Two reviews have been submitted on this manuscript. Both of these reviewers have made several suggestions on how the paper could be improved, but none that could be classified as major. The first reviewer's main points were that a description both typical Ethiopian rainfall and the QPE methods should be included in the main text, and that it would be good to have an idea of how these data are used operationally. The second reviewer's main points were that appropriate references should be cited in the text, and that the conclusions about 3B42 were quite negative despite the fact that this estimator is generally unbiased.

The authors have written extensive replies to both reviews, in which they indicated that they agree with most comments and that the new version of the paper will be modified accordingly. However, I found that a small number of the replies raised some questions. For example, in reply to comment 9 of reviewer #2, the authors state that 1/0.51=1.25, which is simply not true. This difference is caused by the fact that performing linear regression of x on y may not yield a coefficient that is the reciprocal of that resulting from linear regression of y on x. When submitting a revised version of the manuscript, please make sure that these issues do not influence the manuscript.

I have some further minor suggestions for improving the manuscript:
- In Fig. 1, consider using a digital elevation model (DEM) as background rather than a Google Earth image.
- In Fig. 4, consider sorting the gauges according to elevation rather than alphabetically so that the order on the x-axis has a meaning (and as a consequence there is more information in the figure). Alternatively, the elevation of the gauge itself may be used on the x-axis, making the figure more like Fig. 3.
- In Fig. 4c, consider using a logarithmic axis for the bias, because a bias of e.g. 0.25 is as bad as a bias of 4.
- In the new Fig. 5, consider adding interpolated rain gauge data (the "ground truth"), and adjust the color scale of each panel so that they are equal (this makes it much easier to compare the different panels).
- In the Conclusions section, consider including a discussion of why TRMM 3B42 performs worse than the others in terms of RMSE and R2.

When you revise the manuscript, please provide a point-by-point response on how you dealt with these points.

I am looking forward to the revised manuscript.

The authors have incorporated all of the reviewers' comments into the manuscript, and hence the manuscript is nearly ready for publication. I would like to ask the authors, however, to make a modification to Fig.3.

Because you refer to increases of rainfall of 50 and 125 mm per 100 m elevation increase, these numbers should be evident from Fig.3. This is currently not the case. Please switch the axes in this figure (elevation on the x-axis and rainfall on the y-axis; like Figure 2 in the response letter), so that the two relationships mentioned in the caption of the figure are also evident from the figure itself.