1. In your response to reviewer #1, I don’t think it is appropriate to use the Thiessen polygon method to calculate basin-scale precipitation. The SWAT allocate each subbasin with a precipitation gauge (or using the lapse rate and elevation bands), therefore, using the SWAT generated precipitation will be more appropriate.
2. “The impacts of terrace and dam on streamflow are clear”. Please clarify and add convincing reference.
3. Some items have been left out. For example, “Splitting the section “Results and discussion” into two distinctive sections “Results” and“Discussion” would certainly help the authors clarifying their scientific demonstration” In reviewer #1; “more information on check dams” in review #2. Please provide a point to point response to these comments.
4. Please clarify what’s the dominate vegetation for Range-Brush, Forest-mixed, forest-deciduous, etc.
5. A lot of unreadable characters are found in your response. Please do correct them.
6. Check “the annual streamflow is 2.0 mm/yr for study area”. It seems too little for the WeiRiver.
7. One figure showing the observed and simulated runoff/streamflow is enough.

I think the authors addressed most questions from the two reviewers except the following one from reviewer #1:
Another issue is the implicitly presumed stability of the catchment behavior over each of the 2 periods 1960-79 and 1980-2009. A graphic showing annual flow, rainfall (both in mm) and runoff coefficients in each of the 3 nested catchments and intermediary catchments (e.g. the colored areas in figure 2) would provide a first assessment of the possible effects of the land-use changes (as done in Lacombe et al. (2008)). A statistical assessment quantifying change and/or trend significance is also missing (cf. Lacombe et al. (2016) for an example).
In addition, the combined effects of forestation, terraces, grass and dam land on hydrological processes should be discussed.

Specific comments:
1. Figure 1: the subbasin numbers are not necessary.
2. Provide the statistical assessments of the model performance besides Figure 6.

There is no need to shown the HRU numbers in Table 1 as they are spatially shown in Figure 3.