Dear Tyler Davis and co-workers;

I have read your new version with attention. I am however not very much convinced that you indeed replied to the reviewers' main issues as expected. Indeed, the main issue was that the model needed to be validated against observations. You did made progress towards that goal by adding some information on the point-based comparison for San Fransisco area (Figure 3) and by showing a few variables at the global scale. However, since not all variables of the Figure 3 are shown at the global scale, it is rather difficult to have a precise idea of the model still. I do understand that this is not necessarily an easy task to do either.
What I think would be a good addition would be to add other similar sites as the ones of the Figure 3 but in different climate regime. This should not pose major issues since you claim to have the data available for all northern America. You could hence choose other sites with very different climate regimes and show similar responses as the one of Figure 3 and discuss the qualities and drawbacks of your model for the different regions.

What I find really difficult for a potential user is to understand what the model is good at and not good at. From what you have shown, we can see its performance in details in the San Francisco area and a few variables at global scale. What can you tell us for different climate regimes?

With best regards,
Didier Roche

Dear Tyler Davis,

Thank you for the updated version of your manuscript. I appreciate very much the efforts made to better evaluate the model performance. As such, I am mostly happy with the manuscript, albeit for one last request: could you please provide two additional panels on figure 6 with the actual differences between CERES and SPLASH? Such a figure would greatly help to follow the discussion about the differences in the text.
This is not the case for the similar figure 7 where the color coding allows for the differences to be seen readily, but you might want to provide it as well for coherency of the figures. I do not see that last one as a necessity.

With that modification (and the necessary updates into the text to introduce the additional panels) I would be happy to accept the manuscript for final publication.

With best wishes,
Didier Roche

With that modification,

Dear authors,

Thanks for the updated version of the manuscript with the new panels in figure 6 and 7. Could you please provide a little more labels on the color-bars of those two figures? the first one states -12, 0 , 12 and it is not stated or evident whether it is linear, unlike the other panels of the same figure. Same thing for figure 7.

Upon update of those two color-bars, we will have a final version of the manuscript.

Best wishes,
Didier Roche