CALCULATING AND MODELLING CANOPY STOMATAL CONDUCTANCE OF PINUS SYLVESIRIS VAR. MONGOLICA FROM SAP FLOW AND ENVRIROMENTAL DATA IN HORQIN SANDY LAND, CHINA

Mongolian pine plantations bear ecological significance meanings to desertification control and degraded land restoration in arid and semiarid regions. To date knowledge about the water consumption characteristics and physiological response to environmental factors of Pinus sylvesiris var. mongolica (P. sylvesiris) in Horqin Sandy Land, China is quite limited. Thus to better understanding its responses to environmental factors and developing mechanism-based models of forest transpiration can offer a theoretical basis for reasonable afforestation in this region. In this study, the primary objective of this study was to employ a Jarvis-type model of canopy stomatal conductance (g(s)) in P. sylvesiris and use the model to simulate long-term trend variations. During the research, the sap flow rates of six tree samples were monitored under natural conditions from May to October 2015. The volumetric soil moisture content and meteorological factors were obtained at an ecological monitoring tower on site. Results indicated that the obtained sap flow data were applied to the Penman-Monteith equation, which did accurately estimate the seasonal patterns of gs. No time lag occurred during the measurement period. The meteorological variables like vapor pressure deficit (VPD), wind speed (W-s), solar radiation (R-n) and relatively humidity (RH) were significantly correlated with estimated g(s). And Ws factor had the greatest impact on gs values with R-2 were higher than 0.970. Therefore gs values could be directly simulated from Ws data. The Jarvis model was set up to express the variation observed in gs, and better performance was obtained by the model that employs a function of R-n, followed by the model that employs functions of VPD and RH. The results of sequence affecting the accuracy of the model was R-n > VPD > RH. And the results also indicate that models based on different combinations of these three functions generally performed better than those based on Rn, VPD and RH independently. Overall, the proposed Jarvis-type model offers a solution to the above-specified problems and this study could put forward the intentionality reference views and conservation management advice.