Mrs Keyssa Laidi1, Mr Abdelkader Bouahmed1, Mrs Fazia Krouchi1, Mr Federico Vessella2, Mr Bartolomeo Schirone2, Mr Arezki Derridj1
1Mouloud Mammeri University Of Tizi Ouzou, Algeria, , Algeria, 2DAFNE, Tuscia University, Viterbo, Italy
Quercus afares is an endemic tree species of Algeria and Tunisia, occupying a restricted areas where Mediterranean climate rules. It’s considered as a stabilized hybrid originates from two genetically very distant oak species, Q. canariensis Willd. and Q. suber L.
In the context of species distribution models (SDMs), a machine learning (MaxEnt) was employed to predict the potential distribution of the species integrating 20 environmental variables. To test the model accuracy, we used the area under the curve (AUC) of receiver operating characteristic (ROC) method. A Jackknife test, implemented on MaxEnt allowed us to explore the most important environmental variables affecting the species distribution.
The obtained map showed a potential distribution of the species larger than its geographical distribution. Among treated predictors, the most important for the studied species distribution are: annual precipitation, mean temperature of warmest quarter and altitude. The calculated model has a good predictive power (AUC = 0.985).
At broad scale, this study constitutes a valuable tool for reforestation planning and ex-situ conservation of this endemic species in both countries (Algeria and Tunisia). Integrate more variables into the modeling process could improve predictions.
Mr BOUAHMED Abdelkader is an assistant professor at Moumloud Mammeri University (Algeria).
After baccalaureate, he continued his studies at the University of Djelfa in Algeria where he obtained his degrees of engineering and Magister. He continues his PhD studies at the Mouloud Mammeri University in Tizi Ouzou. In his research, he is interested in ecological modeling and the response of species to climate change.
Mr BOUAHMED Abdelkader participate currently to a national research projet (PRFU) intiteled “Appui de la modélisation et de la cartographie à l’évaluation de la dynamique des essences forestières”.