Under the impact of climate change, even the average annual rainfall of Taiwan reaches 2,500 mm, during the drought period, the competition between each goal of water demand is still a problem that needs to be more squarely addressed in agricultural irrigation water sources. Under the conditions of escaping environmental impact, proper use of existing water sources is a major issue today. In order to explore the rationality of groundwater resources for agricultural irrigation, this study takes the irrigation districts under the jurisdiction of the Chang-Hua Irrigation Association as the research area. It is located in the upper reaches of the Zhuoshui River alluvial fan in central Taiwan. The irrigation water source uses canal water supply in addition to rainfall. Farmers in the study area will also pump groundwater for irrigation. This study applied the hydrological conditions in the 1st crop period of 2015 and 2016 to simulate the irrigation district under the scenario of paddy rice and upland crop area ratio of 0.7:0.3 (about 146 ha of rice cultivation and about 63 ha of upland crop farming area). The simulation results showed that the average groundwater usage in the 1st crop period experimental irrigation area in 2015 was 757.75mm, accounting for 35.91% of the total inflow (i. e. the sum of rainfall, canal irrigation water and groundwater). In the case of the rotation area, the groundwater is gradually increased from the No.1 block to over 50% of the total inflow. When the irrigation water supply of the canal is insufficient, the groundwater becomes the main source of irrigation since the No.3 block; the groundwater usage in the 1st crop period of 2016 is 118.3mm, accounting for 8.12% of the total inflow, which is mainly due to the 2016 issue. During the period, the rainfall distribution was relatively average and the canal irrigation water supply was sufficient, so the dependence on groundwater was lower than that in the 1st crop period of 2015. Assuming that the paddy fields are pumped for irrigation without rain, the simulated inflow values and hydrological data during the 1st crop period of 2015 are used for simulation. The results show that when the paddy fields are irrigated by pumping for 0.5 hours without rain. The total outflow of the simulation is similar to the measured out flow. Based on the simulation results of this model, in fact, the study area is the water production area, and the water produced in the 1st crop period is 7.39 million tons, providing irrigation residual water to the downstream irrigation area.