To fly or not to fly; Atmospheric influences on the flight behaviour of Griffon Vulture in France

(1)

To fly or not to fly;

Atmospheric influences

on the flight behaviour of

Griffon Vulture in France

Bachelor Thesis

Dorien van Kranenburg

Faculty of Science

Institute for Biodiversity and Ecosystem Dynamics

Supervisors:

dhr. prof. dr. ir. Willem Bouten

mw. dr. Judy Z. Shamoun-Baranes

(2)

(3)

Abstract

Vultures are one of worlds most successful scavengers, yet the populations all over the world are declining and they have disappeared in many areas. Understanding the movements these animals make due to landscape features and meteorological conditions are essential for predicting their survival in a rapid changing environment. In this thesis a study has been performed on the influences of uplift conditions on the daily movement characteristics of the reintroduced Griffon Vultures in the Grand Causses, France. This was done to gain insight in the relation between updraft conditions, convective velocity and orographic lift, and flight characteristics such as total travelled distance per day, by using the temporal and spatial variation of the different variables. The movement ecology paradigm has been used as a theoretical basis for this research. The focus is on when and where to move under the influence of atmospheric external factors. The uplift conditions, convective velocity and orographic uplift are related to flight characteristics of Griffon Vultures derived from GPS tracking data of the UvA-BiTS system. The data of 6 birds in the months January, February, July and August 2011 is used. For every flight characteristic data is selected and used for statistical generalized linear models. The conclusion drawn from the results is that the total flight duration and total traveled distance per day are influenced by convective velocity. Convective velocity explains around 40% of the variation in these characteristics. The time of first take off on the day and the time of last landing on the day are not influenced by convective velocity and orographic lift. Orographic lift almost never results in a significant influence, this is probably due to the fact that either the orographic lift data is not detailed enough or the GPS-location.

(4)

(5)

1. Introduction

Soaring birds use energy of thermals and orographic uplift during their flight (Pennycuick, 2003). For birds with a relatively high body mass, soaring and gliding is the most energy efficient way of flying. This is because energy costs of flapping increase steeply with body mass, the heavier the bird the more energy it cost to flap (Spaar, 1997). Thus, soaring and gliding birds need a upward air movement caused by e.g. radiation or wind creating thermals and orographic lift (Pennycuick, 1998). Dry areas support the availability of thermals (Duerr et al., 2014) whereas relief supports orographic lift Bohrer et al. (2012). Therefore soaring birds choose flight paths along landscape features that support strong uplift conditions. On the one hand the uplift conditions are determ-ined by these landscape features (Vansteelant et al., 2014) and on the other hand by meteorological conditions. Understanding how these animals travel due to landscape features and meteorological conditions are essential for predicting their survival in a rapidly changing environment and to be able to facilitate habitat management a more effective way (Mandel et al., 2011).

Due to the development of small GPS data loggers it is possible to track birds on multiple scales. This innovation created a new way of studying bird movements (Bouten et al., 2013). Most of the research done with GPS devices focuses on the migration patterns of birds. Vansteelant et al. (2014) found that tail winds explains about 40% of migration performance of the soaring Honey Buzzard. Bohrer et al. (2012) has researched the different preferences of the Turkey Vulture and the Golden Eagle and concluded that the Turkey Vulture preferred the updraft conditions caused by radiation, called convective velocity, and the Golden Eagle on the contrary preferred the updraft conditions caused by wind, named orographic lift. However, daily patterns of foraging flights of colonies are scarcely investigated. Spiegel et al. (2013) has investigated daily flight patterns of Griffon Vultures and the influence of the availability of food in Isra¨el

In this thesis a case study has been performed on the daily movements of Griffon Vultures (Gyps Fulvus) similar to Spiegel et al. (2013) but in the Grand Causses in France. Vultures are one of worlds most successful scavengers, yet the populations all over the world are declining and they have disappeared in many areas. Therefore, various conservation programs have been started (Saran et al., 2014). The colony of Griffon Vultures in the Grand Causses in France has been reintroduced successfully in 1981 (Terrasse et al., 2005). The Griffon Vultures provide several cultural, economic and ecological services in the area (Saran, 2014). They clean up the environment by eating carcasses and save a lot of CO2 emissions when farmers do not have to bring dead animals to processing plants miles away an important ecosystem service when looking at climate change (Morales-Reyes et al., 2015).

Knowing how these birds use different meteorological conditions contributes to predictions of when they fly and how long, on which height and what routes they fly. This is of importance to control the reintroduction in this area (Terrasse et al., 2005) and apply effective habitat manage-ment (Mandel et al., 2011).

First of all, the theory about organismal movements, movement paths of birds and information about the study area will be discussed in the theoretical framework. In order to research the daily flight movements of the Griffon Vulture, tracking data is used and a selection is made for different movements (flight characteristics). Thereafter, the results are presented, discussed and in the end a conclusion is drawn. This all is done for the following research question.

(8)

CHAPTER 1. INTRODUCTION

1.1. Research question

What updraft conditions influence flight characteristics of Griffon Vultures?

1.1.1. sub-research questions

This research question is divided into 4 sub-research questions concerning the different flight characteristics investigated.

1. How do convective velocity and orographic lift influence the time of first take off at a day? 2. How do convective velocity and orographic lift influence the time of last landing of a day? 3. How do convective velocity and orographic influence lift the total time travelled per day? 4. How do convective velocity and orographic influence lift the total traveled distance per day?

1.2. Aim of research

The goal of this research is to gain insight in the relation between updraft conditions, convective velocity and orographic lift, and flight characteristics such as total travelled distance per day, by using the temporal and spatial variation of these different variables. This insight can be used for more detailed research on flight characteristics and for effective habitat management.

1.3. Hypothesis

As the main hypothesis of this research it is stated that: ’Flight characteristics of Griffon Vultures are equally influenced by thermal lift and orographic lift’.

Thermal lift results from interaction of surface radiation and the soil water content. When water is available for evapotranspiration there will be a latent heat flux and a low sensible heat flux. The sensible heat flux is generating convective velocity and thus, thermal lift (Groot, 2015). Orographic lift results from wind that blows air against obstacles and has to move upwards creating lift (Wittebrood, 2015). Griffon Vultures are expected to use both in their daily flight patterns because the area in the Grand Causses has relief to generate orographic lift. The area also has sparsely vegetated areas with a low soil water content allowing the generation convective velocity. The birds roost mostly in the canyons and it is expected that the first take of at a day is dependent on the orographic lift as well as the convective velocity because of the relief and the low water availability of the cliffs they roost. The same accounts for last landing of the day. The total time and distance travelled per day is also equally influenced by orographic lift and thermal lift because their home range contains as much relief as flatter areas on which only convective velocity can provide lift

(9)

2. Theoretical Framework

2.1. Movement Ecology Paradigm

The basis theory of this study can be found in the movement ecology paradigm described by Nathan et al. (2008). This is a conceptual framework focusing on the interplay among four basic mechanistic components of organismal movement. Figure2.1shows a visualization of these components and the external factors: the internal state (why move?), motion (how to move?), and navigation (when and where to move?) capacities of the individual and the external factors (biotic and abiotic) affecting movement. The internal state accounts for the physiological and the psychological state of the focal individual. The drive of the organism to fulfill one or more goals such as foraging. The motion capacity of an individual accounts for its ability to move in various ways or modes. Examples of bird’s motion capacities are flapping and soaring. The navigation capacity accounts for the ability to orient in space and/or time, selecting where and/or when to move. In the case of flight characteristics this can be the time of first take of in the day (Nathan et al., 2008).

In this research there is focused on the navigation capacities of the individual and the external meteorological factors affecting movement. However, that does not mean that the other factors are not influencing the movement path of the Griffon Vulture. It is important for the interpretation of the results which otherwise cannot be explained; Vultures have to forage and the locations where food is found will not always be on an optimal path for soaring and gliding.

(10)

CHAPTER 2. THEORETICAL FRAMEWORK

2.2. Movement Path

In this field of interest some research is done for the navigation capacities of birds and external factors influencing their movement path. (Shepard et al., 2013) shows how physical properties of the environment affect the movement path. Birds as well as other animals use the path with the least cost of transport (COTmin). The COTmin can be achieved when moving with the fluid direction. Fluid is in this case the flow of the air in a certain direction caused either by radiation or wind (Shepard et al., 2013). Therefore wind direction and wind speed are influencing flight speed (Shamoun-Baranes et al., 2003) and thus the movement path of birds.

