Title: Comparing forest sites classifications using two different satellite images and ground measurements in Eastern Turkey

Authors: Ali İhsan Kadıoğulları; Alkan Günlü; Emin Zeki Başkent

Addresses: Faculty of Forestry, Karadeniz Technical University, 61080, Trabzon, Turkey ' Faculty of Forestry, Çankırı Karatekin University, 18200, Çankırı, Turkey ' Faculty of Forestry, Karadeniz Technical University, 61080, Trabzon, Turkey

Abstract: This study aims to determine forest sites using direct, indirect and remote sensing (RS) methods (Landsat 7 ETM+ and IKONOS images) for Ardahan-Yalnzçam Forest of 12149.23 ha with 63 sample plots (1637,8 ha). Four different forest site classes of dry, moderate fresh, fresh and moist were determined in direct method whereas three site index (SI) classes; good (SI = I − II), medium (SI = III) and low (SI = IV − V) were determined in indirect method. Forest sites were also estimated with RS method using supervised classification from Landsat 7 ETM+ and IKONOS images respectively with a 0.77 kappa statistic value and 84.4% accuracy assessments; 0.88 kappa statistic value and 90% accuracy assessments. Spatial analysis functions of geographic information systems (GIS) indicated some important differences between the methods were identified in ground measurement area. The forest sites determined with indirect method indicate that SI I − II cover 213.27 ha, III 600.57 ha and IV − V 502.93 ha and, treeless areas and degraded forest areas 328.67 ha. Whereas direct method defines dry site of 597.35 ha, moist site of 49.86 ha fresh site of 266.06 ha and moderate fresh site of 724.5 ha. Landsat 7 ETM+ classifies dry site of 353.21 ha, moist site of 261.03 ha, fresh site of 479.93 ha and moderate fresh site of 543.6 ha; IKONOS classifies dry site of 284.25 ha, moist site of 307.45 ha, fresh site of 333.01 ha and moderate fresh site of 713.06 ha.

Keywords: forest sites; site classification; geographic information systems; GIS; site index; Landsat; IKONOS satellite images; Turkey; direct sensing; indirect sensing; remote sensing.

DOI: 10.1504/IJGW.2014.058756

International Journal of Global Warming, 2014 Vol.6 No.1, pp.79 - 98

Received: 05 Feb 2013
Accepted: 20 Mar 2013
Published online: 02 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article