Simulation of evapotranspiration in broad-eaved and Korean pine mixed forest in Northeast China's Changbaishan Mountain using inverse Lagrangian dispersion analysis

Quantifying the exchange of water vapor between biosphere and atmosphere requires a detailed understanding of the interaction between canopy structure and local canopy.An inverse Lagrangian dispersion analysis to infer water sources,sinks,and fluxes from measured mean concentration profiles was applied and field-tested in a broad-leaved and Korean pine mixed forest using eddy covariance measurements.The method was based on Localized Near Field(LNF),which coupled with a standard deviation in vertical velocity σ_w(z) and a Lagrangian times cale T_L(z).The results showed that the model had good agreement between modeled and measured daily daytime evapotranspiration,with an average precision of 81 %,i.e.the modeled results were 15 %~25 % higher than that of the measured.Nocturnally,the modeled results were 2-4 times as high as those of the measured,but the reason remained unclear yet.The total calculated evapotranspiration,298.91 mm,was much higher than the measured result,240.33 mm.Finally,the variation of the source/sink with time and depth was also discussed.