Carbon budget of a subtropical montane cloud forest: field measurements and model simulation

Chia-Hsin Wu1, Hou-Sen Chu1, Yue-Joe Hsia1, Shih-Chieh Chang1
1 Department of Natural Resources and Environmental Studies, National Dong Hwa University

P 3.25 in Fluxes between the atmosphere and ecosystems

Poster Session 2 on Tuesday, 16:30-18:00

Carbon budgets of tropical and subtropical montane cloud forests (TMCF) are still not well understood. In this study, Biome-BGC was used to simulate the carbon budgets in a TMCF. The purposes of this study were to test the ability of Biome-BGC in simulating TMCF, and to clarify the gaps between biogeochemical processes and the modelling algorithms.

The Chi-Lan Mountain site (CLM) is a TMCF located in northeastern Taiwan. The dominant specie is yellow cypress (Chamaecyparis obtusa var. formosana). The old-growth forest was harvested around 1960’s. Field measurements at the CLM site included plant biomass, leaf photosynthetic characteristics and CO2 fluxes measured by chamber method. Eddy covariance measurement quantify net ecosystem CO2 exchange (NEE). Daytime NEE was further decomposed into gross ecosystem production (GEP) and ecosystem respiration (ER). Biome-BGC model 4.1.2 was used to simulate carbon fluxes and storage. In this study, meteorological data from 2003 to 2012 was available. Some ecophysiological parameters were derived from own measurements, while others were from default ecophysiological parameter sets. We performed the spin-up run, simulated the 90% harvest, and let the stand grow to current time.

Simulated biomass of leaf, stem, and coarse root are close to the field measurements. Biomass of fine root is overestimated possibly due to the combined parameter of turnover rate of leaf and fine root.

GEP from model simulation and eddy covariance measurement has good agreement in winter than in summer. The overestimation of GEP in summer might result from the high default value of Rubisco activity. A reduction of the Rubisco activity by modification of the model source code could improve the simulation of maximum carboxylation rate of Rubisco but at the same time underestimate GEP. It suggests that there may be some ecophysiological processes in TMCF that need to be modified in Biome-BGC. A possible problem is the stomatal conductance of CO2. Our previous laboratory experiments indicated that photosynthetic capacity of yellow cypress did not decline under foggy condition. In the current model, the wet leaf surface caused by canopy interception will halt transpiration by setting the conductance to be zero, which will strongly limit the simulated CO2 uptake for the trees that have adapted to the cloud forest environment.

We compared simulated soil respiration with observed data from chamber measurements, which shows a 100% overestimation of soil respiration than the observed value. As a result, the ecosystem respiration was also higher than the data from flux tower. From the chamber measurement of soil respiration, we found a negative correlation between soil respiration and soil water content at the CLM site. However, a positive relationship between them was given in the Biome-BGC. A further modification of the model is thus necessary to improve its ability in simulating cloud forest ecosystems.

Letzte Änderung 06.04.2014