Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Do landscape properties influence the migration of Ospreys? : a study of migration patterns using GIS

Peterz, Stina (2007) In Lunds universitets Naturgeografiska institution - Seminarieuppsatser
Dept of Physical Geography and Ecosystem Science
Abstract (Swedish)
Tre fiskgjusar utrustades med sändare innehållande GPS-mottagare inför hösten 2006.
Sändarna gav information om fågelns position varje timme under hela flytten från
Sverige till Afrika. Data från sändarna användes i den här studien tillsammans med
marktäckedata för att ta fram en metod för att undersöka samband mellan landskapet och
olika aspekter av flytten. Fiskgjusen förväntades välja en väg som passerar områden med
stor tillgång på föda och i dessa områden flyga långsammare. Fiskgjusar äter fisk och
därmed är det i sjöar och vattendrag som de hittar sin föda. Två olika metoder användes
för att utvärdera tillgången på vatten i landskapet.
1. Andelen vatten beräknades inom buffertzoner runt flygvägen.
2. Avståndet till närmsta... (More)
Tre fiskgjusar utrustades med sändare innehållande GPS-mottagare inför hösten 2006.
Sändarna gav information om fågelns position varje timme under hela flytten från
Sverige till Afrika. Data från sändarna användes i den här studien tillsammans med
marktäckedata för att ta fram en metod för att undersöka samband mellan landskapet och
olika aspekter av flytten. Fiskgjusen förväntades välja en väg som passerar områden med
stor tillgång på föda och i dessa områden flyga långsammare. Fiskgjusar äter fisk och
därmed är det i sjöar och vattendrag som de hittar sin föda. Två olika metoder användes
för att utvärdera tillgången på vatten i landskapet.
1. Andelen vatten beräknades inom buffertzoner runt flygvägen.
2. Avståndet till närmsta vatten från flygvägen beräknades.
Vägvalet analyserades genom att jämföra den verkliga vägen med antingen simulerade
vägar, eller den kortaste vägen. Vägar simulerades genom att dela upp den verkliga vägen
i olika segment och kasta om dem. Den kortaste vägen skapades genom att dra en rak
linje från start till mål varje dag.
För att undersöka om fiskgjusarna flög snabbare inom områden med mycket vatten
jämfördes flytthastigheten med andelen vatten. Flyttvägen delades upp i segment för
vilka hastighet och andel vatten i buffertzoner runt dem beräknades.
Marktäckedata fanns tillgänglig i 1 km upplösning globalt, 250 och 100 m för Europa och
25 m för Sverige. En jämförelse av olika data visade att metoden för att beräkna avstånd
till vatten var mer beroende av upplösning än metoden för beräkning av andel vatten.
Inga säkra slutsatser kunde dras om fiskgjusarnas beteende under flytten. Det enda
resultat som var signifikant var när påverkan på vägval av avstånd till vatten
analyserades, men eftersom resultatet baserades på 1 km upplösning i data är det
tveksamt om det är tillförlitligt. Inget samband hittades mellan andelen vatten och
flyttväg. Det skulle kunna bero på den grova upplösningen i marktäckedata. Under flera
dagar var andelen vatten 0 % när den beräknades med 1 km upplösning, även i de
områden där det i finare upplösning i data visade sig finnas vatten.
Något samband mellan flyghastighet och andel vatten kunde inte visas. Antagligen
påverkas hastigheten mer av vilken tid på dagen det är. Det verkar fungera bättre att dela
in vägen i längre dagslånga segment istället för att dela upp varje dag. (Less)
Abstract
Three Ospreys were fitted with satellite transmitters including GPS receivers in autumn
2006. These transmitters provide hourly data of the birds’ positions during their
migration from Sweden to Africa. These data, together with land cover data, was used in
this study to develop a method for investigating the influence of landscape properties on
different aspects of bird migration. The Ospreys were assumed to choose a route that
goes through a landscape rich in feeding habitat and in these areas the migration speed
was believed to be slower. Since Ospreys feed on fish their feeding habitat consists of
water. Two different methods were developed to estimate the availability of feeding
habitat in the landscape:
1. The proportion of... (More)
Three Ospreys were fitted with satellite transmitters including GPS receivers in autumn
2006. These transmitters provide hourly data of the birds’ positions during their
migration from Sweden to Africa. These data, together with land cover data, was used in
this study to develop a method for investigating the influence of landscape properties on
different aspects of bird migration. The Ospreys were assumed to choose a route that
goes through a landscape rich in feeding habitat and in these areas the migration speed
was believed to be slower. Since Ospreys feed on fish their feeding habitat consists of
water. Two different methods were developed to estimate the availability of feeding
habitat in the landscape:
