not be the easiest route. "Detours" on the west side are perhaps more

inviting under harsh conditions, in addition to which many Dutch pigeons have

used this route for all other flights.

So this first analysis of the data gathered with the geolocators has given us a

small glimpse of the potential route home of our Barcelona pigeons. Our

experiment is not over yet, because we have two pigeons equipped with a

geolocator at Orange and it is intended that F956, 957 and 958 go to Perpignan.

We will tell all about that in the final Part 3.

Jelle and Ultsje Jellema, Wiebren van Stralen

wiebrenvanstralen@duivensites.nl

paddymac · January 13, 2018

Which route do our pigeons
really take from Barcelona?
Part 2.
In the first part we presented our project using geolocators. How we came to
these devices, why we use them and not GPS loggers and how the first tests
went.
The Barcelona weekend was of course D-day, now it had to happen. Jelle
Jellema had three pigeons basketed with geolocators and had the necessary
confidence in them not to let us down. We all know now that Barcelona 2013
was a very tough race which only the valiant ones came through. The straight
line distance from Barcelona to Nijverdal is 1260 km so under these
circumstances it certainly was no piece of cake. Of the first four arriving
pigeons there were three with geolocators. These are the pigeonâ€™s arrival times
and the provisional National results of the ZLU:
18.58hrs 08-2027103 332nd
National (113th
hens)
19.39hrs 09-1783744 423rd
National (138th
hens)
07.01hrs 08-2027077 847th National (253rd
hens)
From this we can safely conclude that the geolocators do not impede the
pigeons from being able to put up a top performance, a lot different to all the
GPS loggers that we tried earlier. The same weekend Ultsje Jellema sent two
pigeons to Cahors. Ideally both flights would overlap and we could establish the
route taken by the pigeons and we could use the data from Cahors as a
Page
2
reference. Cahors, the cock that Ultsje sent had a number of leg feathers clipped
so that maximum light could fall on the geolocator. Of the two pigeons he sent
one was lost, the other was listed as the 3rd Steggerda pigeon and 18th in the
Frisian Fondclub, 09-1653620 timed at 15.51 won a decent prize from 901
birds. After returning home, the geolocators were read and the data analysed.
The computer programme developed for this makes it a breeze. The data is then
presented as in the picture below:
Figure 1: Two beams with wave patterns.
The top bar is a shrunken view over a longer time period. The area blocked in
red is the part selected (11th and 12th July 2012 in this case), which is then
shown magnified below. The blue pattern shows the light intensity on the day.
The light intensity is shown on the vertical axis, graded on the left starting at 1
and rising to 100,000 lux. Horizontally, the times of the day are shown between
a green and red vertical line. The green and the red line shows that the light
increases (sunrise) and decreases (sunset) from and to around 10 lux (twilight).
The night between them is very clear; the light level is not far above the 1 lux.
From anywhere in the Northern Hemisphere it is known for each day when the
Page
3
sun rises and sets. This is used by fanciers to set the times for neutralization or
hours of darkness.
This data, and the amount of light that the geolocators have saved, makes it
possible to calculate where the pigeon was overnight. A very brief explanation
is under Figure 2.
Figure 2. Showing the earth with the sun to the right.
The bold circle around the earth is the boundary between day and night. To the
left of the line is the night on earth and to the right, where the sun shines, it is
day. The right globe shows the earth about 12 hours later. The sun is still on the
right, but the earth has turned almost 180Â°. Circle B now gives the boundary
between day and night. Now if we want to find the place where the sun rose on
the left and where 12 hours later it rose on the right, then it must be at a point on
both A and B. This position is where the two lines intersect, indicated by a red
asterisk. If we know the times of sunrise and sunset and the date we can
calculate the exact location where these lines intersect and thus the position
around midnight.
Four pigeons, five nights.
In the first part it was indicated that the geolocators have mainly been used in
research on migratory birds such as waders, swallows and terns. These species
move over a greater period of time and therefore there are many observations