Figure 2.2.: Cross-country flight typic-ally involves (i) a soaring phase where the bird gains altitude in a thermal and (ii) a gliding phase

(Shep-ard and Lambertucci,

2013) Another factor for birds is the vertical component

of the air flow. This vertical component originate due to orographic uplift or thermal convection (hereafter thermals). Shepard and Lambertucci (2013) have stud-ied the use of this energy by soaring birds. It is assumed that a bird is able to gain sufficient altitude to glide to its destination, either by encountering frequent thermals and orographic uplifts or gaining enough altitude with one, without flapping. Figure 2.2 shows how birds use thermals to gain height and use gliding to cover dis-tances. Hence, the presence of thermals and orographic uplift is necessary for soaring birds and thus also for the Griffon Vulture to soar and glide to its destination. The height at which birds leave thermals has also to do with the destination the bird has. Yet, thermals and oro-graphic uplift are depending on radiation, wind speed and relief. Relief causes orographic uplift by preventing the horizontal air movement. The air has to go up result-ing in orographic uplift therefore low wind speeds cause fragile orographic lift. On the contrary thermals will be

stronger due to low wind speeds because they are not disrupted by these wind speeds (Shepard and Lambertucci, 2013) (Bohrer et al., 2012).

In this research there is focused on the daily flight movements of Griffon Vultures influenced by the continuously changing weather conditions. Duriez et al. (2014) found that the Eurasian Griffon Vultures are still flying during cloudy conditions although soaring is likely easier on sunny days due to the occurrence of stronger thermals. How this works exactly is still unclear. The differences in flight duration are not studied by (Duriez et al., 2014) as well as other flight characteristics such as average altitude above ground level. This thesis will be an explorative research on getting more into the relation of uplift conditions and the flight characteristics.

(11)

3. Methods

3.1. Data

The data used for this research is provided by the University of Amsterdams Bird Tracking System (Uva-BiTS) and by Groot (2015) and Wittebrood (2015) since this bachelor project is in coopera-tion with their bachelor theses. Uva-BiTS provided tracking data of the Griffon Vultures in France, Groot (2015) provided data about convective velocity in the research area and Wittebrood (2015) provided data about the orographic lift in the research area. Both Groot (2015) and Wittebrood (2015) made a model to predict the lift conditions by using data from the European Centre for Medium-Range Weather Forecasts (ECMWF model). The mathematics describing the airflow in the canyons are very complex, to keep this as simple as possible it is decided to perform the model twice: one model with flow separation and one without flow separation. Flow separation means that the air movement within the canyons does not interact with the air movement outside the canyon. Therefore the flight characteristics are calculated with these 2 data sets. The data from the ECMWF model has a 3 hour resolution. This data is interpolated in order to provide orographic lift and convective velocity per hour on a 100m gridcell size for the whole area. The area boundaries are from 43.888935 to 44.560560 for the latitude and 2.875564 to 3.881436 for the longitude. This results in a maps of 800 by 750 cells. The data is provided for January, February, July and August 2011. An example of the data on the first of January 12 ´O Clock can be found in figure3.1.

It is chosen to perform this research for the winter and summer of 2011 to see the difference in flight characteristics between the most extreme weather conditions. In winter it is expected that birds are more dependent on orographic lift and in the summer it is expected to be the other way around because of the different weather conditions. The first part of the results gives an overview of the data differences in winter and summer. Therefore, the data is processed for summer and winter separately as well as combined to see if there is an overall trend. This is done with the software Matlab 2015a.

(12)

CHAPTER 3. METHODS

3.2. Grand Causses

The study area is part of the Grand Causses. This is Griffon Vulture is reintroduced in this area. The Grand Causses consist of a limestone plateau with cliffs and deep valleys. The plateau is situated at an height ranging from 800 to 1100 meters above sea level (Groc et al., 2007). The valleys or canyons are hollowed out of the limestone and can be 500 meters deep. The north facing slopes in the area are covered with forests whereas the other sides which face the sun are sometimes terraced for cultivation of the land (naturel rgional des Grands Causses’, 2015). The area experiences harsh winters and is subject to temperatures around freezing point and summers with almost no precipitation and strong differences in temperature (Groc et al., 2007). Sheep and goat breeding is a main activity in the area and therefore the sparsely vegetated area is maintained. In order to support the reintroduction of the Griffon Vulture these farmers are allowed to dispose sheep carcasses on special feeding plots (naturel rgional des Grands Causses’, 2015), about 2 hours before sunset (Fluhr and Durier, 2015).

3.3. GPS-tracking

Since June 2010 full-grown Griffon Vultures are tracked with GPS-loggers in the Grand Causses (UvA-BiTS, 2015). Six of the devices have been used in this research: number 140, 141, 210, 212, 225 and 226. These birds were not always present in the area during the time intervals used for this study; bird 140 and 210 lack some data in January and February. The resolution of the data is 30 minutes up to 1 hour in the winter and 5 minutes up to 30 minutes in the summer (UvA-BiTS, 2015). The four months and six birds combined gave over 170.000 GPS locations with accompanying instantaneous speed, altitude, exact time and more information about the time the device needed for calculating distances (fix time) and the number of satellites used for the measurement. Erroneous data had to be removed from the data set. This was done by a check on the fix time and the instantaneous speed. For both criteria a histogram was made and after the highest peek for fix time and after the second peek for instantaneous speed, a threshold point was chosen (AppendixA). For the fix time this was 60 seconds and instantaneous speed above 120 km/h is also removed. Thereafter the data set contained a bit over 140.000 measurements and was appropriate to be used for extracting the flight characteristics.

3.4. Defining time of first take off at a day

The data points that are selected for the first take off of the day had to meet a couple of criteria. The first criterion stated that the speed a bird had on that moment needs to be higher than the threshold speed for a flying bird. To determine this threshold point, a histogram of the instantaneous speed was calculated (Appendix B). The threshold speed has been chosen at 7 km/h; a point between the two peeks from which the first represents movement on the ground and the second one speeds which only can be reached when flying. The second criterion is that the bird was not already flying and the third criterion says that it is the first measurement on the day that has a flight speed above the threshold speed. The points are selected with a loop that can be described as following:

(13)

CHAPTER 3. METHODS

for e a c h d e v i c e

f i n d t h e d a t a o f t h i s d e v i c e o n l y for a l l u n i q u e d a y s

Find t h e d a t a o f t h e c u r r e n t day f o r t h e c u r r e n t d e v i c e and d e t e r m i n e t h e number o f measurements

I n d i c a t e t h a t t h e b i r d i s n o t f l y i n g

if t h e number o f measurements i s e q u a l o r more than 2 , a t i m e can be

s e l e c t e d

for a l l o f t h e s e l e c t e d measurements }

if t h e s p e e d i s h i g h e r than t h e t h r e s h o l d s p e e d and t h e b i r d i s n o t f l y i n g

S t o r e t i m e o f t a k e o f f , GPS l o c a t i o n and t h e u n i q u e day number end

end end end end

The table which is created during the process described above is used to select the correct convective velocity and orographic lift. The GPS location is connected to the value of a 100 meter by 100 meter grid cell in which the GPS location is located. The time on which the convective velocity and orographic lift are selected is not the same as the take off time but a reference point at 12 O’clock on the same day. This reference point is chosen because the convective velocity is increasing during morning hours, therefore selecting the convective velocity on the time of take off would give a positive relation with time of first take off. This process is done with a for loop for every measurement of first take off. At the end of this model a table is produced with the time of first take off, the convective velocity and the orographic lift (AppendixC).

3.5. Defining time of last landing of the day

This flight characteristic is dependent on the same kind of criteria mentioned in ’Defining time of first take off at a day’. However, the criteria are the other way around: the speed has to be lower than the threshold speed, the bird has to be flying already and it has to be the first moment after the last recorded moment of flying on that particular day. The threshold speed is the same as for the time of first take off. The loop used for this characteristic is adjusted from the one described for ’time of first take off on the day’ is. The unique days and measurements have been run from the last to the first value. The if statement is also changed:

if t h e s p e e d i s h i g h e r than t h e t h r e s h o l d s p e e d , t h e b i r d i s n o t f l y i n g and i t i s n o t t h e f i r s t measurement

S t o r e t h e t i m e o f l a n d i n g , GPS l o c a t i o n and t h e u n i q u e day from t h e measurement c h e c k e d b e f o r e

end

After this process the convective velocity and the orographic lift are selected in the same way as for the ’Time of first take off at a day’ (AppendixC).

