Impact of future climate change on watershed-scale precipitation was investigated over Northern California based on future climate projections by means of the dynamical downscaling approach. Thirteen different future climate projection realizations from two general circulation models (GCMs: ECHAM5 and CCSM3) based on four emission scenarios (SRES A1B, A1FI, A2, and B1) were dynamically downscaled to 9-km resolution grids over eight watersheds in Northern California for a period of 90 water years (2010–2100). Analysis of daily precipitation over the eight watersheds showed that precipitation values obtained from dynamical downscaling of the 1981 to 1999 control runs of ECHAM5 and CCSM3 GCMs compared well with the PRISM data. Long-term future trends of annual and seasonal basin-average precipitation were investigated. Although a large variability exists for the projected annual basin-average precipitation within each of the 13 individual realizations, there was no significant long-term trend over the eight study watersheds except for the downward trend in the A1FI scenario. On the other hand, significant upward and downward trends were detected in the seasonal basin-average precipitation except in the winter months (January, February, and March). The trend analysis results in this study indicated the importance of considering seasonal variability, scenario, and model uncertainty.