The GPM Combined Radar-Radiometer Algorithm performs two basic functions: first, it provides, in principle, the most accurate, high resolution estimates of surface rainfall rate and precipitation vertical distributions that can be achieved from a spaceborne platform, and it is therefore valuable for applications where information regarding instantaneous storm structure are vital. Second, a global, representative collection of combined algorithm estimates will yield a single common reference dataset that can be used to “cross-calibrate” rain rate estimates from all of the passive microwave radiometers in the GPM constellation. The cross-calibration of radiometer estimates is crucial for developing a consistent, high time-resolution precipitation record for climate science and prediction model validation applications. Because of the Combined Algorithm’s essential roles as accurate reference and calibrator, the GPM Project is supporting a Combined Algorithm Team to implement and test the algorithm prior to launch. In the pre-launch phase, GPM-funded science investigations led to significant improvements in algorithm function, and the basic algorithm architecture was formulated. This algorithm architecture is largely consistent with the successful TRMM Combined Algorithm design, but it has been updated and modularized to take advantage of improvements in the representation of physics, new climatological background information, and model-based analyses that may become available at any stage of the mission. Post-launch, algorithm physical parameterizations for effects such as the non-uniform beamfilling of the radar footprint by rain and multiple scattering of radar pulses by ice-phase precipitation have been improved. This document presents a description of the GPM Combined Algorithm architecture, scientific basis, supporting ancillary datasets, inputs/outputs, and testing plan.