3.6. Defining total flight duration per day

The total flight duration per day is calculated in three parts. The first part selects the data points from the data set when the bird is flying. This is done for every device and every measurement belonging to that device with this if statement:

(14)

CHAPTER 3. METHODS

if t h e s p e e d i s h i g h e r than t h e t h r e s h o l d s p e e d and t h e b i r d i s n o t f l y i n g S t o r e t h e GPS l o c a t i o n , time , s p e e d and u n i q u e day number

I n d i c a t e t h a t t h e b i r d i s f l y i n g

elseif t h e s p e e d i s h i g h e r than t h e t h r e s h o l d s p e e d and t h e b i r d i s f l y i n g S t o r e t h e GPS l o c a t i o n , time , s p e e d and u n i q u e day number

elseif t h e s p e e d i s l o w e r than t h e t h r e s h o l d s p e e d and t h e b i r d i s f l y i n g I n d i c a t e t h a t t h e b i r d i s n o t f l y i n g anymore

end

At this point all the data is collected when the bird is flying. Then, for the second part, the exactly same thing is done as with the flight characteristics mentioned before. The measurements are connected with orographic lift and convective velocity. However, for this characteristic the convective velocity and orographic lift are selected on about same time the measurement was taken; the hour most close by the measurement time is used.

The last part calculates the time travelled per day. This is done through calculating the time interval with the last and the next measurement. This is done with a loop that can be described as following:

for e a c h d e v i c e

f i n d t h e d a t a o f t h i s d e v i c e o n l y for a l l u n i q u e d a y s

Find t h e d a t a o f t h e c u r r e n t day f o r t h e c u r r e n t d e v i c e and d e t e r m i n e t h e number o f measurements

for a l l o f t h e s e l e c t e d measurements if i t i s t h e n o t t h e f i r s t v a l u e

if t h e f l i g h t measurement b e f o r e i s t h e same a s t h e measurement b e f o r e i n t h e o r i g i n a l d a t a s e t

C a l c u l a t e t h e i n t e r v a l ( 1 ) between t h e s e two measurements and d i v i d e by 2 else The i n t e r v a l ( 1 ) i s z e r o end else The i n t e r v a l ( 1 ) i s z e r o end if I t i s t h e n o t t h e l a s t v a l u e

if t h e f l i g h t measurement a f t e r i s t h e same a s t h e measurement a f t e r i n t h e o r i g i n a l d a t a s e t

C a l c u l a t e t h e i n t e r v a l ( 2 ) between t h e s e two measurements and d i v i d e by 2 else The i n t e r v a l ( 2 ) i s z e r o end else The i n t e r v a l ( 2 ) i s z e r o end C a l c u l a t e t h e t o t a l i n t e r v a l by summing up i n t e r v a l 1 and i n t e r v a l 2 R e c a l c u l a t e t h e n u m e r i c v a l u e t o a hour v a l u e e g . 1 hour and 30 m i n u t e s

i s 1 . 5 h

S t o r e t h e t o t a l i n t e r v a l end

C a l c u l a t e and s t o r e t h e sum o f a l l t h e i n t e r v a l s on t h e c u r r e n t day

C a l c u l a t e t h e mean o f a l l c o n v e c t i v e v e l o c i t y and o r o g r a p h i c l i f t v a l u e s o f f l i g h t moments on t h e c u r r e n t day

end end

Thus, this model creates a table just like the other characteristics with the total flight duration in hours, the mean convective velocity and the mean orographic lift (AppendixD)

(15)

CHAPTER 3. METHODS

3.7. Defining total traveled distance per day

The total travelled distance per day is calculated with the same model as the total flight duration per day. The only difference is that it is necessary to store the flight speed of the measurements. Thereafter it is possible to calculate the travelled distance by summing up the flight speed times the interval for the current day. This is stored as well as the total time and the mean convective velocity and mean orographic lift (AppendixE).

3.8. Performing a generalized linear model

The tables made for every flight characteristics are used as input for a generalized linear model (GLM). For performing a GLM the function fitglm in Matlab is used. The distribution of the residuals is checked in a histogram for a normal distribution and the variable with a coefficients with a probability value higher than 0.05 is removed after which the GLM is performed again. Thereafter the data is visualized in a 3D or 2D scatter plot with the relation found by the GLM (AppendicesA,B,C, D,E).

(16)

(17)

4. Results

The various results of the research are visualized and assessed in this part. First an impression on the data will be presented, thereafter the results will be presented per flight characteristic.

4.1. Data impression

Here an impression of flight behavior of the griffon vultures will be given for the winter and the summer. 4 different types of days will be pointed out: a day with the high maximum convective velocity values and a day with the low maximum convective velocity, a day with high maximum wind speeds and a day with low maximum wind speeds.

(18)

CHAPTER 4. RESULTS

Figure 4.1.: Tracks for the winter, 6th of January, 12 O’Clock, lowest maximum values of convective velocity (0.75 m/s) and for the summer, 16th of July, 12 O’Clock, lowest maximum values of convective velocity (1.25 m/s)

(19)

CHAPTER 4. RESULTS

Figures 4.1 and 4.2show that the Griffon Vulture seems to travel further distances when the higher convective velocity is higher. This is the case when only looking at the winter and when only looking to the summer. The summer has higher values for the convective velocity and therefore the Griffon Vulture travels further in the summer. When the convective velocity is low the birds mostly use the orograhpic lift for flying (fig. 4.1). When the convective velocity is high the birds seem to fly more on places with a high convective velocity in the summer whereas in the winter the birds seems to fly equally above higher orographic lift values and higher convective velocity values (fig. 4.2). However, even on the day with the highest convective velocity in the summer they still seem to use orographic lift: The dark green and purple lines follow the orographic lift in figure4.2.

(20)

CHAPTER 4. RESULTS

Figure 4.2.: Tracks for the winter, 28th of February 12 O’Clock, highest maximum values of con-vective velocity (1.96 m/s) and for the summer 20th of August, 12 O’Clock, highest maximum values of convective velocity (2.82 m/s)

(21)

CHAPTER 4. RESULTS

Figure 4.3.: Tracks for the winter, 14th of January, 12 O’Clock, lowest maximum values for oro-graphic lift (0.15 m/s) from the south east south and for the summer, 5th of August, 12 O’Clock, highest maximum values for orographic lift (0.09 m/s) from the west

(22)

CHAPTER 4. RESULTS

Figure 4.4.: Tracks for the winter, 27th of February, 12 O’Clock, highest maximum values for oro-graphic (9.31 m/s) from the southeast and for the summer, 19th of July, 12 O’Clock, highest maximum values for orographic ( 9.94 m/s) from the east south east

(23)

CHAPTER 4. RESULTS

On the other hand, figure 4.3 and figure 4.4 show that a higher orograhpic lift also seems to result in travelling longer distances by the Griffon Vulture. Because the low orograhpic lift on the 14th of January goes hand in hand with a low convective velocity it is impossible to glide and soar on this day possibly the cause of hardly any fly movements on this day. During the summer, it is suggested that the birds use convective velocity for flying when the orographic lift is almost absent (fig. 4.3). On the other thand, the Griffon Vultures seems use the orographic lift to fly when the convective velocity is not that high. The red line as well as the purple, green, dark blue and light blue line perfectly follow the canyons which create the orographic lift in the summer in figure4.4. The winter in the same figure shows a blue line which is also following the canyons.

4.2. Flight Characteristics

For every flight characteristic a table is presented, table 4.1 for the time of first take of at a day, table 4.2 for time of last landing on a day, table 4.3 for the total flight duration per day and table 4.4 for the total traveled distance per day. This table contains the results for the winter and summer apart from another and a combination of the two. Per season the generalized linear models are performed twice, with and without the influence of canyons. This is done with the 2 predictor variables orographic lift and convective velocity, when one of these was not significant, the generalized linear model is performed again with the variable belonging to the lowest p-value. For example, in table4.1the lowest p-value for the generalized linear model of the winter with canyons the orographic lift has a lower p-value than convective velocity (respectively 0.121 and 0.375), therefore this model is performed again with only orographic lift as predictor variable. The p-value has be lower than 0.05 for a significant influence of the variable, so in this case neither orographic lift nor convective velocity are significant. The intercept represents the response variable when the orographic lift and convective velocity are zero, in this case that is the time of take off in hours after sunrise. The estimated coefficient shows the influence of the predictor variables on the flight characteristics. In the case of table4.1the influence of orographic lift in the generalized linear model of the winter with canyons is -9.67e-3 which means that 1 m/s more orographic lift results in earlier depart of 9.67e-3 hours. The adjusted R2 points out the amount of variation in the response variable that is explained; the maximum of the adjusted R2is one representing 100% explanation of the response variable. In the same case as discussed before the explained variation is barely 1%.

