Abstract: Clarifying ¹⁸O isotope composition of leaf water (δ_L,b) would provide theoretical refe-rence for the study of leaf physiology and forest hydrology. We continuously monitored the concentration of atmospheric water vapor (W_a) and ¹⁸O isotope composition of atmospheric water vapor (δ_v) at the canopy of Platycladus lateralis plantation in the mountain area of Beijing. We analyzed the effects of kinetic fractionation coefficients ε_k1(32%) and ε_k2(28%) on the prediction of δ_L,b by combining the measured leaf water ¹⁸O isotope (δ_x) and δ_L,b of P. lateralis. The results showed that the diurnal variation of W_a was irregular. Atmospheric relative humidity (RH) showed a “V” shape of diurnal variation, and stomatal conductance (g_s) increased first and then decreased at the diurnal scale. W_a, RH, and g_s showed a significant negative correlation with δ_L,b when isotopes approached a steady-state equilibrium around noon. The kinetic fractionation coefficient ε_k1 and ε_k2 were applied to the Craig-Gordon model to predict δ_L,b under the isotopic quasi-steady-state condition. The results showed that the predicted values of ε_k2 approached the observed values of δ_L,b. This result indicated that the application of ε_k2 to the model was more consistent with the change of water isotope concentration in the leaves of P. lateralis in the mountain area of Beijing. These results would improve our understanding of water isotope enrichment model and evapotranspiration resolution model in leaves.