The Modeling And Simulation Of Forest Evapotranspiration

The forest evapotranspiration is an important component not only in water balance, but also in energy balance. It is a great demand for the development of forest hydrology and forest meteorology to simulate the forest evapotranspiration accurately. Meanwhile, it is also a theoretical basis for the management and utilization of water resources and forest ecosystem.Taking the broadleaved Korean pine forest located on Changbai Mountain as an example, this dissertation construct two mechanism models for estimating forest evapotranspiration, model a and model b, based on the aerodynamic principle and energy balance equation. Using the data measured by the Routine Meteorological Measurement System and Open-Path Eddy Covariance Measurement System mounted on the tower in the broadleaved Korean pine forest, the parameters in the models, which are the displacement height d, stability functions for momentumφm and stability functions for heatφh are ascertained. At the same time, the comparison of simulation accuracy among different methods, that include model a, model b, Pruitt method, Businger method, Dyer method and energy balance Bowen ratio method is carried out. From the comparison, we can conclude that the simulation accuracy of model a, 84.06%, is highest, while the simulation accuracy of model b, Pruitt method, Businger method, Dyer method and energy balance Bowen ratio method is 79.47%, 59.85%, 53.81%, 52.07% and 41.15%, respectively.Finally, using the data measured by MAOS—ⅠMicrometeorological Measurement System and the monitoring results of soil moisture content, the total forest evapotranspiration in broadleaved Korean pine forest located on Changbai Mountain from July 1st to September 30th in 2001 is simulated by model a and water balance method. So it can be concluded that the simulation result of model a, 268.91mm, is very close to the result of water balance method, which is 288.18mm. The error between model a and water balance method is 19.29mm, and the relative error is 6.7%.