Abstract: An experiment was conducted to investigate soil evaporation (E), crop transpiration (T), evapotranspiration (ET) and the ratio of evaporation to evapotranspiration (E/ET) of drip-irrigated tomato, which was planted in a typical solar greenhouse in the North China, under different water conditions [irrigation amount was determined based on accumulated pan evaporation (E_p) of 20 cm pan evaporation, and two treatments were designed with full irrigation (0.9E_p) and deficit irrigation (0.5E_p)] at different growth stages in 2015 and 2016 at Xinxiang Comprehensive Experimental Station, Chinese Academy of Agricultural Sciences. Effects of deficit irrigation on crop coefficient (K_c) and variation of water stress coefficient (K_s) throughout the growing season were also discussed. E, T and ET of tomato were calculated with a dual crop coefficient approach, and compared with the measured data. Results indicated that E in the full irrigation was 21.5% and 20.4% higher than that in the deficit irrigation in 2015 and 2016, respectively, accounting for 24.0% and 25.0% of ET in the whole growing season. The maximum E/ET was measured in the initial stage of tomato, while the minimum obtained in the middle stage. The K_c the full irrigation was 0.45, 0.89, 1.06 and 0.93 in the initial, development, middle, and late stage of tomato, and 0.45, 0.89, 0.87 and 0.41 the deficit irrigation. The K_s the deficit irrigation was 0.98, 0.93, 0.78 and 0.39 in the initial, development, middle, and late stage, respectively. The dual crop coefficient method could accurately estimate ET of greenhouse tomato under different water conditions in 2015 and 2016 seasons with the mean absolute error (MAE) of 0.36-0.48 mm·d^-1, root mean square error (RMSE) of 0.44-0.65 mm·d^-1. The method also estimated E and T accurately with MAE of 0.15-0.19 and 0.26-0.56 mm·d^-1, and with RMSE of 0.20-0.24 and 0.33-0.72 mm·d^-1, respectively.