Abstract: The garlic (Allium sativum L.) planting in Guanzhong plain of Shaanxi Province has a common problem of excessive application of nitrogen fertilizers. Establishing model of critical nitrogen concentration dilution curve is an effective approach to assess nitrogen status and optimize nitrogen management in garlic production. Based on 2-year field experiments, the model of critical nitrogen concentration dilution curve was constructed for 2 representative garlic varieties ‘Cangshan’ and ‘Gailiangsuan’ in local area. In 2015-2016, the experiment was carried out in the Institute of Water Saving Agriculture in Arid Areas of China, Northwest A&F University, Yangling, Shaanxi province, Northwest China (108°5′ E、34°19′ N); In 2016-2017, the experiment was carried out in Jintiezhai village of Wugong county, Shaanxi province, Northwest China, (108°16′ E、34°15′N). The experiment was conducted with 6 nitrogen levels: 0 (N0), 60 (N60), 120 (N120), 180 (N180), 240 (N240), 300 (N300) kg/hm2. In 2017-2018, the experiment was also conducted in the Institute of Water Saving Agriculture in Arid Areas for model validation. The plant samples of different treatments were collected 5 times during the study periods, i.e., 15, 30, 40, 50, 70 (or 73) d respectively after garlic returning green in spring. Aboveground dry matter, nitrogen concentration and yield of garlic were determined. With the analysis of critical nitrogen concentration and maximum dry matter in each sample date, the results showed that there existed a negative power function between aboveground maximum dry matter and critical nitrogen concentration of garlic. Using the observed maximum and minimum nitrogen concentration at each sampling date, 2 boundary curves were also determined. With the validation of independent test, the normalized root mean square error of ‘Cangshan’ and ‘Gailiangsuan’ were 9.37% and 7.75%, respectively. It indicated that the simulation performance of these models were proven to be excellent. The nitrogen nutrition index (NNI) calculated based on critical nitrogen concentration dilution curve could directly reflect the real-time nitrogen nutrition status of garlic. The NNI values of N240 treatment varied around 1 in 2-year experiment for 2 garlic varieties, meanwhile this treatment can also obtain the maximum relative yields of garlic scape and bulb. Moreover, the NNI values of N300 treatment were greater than 1, which suggested that the nitrogen concentration in N300 treatment was excessive, resulting in the waste of nitrogen fertilizer. The NNI of 2 garlic varieties were both increased gradually with the raise of nitrogen application level in this study. With the nitrogen level lower than or equal to N240, the significant correlations between the NNI and aboveground dry matters, relative yields of garlic scape and bulb were found. According to the change of NNI value under different nitrogen levels, we suggested the appropriate nitrogen application rate of 240 kg/hm2 for 2 local garlic varieties in this area.