Australia Awards in Indonesia

The Australia Awards are prestigious, transformational scholarships and short courses offered to emerging leaders for study, research and professional development in Australia

02 May 2016

Landscape Scale Carbon Stock Assessment of Tropical Peat Swamp Forests Using an Integrated Field Measurement and Remote Sensing Technique: A Case Study in PT Diamond Raya Timber, Rokan Hilir District, Riau Province, Indonesia

Nunung Puji Nugroho, 2013
The Australian National University

Abstract
Indonesia’s tropical peatland ecosystems constitute one of the largest terrestrial carbon pools in the world. However, anthropogenic and natural disturbances (e.g. logging, drainage, fire, and conversion to other land uses) can easily convert the peatland ecosystems’ status from carbon sinks and stores to carbon sources. Huge amounts of carbon are released to the atmosphere as carbon dioxide (CO2) through peat decomposition and fire. When these emissions are taken into account, Indonesia became one of the largest CO2 emitter worldwide. Despite their significant role in the global climate change, tropical peatland ecosystems (particularly peat swamp forests) are inadequately studied. Thus, the main objectives of this research are to fill the gap in knowledge regarding peat swamp forests and to develop a practical framework for assessing carbon stock at landscape level in peat swamp forests under an active logging company. Research for this thesis has been conducted in the concession forest of PT Diamond Raya Timber (PT DRT) in Rokan Hilir District, Riau Province, Indonesia.

The development of site-specific allometric equations for above-ground biomass and carbon (AGB and AGC) of peat swamp forest were carried out based on 51 and 31 destructively sampled trees, respectively. The carbon fraction used to develop AGC equations were analysed based on 362 subsamples from 31 selected sampled trees using the Walkley-Black method. All the developed AGB and AGC equations have coefficient of determination (R2) of greater than 95%. The lowestR2 values resulted from the simplest models which used only diameter at breast height (DBH) as a predictor. Adding total tree height (H) and wood density (ρ) as a single or a combined variable improved the model predictions. Model 5, which used DBH, H and ρ as predictive variables, provided best performance when estimating the AGB and AGC of the study area. Hence, as long as reliable data are available as input, Model 5 is recommended. The allometric carbon equations developed in this study can be used directly to estimate the AGC of peat swamp forests, which is particularly relevant to carbon accounting under global climate change agreement mechanisms such as REDD+. A performance comparison of the developed allometric equations to the several published equations was conducted to provide a scientific justification on the application of the generalized or other site-specific allometric equations to estimate AGB of trees in peat swamp forests. In general, the developed equations outperformed the published equations used commonly in biomass (carbon) studies.  However, the generalized allometric equations developed by Brown et al. (1989) and Brown (1997) can be applied with the developed site-specific allometric equations, especially for larger DBH classes. This is based on the consideration that these generalized equations were developed using a larger number of samples (> 100 trees) across a wider range of DBH.

The site-specific H-models were developed based on the knowledge that adding H can improve the model performance (from Chapter 4). The models were constructed based on 286 sampled trees using linearized and non-linear functions. The results suggested that the non-linear models outperformed the linearized models based on the statistical parameters and the biological criteria with the FI values of ~91% and the standard error of 3.6 m. All the non-linear models performed equally well with only minor differences. However, Model 7, which was developed based on the modified logistic function, is recommended for estimating H of trees in peat swamp forests considering its statistical performance and its property of passing the point diameter-height (0, 1.3).

The peat swamp forests in the study site have been actively logged by the commercial company (PT DRT) through selective logging. The impacts of this logging practice on forest community composition, structure and biomass/carbon were assessed based on 150 measurement plots. The modified logistic H-D model (Chapter 5) and the AGB and AGC allometric equations with a combined variable of D2H (Chapter 4) were employed to estimate the and the AGB/AGC for all trees enumerated in the plots. The stand attributes comparison between logged and unlogged plots was conducted using non-parametric tests. This study indicated that the species composition, structure and biomass between selectively logged and unlogged forest plots were statistically significantly different, but there are indications that the differences decrease with time since harvest. Selective logging caused more severe impacts on CC and N ha-1 than other stand attributes as indicated by a reduction in the mean values of these attributes of ~30%. Selective logging was expected to cause long-term (more than 20 years) impacts on the S, CC, N ha-1, and BA ha-1. However, the , AGB ha-1 and AGC ha-1in logged forest plots recovered relatively faster after 10 years of logging.

The application of site-specific allometric equations of AGB and were also used in assessing the applicability of two published backscatter-biomass algorithms to estimate the AGB at landscape level in the study site. The selected algorithm was then used in a combination with ALOS-PALSAR HV polarization data. The belowground allometric equation and the root to shoot ratio were adopted to estimate the BGB based on the AGB. The carbon fraction of 0.473 obtained in this study, which was slightly lower than the generic value of 0.5, was used to convert biomass to carbon. The results of the study indicated that the biomass retrieval algorithm developed from the mosaic tropical forests in Malaysia performed better than the one developed from the peat swamp forests in Central Kalimantan of Indonesia. The two BGB estimation methods resulted in relatively similar BGB estimates. The AGB and subsequently the TLTC maps depicted logically the spatial patterns of biomass and carbon across the landscape thus is useful for management purposes, particularly in allocating the resources efficiently for further ground-based investigations (e.g. in supporting the REDD+). The use of ALOS-PALSAR data for biomass estimation is limited by saturation of the backscatter above relatively low biomass levels, with these being most noticeable in moist environments such as those typical of Indonesia. The landscape level AGB ranged from 12.5 to 390.6 Mg DM ha-1 with an average of 213.4 Mg DM ha-1. For the whole concession area, the total AGC, BGC and TLTC were 146, 33 and 179 Mt C.

Share this article on:

Related Article


Back to Top