University of Twente Student Theses
Distribution modelling of the Short-toed Eagle in relation to potential food availability
Niamir, Aidin (2009) Distribution modelling of the Short-toed Eagle in relation to potential food availability.
PDF
3MB |
Abstract: | The main objectives of this research are (1) to predict and explain the distribution of the Short-toed Eagle in Malaga province, Southern Spain, (2) to generate maps of preferred hunting sites and potential available prey for the Eagle and (3) to infer the contribution of prey availability in the Short-toed Eagle distribution model. The spatial distribution of the prey snake species were modelled using multivariate statistical techniques and GIS (Geographic Information System). To select the explanatory environmental variables affecting the species distribution and to find out where the suitable habitat for the selected snake species in Malaga province are, predictive distribution models were created using logistic regression and the environmental favourability function, absence/presence data of the species and a set of independent variables related to bioclimatic, topographic and anthropogenic conditions. A map showing Eagle’s preferred hunting areas was generated by ranking the Corine land cover map using expert knowledge and interviews with local ornithologists. A potential food availability map was generated based on snake prey species distribution, combined with the Short-toed Eagle preferred hunting areas. To assess the predictor variables affecting the Short-toed Eagle distribution in Malaga province, predictive distribution models were created using Maximum Entropy functions, presence data of the species, potential food availability and a set of 7 independent variables related to climatic conditions, topography and NDVI (Normalized Difference Vegetation Index). Presence data (observed nest-locations) were collected during fieldwork in September and October 2008. To indentify a model with the fewest predictors that explained the data satisfactorily, five variables model were selected; minimum temperature in the wettest quarter, NDVI for mid August, precipitation in September, slope and the southness of aspect. This research also revealed that inclusion of the potential available prey in the distribution models did not result in significantly increased AUC (Area Under Curve) compared to the food excluded models. The final predictive model satisfactorily describes the Short-toed Eagle distribution in the Malaga province. This research suggests re-testing the hypothesis using hyper spatio-temporal species distribution data and/or new potential food availability indices. Key words: Short-toed Eagle, Circaetus gallicus, spatial predictive models, Maxent, environmental favourability function, potential food availability, Malpolon monspessulanus, Hemorrhois hippocrepis, Rhinechis scalaris, Malaga province |
Item Type: | Essay (Master) |
Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
Programme: | Geoinformation Science and Earth Observation MSc (75014) |
Link to this item: | https://purl.utwente.nl/essays/92725 |
Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Repository Staff Only: item control page