Water management decisions are crucially dependent on forecast information regarding the possible future evolution of precipitation. In this study Probabilistic Quantitative Precipitation Forecasts (PQPF) based on the 20-member National Centers for Environmental Prediction (NCEP) operational global ensemble forecasts are evaluated and postprocessed. The ensemble forecasts are generated with the NCEP Medium-Range Forecast (MRF) Numerical Weather Prediction (NWP) model at T126 (T62) horizontal resolution, equivalent to a 110 (220) km grid interval out to 3.5 (16) days lead time. Because the model used is not perfect and is not designed to account for model related uncertainty, QPF (first moment) and PQPF (second moment) forecasts based on the raw ensemble output are biased. To reduce these biases, a global calibration is made by defining an adjustment to the QPF value in such a way that the adjusted cumulative forecast distribution over a moving time window (most recent 30-day period) match the corresponding observed distribution, accumulated over the entire continental US. This adjustment, when applied on independent ensemble forecast data, practically eliminates the bias in QPF forecasts (first moment) and substantially reduces the bias in PQPF forecasts, when the results are integrated over the whole contitental US. Local QPF biases, and some bias in PQPF, however, are still present in the globally adjusted ensemble forecasts. Work is underway to carry out a local adjustment based on bias estimates derived for areas centered around each grid point for both the first and second moments. Unlike the raw ensemble, the bias corrected forecasts, and/or PQPF information derived from them will be ready for use in sophisticated probabilistic hydrological models as input describing the expected future evolution of precipitation on the scales of the NWP models.