Estimating above-ground biomass using landsat 8 oli satellite image in pure cedar stands (a case study in Kaş District Directorate of Forestry)

The aim of this master's thesis is estimating the aboveground biomass in pure cedar stands within the planning boundaries of the Kaş Forest Management Directorate, which operates under the Antalya Forest Regional Directorate, using Landsat 8 OLI satellite image. In this context, 618 systematically located sample plots, determined through the update of forest management plans, were utilized. These sample plots, whose coordinates were determined by ground measurements, were transferred to Landsat 8 OLI satellite image. Subsequently, band reflectance, vegetation index values and texture values of the sample areas were calculated. Multiple regression analysis was used to examine the relationships between band reflectance, texture and vegetation indices features and their modeling with above-ground biomass quantities was demonstrated. The analysis of the findings revealed that the regression equation using Band2, Band5, and Band7 as independent variables had a relationship with aboveground biomass (R²=0,186- Sy,x =78,5729), while the regression equation using NBR, VIS123, SARVI, and ND32 as independent variables showed a relationship with aboveground biomass (R²=0,181- Sy,x=77.5853). Additionally, strong relationships were identified between aboveground biomass and regression models using texture features as independent variables, including Band2_3x3_COR, Band3_5x5_M, Band4_7x7_CON, Band5_9x9_HOM, Band6_9x9_DIS, and Band7_9x9_ENT, yielding an adjusted R²= 0.453 and Sy,x=66.9189. The studies reviewed demonstrate that texture features provide results with high accuracy in aboveground biomass estimation.