1. The proportion of water within buffer areas around the route was calculated.
2. The distance to nearest water body from the route was calculated.
The choice of route was analysed by comparing the real route to either simulated routes
or the shortest route. The comparison was made for each day. Routes were simulated by
rearranging the different segments of the real route. The shortest route was drawn as a
straight line from the first to the last position of the day.
The relation between migration speed and the proportion of water in the landscape was
tested. The route was divided into different segments and the proportion of water and
speed for each segment was calculated.
Land cover data was available in 1 km resolution globally, 250 and 100 m for Europe and
25 m for Sweden. An evaluation of land cover data showed that the calculation of
distance to water was more resolution dependent than calculating the proportion of water.
No definite conclusions about the Osprey’s behaviour could be drawn. The only
significant result was from the comparison between distance to water and choice of route,
but since this was done with 1 km resolution it is not very reliable. When the proportion
of water was calculated to analyse the migration route no relationship was found. This
could be because of the coarse resolution in data. On several of the days the proportion of
water was 0 % when the 1 km resolution in data was used, even in areas where there was
water represented in finer resolution data.
Migration speed did not seem to be dependent on the proportion of water. Only when the
days were not divided into segments there is some indication that there is a relationship.
The time of day probably affects the length of the segments to a greater extent than
proportion of water. (Less)
Please use this url to cite or link to this publication:
author
Peterz, Stina
supervisor
organization
year
type
H1 - Master's Degree (One Year)
subject
keywords
ospreys, bird migration, geography, physical geography, GIS
publication/series
Lunds universitets Naturgeografiska institution - Seminarieuppsatser
report number
140
language
English
id
1938002
date added to LUP
2011-05-02 09:18:04
date last changed
2011-12-14 10:24:00
@misc{1938002,
  abstract     = {{Three Ospreys were fitted with satellite transmitters including GPS receivers in autumn
2006. These transmitters provide hourly data of the birds’ positions during their
migration from Sweden to Africa. These data, together with land cover data, was used in
this study to develop a method for investigating the influence of landscape properties on
different aspects of bird migration. The Ospreys were assumed to choose a route that
goes through a landscape rich in feeding habitat and in these areas the migration speed
was believed to be slower. Since Ospreys feed on fish their feeding habitat consists of
water. Two different methods were developed to estimate the availability of feeding
habitat in the landscape:
1. The proportion of water within buffer areas around the route was calculated.
2. The distance to nearest water body from the route was calculated.
The choice of route was analysed by comparing the real route to either simulated routes
or the shortest route. The comparison was made for each day. Routes were simulated by
rearranging the different segments of the real route. The shortest route was drawn as a
straight line from the first to the last position of the day.
The relation between migration speed and the proportion of water in the landscape was
tested. The route was divided into different segments and the proportion of water and
speed for each segment was calculated.
Land cover data was available in 1 km resolution globally, 250 and 100 m for Europe and
25 m for Sweden. An evaluation of land cover data showed that the calculation of
distance to water was more resolution dependent than calculating the proportion of water.
No definite conclusions about the Osprey’s behaviour could be drawn. The only
significant result was from the comparison between distance to water and choice of route,
but since this was done with 1 km resolution it is not very reliable. When the proportion
of water was calculated to analyse the migration route no relationship was found. This
could be because of the coarse resolution in data. On several of the days the proportion of
water was 0 % when the 1 km resolution in data was used, even in areas where there was
water represented in finer resolution data.
Migration speed did not seem to be dependent on the proportion of water. Only when the
days were not divided into segments there is some indication that there is a relationship.
The time of day probably affects the length of the segments to a greater extent than
proportion of water.}},
  author       = {{Peterz, Stina}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Lunds universitets Naturgeografiska institution - Seminarieuppsatser}},
  title        = {{Do landscape properties influence the migration of Ospreys? : a study of migration patterns using GIS}},
  year         = {{2007}},
}