(24)

CHAPTER 4. RESULTS

4.2.1. Time of first take off at a day

Table 4.1.: Generalized linear regression models for first take off time on the day for the different seasons and combined seasons as observed from GPS-tracking and as predicted from orographic lift and convective velocity reference points on 12 O’Clock on the same day as the observation

Season Model Predictor Estimated

Coefficient p-value

Adjusted R2 Intercept 0.201 6.57e-20

Orographic Lift -9.67e-3 0.121 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity -1.94e-2 0.375 0.0090 Intercept 0.202 4.03e-20 Orographic Lift -1.56e-2 0.244 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity -2.11e-2 0.334 0.097 Intercept 0.185 0 1 + Orographic Lift With

Canyons Orographic Lift -1.01e-2 0.106 0.010 Intercept 0.185 0

Winter

1 + Orographic Lift

Without

Canyons Orographic Lift -1.59e-2 0.235 0.0026 Intercept 0.202 4.03e-20

Orographic Lift -1.56e-2 0.244 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity -2.11e-2 0.334 0.0023 Intercept 0.134 2.62e-21 Orographic Lift -2.13e-2 4.11e-2 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 1.24e-3 0.876 0.0070 Intercept 0.185 0 1 + Orographic Lift With

Canyons Orographic Lift -1.59e-2 0.235 0.010 Intercept 0.136 0

Summer

1 + Orographic Lift Without

Canyons Orographic Lift -2.13e-2 4.06e-2 0.010 Intercept 0.189 0

Orographic Lift -2.52e-4 0.94 1+ Orographic Lift

+ Convective Velocity With

Canyons Convective

Velocity -2.50e-2 8.05e-05

0.029

Intercept 0.188 0 Orographic Lift -1.89e-2 2.46e-2 1 + Orographic Lift

+ Convective Velocity

Without

Canyons Convective

Velocity -2.51e-2 6.35e-05

0.039

Intercept 0.154 0 1 + Orographic Lift With

Canyons Orographic Lift -1.45e-3 0.648 -0.0017 Intercept 0.153 0

Winter and Summer

Canyons Orographic Lift -1.87e-2 2.82e-2 0.0080

(25)

CHAPTER 4. RESULTS

4.2.2. Time of last landing of the day

Table 4.2.: Generalized linear regression models for last landing time on the day for the different seasons and combined seasons as observed from GPS-tracking and as predicted from orographic lift and convective velocity reference points on 12 O’Clock on the same day as the observation

Coefficient p-value

Adjusted R2 Intercept -0.124 1.82e-09

Orographic Lift 2.40e-3 0.620 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 3.07e-2 0.166 0.0020 Intercept -0.122 2.14e-09 Orographic Lift 1.95e-2 8.79e-2 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 2.73e-2 0.215 0.0200

Intercept -9.88e-2 3.09e-32 1 + Orographic

Lift

With

Canyons Orographic Lift 2.91e-3 0.548 -0.0043 Intercept -9.98e-2 3.25e-33

Winter

1 + Orographic Lift

Without

Canyons Orographic Lift 2.11e-2 6.40e-2 0.016 Intercept -0.106 3.91e-09

Orographic Lift 4.54e-3 0.458 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 1.78e-3 0.868 -0.0042 Intercept -0.108 1.51e-09 Orographic Lift 3.85e-3 0.630 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 2.80e-3 0.791 -0.0052 Intercept -0.103 0 1 + Orographic Lift With

Canyons Orographic Lift 4.71e-3 0.434 -0.0012 Intercept -0.104 0

Summer

Canyons Orographic Lift 3.97e-3 0.618 -0.0023 Intercept -0.104 2.05e-21

Orographic Lift -2.10e-3 0.607 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 1.07e-3 0.881 -0.0037 Intercept -0.103 2.30e-21 Orographic Lift -1.13e-2 8.47e-2 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 7.94e-4 0.911 0.0021 Intercept -0.103 0 1 + Orographic Lift With

Canyons Orographic Lift -2.04e-3 0.616 -0.0016 Intercept -0.102 0

Winter and Summer

(26)

CHAPTER 4. RESULTS

4.2.3. Total flight duration per day

Table 4.3.: Generalized linear regression models for total flight duration per day for the different seasons and combined seasons as observed from GPS-tracking and as predicted from the mean orographic lift and mean convective velocity the bird experienced during their flight movements on a day

Coefficient p-value Adjusted R2 Intercept 1.82 9.66e-05 Orographic Lift 0.283 0.147 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 0.917 0.128 0.014 Intercept 1.84 9.40e-05 Orographic Lift 0.147 0.747 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 0.888 0.143 0.0019 Intercept 1.82 9.48e-05 Winter 1 + Convective Velocity Convective Velocity 0.899 0.137 0.0071 Intercept 7.06 3.74e-22 Orographic Lift -0.162 0.770 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 0.956 06.17e-2 0.0047 Intercept 7.00 1.55e-21 Orographic Lift -0.852 0.515 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 0.995 5.36e-2 0.0057 Intercept 7.06 2.99e-22 Summer 1 + Convective Velocity Convective Velocity 0.955 06.17e-2 0.0074 Intercept -3.59e-2 0.93 Orographic Lift 0.21 0.41 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 5.77 0 0.37 Intercept -0.051565 0.89582 Orographic Lift -0.21527 0.72523 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 5.7723 0 0.36 Intercept -3.76e-2 0.92 Winter and Summer 1 + Convective Velocity Convective Velocity 5.76 0 0.37

(27)

CHAPTER 4. RESULTS

4.2.4. Total traveled distance per day

Table 4.4.: Generalized linear regression models for total travelled distance per day for the different seasons and combined seasons as observed from GPS-tracking and as predicted from the mean orographic lift and mean convective velocity the bird experienced during their flight movements on a day

Coefficient p-value Adjusted R2 Intercept 79.62 4.23e-4 Orographic Lift 10.55 0.266 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 49.63 9.13e-2 0.012 Intercept 80.29 4.16e-4 Orographic Lift 4.38 0.843 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 48.65 9.94e-2 0.0048 Intercept 79.91 4.07e-4 Winter 1 + Convective Velocity Convective Velocity 48.99 9.56e-2 0.010 Intercept 200.51 2.20e-09 Orographic Lift -4.01 0.880 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 137.15 4.64e-08 0.080 Intercept 195.70 6.58e-09 Orographic Lift -66.73 0.288 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 140.23 2.81e-08 0.083 Intercept 200.64 2.02e-09 Summer 1 + Convective Velocity Convective Velocity 137.11 4.47e-08 0.083 Intercept -33.04 6.09e-2 Orographic Lift 8.69 0.46 1+ Orographic Lift + Convective Velocity With Canyons Convective Velocity 292.41 0 0.42 Intercept -34.12 5.42e-2 Orographic Lift -15.61 0.570 1 + Orographic Lift + Convective Velocity Without Canyons Convective Velocity 292.97 0 0.42 Intercept -33.11 6.02e-2 Winter and Summer 1 + Convective Velocity Convective Velocity 292.32 0 0.42

(28)

(29)

5. Discussion

The results will be discussed in this part of the report. This will be done for every flight charac-teristic as well as for the different results caused by the different calculated orographic lift (with or without canyons). The first section of the results, Data Impression, will also be discussed because this gives a good indication of the different effects of season and convective velocity and orographic lift.

5.1. Data Impression

In general, the Griffon Vulture flies more when the convective velocity is higher (fig 4.1, 4.2). This is probably due to the fact that the landscape consists of large relatively flat areas on the plateau (Groc et al., 2007)). When the birds are not flying along the canyons they depend on the convective velocity to be able to soar and glide through the landscape (Shepard and Lambertucci, 2013). However this does not mean that the Griffon Vulture never uses the orographic lift. The tracks are sometimes following the places with orographic lift closely (fig4.1,4.4,4.2).

The differences between winter and summer are very well visible. The maximum distance during the winter is not as far as the shortest distance flown in the summer (fig. 4.1,4.2,4.3,4.4). This is can be caused by the fact that food is widely available during the winter because of the severe conditions more animals die (Fluhr and Durier, 2015) and in the summer they have to search harder for it, resulting in longer travel distances. Another reason for this difference is the amount of uplift in the winter and summer. The orographic lift is about the same in its extremes (fig. 4.3, 4.4), but the convective velocity is higher during the summer the highest maximum value in the winter is 1.96 m/s and in the summer the lowest maximum values are already 1.25 m/s (fig4.1, 4.2). This is caused by the higher radiation in the summer (Groot, 2015).

