Under the Watershed Allied Telemetry Experimental Research (WATER) project, a significant amount of snow size data was collected from March to April 2008. However, because of limited observation data for the Qinghai-Tibet Plateau, the modeling behavior was not satisfactory. This paper demonstrates characteristics of the snow drop size distribution (SSD) in this region. The experimental area is located in the northeastern part of the Qinghai-Tibet Plateau. The Heihe River Basin, which is the second largest interior river basin in China and is located on the northern slopes of the Qilian Mountains, was selected as the simulation region. This basin ranges from approximately 5,000 m to 1,000 m in elevation. A new generation Parsivel disdrometer, the OTT Parsivel, was used for measurements. Four data sets were compiled to determine the average distributions for four different snowfall rates. The characteristics of the snow particle size distribution in the mountainous area were analyzed. Similar to the raindrop distribution, there was a multi-peak structure. Most peaks appear in the D < 2 mm region (D: diameter of the snow drop size). An M-P distribution and a Г distribution were developed based on the precipitation data observed in Qilian mountainous area. We found that the Г distribution has a better fit than the M-P distribution for the actual distribution. In addition, we observed that the intercept parameter (N0) and the slope parameter (Λ) correlate well with the shape parameter (μ). The disdrometer data can also be used to model the reflectivity factor (ZH) and differential reflectivity factor (ZDR). The radar reflectivity (ZHH, ZVV) and differential reflectivity (ZDR) were modeled in order to facilitate understanding of the connections between radar and ground measurements, and were used to support work for the improvement of rainfall estimates by polarimetric radar. Rain rate estimation using radar measurements was based on empirical models, such as the Z-R relationship and R(ZH, ZDR) in the Qilian mountainous areas. The relationship of R=0.017×100.079×Z_H-0.022×Z_DR is better than R=0.019×100.078×Z_H for estimating R (melted snow). The normalized errors (NE) of R(ZH) and R(ZH, ZDR) are 13.22% and 5.20%, respectively.