Estimation of Wood Production Spatial Distribution in the High Forest on Igman Mountain

Abstract

UDK: 630*52/*56:528.8(234.422 Igman)
 Information about quantitative and qualitative forest attributes are the base for successful forest planning and management. Forest inventories collect number of data used for different estimations from large (management unit level) to small (forest stand) scales. Then, control sampling has to be done in order to confirm regularity of terrestrial work. Such sample becomes data source too. Recent approach for forest characterization includes all available information as sources for additional non-standard insight. Here were used available data about wood volume and increment from control sample for high forest on mountain Igman. Also, recent Landsat TM image from vegetation period was available and used in this research. Here is applied k nearest neighbor’s estimation method. Five nearest neighbors and Euclidian distance is chosen for estimation and mapping. Biases for all forest attributes were non-significant. Obtain results show non significant differences between means and observed and estimated distributions of wood volume and increment. It is estimated higher mean wood volume and increment of broadleaves while means for conifers and totals are lower. That higher wood volume and increment is estimated in all diameter classes for broadleaves while lower quantities are estimated for conifers. Spatial mapping presents distribution of wood volume and increment respecting variability of vegetation in high forest on Igman.