In the last three decades, remote sensing has rapidly explored various fields of applications. One of the challenges is applying remote sensing data for estimating global precipitation (rainfall/snowfall) particularly over regions where traditional observation techniques cannot provide any information. In this project, satellite based multi-spectral microwave-based cloud information from Advanced Microwave Sounding Units (AMSU) is used for snowfall rate estimation. Ground surface and meteorological information, such as topography, temperature, relative humidity, and wind speed and direction, are used in conjunction with remote sensing data to improve snowfall detection and estimation. Snowfall pixels are detected when the surface temperature (Ts) is less than zero degrees Celsius. An Artificial Neural Networks (ANN) based algorithm is under development to estimate snowfall rate at daily 0.25° x 0.25° resolutions in this study. Daily snowfall rate observations only for the snowfall pixels used for training and validating the model are from SNOwpack TELemetry (SNOTEL) stations due to their good coverage over the western United States. A daily snowfall observation is the difference between the Snow Water Equivalent (SWE) for the same and the previous day when the maximum daily surface temperature is less than 0º Celsius. The preliminary results demonstrate that using combination of two high microwave frequencies (AMSU-MW with 89GHz and 150GHz) improves the snowfall rate estimates.