ESTIMATION TREE DENSITY AS OBJECT-BASED IN ARID AND SEMI-ARID REGIONS USING ALOS Hadi Fadaei* Rikie Suzuki and Ram Avtar Japan Agency for Marine-Earth Science and Technology (JAMSTEC) Research Institute for Global Change (RIGC), 3173-25 Showa-machi, Kanazawa-ku, Yokohama-shi, Kanagawa-ken 236-0001 Japan Email: fadaei@jamstec.go.jp, rikie@jamstec.go.jp, avtar@jamstec.go.jp KEY WORDS: Tree density, object-based, arid and semi-arid, ALOS satellite and support vector machine (SVM) ABSTRACT: Object-based approaches, which operate at the scale of real-world objects rather than pixels, offer a means of analyzing earth observation data in a realistic context and integrating associated ancillary information to support real-world applications. Arid and semi-arid lands, including desert, scrubland, grassland and savanna, cover about 40%, or an estimated 58.5 million km2, of the terrestrial surface. Iran is located in the mid-latitude belt of arid and semi-arid regions cover more than 60% of the country. In this research three case studies were investigated as object-based that are locate in the arid and semi-arid regions. In this research two sensor of ALOS satellite including AVNIR-2 that has multispectral band with 10m resolution on the ground and PRISM that has panchromatic band with 2.5m resolution on the ground have been used. The objective of this research is investigated tree density as object-based using image segmentation and support vector machine (SVM) that is supervised classification with four kernel types including linear, polynomial, radial basis function (RBF), and sigmoid have been applied. Then the results of object-based with ground sampling data on each case study have been evaluated. Pan sharpening image from high-resolution and low resolution has been created. On image pan sharpening sampling plot and sub-sampling plot incorporating it for analyzing were selected. The result of object based for tree density for each case study should be seen and also evaluated with ground sampling data. 1. INTRODUCTION Biodiversity conservation has become an increasingly important issue in forest management, with forest managers now having to include biodiversity considerations within existing management plans (Ozdemir et al., 2008). The arid and semi-arid areas of northeast Iran consist of about 3.4 million ha populated by two main tree species. One is the broad-leaved Pistacia vera and the other is the conifer Juniperus excelsa subsp. Polycarpos, which Iranians know as Persian juniper and the third case study that is located in other vegetation zone but has same climate condition as above species called Cupressus sempervirence var. horzontalis belong to Cupressus family located in the north of Iran. The morphologically of this spices is typically 20 - 30m high, with the crown diameter of 2 - 3m and fruit of this species is wooden with spherical shape. Fisher and Andrew (1995) investigated the status and ecology of Juniperus excelsa subsp. Polycarpos woodland in the northern mountains of Oman. Juniperus excelsa M.-Bieb. subsp. Polycarpos (K. Koch) Takhtajan is found from Turkey and Afghanistan eastward and southward to Iran (Fisher and Andrew, 1995). This species generally grows at elevations 1500 to 2900 m above sea level, but is sometimes found in other ranges. J. excelsa subsp. Polycarpos woodland, and even small increases in climatic stress could imperil the present status of these woodlands (Fisher and Andrew, 1995). In northeast Iran, junipers form open woodlands, with a maximum tree density of approximately 150 trees per hectare (pers. obs., Andrew et al., 1996). Pistachio (Pistacia vera), a deciduous-broadleaved species, which is one of the ecologically most important native species of Iran (Yıldız et al., 1998; Safari and Alizadeh, 2007; Miraliakbari and Shahidi, 2008). Pistachio is thought to have been cultivated for 3000–4000 years in Iran. Currently, Iran is the world’s largest producer of pistachio, followed by the United States, where most pistachio production occurs in California. In 2004, Iran and the United States were responsible for 44% and 13% of global pistachio production, respectively (Razavi, 2006). Pistachios are one of the most important non-oil exports from Iran (Fadaei et al., 2007). Natural stands of pistachio are not only environmentally important, but are also genetically important as seed storage for pistachio production in orchards. C. sempervirens is not of commercial importance as a timber tree, although its wood is much utilized locally (PhilIips, 1948). Yellowish or pale brown in hue and close-grained, it has numerous, fine, inconspicuous medullar rays but rather more distinct growth rings (Elwes and Henry, 1906-1913; Dallimore and Jackson, 1966). It gives off a penetrating odour, which, although agreeable to human beings, is obnoxious to insects, and is reputed to keep moths away (Brown and Laurie, 1968). Light in weight, it is moderately hard and exceedingly durable, and it lasts indefinitely under water. However, hot, dry weather conditions in the region can make forest inventory work on the ground difficult. Moreover, ground surveys require much time, labor, and money, even when using GPS equipment. In contrast, the use of remote-sensing data for performing forest inventories in arid and semi-arid areas is more cost-effective, less time- consuming, and less labor-intensive (Fadaei and Kolahi, 2008). Presently, natural resource managers are beginning to use remotely sensed satellite imagery to calculate the density, species, and structure states of forests (Moskal et al., 2003). 2. MATERIAL AND METHOD 2.1 Study area This research was carried out in the three case studies, both of them located in the northeast vegetation zone of Iran that called the Irano-Touranian zone, and the third case study is Zarbin Proceedings of the 4th GEOBIA, May 7-9, 2012 - Rio de Janeiro - Brazil. p.668 668