Abstract: Abstract: Reference crop evapotranspiration (ET0) is an important part of water cycle and water balance. Accurate estimation of ET0 becomes vital in planning and optimizing irrigation schedules and irrigation systems management. Numerous methods have been proposed for estimating ET0, among which Penman-Monteith (P-M) model recommended by Food and Agriculture Organization of the United Nations (FAO) in 1998 since it provids the highest accurate results across the world wherever in an arid or humid environment. But the main problems computing ET0 by P-M model are its complicated nonlinear process and requirements of many climatic variables. Thus, there is an urgent need to develop a much simpler and more appropriate model in areas with limited data, such as the Hargreaves-Samani (HS) method, which is modified in this paper. In order to obtain more accurate HS model of ET0 in hilly area of central Sichuan Basin, 3 parameters(including temperature index, temperature coefficient and temperature constant) were calibrated based on Bayesian Theory using daily meteorological data from 1953 to 2002. Then, the daily and monthly ET0 from 2003 to 2013 were computed by the modified HS model, original HS model and PM model. Compared with the value recommended by FAO (the temperature index, temperature coefficient and temperature constant are 0.0023, 0.5 and 17.8, respectively), the 3 parameters obtained from the modified HS model were much smaller, and increased by latitude (0.00213、0.46 and 16.5 in north zone, 0.00217, 0.44 and 16.36 in central zone, 0.00212, 0.44 and 16.21 in south zone). The ET0 calculated by PM model was taken as the standard, the relative error of modified HS model decreased form 14.2%-60.9% to -1.1%-33.4% in north, 40.6%-92.6% to 16.9%-61.1% in central, 31.3%-96.0% to 8.5%-64.4% in south, and 32.1%-82.7% to 9.5%-52.6% in whole hilly area of central Sichuan Basin. Through the correlation analysis, the slope of regression curve between ET0 calculated by PM model and ET0 calculated by modified HS model were 1.16 in north(R2 = 0.91), 1.02 in central (R2 = 0.88), 0.99 in south(R2 = 0.88), and 1.13 in whole hilly area of central Sichuan Basin(R2 = 0.91), respectively. The trend analysis based on monthly ET0 showed that the trend of ET0 based on modified HS model was the same as that from PM model, which performed as the downwards Quadratic parabola in one year but increasing slightly in years from 2003 to 2013. The relative error of ET0 obtained from the modified and original HS model was above zero, and the former was smaller. Therefore, the modified Hargreaves-Samani model based on Bayesian Theory is more adaptive and accurate, and can be used as a simple method for the calculation of ET0 in hilly area of central Sichuan Basin.