Abstract: To promote the development of technology modernization and irrigation water management modernization in Jiangxi Province and provide more accurate and timely crop water demand forecast data support for real-time irrigation decision-making, this paper develops a daily rice water demand forecast and web-based release system in Jiangxi Province. The system uses the relational data management system MySQL to obtain the parameters and basic information required for the calculation of the rice water demand forecast model by using calibrated Hargreaves-Samani（HS）, Blaney-Criddle（BC） and McCloud（MC） models, and crop coefficient to predict the ETc value of crop evapotranspiration at 26 meteorological stations in Jiangxi Province in the next 7 days. Users can login to the website to check ETc forecasts for any site, any stage of rice growth, and any model. The website page is simple and easy to use. In general, the four calibrated models have good forecast accuracy and can be used to forecast the water requirements of rice in the whole province, laying a scientific foundation for irrigation decision-making toward water saving.