Study On The Model And Mechanisms Of Water Coupled Carbon In The Vegetation-atmosphere Interface Of Evergreen Broad-leaved Forest In Tiantong

The vegetation-atmosphere interface is the exchange space of water and CO2. Water and CO2 are the main components of water and carbon fluxes of the terrestrial ecosystem, and are controlled by canopy stoma. Therefore, canopy stoma is the connected point of water and carbon of the territal ecosystem. Canopy stoma is sensitive to environmental factors, and the ecological process that canopy controlled and adjusted the water and carbon fluxes. How to model the fluxes of water and carbon of the canopy is a scientific issues that many scientists focused on.This study was conducted in the evergreen broad-leaved forest in Tiantong National Forest Park, and measured sap flow, meteorological and hydrological factors, leaf carbon isotope of canopy, and water use effiency, based on coupling of sap flow and leaf carbon isotope, the coupling model of water and carbon model is created, and the process of water and carbon coupled is clarified.1 Permanent plot study was carried out from the meteorological data from June 2009 to august 2010 in Forest Ecosystem Observation and Research station in Tiantong. Total rainfall outside the forest was 2316 mm, and the throughfall, stemflow, and canopy interception accounted for 81.7%,2.3% and 16.0% of the total rainfall respectively. The through fall and stem flow had significant linear relationships with rainfall (P<0.01). The canopy interception rate showed significantly negative correlation with the rainfall, rainfall duration, rainfall intensity, and relative humidity during rainfall (P<0.01), but positive correlation with wind velocity (P<0.01).2 Saturated hydraulic conductivity (Ks) increased significantly in the 0-60 cm layers with the vegetation succession (p<0.001). The average Ks in the 60 cm soil profile significantly increased from the bare land, Lithocarpus glaba+Laroptahon chenese shrub, Pinus massonian forest, Schiima superba+Pinus massoniana forest, Schima superba forest to Castanopsis fargesii forest (df=5,F=10.69, p=0.002). In the shrub stage, the average Ks had made significant difference to bare land (p<0.05), in the Schima superba forest, the average Ks had been increased significantly(p<0.05), in the climax stage, the average Ks was 3.28 mm min-1 and reached maximum value. Soil bulk density, non-capillary porosity, and silt content were the key factors that affected Ks. Soil organic matter (SOM) was also increased with vegetation succession and positively correlated to Ks significantly (p<0.01). Castanopsis fargesii forest was able to resist the storm intensity of 10 years recurrence. Schima superba forest was able to resist the storm intensity of 1 year recurrence. However, overland runoff occurred easily in the early successional stages, such as, in the bare land, Lithocarpus glaba+Laroptahon chenese shrub, Pinus massonian forest.3 In the Castanopsis fargesii community, the values of transpiration varied clearly from 5.07 (g m-2 s-1) to 84.28 (g m-2s-1), Maximum value was 18 times greater than minimum value, which indicated that trees with different diameters varied obviously in water using straggles, contribution, functions in the same community. Trees with larger diameter had higher value of transpiration because of their larger sapwood area. Furthermore, Tall trees occupied superior spaces and obtained more resources than any other tree species. These conditions ensured tall trees had greater sap flow. Although the value of transpiration of Catanopsts fargesii was greater than that of Schima superba attributed to larger diameter and sapwood area of Catanopsts fargesii, the sap low velocity of Schima superba was higher than Catanopsts fargesii because the stoma size of Schima superba was bigger than Catanopsts fargesii. Sap flow of Catanopsts fargesii and Schima superba varied widely associated with weather conditions. In the sun day, sap flow regularly changed with the solar radiation, in the cloudy day, sap flow fluctuated with solar radiation. However, in moderate rain, sap flow decreased significantly, and in the storm weather condition, such as October 1,2009, the value of sap flow declined to 0-0.5 g m-2 s-1, a lowest value in the same month. Annual change of sap flow and transpiration of Catanopsts fargesii and Schima superba was consistent with seasonal change. In winter from December to February, sap flow and transpiration was at a low lever, then became increasing in March, till to July and August, sap flow and transpiration reached maximum value. After September, sap flow and transpiration began decreasing. Photosynthetic active radiation (PAR) and vapor deficient (D) driving force of canopy transpiration.4 The values of evapotranspiration were closed estimated by the method of water balance and Penman-Metainth Equation in Catanopsts fargesii community. During the study period from June, 2009 to August,2009, evapotranspiration, soil water content change, and run off accounted for 72.1%,20.2%, and 6.7% of total rainfall. Canopy transpiration accounted for 81.4% of evapotranspiration, which