LAUSR

ABSTRACT Forests and afforestation have significant importance, given the present course of World’s climate resulting from our insensitive behaviours, policies, ambitions, etc. Turkish Forest Service has departments, sole purposes of which are to combat desertification through afforestation, flood control, and watershed rehabilitation works. Majority of the works is carried out in the semi-arid central part of the country, which is known as the Irano-Turanian phytogeographic region. Hydroelectric dam projects, which have extensively been conducted to produce energy and to harness the rivers when needed, have been at the centre of these efforts in the region. Afforestation works, which have long been used to safeguard dam reservoirs, have always been associated with such projects because they are implemented even prior to water deposition. They along with the country’s managed forests are looked after as the natural forest cover. Improvements in remote sensing at this point enabled the agencies to diversify their efforts in monitoring the natural resources and strengthening the decision makers’ hand in management preferences in both global and regional scales. There are various data produced through the classification of satellite imagery representing the existence or the percentage of forests in any location on Earth. In this study, Global Phased Arrayed L-band Synthetic Aperture Radar PALSAR-2/PALSAR mosaic and forest/non-forest maps of Japan Aerospace Exploration Agency (JAXA) and Global Forest Cover Change (GFCC) thematic image tiles of National Aeronautics and Space Administration (NASA), which are two of the widely accepted such data, were compared to one another and to our control results produced by a pixel-based classification method to assess the performance of afforestation works around seven major hydroelectric dam reservoirs situated within the central part of Turkey. The results showed that Global PALSAR-2/PALSAR forest/non-forest maps were somewhat capturing the trend, but overly exaggerating the ground facts up to 1064%. GFCC epochs, on the other hand, did not yield sound results in this semi-arid part of Turkey. Latter data could not even capture the successfully afforested sites.