Small Footprint Full-Waveform Metrics Contribution to the Prediction of Biomass in Tropical Forests

摘要: We tested metrics from full-waveform (FW) LiDAR (light detection and ranging) as predictors for forest basal area (BA) aboveground biomass (AGB), in a tropical moist forest. Three levels of are tested: (i) peak-level, based on each return echo; (ii) pulse-level, the whole signal emitted pulse; (iii) plot-level, simulating large footprint dataset. Several have significant correlation, with two predictors, found by stepwise regression, particular: median distribution height above ground (nZmedian) fifth percentile total pulse intensity (i_tot5th). The former contained most information explained 58% 62% variance AGB BA values; regression left us four respectively, explaining 65% 79% variance. For BA, were standard deviation, (i_totstdDev, i_totmedian i_tot5th) nZmedian, whereas AGB, only last used. plot-based metric showed that echo count (HOMTC) performs best, very similar results expected. Cross-validation allowed analysis residuals model robustness. discuss our considering specific case scenario complex structure high degree variability terms biomass.