Interactive comment on “ Quantifying the impacts of human water use and climate variations on recent drying of Lake Urmia basin : the value of different sets of spaceborne and in-situ data for calibrating a hydrological model

Comment#1: The results from natural simulation in the manuscript shows a negative TWSA trend (Page 21, Lines 11-13), especially in 2009-2013; to what do the authors attribute this declining trend? Does any of the climate variables, such as precipitation and temperature, over the region show a similar declining trend? How much of the negative TWSA trend can be explained by the changes in climate variables? Even though the manuscript title says “climate variations”, discussion regarding this part is currently too brief.

This manuscript uses WaterGAP Global Hydrology model to quantify the effects of human water use on inflow to Lake Urmia, lake water volume, and groundwater. The model was manually calibrated 4 for time using different observation data sets (remote sensing of irrigated area, monthly total water storage anomaly, insitu observations of stream flow, and groundwater levels from 284 wells). Strengths of the work include a focus on the pressing problem or Lake Urmia decline and identification of the effects on groundwater. With these strengths, there are also several issues that I feel need to C1 be addressed to accept this manuscript for publication.
1. Is the finding that humans affected lake decline new? There have been several recent studies that report this finding (Alborzi et al. 2018;Chaudhari et al. 2018;Shadkam et al. 2016) and some of these studied used the same model inputs as this work and also report groundwater changes. What is new in this work?
2. Is a global hydrologic model appropriate for a basin level analysis? The description of how the model simulates relevant processes is scant. Given resonance times, how appropriate is the temporal spacing (daily) relevant to the spatial grid size? Is it computationally efficient to run a global model for 15 or 20 grid cells of interest? There needs to be a much stronger justification for why the modeling and calibration methods are the correct approaches to use to answer the motivating questions.
3. There is a lot of focus in the text on the multiple calibration variants run with different input data sets. What was learned from this activity? How do those results effect Lake Urmia management?
4. Also, what could one potentially learn from 4 model calibrations that use different calibration data and yield four different models?
5. What are the limitations of this study?
6. The discussion of uncertainty in the results needs to go much deeper and be more specific. This uncertainty is real and likely plays a large role in the interpretation of the results.
7. I found the writing difficult to follow in numerous places, particularly the results section. There are lots of acronyms, run-on sentences, and text that digresses from the section headers or topic sentences of paragraphs. The writing here made it difficult for me to see the main results and findings of the work.
Overall I recommend decline for publication.

C2
Additional line-by-line comments include: pp. p. 7. What is total water storage anomaly (TWSA)? This term seems rather central to the paper. Please explain.
p. 8, lines 13-18. This method of applying (1 -return flow multipliers) to the abstractions to estimate consumptive use assumes that water is used by only one water user. Is this a realistic assumption? If the return flow is used by another agricultural user and then again by a 3rd or 4th user, the basin-wide consumptive use fraction will be much different than the values reported. The large grid size magnifies this error. p. 20, lines 1-5. I would expect to see better calibration with more observational data (i.e., streamflows and lake levels).
p. 20, line 22. I'm confused. The scenario "with reservoirs but without human water use" does not fit either of the two scenarios described in the prior sentence.
p. 20, line 24. What is meant by anomalies? This term has still not been defined.
p. 21, line 17, "The lower lake water loss. . .." What are the loss terms besides evaporation? How are these other loss terms smaller when inflow is larger? Explain.
p. 21, lines 20-23. I don't follow this explanation. There are too many NAs in this sentence. What causes the difference between the naturalized and anthropogenic scenarios?
p. 21, lines 25-28. I don't follow. What is the connection between the first part of the sentence and the second part? . What is being shown in panels A, B, and D? The y-axis labels were mentioned in the text but never explained. Figure A1a. The color scheme makes it difficult to differentiate grid cells. Use only three colors to differentiate the 3 types of storage. How can storage volume be negative? Data availability. I don't follow. If the authors do not have permission to share the data, then how can they share by author request? The HydroSat site underwritten by the University of Stuttgart is neat. What is the original source data for Urmia? Also, there is no water storage anomaly data for Urmia.