Akademik Veri Yönetim Sistemi
Tezin Türü: Doktora
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Fen Edebiyat Fakültesi, Biyolojik Bilimler Bölümü, Türkiye
Tezin Onay Tarihi: 2007
Tezin Dili: İngilizce
Öğrenci: Uğur Zeydanlı
Danışman: CEMAL CAN BİLGİN
Özet:Throughout the past century, vegetation scientists have been studying plant communities to develop classification standards for global mapping purposes. In Turkey, although there are several vegetation classification schemes in use by botanists, foresters or geographers, none is widely accepted by all for vegetation mapping. In this study, a hierarchical, ecologically meaningful, physiognomic-floristic classification system was adapted and developed for terrestrial vegetation in Turkey. The system has eight hierarchical levels, with the alliance and the association as its lowest two floristic levels, and the classes are designed to be easily detected in the field or indirectly through remote sensing. To test of its utility, a supervised vegetation classification of the whole Mediterranean Region of Turkey was carried out using Landsat ETM images. The accuracy of the classification ranged between 55% and 69% depending on the level of hierarchy. A further test for its ecological utility was carried out by comparing identified vegetation classes with breeding bird communities derived from data gathered through field observations at 193 ecological communities at 83 sites. The proposed classification scheme has proven to be reasonably accurate when widely available satellite imagery is used and ecologically meaningful as shown by a high concordance with observed bird community patterns. It is suggested that this new system can be safely applied to other regions of Turkey for purposes of vegetation mapping, species habitat modeling, and nature conservation if proper image set and ancillary data is used.