available over that long period containing many nights. So we used a device that
Page
4
is not specifically designed for racing pigeons, which is a shortcoming. This did
not allow us to show anything with the data from Ruffec, because it was just a
one day race and the pigeons did not have to stay out overnight. For Barcelona
we knew definitely that the pigeons would be out overnight and also Cahors the
heavy flight conditions made for an overnight stay. This information we could
use to make our calculations, based on the assumption that the pigeons would
not move between sunset and sunrise (although this is not necessarily a fact). In
addition, there must be sufficient light received, otherwise it is only possible to
estimate the latitude as it is a lot harder with longitude. For us this was the most
interesting aspect as we wanted to know whether the pigeons went along the
west coast of France to fly home or along the eastern side, so we needed to take
the risk. Below we describe each pigeon and the results we now have:
F956, trip with two nights.
On Sunday morning at 7 am this pigeon was back home, so for the experiment
this was perfect because she was twice overnight and therefore can be
calculated with more certainty. The luminosity which has been recorded is
shown in Figure 3 but it is not of the intensity shown in the loops in Figure 1.
The moments of liberation and homecoming are marked with a red arrow,
which are clearly distinguishable. It is also striking that the light pattern on day
one is much more stable than the second day. Only in the afternoon of July 6
there appear more lines which could mean that the pigeon on the first day
travelled a shorter distance.
Figure 3: intensity of F956 during the period July 5 / July 7.
Page
5
F957, Second to the loft.
A smooth but hard trip for this hen that was home just after 7.30pm on
Saturday, she was a little over 40 minutes later than her loft mate who came
first. Even with her you can see the wave pattern and a clear peak at
homecoming and at liberation. On July 7, she is already in the loft, which means
the sunset is less well defined; we have manually marked this (solid red line at
6
th July evening instead of dotted line). The measurements in the afternoon
clearly increased between three and seven hours, producing a clearer footprint.
Figure 4: Intensity of F957, morning flight July 5
th to evening July 6
th
.
F958, first to the loft.
This was the first hen to arrive with a geolocator at 18:58 and is the 3
rd pigeon
in NIC Borne. It has a much more stable light pattern than F957, both on the
first and on the second day. Striking on her first flying day is the brightness
reduced at night, just before 2100 pm, she would then be roosting (almost 12
hour flight). On the second day there is a dip in brightness around 1400 hours
(1600 hours our time), perhaps a drink break? We cannot retrieve the intensity
measurements or the temperature.
Page
6
Figure 5: Intensity of F958, the first home on July 6.
The F960 at Cahors, a different picture on the same day.
With this cock of Ultsje we first trimmed the leg feathers so that they would
keep the geolocator free from the shade as much as possible. The fruit of this is
shown by Figure 6. While the three hens at Barcelona show intensity values
between 1,000 and 10,000 lux, this is very different for F960. The light levels
are truncated at 100,000 but during the afternoon of July 5 they were probably
still higher. It is also a much less erratic pattern, although of course we are
dealing with a more constant airspeed and altitude. The liberation at 14:00 (12h
UTC) is clearly visible, as well as the return the next day at 15:15 (13:15
standard time).
Figure 6: Light intensity measured during Cahors F960 on 5
th and 6
th July.
Page
7
For F960 there is more data available because Ultsje Jellema has been in the
habit of weighing the pigeons. Regularly the weights of the pigeons are
recorded both at basketing and returning so the condition gradient becomes
visible. Pigeons being prepared for basketing put on some weight and pigeons
generally lose weight before they arrive at home. The amount of weight loss
also says something about the deviation of the flight line after all the more
distance a pigeon travels the more fat they will use to get home. These
measurements have now been available for several years and have been already
developed in a graph which is shown in figure 7. As the pigeons have to travel
more miles, the percentage weight loss is greater equally if a pigeon is flying a