5.2. Time of first take off on the day

The time of first take off on the day is significant for the combination winter and summer without the influence of canyons. The other generalized linear models are neither significant for orographic lift and convective velocity together nor for the orographic lift on itself (table4.1). This can be caused by the fact that only one reference point on the day is used instead the whole curve of convective velocity and orographic lift during daytime. The total amount of convective velocity and orographic lift could represent the daily conditions better than one reference point at 12h O’Clock. Another shortcoming of the model is that the first take off implies that the bird is taking off for foraging flights, however the short flight movements are not deleted out of the data selection. This causes some additional noise in the table used for the generalized linear model. It may otherwise be caused by alternative influences on the movement path described by (Nathan et al., 2008). The internal state can influence the time of take off as well as the uplift conditions; if the bird is not hungry yet, why should it move? This incentive is different for the winter and summer as already mentioned. However, the intercepts indicating the time birds take off when there is no influence of convective velocity or orograhpic lift is always between 1 hour and 20

(30)

CHAPTER 5. DISCUSSION

5.3. Time of last landing on the day

The time of last landing on the day is just like the time of first take off neither significant for orographic lift combined with convective velocity nor significant for the orographic lift on its own when looking at only the winter or the summer (table 4.2). Yet, the generalized linear model for the winter and summer without canyons is significant for the orographic lift. Because of the fact that the procedure to select data is done on the same principles as the time of first take off, this characteristic has the same short-comings: the reference point could be replaced with a total amount for the day and the short flight movements are not excluded from the selection. The internal state as described in Nathan et al. (2008), is an alternative major influence on the movements of the Griffon Vulture. The reintroduction program provides food on artificial feeding sites 2 hours before sunset (Fluhr and Durier, 2015). That influences the time of last landing, which is always a bit earlier than one hour before sunset as described by the intercept in table4.2.

5.4. Total flight duration per day

The total flight duration per day is significant for the summer on itself with convective velocity as predictor variable. This model is significant but not explaining much of the variation in total flight duration: only 0.7% (table4.3). This can be due to alternative influences on the movements of the Griffon vulture (Nathan et al., 2008). The internal state and other external influences can probably explain a larger part of the variation in total flight duration. However, for this generalized linear model the mean value of orographic lift and convective velocity is used. This parameter generalizes the extremer values of orographic lift and convective velocity. Thus, the range of these mean values is not very broad and the effect will be smaller. The reason that the summer results in a significant model can be due to the difference in amount of data for the winter and the summer. The birds fly more during the summer and the resolution of the data is higher than for the winter data. In the winter flights shorter than 30 minutes are not necessarily detected in the data when it is precisely between measurements while in the summer measurements are taken about every 5 minutes. So, when combining the two season the results change due to a higher amount of variation. The influence of convective velocity on the total flight duration is significant and explains 37% of the variation, whereas the intercept and orographic lift are far from being significant.

The difference between summer and winter with respect to the total time the Griffon Vultures fly, may be explained by the difference of food availability as already mentioned. In the winter the intercept of the model is about 1.8 hours and in the summer this is about 7 hours (table: 4.3.

5.5. Total traveled distance per day

The total traveled distance per day is either for the summer significant, for the winter and for a combination of summer and winter when looking at the convective velocity (table 4.4). This is remarkable because the time calculated for the total flight duration is used with the instantaneous speed to calculate the total traveled distance. The values for instantaneous speed have a larger uncertainty than the values for time. However, the ranges of distances are larger than those for time probably causing the lower p-values.

Furthermore, this flight characteristic is exactly the same selected from the original data set therefore the causes of uncertain models are also the same. This also accounts for the differences between summer and winter with respect to the total distance traveled on average: in the winter the basic distance determined by the intercept of the models is about 80 kilometres and for the summer this is about 200 kilometres (table4.4).

(31)

CHAPTER 5. DISCUSSION

5.6. Orograhpic lift

The orographic lift never has a significant influence in a generalize linear model for explaining the flight characteristics in only the summer or the winter. Yet, when looking at the visualized tracks, this is a bit strange. The tracks are following the places with orographic lift closely for parts of the total track (fig. 4.1, 4.2, 4.4). This can be due to the resolution off the orographic lift data. The grid cell size is 100 by 100 meters and in canyons the air flows are changing on a much smaller scale (Wittebrood, 2015). When the orographic lift would be calculated with higher resolution and a more complicated algorithm that describes this flow better, it is possible that the orographic lift will become significant for the total travelled distance and total flight duration per day for only the summer or the winter. Along with that it is also possible that the GPS locations are not determined exact enough and fall in a different grid cell than in the grid cell where the bird was in reality. Therefore it would have been an option to check the 8 cells around the initially selected grid cell and take the cell with the highest value. Unfortunately time ran out so this is a recommendation for further research.

5.7. Recommendations

Since this research is only explorative, it would be of great value if more research would be done on flight characteristics of the Griffon Vulture. First of all, the orographic lift used for the analysis should be calculated in more detail to be able to perform a detailed analysis. In this research this was not the case and that could be the reason orographic lift is found to be insignificant. Also, as just mentioned, it is recommended to check the 8 cells around the initially selected grid cell for higher values of orographic lift and replace the initially cell for the cell with the highest value. Furthermore, these four flight characteristics are only a subset of all the characteristics that can be researched e.g. mean and maximum flight altitude, above which land use they spend the most time. Nevertheless, it is recommended to take into account that uplift conditions are not the only factor influencing these characteristics.

(32)

(33)

6. Conclusions

To come back on the research question that this report aims to answer: What updraft conditions influence flight characteristics of Griffon Vultures? and the different flight characteristics that are studied.

The first take off time at a day is not significantly influenced by orographic lift and convective velocity. This can be due to short-comings of the data selection and the quality of the orographic lift data. However, the internal state of the Griffon Vulture also influences the time of first take off on the day.

Furthermore, the last landing of the day is also not significantly influenced by orographic lift and convective velocity. The same short-comings of the data selection procedure as for the first take off time are a possible cause of this. Nevertheless, if this is not due to the short-comings of the model the other incentives for movement as described by Nathan et al. (2008) are probably having more influence than the uplift conditions.

The total flight duration per day is significantly influenced by convective velocity, the amount of variation it explains is 37% when looking at the combined seasons. The other variation is more likely to be explained by the internal state and other abiotic influences. The orographic lift has not a significant influence for any of the generalized linear models. The amount of data present for the winter can be the cause of the insignificance of the convective velocity in the winter.

The total traveled distance is significantly influenced by convective velocity as well as the total flight duration per day. The variation of the different distances that is explained by convective velocity is 42%. The orographic lift is not influencing the total traveled distance in any of the generalize linear models. The other variation in total traveled distance is likely to be explained by the internal state and other abiotic influences.

(34)

(35)

7. Acknownledgement

I would like to thank Willem Bouten for his remarks and supervising during this research, Ivana Post for the spelling checks, Jim Groot and Marnix Wittebrood for thinking along and sharing their bachelor theses and Wouter Vansteeland for the explanation of generalized linear models. Also, I would like to thank Julie Fluhr and Olivier Durier for thinking along, discussing my research and skype conversation.

(36)

(37)

Bibliography

Bohrer, G., Brandes, D., Mandel, J. T., Bildstein, K. L., Miller, T. A., Lanzone, M., Katzner, T., Maisonneuve, C., and Tremblay, J. A. (2012). Estimating updraft velocity components over large spatial scales: contrasting migration strategies of golden eagles and turkey vultures. Ecology Letters, 15(2):96–103. 1,4

Bouten, W., Baaij, E. W., Shamoun-Baranes, J., and Camphuysen, K. C. (2013). A flexible gps tracking system for studying bird behaviour at multiple scales. Journal of Ornithology, 154(2):571–580. 1

Duerr, A. E., Miller, T. A., Lanzone, M., Brandes, D., Cooper, J., O’Malley, K., Maisonneuve, C., Tremblay, J. A., and Katzner, T. (2014). Flight response of slope-soaring birds to seasonal variation in thermal generation. Functional Ecology. 1

Duriez, O., Kato, A., Tromp, C., Dell’Omo, G., Vyssotski, A. L., Sarrazin, F., and Ropert-Coudert, Y. (2014). How cheap is soaring flight in raptors? a preliminary investigation in freely-flying vultures. PloS one, 9(1):e84887. 4

Fluhr, J. and Durier, O. (2015). Oral conversation. 9th of June 2015. 6,23, 24

Groc, S., Delabie, J. H., C´er´eghino, R., Orivel, J., Jaladeau, F., Grangier, J., Mariano, C. S., and Dejean, A. (2007). Ant species diversity in the grands causses(aveyron, france): in search of sampling methods adapted to temperate climates. Comptes rendus biologies, 330(12):913–922. 6, 23

Groot, J. (2015). Convective velocity, a prerequisite of soaring birds Bachelor Thesis. 2,5,23 Mandel, J. T., Bohrer, G., Winkler, D. W., Barber, D. R., Houston, C. S., and Bildstein, K. L.

