Multi-source Remote Sensing Quantification Of Forest Biomass Based On SAR Polarization Decomposition And TM Data

Forest biomass is the result of accumulation in the long-term production and metabolism of forest ecosystem,and it is the energy basis and material source of forest ecosystem operation.It includes the biomass of trees and the biomass of undergrowth layer,which includes the total weight of roots,stems,leaves,flowers,seeds and litter.Usually expressed in terms of dry matter mass or energy accumulated per unit area or per unit time.Influenced by photosynthesis,respiration,death,harvest and human activities,it is a comprehensive result of forest succession,human activities,natural disturbance,climate change and air pollution,and is an important index to evaluate the structure and function of forest ecosystem.There are many methods to estimate forest biomass.This paper USES multi-source remote sensing data to estimate forest biomass on a large regional scale,which is advanced and practical.The accurate estimation of forest biomass is helpful for people to monitor the growth state of forest and grasp the change law of forest,which provides theoretical basis for rational protection,management,utilization of forest resources and creation of stable and high yield forest ecosystem.Regional scale accurate forest biomass estimation has great significance in understanding current forest status and scientific forest management.This study aimed to quantify regional scale forest biomass through the polarimetric SAR and LANDSAT 5 TM.Firstly,SAR data was polarized by polarization decomposition.Then 51 parameters from polarization decomposition parameters and 6 TM bands were used as predictor variables with forest biomass W as response variables.Two methods were implemented for model construction:1)Stepwise regression.The final model includes two variables with R2 of 0.534,prediction accuracy of 67.51%and RMSE of 43.21 t/ha;2)Optimal subset method.Bootstrap was applied to select 5 parameters out of 9 to build the model.We used cross-validation for model validation.The final model has R2 of 0.7682,prediction accuracy of 88.32%,prediction RMSE as 14.98 t/ha,test accuracy of 86.21%,test RMSE of 19.14 t/ha and CP index of 5.2495.The total biomass in Chicheng is 256.5045 t/ha.We used optimal subset method to build forest biomass estimation model and acquired forest biomass distribution map.The results showed that C band polarimetric SAR and optical LANDSAT5 TM data can get accurate estimation of forest biomass.