Comments submitted by John Dunne, National Oceanic and Atmospheric Administration, Department of Commerce (DOC)

This is a very impressive model analysis and synthesis document that does a great job quantifying uncertainties in watershed response. However, while it comes to the conclusion that the uncertainties in individual input climate change projections are the fundamental limitations of the study, these implications are only weakly described in the document in section 7.1 with "For most measures in most watersheds, there is a substantial amount of variability between predictions based on different downscaled climate products. This reflects our uncertainty in predicting future climate, especially the future joint distribution of precipitation and potential evapotranspiration that is fundamental to watershed response, and reinforces the need for an ensemble approach for evaluating the range of potential responses." Given that regionally downscaled climate models do not add considerable value to the global simulation (e.g. Racherla et al., The added value to global model projections of climate change by dynamical downscaling: A case study over the continental U.S. using the GISS-ModelE2 and WRF models. J. Geophys. Res.: Atmos., 117, http://dx.doi.org/10.1029/2012JD018091), the underlying uncertainties can only be reduced by improving the global scale climate simulations. Certainly the currently stated need for further ensembling of current global models has it's role in obtaining a "best estimate given present uncertainty", but for actually reducing uncertainty, it seems like this document should more strongly advocate for improvements in the global climate models both in section 7.1 and in the executive summary.