indicated that transpiration was the main component of evaportranspiration. The maximum value of Catanopsts fargesii canopy was 7.2 mm d-1 in August, and minimum value was 0.52 mm d-1 in February. Photosynthetic active radiation (PAR) and vapor deficient (D) driving force of canopy transpiration. The results of this study suggested that the climax forest, Catanopsts fargesii community had good function of intercepting rainfall and reforming runoff.5δ13C1, of six tree species varied from-28.45‰to-32.49‰, and water use effecint (WUE) varied from 3.5-4.7 mmolCO2·mol-1 in the Schima superba community. The WUE order of six tree species was Cyclobalanopsis myrsinae> Schima superba> Castanopsis sclerophylla> Catanopsts fargesii> Lithocarpus glaba> Castanopsis carlesii. The WUE of the dominant tree species collaboratly changed with the sequence of the forest succession. WUE of Lithocarpus glaba decreased, Catanopsts fargesii and Schima superba increased as the succession process. However,, WUE of Schima superba began to decreasing, and WUE of Catanopsts fargesii was still increased in the late ssuccessional stage. These results were so interesting that verified the reasonableness of the forest succession series in Tiantong National Park. WUE was correlated to habitation. On the ridge, WUE of Cyclobalanopsis sessilifolia was greatest, and lower values of Catanopsts fargesii and Schima superba increased as the succession process. However, In the shrub stage, where forest was serious disturbed, Lithocarpus glaba had higher WUE value, lower value of Catanopsts fargesii and Schima superba. However, in the sites of evergreen broad-leaved forest, where soil water conditions were better, WUE values of Catanopsts fargesii and Schima superba were greater than Lithocarpus glaba.6 The mean value ofδ13C under the gradients of 8 m canopy was from-27.57‰to-31.96‰, and showed Martinelli "canopy isotope depleted effect". The decreased value was-3.1‰, and this value was closed to-3.6% in the canopy of tropical rain forest and boreal forest. We tested this hypothesis in the subtropical evergreen broad-leaved forest in Eastern China. The mean of WUE along canopy was from 3.29 to 5.25 mmolCO2 mol-1. The order of WUE was upper canopy>middle canopy> low canopy. WUE along canopy temperately changed with the season change. The maximum value of WUE was 5.25 mmolCO2 mol-1, and appeared in May. Net photosynthetic rate (Pn) and Nmass varied significantly Along the gradient of canopy (p<0.01). Pn of the upper canopy was higher than lower canopy. Pn increased from January to July, and obtained maximum value (15.4μmolCO2 m-2 s-1) in July. After September, Pn began decreasing. Nmass showed same trend as Pn, and maximum value (2.8%) also appeared in July. From December to Januray, Nmass was lower than any othe month.δ13C, WUE, and Pn were positive correlated to Nmass along the gradient of canopy.7 Based on sap flow and△13C, The first step, The model of water coupled carbon was constructed in each canopy gradient of Catanopsts fargesii community: then, according to LAI of each canopy Catanopsts fargesii communityIn order to test accuracy of model, we used Li-6400 and measured the Pn in the upper canopy,4 m,6m layers along the gradients of canopy, and calculated Pn of canopy by big leaf model. The predicted data fitted the measured data very well, which indicated the model had good performance (F=46.89, P<0.0001).We inputted the data ofδ13C, sap flow density and LAI in August,2010 into the model, and obtained the data of Pn. The diurnal change of Pn accosiated with the solar radiation, before 8:00 AM, the value of Pn is very low, and closed to 0. After 8:00 AM, Pn increased quickly, and achieved maximum value at 10:00 AM. After this time, Pn began decreasing, after 18:00 PM, Pn declined to 0 levels. Air temperature (T) was linearly correlated to Pn, PAR, and D fitted Pn significantly as logarithmic curve. Moreover, canopy transpiration (Tr) was also correlated to Pn significantly. When Tr< 200 g m-2s-1, Pn increased with Tr increasing, when Tr> 200g m-2s-1, Pn declined with Tr increasing.8 Based on the daily meteorological data from 1954 to 2009 in Tiantong region, net primary productivity (NPP)model of Zhou Guangsheng was used to study effect of climate change on NPP of evergreen broad-leaved forest. The result showed that, (1) Monthly average values of air temperature, precipitation, and reference evaportranspiration (ET0) clearly increased in July and August, and the annual trends of average air temperature, precipitation, and ET0 increased significantly in Tiantong region in recent 60 years (P＜0.001). (2) Annual average NPP of 56 years was 12.196 t·hm-2·a-1, and increased significantly in recent 60 years (P<0.0001). (3) In the future, if air temperature increases 2℃and precipitation increases 20%, NPP will increase 15.9% in this region, if air temperature increases 2℃and rainfall decreases 20%, NPP will decrease 4.9%, if air temperature increases 2℃and rainfall is not changed, NPP will increase 5.5%. (4) Annual average precipitation, ET0 and air temperature were the main meteorological factors that affected NPP in this region.