Detecting tropical forest biomass dynamics from repeated airborne Lidar measurements
Abstract. Reducing uncertainty of terrestrial carbon cycle depends strongly on the accurate estimation of changes of global forest carbon stock. However, this is a challenging problem from either ground surveys or remote sensing techniques in tropical forests. Here, we examine the feasibility of estimating changes of tropical forest biomass from two airborne Lidar measurements acquired about 10 yr apart over Barro Colorado Island (BCI), Panama from high and medium resolution airborne sensors. The estimation is calibrated with the forest inventory data over 50 ha that was surveyed every 5 yr during the study period. We estimated the aboveground forest biomass and its uncertainty for each time period at different spatial scales (0.04, 0.25, 1.0 ha) and developed a linear regression model between four Lidar height metrics and the aboveground biomass. The uncertainty associated with estimating biomass changes from both ground and Lidar data was quantified by propagating measurement and prediction errors across spatial scales. Errors associated with both the mean biomass stock and mean biomass change declined with increasing spatial scales. Biomass changes derived from Lidar and ground estimates were largely (36 out 50 plots) in the same direction at the spatial scale of 1 ha. Lidar estimation of biomass was accurate at the 1 ha scale (R2 = 0.7 and RMSEmean = 28.6 Mg ha−1). However, to predict biomass changes, errors became comparable to ground estimates only at about 10-ha or more. Our results indicate that the 50-ha BCI plot lost a~significant amount of biomass (−0.8 ± 2.2 Mg ha−1 yr−1) over the past decade (2000–2010). Over the entire island and during the same period, mean AGB change is −0.4 ± 3.7 Mg ha−1 yr−1. Old growth forests lost biomass (−0.7 ± 3.5 Mg ha−1 yr−1), whereas the secondary forests gained biomass (+0.4 ± 3.4 Mg ha−1 yr−1). Our analysis demonstrates that repeated Lidar surveys, even with two different sensors, is able to estimate biomass changes in old-growth tropical forests at landscape scales (>10 ha).