Estimating carbon storage of forest floor components at varying altitudes in tropical forest of Pahang, Malaysia

• Main Author: Vijayanathan, V. Jeyanny
• Format: Thesis
• Language: English
• Published: 2014
• Online Access: http://psasir.upm.edu.my/id/eprint/67769/1/fp%202015%2091%20ir.pdf; http://psasir.upm.edu.my/id/eprint/67769/
• Institution: Universiti Putra Malaysia

Increasing atmospheric carbon dioxide (CO2) concentrations at alarming rates has triggered the need to conserve and monitor carbon (C) stocks for climate change mitigation. Tropical forests are important carbon sinks which are dynamic due to topographic variations, biomass components, forest floor quality, decomposition processes and spatial variation. Precise and reliable estimation of C stocks and its confounding processes that release/store C are still absent in tropical montane and lowland forests in Malaysia. The objectives of this study were i) to quantify above and belowground biomass C stocks in a lowland forest and montane forest with varying topography; ii) to determine the potential indicators (i.e. litter and duff decomposition rate, forest floor component and properties, soil CO2 fluxes and C:N ratios) of soil C storage and iii) to determine the spatial variability of litter, soil C, C:N, and forest floor component depths of a tropical lowland forest and tropical montane forest with varying topography. A systematic design of 10 m x 10 m plots was established for soil (0- 15 cm depth), litter and aboveground biomass sampling along three slope positions at the montane forest and one plot in the lowland forest. Basic soil characteristics and botanical distribution were determined. A litter bag study and soil CO2 flux measurements were conducted for 480 days in the montane and lowland forest.Forest floor materials were carbon dated and segregated for precise bulk density and carbon fraction measurements. Soil C, C:N, litter depth and various decomposing layers were explored using geostatistics to determine spatial variability. Litter and soil carbon stocks were significantly higher (3 and 5-fold) in the montane forest compared to the lowlands. The aboveground biomass ranged from 100 to 120 Mg C ha-1 and was the most dominant pool (> 40%) for all sites. The decomposition decay rate constant, k ranged from -0.002 to – 0.004 day-1 for the tropical montane and lowland forest. Lowlands showed increased mass loss and significant linear regression relationships between mass loss and litter quality except for C, lignin and cellulose. Soil CO2 fluxes were higher in the lowlands and positively correlated with decomposition and water filled pore space (WFPS). Duff (hemic + sapric) segregation resulted higher bulk density values (0.2) compared to litter (0.04) and revealed a more precise carbon fraction for litter (0.43) and duff (0.55) to be used for forest floor C stocks predictions utilizing significant linear regressions. Duff may reside up to 60 years. Duff decomposition is impeded even at higher temperatures in Forest Research Institute Malaysia campus and retained most of its carbon, nitrogen and lignin. Relationship between soil C:N and C were strong for all plots. Soil total C, C:N, and litter depth exhibited spatial variability at both forest types. Similarly, the litter, hemic and the total forest floor depth fractions confirmed spatial variations. Most variables exhibited a strong spatial dependence with the exception of C:N at the sideslope, litter depth at Jengka VJR and hemic depth at the summit (moderate). Surface maps for total C, C:N, litter depths, hemic and total forest floor depth showed distinct spatial clustering and displayed acceptable accuracy of interpolated values. Forest floor in the montane forest acts as an important C stock in the tropical forest which needs to accounted precisely for national C accounting. Research and developments in monitoring C stocks in montane forests within national and regional areas must be explored to avoid undestimation.