| dc.description.abstract | Perum Perhutani is one of the state companies that has a long history in managing forest area in the most populated island, Java. However, in line with the population and economical growths, the company is frequently facing problem on forest protection, particularly on wood stealing, forest encroachment, forest fire etc. Thus, the company needs to establish a tool that may support forest management through establishment of forest inventory. In the forest planning, forest inventory had been playing an important role. Forest inventory is usually performed to acquire data and information of the forest resources potency through three methods, that are terrestrial, remote sensing and combination of the mentioned two methods (e.g., two-phase or double sampling technique). Generally, double sampling technique employs remote sensing data obtained by air mapping or satellite imaging. Among these two image sources, satellite images had been more preferable since they have shorter temporal resolution, digital format, better spatial resolution and to some extent are cheaper. To facilitate forest survey, forest inventory usually uses aid tools such as tree volume table or stand volume table. In this research the author focused on the establishment of stand volume table of teak forest using high resolution satellite images. The model is then referred to as satellite volume table. The objective of this study is to establish stand volume of teak forest (Tectona grandis Linn.f) using Ikonos images in KPH Jatirogo Perum Perhutani Unit II, East Java. Ikonos data used were recorded on 9 October 2006. The softwares used were ArcView GIS 3.2, Erdas Imagine 9.1 and Minitab 14. For ground ground surveys, the equipments used were GPS, brunton compass, phi-band, haga hypsometer and digital camera. The study encompass following steps, i.e., images pre-processing, ground sampling and data analysis. The study results showthat the crown diameter means derived from Ikonos image (Dc) have high correlation with clear bole tree height for estimating (merchantable stand volume (Vbc), having correlation coefficient of 0,824. Based on statistical analysis, model evaluation, and model simplicity, the best model selected for predicting stand potency is Vbc = - 19,2 + 25,2 Dc, with coefficient of determination of 83,2%. Using the double sampling technique, the stand volume per unit area are approximately 136,331 m3/ha, having variance of the mean of 164,918 (m3/ha)2. At 95% of confidence level, the stand volumes are ranging from 109,979 m3/ha to 162,683 m3/ha with 19,33% of sampling error. The study also shows that there is a high consistency between the crown diameter measured from ground and Ikonos having coefficient of determination of 95,2%. Keywords: stand potency, double sampling technique, ikonos | id |
