Forest Aboveground Biomass Estimation Using Lidar And Rapideye Data

Forest biomass is closely related to human activities and has strong capacity in forest ecosystem carbon sequestration.The research on forest aboveground biomass has significance for ecological environment and climate change.Remote sensing technology with its unique characteristics in data collectting,large area coverage and digital format,has become an important tool for forest biomass estimation.Although optical and radar images are often used to biomass estimation,their spectral saturation problems,caused by complex structure and large biomass density,result in poor estimation precision.Lidar is an active remote sensing technology,which has a rapid development in recent years.By penetrating the forest vegetation canopy,laser pulse can obtain the three-dimensional structure,solve the problem of data saturation effectively and improve data estimation accuracy.Previous researches have explored the potential to integrate lidar and optical data in aboveground biomass(AGB)estimation,but how different data sources,vegetation types,and modeling algorithms influence AGB estimation is poorly understood.This research takes the Tomé-A?u as study area,conducts a comparative analysis of different data sources(i.e.,lidar,RapidEye and their combination)and modeling approaches(i.e.,linear,nonlinear,random forest,and support vector regression – SVR)in improving AGB estimation under stratification and non-stratification conditions.RapidEye-based spectral responses and textures,lidar-derived metrics,and their combination were used to develop AGB estimation models.The results indicated that(1)overall,RapidEye data are not suitable for AGB estimation,but when AGB falls within 50-150 Mg/ha,SVR based on stratification of vegetation types(e.g.,secondary forests and agroforestry)provided good AGB estimation with relative root mean squared error of 35.0%.(2)Lidar data provided more stable and better estimations than RapidEye data no matter which modeling approaches were used.Stratification of vegetation types cannot improve estimation accuracy.(3)The combination of lidar and RapidEye data cannot provide better performance than lidar data alone.(4)AGB ranges affect the selection ofthe best AGB models,and a combination of different estimation resultsfrom the best model for each AGB range can improve AGB estimation.(5)This research implies that an optimal procedure for AGB estimation for a specific study exists,depending on the careful selection of data sources,modeling algorithms,forest types,and AGB ranges.