Abstract
Background: Extensive forest inventory data is available from commercial timber companies. For this study, over 20,000 plots were compiled for North, East and West Kalimantan provinces, with more than 17,000 of these exceeding our quality assurance tests. This study aimed to: (1) explore the potential use of existing permanent sample plots and forest inventory data established and measured by timber concessions; (2) assess uncertainties of aboveground biomass (AGB) estimates using various allometric models; (3) analyse the dynamics of AGB in logged-over dipterocarp forests; (4) analyse AGB stocks and emission factors in tropical dipterocarp ecosystems. Methods: Two types of forest monitoring datasets measured by timber companies in Indonesia were compiled and assessed in this study: permanent sample plots (PSPs) (24 1-ha plots), and the overall periodic timber inventory (OPTI) (17,301 plots). We compared various allometric equations for estimating AGB of the plots and developed a simple AGB equation using basal area (BA) as predictor. We further evaluated the AGB growth and mortality using the PSP plots. Results: We found that the model using only tree diameter (D) as a predictor variable tended to be unbiased when aggregating the estimates at larger plots. We also found that BA per hectare could explain the variation of AGB at plot level (adjusted r2 = 0.911; root mean square error [RMSE]: 27.8). We overlaid the OPTI plot with the land cover map and estimated the mean AGB of the associated land cover classes. The mean AGB of primary dryland forest, secondary dryland forest and bush classes were 281.1 + 4.0 Mg/ha, 231.5 + 1.7 Mg/ha and 179.0 + 5.0 Mg/ha, respectively. Nine years after logging, the mean AGB is still lower than the mean AGB two years after logging. The growth rate (2.5%) was still lower than the mortality rate (3.1%), and recruitment (0.2%) did not occur until seven years after logging. Conclusions: The results of this study suggest that the existing forest monitoring data should be incorporated into the carbon accounting system at district, province and national level to improve the estimation of forest biomass and emission factors related to forest degradation and deforestation. However, there is a need for data quality assessment prior the analysis and a standardised platform for nation-wide forest inventory database is therefore required.
Spatial distribution of forest aboveground biomass estimated from remote sensing and forest inventory data in New England, USA
