Data of 51 distribution points in the country and average day and night temperature difference (bio2), seasonal change of temperature (bio4), maximum temperature in the warmest month (bio5), average temperature in the wettest season (bio8), average temperature in the driest season (bio9), precipitation in the driest month (bio14), precipitation seasonality (bio15), precipitation in the warmest season (bio18), altitude (elev), slope (), Eleven environmental factors of aspect were imported into MaxEnt version 3.4.4 to establish a model. The parameters are set as follows: 25% of the distribution points of soft comfrey were randomly selected for model testing, and 75% of the distribution points were used for model training. The element types were Quadratic features, Threshold features, and Hinge features. The number of repeated runs was 10 times. Other parameters were kept at default. The simulation results were output in ASC file format. The Jack-knife method was used to evaluate the contribution of various environmental variables, and the area under the receiver operating characteristic curve (ROC)(AUC value) was used to evaluate the accuracy of the prediction results of the MaxEnt model. The closer the AUC value is to 1, the higher the accuracy of the model, the more reliable it is. When the AUC value is between 0.9 and 1.0, the model prediction result is the best, when the AUC value is between 0.8 and 0.9, the model prediction result is better, when the AUC value is between 0.7 and 0.8, the model prediction result is average, and when the AUC value is less than 0.6, the model prediction fails. The mean area under the ROC curve (AUC) of Arnebia rugosa was 0.960 and the standard deviation was 0.022, which indicated that the prediction model had excellent accuracy and the prediction results of potential suitable areas of Arnebia rugosa were extremely reliable. The results of the potential suitable areas of Arnebia japonica predicted by the MaxEnt model under different periods and scenarios were imported into ArcGIS software, and the suitable distribution areas of Arnebia japonica were divided into four levels using the Reclass tool: unsuitable area (0-0.07), low suitable area (0.07-0.3), medium suitable area (0.3-0.6), and high suitable area (0.6-1.0), and the area of each suitable area was counted respectively.