GENERAL COMMENTS:

The present paper is on “Sustainable use of water resources…”.  The topic/paper is interesting but needs a number of modifications before it can be accepted.

“Sustainable/ sustainability” is mentioned several times in the text (and the title). In my understanding sustainable water resources management is not simply supplying an ever growing demand (that’s a really old, outdated approach from the last century) but an important part is water demand management. In the paper there is nothing about water demand management, but in one of their scenarios the authors simply interpolate past/observed water use/demand into the future.

Overall, the wording (or definitions) needs to be reconsidered. The authors define “sustainable/ sustainability” as state where "GWR>GWW"; i.e. "groundwater recharge" is higher than "groundwater withdrawal". A state where GWW/GWR ratio is higher than "1" is considered “unsustainable”. The authors consider GWR only; there is no information (to my understanding) on the groundwater level or volume. The authors assume that the whole water demand can be withdrawn, e.g. stating that "The rate of GWW to GWR is greater than 2..." (p. 7, lines 195/196). Without information on the groundwater level or volume, how can they say that the water volume demanded can be withdrawn? In my opinion the whole text needs to be changed (I will define demand on groundwater here as “GWD”): the authors can only calculate what volume can be “sustainably” withdrawn from groundwater. In cases where “GWD>GWR” water withdrawals will become unsustainable, i.e. water demand on groundwater resources is larger than groundwater recharge, meaning the groundwater level or volume will decline. The (or one version for) correct wording would be that “Only 50% of the water demand can be sustainably withdrawn…” instead of "The rate of GWW to GWR is greater than 2..." (p. 7, lines 195/196). I think the whole text (and Tables/Figures) needs to be changed in this direction.

It is not clear if the results (e.g. “GWW/GWR”) represented (e.g. Fig. 2) a calculated on an annual basis and then averaged over the entire (30 years) period or if it is calculated for the entire (30 years) period. If it is calculated on an annual basis and averaged afterwards, how is groundwater storage, e.g. in a very wet year GWR may be (much) higher than GWD and increase availability (increased groundwater level or volume) in the next year, considered? Also, to my knowledge, there is no groundwater flow between sub-basins in SWAT. In reality a sub-basin (region) with high groundwater withdrawals, resulting in declining groundwater levels, may receive groundwater inflow from neighbouring sub-basins (regions) without groundwater withdrawals, i.e. high(er) groundwater levels.

The paper builds on work of Mahmoodi et al. (2020a) (referenced as Mahmoodi, N., Kiesel, J., Wagner, D.P., and Fohrer, N.: Water use systems and soil and water conservation methods in a hydrological model of an Iranian Wadi system. J. Arid Land, 12, 545–560, https://doi.org/10.1007/s40333-020-0125-3, 2020a.), describing the set-up and calibration/validation of a SWAT model to the Halilrood river basin.

In a simplified way the water balance can be given as:

SW(t) = SW(t-1) + PRECIP(t) - SR(t) - ET(t) - PERC(t) – SSF(t)

where SW(t) is the soil water content at time step t, PRECIP is precipitation, SR is surface runoff, ET is evapotranspiration, PERC is percolation, and SSF is subsurface flow (with the last two variables describing groundwater-/subsurface part of the system, this part may be differentiated differently depending on the complexity of the approach/model). The important point is that there are some variables that are observed usually, i.e. PRECIP and river flow (combining SR and SSF), while for other variables seldom or only for very few locations observed values are available, e.g. ET and PERC. Assuming (for simplicity) constant SW and known PRECIP and river flow, potential and actual evapo(-transpi-)ration and PERC are unknown. Depending on the parametrisation of the model potential and actual evapo(-transpi-)ration and PERC can differ strongly. In a model calibrated/validated on stream flow only there is high uncertainty if potential and actual evapo(-transpi-)ration and PERC are simulated correctly - here either observed values, e.g. lysimeter measurements (point information), or from other sources, e.g. LAI or actual evapotranspiration from GLEAMS or MODIS (areal information) should be used to validate the model results. In the paper of Mahmoodi et al. (2020a) such a validation is not carried out and therefore I wonder how the authors can prove the reliability of their SWAT parametrisation. For instance in Table 4 (Selected parameters for calibration in the SWAT model) Mahmoodi et al. (2020a) give "EVRCH Reach evaporation adjustment factor from 0.5 to 0.8", to my understanding this means that the potential evaporation is reduce strongly by 20 to 50% (i.e. more water may percolate, become groundwater recharge)!

As given in Mahmoodi et al. (2020a, p. 551):

“WYLD=SURFQ+LATQ+GWQ–TLOSS, (1)

where WYLD is the water yield (mm); SURFQ is the surface runoff (mm); LATQ is the lateral flow contribution to stream (mm); GWQ is the groundwater contribution to stream flow (mm); and TLOSS is the transmission losses (mm), i.e., water loss via transmission through the bed of the channels.” Transmission losses in semi-arid environments/rivers can be very high, introducing another uncertain variable in the simulations.