(2011). Migration path annotation: cross-continental study of migration-flight response to environmental conditions. Ecological Applications, 21(6):2258–2268. 1

Morales-Reyes, Z., Pérez-Garc´ıa, J. M., Moleón, M., Botella, F., Carrete, M., Lazcano, C., Moreno-Opo, R., Margalida, A., Donázar, J. A., and Sánchez-Zapata, J. A. (2015). Supplanting eco-system services provided by scavengers raises greenhouse gas emissions. Scientific reports, 5. 1

Nathan, R., Getz, W. M., Revilla, E., Holyoak, M., Kadmon, R., Saltz, D., and Smouse, P. E. (2008). A movement ecology paradigm for unifying organismal movement research. Proceedings of the National Academy of Sciences, 105(49):19052–19059. 3, 23, 24,27

naturel rgional des Grands Causses’, P. (2015). The causses.

http://www.parc-grands-causses.fr/en/discover/landscapes/causses. last acessed on 20-6-2015. 6

Pennycuick, C. (1998). Field observations of thermals and thermal streets, and the theory of cross-country soaring flight. Journal of Avian Biology, pages 33–43. 1

Pennycuick, C. (2003). The concept of energy height in animal locomotion: separating mechanics from physiology. Journal of theoretical biology, 224(2):189–203. 1

(38)

BIBLIOGRAPHY

Shamoun-Baranes, J., Baharad, A., Alpert, P., Berthold, P., Yom-Tov, Y., Dvir, Y., and Leshem, Y. (2003). The effect of wind, season and latitude on the migration speed of white storks ciconia ciconia, along the eastern migration route. Journal of Avian Biology, 34(1):97–104. 4

Shepard, E. L. and Lambertucci, S. A. (2013). From daily movements to population distributions: weather affects competitive ability in a guild of soaring birds. Journal of The Royal Society Interface, 10(88):20130612. 4,23

Shepard, E. L., Wilson, R. P., Rees, W. G., Grundy, E., Lambertucci, S. A., and Vosper, S. B. (2013). Energy landscapes shape animal movement ecology. The American Naturalist, 182(3):298–312. 4

Spaar, R. (1997). Flight strategies of migrating raptors; a comparative study of interspecific variation in flight characteristics. Ibis, 139(3):523–535. 1

Spiegel, O., Harel, R., Getz, W. M., and Nathan, R. (2013). Mixed strategies of griffon vul-tures(gyps fulvus) response to food deprivation lead to a hump-shaped movement pattern. Movement ecology, 1(5). 1

Terrasse, M., Houston, D., and Piper, S. (2005). Long term reintroduction projects of griffon gyps fulvus and black vultures aegypius monachus in france. In Proceedings of the International Conference on Conservation and Management of Vulture Populations, pages 98–107. 1

UvA-BiTS (2015). Griffon vulture in the grands causses. http://www.uva-bits.nl/project/griffon-vulture-in-the-grands-causses/. last accessed on 20-6-2015. 6

Vansteelant, W., Bouten, W., Klaassen, R., Koks, B., Schlaich, A., van Diermen, J., van Loon, E., and Shamoun-Baranes, J. (2014). Regional and seasonal flight speeds of soaring migrants and the role of weather conditions at hourly and daily scales. Journal of Avian Biology. 1 Wittebrood, M. (2015). Orographic lift and combined landscape dynamics Bachelor Thesis. 2,5,

25

(39)

A. Data Preperation

A.1. Histograms

Figure A.1.: Fixtimes of the raw dataset derived from UvA-BiTS for the January, February, July and August 2011 for the devices 140, 141, 210, 212, 225, 226

(40)

APPENDIX A. DATA PREPERATION

Figure A.2.: Instantaneous speeds of the raw dataset derived from UvA-BiTS for the January, February, July and August 2011 for the devices 140, 141, 210, 212, 225, 226

A.2. Matlab Script

%% Get Data From UvA BiTs and ECMWF Model

% P a r t o f t h e b a c h e l o r t h e s i s by D o r i e n van Kranenburg % To F l y o r n o t t o f l y , a t m o s p h e r i c i n f l u e n c e s on t h e f l i g h t b e h a v i o r o f % t h e G r i f f o n V u l t u r e i n F r a n c e c l c c l e a r a l l c l o s e a l l %% j a v a add path j a v a a d d p a t h (’D: \ D o r i e n \ B a c h e l o r P r o j e c t \ p o s t g r e s q l −9.4 −1201. j d b c 4 . j a r ’) %% add path M a t l a b 2 G o o g l e E a r t h t o o l b o x addpath (’D: \ D o r i e n \ B a c h e l o r P r o j e c t \ g o o g l e e a r t h ’) %% add path Openearth t o o l b o x

addpath ( g e n p a t h (’D: \ D o r i e n \ B a c h e l o r P r o j e c t \ o p e n e a r t h ’) ) % %% G e t t i n g t h e d a t a from UvA−BiTS

username = ; % put y o u r u s e r name and p a s s w o r d h e r e . p a s s w o r d = ;

KDevice = [ 1 4 0 , 1 4 1 , 2 1 0 , 2 1 2 , 2 2 5 , 2 2 6 ] ; % [ 2 1 2 , 2 1 9 , 225 2 2 6 ] s t a r t t i m e = ’ 2011−01−01 0 0 : 0 0 : 0 0 ’;

s t o p t i m e = ’ 2011−03−01 0 0 : 0 0 : 0 0 ’;

[ Times , GPSData , SpeedTime , CheckData , SpeedData , AccuracyData , . . .

(41)

C o n d i t i o n D a t a ] = G e t B i t s D a t a ( username , password , . . . KDevice , s t a r t t i m e , s t o p t i m e ) ;

s t a r t t i m e = ’ 2011−07−01 0 0 : 0 0 : 0 0 ’; s t o p t i m e = ’ 2011−09−01 0 0 : 0 0 : 0 0 ’;

[ Times1 , GPSData1 , SpeedTime1 , CheckData1 , SpeedData1 , AccuracyData1 , . . . C o n d i t i o n D a t a 1 ] = G e t B i t s D a t a ( username , password , . . .

KDevice , s t a r t t i m e , s t o p t i m e ) ; % c o n c e n a t i n g w i n t e r and summerdata

Times = [ Times , Times1 ] ;

GPSData = [ GPSData , GPSData1 ] ;

SpeedTime = [ SpeedTime , SpeedTime1 ] ;

CheckData = [ CheckData , CheckData1 ] ;

SpeedData = [ SpeedData , SpeedData1 ] ;

AccuracyData = [ AccuracyData , AccuracyData1 ] ;

C o n d i t i o n D a t a = [ C o n d i t i o n D a t a , C o n d i t i o n D a t a 1 ] ;

% Time g i v e s [ ID , Year , Month , Day , Hour , Minute , Second , numTime ] ; % GPSData g i v e s [ ID , Lat , Long , Alt , TSpeed ]

% CheckData g i v e s [ NrSat , F i x t i m e ]

% SpeedData g i v e s [ ID , XSpeed , YSpeed , ZSpeed , CSpeed3d , CSpeed2d , TSpeed ] % AccuracyData g i v e s [ HAccuracy , VAccuracy , SpAccuracy ]

