Assessing the effect of ensemble learning algorithms and validation approach on estimating forest aboveground biomass: A case study of natural secondary forest in Northeast China

Landsat 8 OLI imagery (path/row: 117/28) processed to L1T level was acquired on September 13, 2015 (LC81170282015256LGN01). The preprocessing of Landsat 8 OLI imagery data includes three steps: radiometric calibration, atmospheric correction and topographic correction. The Fast Line-of-Sight Atmospheric Analysis of Spectral Hypercube (FLAASH) radiative transfer model was employed for atmospheric correction (Safari et al. 2017). The topographic correction was performed using the well-known Sun Canopy Sensor + C correction (SCS + C) approach provided by extension tool of “Topographic Correction V5.3 4 S1”.