Dear authors,
Your revised version of the manuscript addresses well most comments from the reviewers. I appreciate your effort in preparing a global sensitivity analysis, and editing your manuscript with a professional proof-reading service. This new version reads much better than the previous versions.
As pointed out by the reviewers, the model makes a number of assumptions that may need additional support from empirical studies. It is also highly parameterized, and from this version it is hard to know what would be the effect of parameter uncertainty in predictions.
I also have reservations on your use of a simple Euler scheme to solve the system of partial differential equations. It is well known that this method has serious issues with numerical stability and large approximation errors.
From my point of view, these are important issues of the current implementation of your model. However, GMD is a scientific forum for the description of models, and we are open to publish model descriptions that may be re-evaluated and improved in the future. So I will accept your current model description, but I would appreciate if you make two modifications to the current version:
- Please add a paragraph at the end of your Discussion identifying potential limitations and areas of improvement in your model. For example, the high number of parameters is one potential limitation for the use of this model in other ecosystems, and it is an issue that may translate in high uncertainties in predictions. Other implicit assumptions identified by the reviewers could also be mentioned here.
- Avoid very definite statements in the Abstract and the Conclusions about the 'need for biogeochemical models to comprehensively consider the P cycle'. Although, we all would agree that the P cycle is very important for global scale biogeochemistry, your study does not address directly this topic. You do not provide evidence that a global scale model would not be able to predict effects of global change on C sequestration globally. You have some interesting results for a gradient in Sweden, but this is far from a modeling study that compares models with and without P cycle at the global scale. So, please re-phrase or remove these definitive statements from your manuscript.

Best regards,
Carlos A. Sierra