little it does not need to consume too much energy. In particular, the steepness
of the line can be used to say 1) whether a pigeon is economical in its energy
use or not, and 2) how far a pigeon has been flying if you know what the weight
loss in % has been.
Figure 7: weight loss versus distance,
the result of three years of weight measurements U. Jellema
When F960 was basketed this cock weighed 480 grams and on return a weight
of 400 grams, which means that 17% of its weight at basketing has been
consumed during the flight. Of course this is not an exact science, because
while staying in the basket one pigeon might eat and drink more than another,
causing weight in the digestive tract. With this weight and the early prize (direct
Page
8
flight) we can deduce from the figures that this pigeon must have flown
between 900 and 1100 km, roughly. This seems to be fairly accurate.
Probable overnight place.
Previously this technique required all kinds of complicated calculations to
adjust and determine the location nowadays this happens in the background
within the program. The key question is, where were the pigeons themselves
during their stay? That is, given the low accuracy of this technique it cannot be
said with certainty. As indicated above, the latitude especially is pretty accurate
and clearly 5Â° east was not the latitude for all three pigeons. 5Â° east is important
because almost the entire Rhone Valley is at that latitude. In the table below are
the probable coordinates displayed for the 4 pigeons. The latitude is a lot more
secure than the longitude, which may differ just one or two points, meaning that
the pigeon may have been found 50 km to the north or south. Furthermore the
distance â€œas the crow fliesâ€ is shown in the last column. This shows that the
pigeons fly more mileage than we account for, for all four pigeons deviate from
a straight line, F960 is still the least on the first flying day. That is also the
closest to an average rate of speed of a pigeon with moderate headwind.
Table 1: calculated on eight sites and perpendicular distance from liberation to night
landing place.
Of course, these coordinates can also be plotted on a map, and we have done
that in the final image.
GEOLOCATOR NR. FLIGHT LATITUDE LONGITUDE DISTANCE TO NIGHT LANDING
PLACE
F956 BARCELONA 51 3,7 1090 km (2nd Night)
F957 BARCELONA 47 1,7 633 km in 11 hours (57 km/h)
F958 BARCELONA 49 1,2 835 km in 13 hours (64 km/h)
F960 CAHORS 48 3,7 480 km in 7 hours (70 km/h)
Page
9
Figure 8: Markers of the Barcelona birds night spots (A, B, C), the blue marker shows
the night spot of the Cahors cock. At the top are the two lofts of Jellema Sr. and Jr. and
under the planes the liberation points Barcelona and Cahors
It is striking that the F960 from Cahors has picked the easternmost night spot.
This pigeon is on the same latitude as the F956 and we know that she was at
home on July 7th at 7 pm and may therefore have been sitting just two hours
from the loft. This means that they are already close to the Belgian border.
Where this pigeon has been the first night is to less well known and we
Page10
therefore prefer to omit it until we have better information. The fact is that a lot
of pigeons came home much later than Jelleâ€™s three pigeons, since all three have
earned a neat position on the result. On Monday they were still timing in
Belgium and also in (Northern) Netherlands with less than 25% home. These
pigeons have spent the night further south than the pigeons in this experiment
and it is obvious that they remain "stuck" in the Rhone Valley or that they have
gone around the Pyrenees for their first night. The map above shows that all
pigeons have taken the west coast of France route to go home, something that
no one had expected in advance. Depending on the location of the home loft, it
can be better for the pigeons to take the east side but with the Mistral present it
may not be the easiest route. "Detours" on the west side are perhaps more
inviting under harsh conditions, in addition to which many Dutch pigeons have
used this route for all other flights.
So this first analysis of the data gathered with the geolocators has given us a
small glimpse of the potential route home of our Barcelona pigeons. Our
experiment is not over yet, because we have two pigeons equipped with a
geolocator at Orange and it is intended that F956, 957 and 958 go to Perpignan.
We will tell all about that in the final Part 3.
Jelle and Ultsje Jellema, Wiebren van Stralen,
wiebrenvanstralen@duivensites.nl