CO2 exchange of a tropical rainforest at Pasoh in Peninsular Malaysia

The eddy covariance method was used to observe carbon dioxide flux (Fc) for 3 years over an old-growth tropical rainforest at Pasoh in Peninsular Malaysia. The average nighttime Fc and NEE were 3.6 and 4.7 μmol m−2 s−1, respectively, when friction velocity (u*) ≥ 0.2, while the average soil respiration rate measured occasionally with the chamber method was estimated to be 4.1 μmol m−2 s−1 if we used the relationship between volumetric soil water content, or 3.8 μmol m−2 s−1 if we used the pure average of nine observations. The comparison of Fc, NEE and soil respiration strongly suggests that nighttime CO2 efflux was underestimated even when the CO2 storage term is considered. The correction of Fc data for low u* did not correct this underestimate, and it was strongly suggested that this forest was not a strong carbon sink, as raw Fc data, even with the u* filtering, implied. Monthly average daytime diurnal changes in CO2 exchange over the canopy were fairly constant from 2003 to 2005 despite fluctuations in soil moisture, solar radiation, air temperature and vapor pressure deficit between dry and wet periods. An obvious inhibition of canopy CO2 exchange in the afternoon coupling with increases in VPD and air temperature was observed year-round. In the case of nighttime NEE, a slight increase in wet period was observed. Diurnal pattern of CO2 exchange revealed that obvious restriction of canopy photosynthesis in the afternoon was occurred in this forest irrespective of soil moisture. Seasonal and annual patterns of CO2 exchange reveal that one of the main causes which induce the difference in NEE pattern between Amazonian and Southeast Asian rainforests was not the difference in daytime photosynthesis but the difference in ecosystem respiration related with dry and rainy seasons. The decrease of CO2 uptake (increase of NEE) at wet period was also observed at this site, but did not form an obvious seasonality such as in Amazonian forests.