The Golden Eagle and the Bonelli’s Eagle on the Island of Crete, Greece

Nesting habitat selection & breeding distribution of two sympatric insular eagle populations: The Golden Eagle (Aquila chrysaetos) and the Bonelli’s Eagle (Aquila fasciata) on the island of Crete (Greece)

1Xirouchakis

Stavros, 2Solanou Maria & 1Georgopoulou Elisavet

1Natural History Museum of Crete, University of Crete, University Campus (Knossos) , Heraklion 71409, Crete, Greece 2 Biology Department, University of Crete, University Campus (Voutes), Heraklion 70013, Crete, Greece

Eagles of Palearctic: Study and Conservation II International Scientific and Practical Conference 7–10 September 2018 Park-Hotel Lake Aya, Katun village, Altai Kray, Russia

Aquila chrysaetos (6 subspecies)

nesting resident wintering

Aquila fasciata

Study area & Species  Geographical isolation: 5 millions years ago  Glacial period: 2,5 millions – 12.000 years ago  Human presence 8.000 years ago  350 Birds (90 breeding)  18 mammals  11 reptiles  150 land snails  Invertebrates (10-50% endemic)  1.600 plants

No territories Crete/ Greece

No pairs/ inds

Aquila chrysaetos

27/150

22-25/ 60-80

R

EN

Aquila fasciata

25/ 140

17-20/

EN

VU

Species

Europe

Greece

Aquila chrysaetos

Nest construction: Χ Egg-laying & incubation: 0 Egg-hatching & chick rearing: *

Aquila fasciata

Breeding & Foraging habitat

Diet

Nest site requirements

Aims     

To identify the nesting areas of the species on the island of Crete To investigate the factors that affect nest-site selection To construct predictive breeding distribution maps To delineate the potential suitable nesting habitat for the species To asses nesting habitat spatial overlap & compare nest-site variables

Methods 1. Locate all eagle nest sites (fieldwork) 2. Spatial mapping of eagle nest sites (GIS geodatabase) 3. Assessment of nest-site environmental variables (GLM logistic regression model) 4. Prediction maps (Species Distribution Model) 5.

Breeding habitat overlap (Quantum, SAGA - GIS software)

6. Univariate analysis & comparison of nest-site habitat variables (R software)    

Selection of environmental variables (biological meaning) Proper statistical analysis (autocorrelation) Model selection (stepwise procedure) SDM evaluation (MaxEnt)

“eagle” pixels = 400X400m grids containing an eagle nest VIFi= 1/(1-Ri2) Bioclimatic variables http://www.worldclim.org/bioclim

BIO1= Annual Mean Temperature BIO2= Mean Diurnal Range (Mean of monthly (max temp - min temp)) BIO3= Isothermality (BIO2/BIO7) (* 100) BIO4= Temperature Seasonality (s.d.*100) BIO5= Max Temperature of Warmest Month BIO6= Min Temperature of Coldest Month BIO7= Temperature Annual Range (BIO5-BIO6) BIO8= Mean Temperature of Wettest Quarter BIO9= Mean Temperature of Driest Quarter BIO10= Mean Temperature of Warmest Quarter BIO11= Mean Temperature of Coldest Quarter BIO12= Annual Precipitation (mm) BIO13= Precipitation of Wettest Month BIO14= Precipitation of Driest Month BIO15= Precipitation Seasonality (CV) BIO16= Precipitation of Wettest Quarter BIO17= Precipitation of Driest Quarter BIO18= Precipitation of Warmest Quarter BIO19= Precipitation of Coldest Quarter

Disturbance variables (DEM) dist_pop= distance from the nearest inhabited area dist_road= distance to the nearest road dist_urban= distance from the nearest urban settlement dist_water= distance from the nearest water body

Landscape variables (DEM & http://) altitude= elevation of nest site aspect= orientation of nest-site slope= % slope of the nesting cliff geo_index= rock type of nesting cliff hnv= % of “High Nature Value” farmland in “eagle” pixels landuse_index= land use type in “eagle” pixels livestock= No. of sheep & goat in “eagle” pixels

MaxEnt SDM Evaluation Area under ROC curve (AUC) ROC is Sensitivity by (1- Specificity)=(FPR)

True Positive Rate

1

AUC > 0.5 Higher Predictive Power AUC = 0.5 Random Chance AUC < 0.5 Worse than Random

0 0

False Positive Rate

Sensitivity= % of presences correctly predicted Specificity= % of absences correctly predicted

1

Results Aquila chrysaetos Coefficients:

Aquila fasciata Estimate

(Intercept)

-5.621e-01

-1.52

Slope

2.472e-02

Distance to urban areas Distance from water bodies Land use

Variable

Coefficients:

z value Pr(>|z|) 0.12772

Estimate z value Pr(>|z|)

(Intercept)

-0.313

-0.75

0.45429

2.81 0.00492 **

Altitude

-0.001

-2.42

0.0154 *

1.517e-04

1.95

0.0515.

Slope

0.034

2.89

-6.710e-04

-2.05

0.04002*

0.00386 **

Land use

-0.134

-2.13 0.03332 *

-2.042e-01

-2.78

0.0055**

Mean ± se

range

Wilcox test value (W)

P-value

Aquila chrysaetos

Aquila fasciata

Aquila Chrysaetos

Aquila Fasciata

715±44 m

372± 28 m

135-1387 m

71-912 m

428

<0.0001***

27.5±2.1

25.6±1.9

1-60

1-49

1089

0.60

Distance from urban areas

2±0.15 km

2.6± 0.24

0.4-11.7 km

0.4-10.8 km

866

0.03*

Distance from water

451±86 m

379±92m

0.1-3 km

0.1-4 km

1003

0.224

Altitude Slope

Aquila chrysaetos

AUC = 0.95

Altitude Slope Substrate Land use & HNV farmland Livestock Distance from roads (& humans) Mean Temp of Driest Quarter

Aquila fasciata

AUC = 0.93

Altitude Slope Land use Distance from roads Distance from water bodies Mean diurnal temp range

Aquila chrysaetos Aquila spp. breeding distribution

Aquila fasciata

Species

Km2

% Crete

A. chrysaetos

1282.6

15.5

A. fasciata

1900.6

23

Aquila spp.

647.2

7.8

Conclusions • • • • • • •

Both Aquila species select vertical cliffs for nesting within pastoral areas A. fasciata breeds at lower altitude & closer to the sea (prey species) A. chrysaetos selects inland mountains ( limestone substrate, livestock) Weather conditions – Rural practices (carrion, prey behaviour) Variable tolerance to human presence (territory abandonment) Habitat alteration & land use changes (agriculture, tourism, energy) Home range & space use (telemetry data)

Благодарю за ваше любезное внимание