Exploring Bamboo Forest Aboveground Biomass Estimation Using Sentinel-2 Data

Bamboo forests,due to rapid growth and short harvest rotation,play an important role in carbon cycling and local economic development.Accurate estimation of bamboo forest aboveground biomass(AGB)has garnered increasing attention during the past two decades.However,remote sensing–based AGB estimation for bamboo forests is challenging due to poor understanding of the mechanisms between bamboo forest growth characteristics and remote sensing data.The objective of this research is to examine the remote sensing characteristics of on-year and off-year bamboo forests at different dates and their AGB estimation performance.This research used multiple Sentinel-2 data to explore AGB estimation of bamboo forests in Zhejiang Province,China,by taking into account the unique characteristics of on-year and off-year bamboo forest growth features.Combining field survey data and Sentinel-2 spectral responses(spectral bands and vegetation indices)and textural images,random forest was used to identify key variables for AGB estimation.This study mainly discusses from the following aspects:1.Extract off-year and on-year bamboo distribution.First,Preprocess the Sentinel-2 data then the seasonal index SBI was constructed according to the spectrum in different periods of bamboo forest,combined with matched filter and use decision tree to classify the distribution information of bamboo forest in the reseach area.2.Extract the estimated variables of biomass of sentinel-2 data as independent variables,and to take the biomass data of bamboo in sample plots as dependent variables.Based on non-stratification and stratification to exploration of the relationship between remote sensing variables and biomass,two modeling methods random forest(RF)were adopted.After the accuracy evaluation of the model,the model with the highest accuracy was selected to make the distribution map of bamboo forest biomass.The main conclusions are as follows:1.The on-year and off-year bamboo forests have considerably different spectral signatures,especially in the wavelengths between Red Edge 2 and NIR2(740 nm – 865 nm),making it possible to separate on-year and off-year bamboo forests;2.On-year bamboo forests have similar spectral signatures although AGB increases from as small as40 Mg/ha to as high as 90 Mg/ha,implying that optical sensor data cannot effectively model AGB;3.Off-year bamboo AGB have significant relationship with red and SWIR in April image,or red edge2 in July,but the AGB saturation problem makes its poor estimation accuracy;4.For bamboo forest AGB,stratification of on-year and off-year bamboo forests has limited effects in improving overall AGB estimation,and non-stratification using April image is recommended.However,stratification indeed considerably improved off-year bamboo AGB estimation,but cannot improve the on-year bamboo AGB estimation;5.The Sentinel-2 data cannot solve the bamboo AGB data saturation(AGB greater than 70 Mg/ha)problem,the similar problem as other optical sensor data such as Landsat.More research should beexplored in the future to integrate multiple source data – remotely sensed data(e.g.,lidar,optical sensor data)and ancillary data(soil,topography),in AGB modeling for improve bamboo forest AGB estimation.