Dynamics And Controls Of Carbon Exchange Of A Tropical Montane Rain Forest At Jianfengling, China

Tropical montane rain forest at Jianfengling is one of the most typical rain forests in China. We used two independent approaches, biometry and micrometeorology, as well as CO2/H2O profile to determine the net ecosystem exchage (NEE) and to explore the environmental controls of tropical montane rain forest at Jianfengling, Hainan Island. The results show as follows:1. Annual mean air temperature has been rising in recent 26 years, and annual mean minimum temperature had a significant increase but mean maximum temperature increased slightly. Diurnal courses of air temperature showed upside down U shape, and the minimum value occurred in 6:00-8:00, the maximum value appeared near 14:00. The precipitation at Jianfengling is rich, but varied distinctly from dry season to wet season. The prevailed wind came from southern, and the wind speed profile displayed a highly spatial-temporal variability and which had a positive relationship with height below canopy. Diurnal courses of solar radiation, net radiation and photosynthetic active radiation were the same as air temperature, and the daily peak appeared at 12:00, while at night the solar radiation and PAR were near zero. Annual changes showed single-peak trend, monthly peak appeared in April, and monthly valley value was in February.2. Biometric inventories of 25 years from 1983 indicated that the forest was either a source or a modest sink of carbon. The biomass density varied between 397.05±57.92 and 502.35±96.32 Mg·ha-1, and averaged 453.13±80.06 Mg·ha-1; while carbon density varied between 201.43±29.38 and 254.85±48.86 Mg C·ha-1, and averaged 230.84±40.61 Mg C·ha-1. The biomass of trees with dbh> 10 cm comprised 96% of the total above-ground biomass (AGB) for all trees with dbh> 1.0 cm; and the biomass of trees with dbh> 45 cm contributed 32% of the total AGB, although the number of stems accounted for just 1% of the total stems. The carbon source or sink based on ground measurements indicated that plot P9201 was a net sink at a rate of 2.13±0.46 Mg C ha-1yr-1, while the plot P8302 with a rate of 0.49±0.17 Mg C ha-1yr-1.3. Based on four years of eddy covariance data for CO2 fluxes in tropical montane rain forest we found that the net ecosystem exchange of carbon dioxide (NEE) varied between-1.89 and-2.96 Mg C-ha-1·yr-1(a positive flux indicates carbon loss by the forest, a negative flux indicates carbon gain) from the year of 2006 to 2009, and averaged-2.36±0.42 Mg C-ha-1·yr-1, which showed a moderate sink of this forest. The diurnal pattern of flux showed a significant transition from daytime to nighttime, this similar trend is also found in the course of interannual and seasonal (wet and dry season) variability of flux. The forest was found productive with the gross ecosystem production (GEP) of 19.19±2.23,19.19±2.46,21.38±2.27,19.04±2.23 Mg C·ha-1·yr-1 from 2006 to 2009 respectively, and averaged GEP is 19.70±2.30 MgC·ha-1·yr-1.4. Inventory-based method has been used as major traditional means of addressing net ecosystem carbon exchange of an ecosystem over:multiple years. In recent years the eddy covariance technique has emerged as an alternative way to assess ecosystem carbon exchange. Although these two methods have its merits and limitations, the results in this study produced from eddy covariance technique did well match the estimates of net ecosystem productivity produced with the ground-based inventory. The study showed that we can remedy bias errors of eddy covariance by conducting comparative studies with biomass inventories to constrain flux estimates. The eddy covariance method is particularly adept as studying ecosystem physiology. Specifically, it can be used to quantify how CO2 exchange rates of whole ecosystems respond to environmental perturbations, and when paired systems are applied the method can be used to assess management questions such as the effects of disturbance and plant functional type. Inventory studies of biomass change produce estimates of annual net primary productivity in a variety of temporal and spatial scales. So, by conducting comparative studies in future research at Jianfengling tropical montane rainforest will have a good prospect and production. Overall, the maintance of flux tower and post-field data processing (including quanlity analysis, quanlity control, and data gap filling) are the key steps on deciding whether the technique of eddy covariance can evaluate accurately carbon exchange of ecosystems or not. But, our results estimated from four years of eddy covariance showed that a significant accumulation of carbon (2.36±0.42 Mg C·ha-1·yr-1) at Jianfengling tropical montane rain forest was found, this estimate was independently confirmed through biometric methods which also showed the forest to be gaining carbon at a rate of 2.13±0.46 Mg C ha-1yr-1.Our study provide a good case study in the situation that the results produced from eddy covariance technique did not well match the estimates of net ecosystem productivity produced with the ground-based inventory. Undisturbed tropical rain forests have been historically presumed to contribute little to changes in atmospheric carbon dioxide. But our results indicted that this forest was a carbon sink based on the eddy covariance technique and the ground-based inventory. These results indicated that undisturbed tropical rain forests still have the potential to accumulate carbon from atmosphere. Quadratic relationships were found between the strength of carbon sequestration and heavy rainstorms and dry months. Precipitation and evapotranspiration are two major factors controlling carbon sequestration in the tropical mountain rain forest. Besides that, we also found a strong relationship between disturbance (such as typhoon) and the dynamics of carbon exchange.