Furthermore, the paper builds on work of Mahmoodi et al. (2020b) (referenced as “Mahmoodi, N., Wagner, D.P., Kiesel, J., and Fohrer, N.: modeling the impact of climate change on streamflow and major hydrological components of an Iranian Wadi system. Water Clim. Change, https://doi.org/10.2166/wcc.2020.098, 2020b.”): "The G-RCM CSIRO-SMHI was chosen since it represented the median model of the major hydrological components (Mahmoodi et al., 2020b)." The authors write that they follow the argumentation of Tebaldi and Knutti (2007) and Thober and Samaniego (2014). However, reading the paper of Tebaldi and Knutti (2007) I only find that they discuss using (weighted) averages of a number of climate models, based on the idea that the performance can be improved by averaging or combining results from multiple models. But this is that first a number of climate models are run and then the results are averaged (in the case of impact models the input from a number of climate models are used and then results of the impact models are averaged). In the study of Thober and Samaniego (2014) a number of (meteorological) indices are used to select regional climate models, i.e. reduce the number of impact model runs. In case such an analysis (to select G-RCM CSIRO-SMHI) was carried out in the present study this analysis and its results need to be described (e.g. in the Suppl. Material), otherwise the results are not replicable (I also need to mention that Thober and Samaniego (2014) reduce the number but they do not propose to use results of only ONE climate model). In the paper of Mahmoodi et al. (2020b) I don*t find any information that “G-RCM CSIRO-SMHI … represented the median model of the major hydrological components”.

The use of only one climate scenario restricts the value of the paper – for a more wet or dry (climate) future results could be very different. Such results would be needed to enable more robust decisions on future water resources management, especially as the authors state that effects of climate change on surface/groundwater resources are much more significant than future water demand/use.

When reading the paper Mahmoodi et al. (2020b) I also found that minimum and maximum elevation for the river basin shown in Figures 1 (Mahmoodi et al. (2020b) and the present study) are different - please explain.

In the whole text please use “streamflow” instead of “flow” when referring to river flow, otherwise readers could be confused as the paper is on surface water and groundwater. The results given for streamflow (IHA, VRA) are all for the outlet of the basin?

SPECIFIC COMMENTS:

1. Introduction:

Page 1; Lines 23: “…alteration caused by natural or anthropogenic activities…” I am not sure what is “natural … activities”; rethink formulation.

Page 2; Lines 40/41: “…precipitation... rainfall...”; please use either “precipitation” or “rainfall” in the whole text

Page 3; Line 68: “Further aggravation will put increasing pressure on the...”; “aggravation” of what (rethink formulation)?

2. Materials and methods:

Page 3; Lines 80/81: “…increased over the last years at the outlet of the basin during the past 33 years...”; reformulate “…increased over the last years … during the past 33 years...”

Page 4; Lines 95/96: “Good performance for modeling daily streamflow values was achieved judged by a multi-metric approach including NSE (0.76 and 0.54)...”; according to Moriasi et al. (2007: Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE, 50(3): 885–900) an NSE of 0.54 is considered as “satisfactory”.

Page 4; Lines 114 and 116 (also line 459 “Statistics Center of Iran”): on page 114 it is "Statistics Cerner of Iran"; on page 116 it is "Statistical Center of Iran" - The homepage (https://www.amar.org.ir/english) gives the translation/name "Statistical Centre of Iran", please use the official translation/name given on the homepage.

Page 4; Lines 123/124: "To meet the future domestic, agricultural and industrial water demand, increases in the number of wells and qanats are linearly extrapolated...", so you assume that all wells and qanats have the same (water) yield? Please justify this assumption.

3. Results:

Page 7; Line 199: “...unsustainable subbasins (GWW<GWR)...”, unsustainable is “(GWW>GWR)”; however, I suggest to change the wording (see “GENERAL COMMENTS”) and this part needs to be rewritten anyway…

Page 7; Lines 207/208: “... Among these 56 unsustainable sub-basins, GWW/GWR ratio is higher than 5 in 42 sub-basins....”, this means “only 20% of the water demand can be supplied sustainably”?; change the wording (see “GENERAL COMMENTS”) - all the results presented need to be changed accordingly.

Page 8; Lines 211/212: “...it drops from 385 (106 m3 yr-1) in model setup period to 172 (106 m3 yr-1)...”, what is the unit of the first numbers (385; 172)?

Page 8; Lines 217: "3.1" is "Groundwater sustainability"; therefore I suggest to call "3.2" "Streamflow alteration",… give a subtitle that’s clearly points to Streamflow/Surface water…

4. Discussion:

Page 10; Line 282: “…substantial deficits in discharge during...”; is this “groundwater recharge” or “river discharge”?

