210
N. Novita et al.
Soil Carbon
Peat depth and sample cores were extracted from three sampling points near the
center of each circular plot. We measured peat depth by forcing a steel open-faced peat
auger downward through peat horizon to the organic-mineral transitional horizon. An
extension handle was used if peat depth was greater than 1 m. Measurements of soil
bulk density and C concentration at various peat depths were taken to determine the
soil C stocks (Donato et al. 2011; Kauffman et al. 2011). The core was systematically
divided into the depth intervals of 0–15, 15–30, 30–50, 50–100, and >100 cm (if
mineral layers were not detected before 100 cm depth). Samples of a known volume
were carefully placed in whirlpak bags and transported to the Bogor Agricultural
University for laboratory analysis. Soil samples were dried to a constant mass at
60 °C and weighed to obtain soil bulk density. The organic carbon content of soil
samples was determined via the induction furnace method using a carbon–nitrogen
analyzer (LECO Corporation, St. Joseph, Michigan, USA). The induction furnace
with an elemental carbon analyzer is considered the most reliable method compared to
other C determination methods, e.g., wet combustion and loss of ignition (Kauffman
and Donato 2012) in tropical peat soils (Farmer et al. 2014). The soil carbon stocks
were calculated as the products of bulk density and carbon concentration combined
with plot-specific peat depth.
Statistical Analyses
Microsoft Excel and IBM SPSS were used for data analysis with a probability
threshold (p value) of 5%. The residual values of soil properties (bulk density, C, N,
and C/N ratio), ecosystem C stocks, and their log transformations were not normally
distributed based on the Shapiro–Wilk test. Differences in soil properties, biomass,
and C among ecosystems (primary forest, secondary forest, and oil palm plantation) were tested using the non-parametric Kruskal–Wallis test. If the result was
significant, a pairwise comparison test was applied to determine which means were
significantly different.
Results
Ecosystem Structure and Composition
The distribution of tree diameters described a typical reverse J-shaped curve in all
forest sites where small trees were most abundant (Fig. 10.4). Within forest sites, there
was considerable site-to-site variation in the distribution of tree diameter classes and
therefore forest structure. In general, the range in diameters was larger in Beguruh and
N. Novita et al.
Soil Carbon
Peat depth and sample cores were extracted from three sampling points near the
center of each circular plot. We measured peat depth by forcing a steel open-faced peat
auger downward through peat horizon to the organic-mineral transitional horizon. An
extension handle was used if peat depth was greater than 1 m. Measurements of soil
bulk density and C concentration at various peat depths were taken to determine the
soil C stocks (Donato et al. 2011; Kauffman et al. 2011). The core was systematically
divided into the depth intervals of 0–15, 15–30, 30–50, 50–100, and >100 cm (if
mineral layers were not detected before 100 cm depth). Samples of a known volume
were carefully placed in whirlpak bags and transported to the Bogor Agricultural
University for laboratory analysis. Soil samples were dried to a constant mass at
60 °C and weighed to obtain soil bulk density. The organic carbon content of soil
samples was determined via the induction furnace method using a carbon–nitrogen
analyzer (LECO Corporation, St. Joseph, Michigan, USA). The induction furnace
with an elemental carbon analyzer is considered the most reliable method compared to
other C determination methods, e.g., wet combustion and loss of ignition (Kauffman
and Donato 2012) in tropical peat soils (Farmer et al. 2014). The soil carbon stocks
were calculated as the products of bulk density and carbon concentration combined
with plot-specific peat depth.
Statistical Analyses
Microsoft Excel and IBM SPSS were used for data analysis with a probability
threshold (p value) of 5%. The residual values of soil properties (bulk density, C, N,
and C/N ratio), ecosystem C stocks, and their log transformations were not normally
distributed based on the Shapiro–Wilk test. Differences in soil properties, biomass,
and C among ecosystems (primary forest, secondary forest, and oil palm plantation) were tested using the non-parametric Kruskal–Wallis test. If the result was
significant, a pairwise comparison test was applied to determine which means were
significantly different.
Results
Ecosystem Structure and Composition
The distribution of tree diameters described a typical reverse J-shaped curve in all
forest sites where small trees were most abundant (Fig. 10.4). Within forest sites, there
was considerable site-to-site variation in the distribution of tree diameter classes and
therefore forest structure. In general, the range in diameters was larger in Beguruh and