% C o n d i t i o n D a t a g i v e s [ P r e s s , Temp , L o c a t i o n , D i r e c t i o n , A l t A g l ] c l e a r Times1 GPSData1 SpeedTime1 CheckData1 SpeedData1 AccuracyData1

c l e a r C o n d i t i o n Data1 %% c l e a n i n g up GPSdata N r S t o r e = 1 ; NrStoreA = 1 ; % c h e c k e n op s a t e l i t e s , f i x t i m e en i n s t a n t a n i o u s s p e e d NrSat = CheckData ( 1 , : ) ; FixTime = CheckData ( 2 , : ) ; CSpeed2d = SpeedData ( 6 , : ) ; f i g u r e h i s t( FixTime , 8 0 ) % d e l e t e FixTime > 25 x l a b e l(’ f i x t i m e i n s e c o n d s ’) y l a b e l(’ f r e q u e n c i e s ’) f i g u r e h i s t( CSpeed2d , 3 2 0 ) % d e l e t e CSpeed > 120 km/h x l a b e l(’ I n s t a n t e n e o u s s p e e d i n km/h ’) y l a b e l(’ f r e q u e n c i e s ’) x l i m ( [ 0 2 0 0 ] ) y l i m ( [ 0 2 0 0 0 ] ) f o r i n d e x = 1 : l e n g t h( FixTime )

i f ( FixTime ( i n d e x ) < 5 0 ) && ( CSpeed2d ( i n d e x ) < 1 2 0 ) I n d e x P r o p e r D a t a ( N r S t o r e ) = i n d e x ; N r S t o r e = N r S t o r e +1; e l s e IndexNonProperData ( NrStoreA ) = i n d e x ; NrStoreA = NrStoreA +1; end end % v a r i a b l e s / m a t r i x e s i n c l e a n e d v e r s i o n : Day = Times ( 4 , : ) ; numTime = Times ( 8 , : ) ;

% TimesA = Times ( : , IndexNonProperData ) ;

% DayA = Day ( IndexNonProperData ) ;

(42)

% GPSDataA = GPSData ( : , IndexNonProperData ) ; % SpeedTimeA = SpeedTime ( : , IndexNonProperData ) ; % CheckDataA = CheckData ( : , IndexNonProperData ) ; % SpeedDataA = SpeedData ( : , IndexNonProperData ) ; % ISpeedA = CSpeed2d ( IndexNonProperData ) ; % AccuracyDataA = AccuracyData ( : , IndexNonProperData ) ; % ConditionDataA = C o n d i t i o n D a t a ( : , IndexNonProperData ) ; % FixtimeA = FixTime ( IndexNonProperData ) ;

% NrSatA = NrSat ( IndexNonProperData ) ; % CSpeed2dA = CSpeed2d ( IndexNonProperData ) ;

Times = Times ( : , I n d e x P r o p e r D a t a ) ; Day = Day ( I n d e x P r o p e r D a t a ) ; numTime = numTime ( I n d e x P r o p e r D a t a ) ; GPSData = GPSData ( : , I n d e x P r o p e r D a t a ) ; SpeedTime = SpeedTime ( : , I n d e x P r o p e r D a t a ) ; CheckData = CheckData ( : , I n d e x P r o p e r D a t a ) ; SpeedData = SpeedData ( : , I n d e x P r o p e r D a t a ) ; I S p e e d = CSpeed2d ( I n d e x P r o p e r D a t a ) ; AccuracyData = AccuracyData ( : , I n d e x P r o p e r D a t a ) ; C o n d i t i o n D a t a = C o n d i t i o n D a t a ( : , I n d e x P r o p e r D a t a ) ; F i x t i m e = FixTime ( I n d e x P r o p e r D a t a ) ; NrSat = NrSat ( I n d e x P r o p e r D a t a ) ; CSpeed2d = CSpeed2d ( I n d e x P r o p e r D a t a ) ; % Save d a t a

s a v e(’ Times . mat ’, ’ Times ’, ’−mat ’) ;

s a v e(’ Day . mat ’, ’ Day ’, ’−mat ’) ;

s a v e(’ numTime . mat ’, ’ numTime ’, ’−mat ’) ;

s a v e(’ SpeedTime . mat ’, ’ SpeedTime ’, ’−mat ’) ;

s a v e(’ GPSData . mat ’, ’ GPSData ’, ’−mat ’) ;

s a v e(’ CheckData . mat ’, ’ CheckData ’, ’−mat ’) ;

s a v e(’ SpeedData . mat ’, ’ SpeedData ’, ’−mat ’) ;

s a v e(’ I S p e e d . mat ’, ’ I S p e e d ’, ’−mat ’) ;

s a v e(’ AccuracyData . mat ’, ’ AccuracyData ’, ’−mat ’) ;

s a v e(’ C o n d i t i o n D a t a . mat ’, ’ C o n d i t i o n D a t a ’, ’−mat ’) ; % d e f i n i n g t h r e s h o l d s p e e d f i g u r e h i s t( I S p e e d , 3 2 0 ) % d e l e t e CSpeed > 120 km/h x l a b e l(’ I n s t a n t e n e o u s s p e e d i n km/h ’) y l a b e l(’ f r e q u e n c i e s ’) x l i m ( [ 0 2 0 0 ] ) y l i m ( [ 0 2 0 0 0 ] ) %% G e t t i n g t h e d a t a from ECMWF s t a r t t i m e = ’ 2011−01−01 0 0 : 0 0 : 0 0 ’; s t o p t i m e = ’ 2011−03−01 0 0 : 0 0 : 0 0 ’;

[ B i r d s ]= GetMeteoData ( username , password , s t a r t t i m e , s t o p t i m e ) ; MeteoTime = B i r d s . Time ; Wdir = B i r d s . Wdir ; WindSp = B i r d s . windsp ; BLH = B i r d s . b l h ; SSRad = B i r d s . s s r d ; P r e c i p = B i r d s . t p ; c l e a r B i r d s s t a r t t i m e = ’ 2011−07−01 0 0 : 0 0 : 0 0 ’; s t o p t i m e = ’ 2011−09−01 0 0 : 0 0 : 0 0 ’;

[ B i r d s 1 ]= GetMeteoData ( username , password , s t a r t t i m e , s t o p t i m e ) ; MeteoTime1 = B i r d s 1 . Time ;

Wdir1 = B i r d s 1 . Wdir ;

WindSp1 = B i r d s 1 . windsp ;

BLH1 = B i r d s 1 . b l h ;

(43)

SSRad1 = B i r d s 1 . s s r d ; P r e c i p 1 = B i r d s 1 . t p ;

c l e a r B i r d s 1

MeteoTime = [ MeteoTime ; MeteoTime1 ] ;

Wdir = [ Wdir ; Wdir1 ] ;

WindSp = [ WindSp ; WindSp1 ] ;

BLH = [ BLH ; BLH1 ] ;

SSRad = [ SSRad ; SSRad1 ] ;

P r e c i p = [ P r e c i p ; P r e c i p 1 ] ;

c l e a r MeteoTime1 Wdir1 BLH1 SSRad1 P r e c i p 1

s a v e(’ MeteoTime . mat ’, ’ MeteoTime ’, ’−mat ’) ;

s a v e(’ Wdir . mat ’, ’ Wdir ’, ’−mat ’) ;

s a v e(’ WindSp . mat ’, ’ WindSp ’, ’−mat ’) ;

s a v e(’BLH . mat ’, ’BLH ’, ’−mat ’) ;

s a v e(’ SSRad . mat ’, ’ SSRad ’, ’−mat ’) ;

s a v e(’ P r e c i p . mat ’, ’ P r e c i p ’, ’−mat ’) ;

f u n c t i o n [ Times , GPSData , SpeedTime , CheckData , SpeedData , AccuracyData , . . . C o n d i t i o n D a t a ] = GetBitsDataB ( username , password , KDevice , . . .

s t a r t t i m e , s t o p t i m e ) %% D e s c r i p t i o n

% t h i s f u n c t i o n g e t s d a t a from t h e Uva−B i t s p u b l i c d a t a b a s e . The t a b l e i t % d e r i v e s i n f o r m a t i o n from i s ’ g p s . e e t r a c k i n g s p e e d l i m i t e d t ’

%% I n p u t

% username , password , KDevice , s t a r t t i m e , s t o p t i m e

% f o r making c o n n e c t i o n a username and p a s s w o r d i s n e e d e d .

