Abstract: The theoretical basis of a model for estimation of evapotranspiration of greenhouse eggplants in North China was constructed for the development of optimized irrigation systems.Because irrigation of vegetables in North China has been much higher than crop evapotranspiration, severe deep soil water leakage has occurred, resulting in soil salinization and compaction.This has had a significant impact on the yield and quality of vegetables, resulting in larges waste of water resources.An evapotranspiration model was needed to not only guide irrigation to improve water use efficiency of vegetables, also lay the foundation for the application of agricultural information technology.It can be used to alleviate over-exploitation of groundwater and reduce frequent occurrences of diseases and pests in vegetables due to large amount irrigation.Therefore, it was necessary to establish a model that can accurately simulate vegetable evapotranspiration under greenhouse conditions.In this experiment, the effect of irrigation quota on yield and water use efficiency of eggplant in greenhouse was investigated by setting watering quotas of 15 mm (W1), 22.5 mm (W2), 30 mm (W3) and 37.5 mm (full irrigation quota, CK).The fields were irrigated to ensure sufficient water supply at seeding, flowering and maturity stages, at which the soil moisture contents were kept respectively at 70%, 80% and 70% of field capacity. Based on the modified Penman-Monteith equation and by analysis of relationship between crop coefficient and leaf area index of CK treatment, a model for the evapotranspiration based on meteorological data and leaf area index was established.The model was validated by measured evapotranspiration data under W1, W2 and W3.The results showed that average relative errors between the measured evapotranspiration for W1, W2 and W3 and the simulated values by the model were respectively 17.81%, 18.31% and 17.97%. This showed that the modified Penman-Monteith equation accurately estimated reference crop evapotranspiration under greenhouse conditions.Crop coefficient (K_c) had a significant linear regression with leaf area index (LAI), K_c=0.21 LAI+0.199 1(P < 0.05), indicating that it was possible to determine crop coefficient from leaf area index.The simulated results showed on significant difference in yield (P > 0.05), and significant difference in WUE (P < 0.05) between W2 and CK.WUE of W2 was 31.59% higher than that of CK, indicating that W2 addressed both yield and WUE.Crop coefficients of eggplants at seedling stage and flowering to fruiting period and ripening stage under W2 treatment were respectively 0.21-0.46, 0.62-0.94 and 0.70-0.92.The study suggested that the new model was applicable in estimating actual evapotranspiration of facility eggplants.It calculated crop coefficient that was of practical value in water-saving irrigation.The research was significant for guiding agricultural precision irrigation.