Sensitivity analysis of model parameters is a very important step in the process of model regionalization application. By applying the statistical analysis on a vast amount of model outputs against the random combinations of the model parameters, the sensitivity index was first calculated for each parameter in the Microwave Emission Model of Layered Snowpacks (MEMLS). The result shows that the sensitivity of the output of MEMLS for each parameter at 37GHz is higher than that at 18.7GHz and the sensitivity is the highest for the physical temperature, and followed by the snow grain sizes. Based on the difference of the sensitivity index of different parameters, we constructed the indicator factor which can reflect the main factor leading to the brightness temperatures variation. It is of significance in understanding the microwave radiation characteristics of snow, and it is also of great benefit to cognize the microwave radiation mechanism by the further design of experiments.