% The d e v i c e s from which t h e d a t a h a s t o be i m p o r t e d a r e d e t e r m i n e d i n a % v e c t o r c a l l e d KDevice

% s t a r t t i m e h a s t o be s p e c i f i e d e g : ’2012 −07 −01 0 0 : 0 0 : 0 0 ’ % s t o p t i m e h a s t o be s p e c i f i e d e g : ’2012 −09 −01 0 0 : 0 0 : 0 0 ’ %% Output

% The f u n c t i o n c l u s t e r s t h e d a t a from t h e d a t a b a s e e e e c o l o g y s p e e d l i m i t e d % Times = [ ID ; Year ; Month ; Day ; Hour ; Minute ; Second ; numTime ] ;

% GPSData = [ ID ; Lat ; Long ; A l t ] ; % SpeedTime = [ TSpeed ; Time ] ; % CheckData = [ NrSat ; F i x t i m e ] ;

% SpeedData = [ ID ; XSpeed ; YSpeed ; ZSpeed ; CSpeed3d ; CSpeed2d ] ; % AccuracyData = [ HAccuracy ; VAccuracy ; SpAccuracy ] ;

% C o n d i t i o n D a t a = [ P r e s s ; Temp ; D i r e c t i o n ; A l t A g l ] ; %% G e t t i n g t h e d a t a from UvA−BiTS NewStart = 1 ; f o r k =1:l e n g t h( KDevice ) %% Read d a t a d i r e c t l y from t h e d a t a b a s e %% c o n n e c t u s i n g j d b c

conn = d a t a b a s e (’ e e c o l o g y ’, username , password , . . .

’ o r g . p o s t g r e s q l . D r i v e r ’, s t r c a t (’ j d b c : p o s t g r e s q l : / / p u b l i c . ’ , . . . ’ e−e c o l o g y . s a r a . n l : 5 4 3 2 / e e c o l o g y ? s s l f a c t o r y=o r g . p o s t g r e s q l . ’ , . . . ’ s s l . N o n V a l i d a t i n g F a c t o r y& s s l =t r u e ’) )

%% s e t db p r e f s .

% s e t d a t a b a s e p r e f . t h i s d e t e r m i n e s how t h e r e s u l t s a r e r e t u r n e d . % ’ s t r u c t u r e ’ means you can r e f e r e n c e t h e f i e l d s by t h e same names a s % t h e d a t a b a s e t a b l e s u s e . ( s e e doc s e t d b p r e f s ) s e t d b p r e f s (’ DataReturnFormat ’,’ s t r u c t u r e ’) ; %% p a r a m e t e r s f o r t h e t r a c k q u e r y % s e l e c t d e v i c e s d e v i c e = num2str( KDevice ( k ) ) ; % s t a r t t i m e i s p a s s e d a s a s t r i n g p a r a m e t e r t o s q l s o i t s h o u l d by

(44)

APPENDIX A. DATA PREPERATION % put t h e s i n g l e q u o t e s around i t . s t a r t = s t r c a t (’ ’ ’ ’, s t a r t t i m e ,’ ’ ’ ’) ; s t o p = s t r c a t (’ ’ ’ ’, s t o p t i m e , ’ ’ ’ ’) ; % d a t a = s t r c a t ( ’ ’ ’ ’ , data , ’ ’ ’ ’ ) ; % s q l q u e r y . s e l e c t from t h e g p s u v a t r a c k i n g t a b l e % a l l t h e r e c o r d s i n between t h e s t a r t and s t o p t i m e and % s o r t them on t h e d a t e t i m e f i e l d . s q l = s t r c a t (’ s e l e c t t . d e v i c e i n f o s e r i a l , ’, . . . ’ d a t e p a r t ( ’ ’ y e a r ’ ’ : : t e x t , t . d a t e t i m e ) AS y e a r , ’ , . . . ’ d a t e p a r t ( ’ ’ month ’ ’ : : t e x t , t . d a t e t i m e ) AS month , ’ , . . . ’ d a t e p a r t ( ’ ’ day ’ ’ : : t e x t , t . d a t e t i m e ) AS day , ’ , . . . ’ d a t e p a r t ( ’ ’ hour ’ ’ : : t e x t , t . d a t e t i m e ) AS hour , ’ , . . . ’ d a t e p a r t ( ’ ’ minute ’ ’ : : t e x t , t . d a t e t i m e ) AS minute , ’ , . . . ’ d a t e p a r t ( ’ ’ s e c o n d ’ ’ : : t e x t , t . d a t e t i m e ) AS s e c o n d , ’ , . . . ’ t . l a t i t u d e , t . l o n g i t u d e , t . a l t i t u d e , t . p r e s s u r e , ’ , . . . ’ t . t e m p e r a t u r e , t . s a t e l l i t e s u s e d , t . g p s f i x t i m e , ’ , . . . ’ t . p o s i t i o n d o p , t . h a c c u r a c y , t . v a c c u r a c y , t . x s p e e d , ’ , . . . ’ t . y s p e e d , t . z s p e e d , t . s p e e d a c c u r a c y , t . l o c a t i o n , ’ , . . . ’ t . s p e e d 3 d , t . s p e e d 2 d , t . d i r e c t i o n , t . a l t i t u d e a g l ’ , . . . ’ from g p s . e e t r a c k i n g s p e e d l i m i t e d t ’ , . . . ’ where ( d a t e t i m e >= ’, s t a r t , ’ ) ’ , . . . ’ AND ( d a t e t i m e <= ’, s t o p , ’ ) ’ , . . . ’ AND ( d e v i c e i n f o s e r i a l = ’ , d e v i c e , ’ ) ’, . . . %) ; ’ ORDER BY d a t e t i m e ’ ) ; %% run a q u e r y % g e t a ’ d a t a b a s e c u r s o r ’ % t h i s s e n t s t h e s q l command t o t h e d a t a b a s e and r e t u r n s a % ’ c u r s o r ’ t h a t can be u s e d t o r e t r e i v e t h e d a t a i n t h e t a b l e . c u r s = e x e c ( conn , s q l ) ; % t h i s f e t c h command a c t u a l l y t r a n s f e r s t h e d a t a from d a t a b a s e t a b l e % t o matlab c u r s = f e t c h ( c u r s ) ; %% g e t t h e d a t a from t h e c u r s o r % b e c a u s e p r e v i o u s l y s e t d b p r e f s i s s e t t o ’ s t r u c t u r e ’ d e v i c e s % i s a s t r u c t w i t h a r r a y f i e l d s t h a t have t h e same name a s t h e % d a t a b a s e columns . t o g e t t h e column d e v i c e i n f o s e r i a l from % t h e t a b l e a s an matlab a r r a y t y p e t r a c k s . d e v i c e i n f o s e r i a l t r a c k s = c u r s . Data ; c u r s . m e s s a g e %% c l o s e t h e d a t a b a s e c o n n e c t i o n % ( u n l e s s r u n n i n g more q u e r i e s ) c l o s e( conn ) ; %% copy d a t a D e v i c e L e n g t h = l e n g t h( t r a c k s . d e v i c e i n f o s e r i a l ) ; i n d e x T r a c k s = 1 ; f o r i n d e x = NewStart : ( NewStart + D e v i c e L e n g t h −1) ID ( i n d e x ) = t r a c k s . d e v i c e i n f o s e r i a l ( i n d e x T r a c k s ) ; Year ( i n d e x ) = t r a c k s . y e a r ( i n d e x T r a c k s ) ; Month ( i n d e x ) = t r a c k s . month ( i n d e x T r a c k s ) ; Day ( i n d e x ) = t r a c k s . day ( i n d e x T r a c k s ) ; Hour ( i n d e x ) = t r a c k s . hour ( i n d e x T r a c k s ) ; Minute ( i n d e x ) = t r a c k s . minute ( i n d e x T r a c k s ) ; Second ( i n d e x ) = t r a c k s . s e c o n d ( i n d e x T r a c k s ) ; Long ( i n d e x ) = t r a c k s . l o n g i t u d e ( i n d e x T r a c k s ) ; Lat ( i n d e x ) = t r a c k s . l a t i t u d e ( i n d e x T r a c k s ) ; A l t ( i n d e x ) = t r a c k s . a l t i t u d e ( i n d e x T r a c k s ) ; P r e s s ( i n d e x ) = t r a c k s . p r e s s u r e ( i n d e x T r a c k s ) ; Temp( i n d e x ) = t r a c k s . t e m p e r a t u r e ( i n d e x T r a c k s ) ; NrSat ( i n d e x ) = t r a c k s . s a t e l l i t e s u s e d ( i n d e x T r a c k s ) ; F i x t i m e ( i n d e x ) = t r a c k s . g p s f i x t i m e ( i n d e x T r a c k s ) ; PosDop ( i n d e x ) = t r a c k s . p o s i t i o n d o p ( i n d e x T r a c k s ) ; HAccuracy ( i n d e x )= t r a c k s . h a c c u r a c y ( i n d e x T r a c k s ) ;

To fly or not to fly; Atmospheric influences on the flight behaviour of Griffon Vulture in France