Due to complex topography and high elevation drop, rainfall stations in southwest China are unevenly distributed. Since precipitation data of these stations are not well representative, rainfall forecast in the southwest region isn’t accurate enough, resulting in huge losses of lives and property every year. The purpose of this paper is to explore the general idea of precipitation interpolation method for mountainous area, and to enhance forecast precision. Lijiang is a typical mountainous city in southwest China with a maximum altitude drop of 4 516 m. According to monthly rainfall data of four decades at 40 rainfall stations in Lijiang, we investigated the optimum interpolation method among inverse distance weight method (IDW), ordinary Kriging method, spline function method, and trend surface method. The analysis result of interpolation errors of six checkout stations shows that IDW is the optimum method for the study region, with the relative error stabilizing at 7.8% when power exponent exceeds 8; Kriging method could also achieve smooth interpolation; spline function method is the most inferior. The research results could offer reference for rainfall forecast in mountainous areas of southeast China.