Technical Abstract: To date, seasonal climate forecasts for precipitation lack useful skill for much of the contiguous U.S. This leaves major agricultural areas with questionable climatic guidance for decision support. This presentation will report on work in progress intended to produce monthly, location-specific probabilistic guidance for precipitation on monthly to seasonal scales, with better skill than WMO-standard 30-year climatologies, for U.S. locations with relatively weak ENSO signals. One of the concomitant goals is to develop an easily applicable method for generating such guidance to use with crop or hydrologic models, to facilitate transfer to application communities. The emphasis is on prediction of significant variations from “normal”, but that necessarily requires a definition of “normal” that accounts for the recent rapid changes in climate. Approaches under investigation for new definitions of "normal" include Livezey et.al's linear detrending approach and data-driven spline fits. If the work is successful, the method will provide probabilistic guidance suitable for inclusion in risk-based decision support tools for a wide range of applications.