Alfalfa plays an important role in dairy farms for economic and ecological reasons in northwest China. In order to manage the complex relations between alfalfa growth and environmental issues accurately and timely, the alfalfa simulation model has been widely studied. An alfalfa growth simulation model based on water and nitrogen factors (ALFSIM-WN) was proposed. The model was divided into 3 sub-models, including crop dynamic sub-model, water balance sub-model and nitrogen balance sub-model using modular design method. It took alfalfa in Ningxia irrigation area of Yellow River as research object and could simulate the growth dynamics of alfalfa under different water supply and variable fertilization. It also could estimate the yield of alfalfa. In order to verify the accuracy and applicability of ALFSIM-WN, experiments were carried out at grassland experiment area of Maosheng Grass Limited Company in Yinchuan (northwest China, 3855N, 10606 E). The experimental alfalfa was Medicago sativa No.7. The irrigation methods included surface irrigation and subsurface drip irrigation. A split plot was designed as 2 experimental treatments. The 1st treatment was the irrigation amount which was divided into 5 irrigation levels including the surface irrigation (1 199 mm) and the subsurface drip irrigation (525, 600, 675, 750 mm). The 2nd treatment was nitrogen application rate which included 4 nitrogen application levels (0, 60, 120, 180 kg/hm2). A total of 17 treatments were designed in experiment. Different irrigation levels were carried out in batches according to different cuts and different growth period of alfalfa. Different nitrogen application levels were carried out in different cuts. Through 2 years experiments (2016-2017), model parameters had been determined using data from 2016, and growth parameters of alfalfa (such as growth period, leaf area index, soil moisture dynamics and yield) had been predicted based on gathering meteorological date, soil data and field management data of 2017. The comparison between simulated and observed was taken in this paper. The results showed that the simulated values by the model was in agreement with the trend of the observed values under different irrigation and nitrogen treatments, such as growth period, leaf area index, soil water content and yield of the alfalfa. Alfalfa in Ningxia irrigation area of Yellow River could harvest 3 to 4 times per year, the mean relative error of the growth period of alfalfa in 4 cuts simulated by the model was between 1.9% and 5.7%, which was consistent with the local growth period for forage. In addition, through analyzing the 4 cuts data in 2017, the mean relative error of leaf area index simulated by the model was between 2.3% and 17.6%, the mean relative error of soil water content simulated by the model was between 2.3% and 17.6%, and the mean relative error of yield simulated by the model was between 1.7% and 16.2%. The root mean square error of the leaf area index was between 0.09 and 0.44, the root mean square error of soil water content is between 0.009 and 0.039 cm3/cm3, and the root mean square error of yield was between 0.3 and 2.3 t/hm2. Therefore, the ALFSIM-WN had higher simulation accuracy in simulated growth period, leaf area index, soil water content and yield, indicating that the model has good application in Ningxia irrigation area of Yellow River and can be used as an effective simulation tool for alfalfa growth in forage planting. [ABSTRACT FROM AUTHOR]