The clarity and organization of the manuscript have been improved after the revision made by the authors. However, the authors did not take all the comments from both reviewers into account and explain their rationale and actions well. It is for this reason that I do not suggest publications in ACP unless the authors can clearly explain what is the information embedded in the ecosystem-specific scaling coefficient when up-scaling the leaf-level model to ecosystem-scale GPP.
1. The soil drying process can be considered by the leaf-level gas exchange model proposed elsewhere (Hari et al., 2017) through the model parameter λ (i.e., the cost of transpiration; a measure of water use efficiency). Based on the reply to previous review, however, only a “mean” optimal value (inferred from a large corpus of data) is used. This suggests that how leaf-level gas exchange is impacted by the soil water status is not explicitly taken into account.
2. The up-scaling is mainly done through LAI of the dominant species (i.e., Scot pine). However, the ecosystem-level GPP can be also impacted by the activities of understory species and the soil. When adding an extra parameter (i.e., ecosystem-specific scaling coefficient) to upscale leaf-level processes to ecosystem level, this scaling coefficient then contains information associated with all the possible contributions other than leaf-level processes from single species. This explains why both reviewers suggested that the ecosystem-specific scaling coefficient should be reported and further discussed especially when the soil-drying effects are not included in the leaf-level model.
1. p2, Line 24-30: Duursma et al. (2009) also did not consider drying soil effects in their model.
2. Eqs. (1) and (3): Is E the transpiration rate or efficiency of photosynthetic light and carbon? According to Hari et al. (2017), it seems like that E is the same as b. What is difference between Eq (1) and (3)? Is b=a4S=E or there is time up-scaling from daily p (i.e., Eq (1)) to annual p (i.e., Eq. (3))?
3. p5, Line 14-20: This is why the ecosystem-specific scaling coefficient should be reported and carefully discussed.
4. Fig. 3 A): Should it be the onset dates of photosynthesis not the onset and cessation dates of …..?
5. p6, Line 22-29: Is the residual calculated as the difference between the measured and calculated GPP? If so, the same order of residual as measured GPP suggests significant deviation between predicted and measured GPP. This may not be surprised especially when the variation of GPP is highly non-linear with respect to the environmental factors (e.g., T, CO2, PAR and soil water status). That is, the residual for a specific environmental category can be still impacted by other environmental factors.