Page 11; Line 285: “…predicted unsustainability of groundwater could...”; correct “…predicted unsustainability of groundwater use could...”?

Page 11; Line 287: “…could lead to a higher groundwater withdrawal in summer season...”; correct “…could lead to a higher demand on groundwater in summer season...”?

5. Conclusion:

Needs to be adapted according to “GENERAL COMMENTS”.

Supplementary material

- in Figure S1 units (e.g. for “Group 1” and “Group 2” the for x-axis "year" and for the y-axis "m³ s^-1") should be given

- Figure S2 needs a better description to understand what is shown (e.g. the title states “Distribution of annual values…” but shown are monthly values; or is it - the single dots - the 30 monthly values of the 30-years per period, e.g. for January?); also here units are missing. The results shown are scattered (not ordered) – is this because the results are given according to their temporal occurrence (i.e. the year)? If yes, why a different x-axis when compared to Fig. S1?

Review to HESS-2020-599

General comments

General comments - Reviewer

The topic covered fully corresponds to the aims and scope of the Hydrology and Earth System Sciences. Moreover, it covers aspects of hydrology and management. Though, the manuscript, in its present form, has a few weaknesses. Appropriate revisions to the following points should be undertaken to improve the readability and increase the interest for a general audience.

The title should better describe the study. Sustainable water use is not the subject of this study. I would rather say potential future climate impacts on water resources. At the end of the introduction provide a statement on the novelty of the study. At the moment it is not clear enough. Two sections of the manuscript need improvements. More specifically 2.2 Hydrological model and 2.3 Future climate change simulation sections. These two sections are parts of previously published work and they are confusing in relation to this research effort. Authors should better justify the selection of the G-RCM. I strongly recommend a flow chart of the methodology followed. Please provide as supplementary table the IHA Non-Parametric RVA Scorecard. Methodology limitations should be mentioned. Discussion section should be extended, as this is the main contribution to future researchers. Explain and discuss why the streamflow time series was generated for the period from 1993 to 2009 and 1979-2009. The presented results are influenced by this choice and it should be motivated. Moreover, at least a minimal discussion should be provided on expected differences (if any) when using different reference periods. Please contrast your own findings with those of previous literature about hydrological studies and climate change in the area (e.g. https://doi.org/10.1029/2008WR007615; https://doi.org/10.1186/s40068-019-0135-3; https://doi.org/10.1007/s10666-019-09665-x). Uncertainty of the model outputs should be also in discussion.    

Specific Comments

Comment 1

Page 1, Line 11-13“A hydrologic model is used to assess streamflow and groundwater recharge of the Halilrood Basin on a daily time step under different scenarios over a model setup period (1979-2009) and for two future scenario periods (near future: 2030–2059 and far future: 2070-2099).”

Please specify the number of the different scenarios.

Comment 2

Page 3, Line 92-94

“Based on representative climatic conditions, an eight-year period was used for model calibration (1995-2003) and a six-year period for validation (2004–2009).”

Please specify which are the representative climatic conditions. Are these observations? If yes please provide the reference and the source of the data. 

Comment 3

Page 3, Line 88-89:

“The Soil and Water Assessment Tool (SWAT, Arnold et al., 1998; Arnold et al., 2012) was used to simulate the streamflow of Halilrood River between 1993 and 2009 on a daily time step.”

Perhaps, authors mean that the outputs of the SWAT model were used in this study. This description is poor and confusing. Most of the work presented in this section, has been carried out in Mahmoodi et al. (2020a). Please consider to rewrite this section by clarifying the work you did for this research effort. Moreover, please specify if  SWAT parametrization using the representative climatic conditions is sufficient when using the CSIRO-SMHI. In Mahmoodi et al. (2020b) it is mentioned that “Further details on model parameterization and performance are available in Mahmoodi et al. (2020)”. I couldn’t find this study on the internet.

Mahmoodi N., Kiesel J., Wagner D. P. & Fohrer N. 2020 Water use systems and soil and water conservation methods in a hydrological model of an Iranian Wadi system. J. Arid Land. 12 (4), 1–16

Comment 4

Page 4, Line 110-112:

 “The calibrated and validated SWAT model was run with this climate model output to simulate groundwater recharge and streamflow for the model setup period (1979-2009) and two future periods (near future: 2030-2059 and far future: 2070-2099)”.

SWAT model was run in previous work (Mahmoodi et al., 2020b). Please, provide more information about the model outputs. Perhaps a table with descriptive statistical analysis of the observational streamflow data, together with the model outputs (e.g. streamflow for the three different periods) would help readers to follow the logic of the paper.

Comment 5

Page 4, Line 104-107“For the present impact study, following the argumentation that errors level out and a projection can be better represented through averaging, i.e., taking the mean, median, or weighting (Tebaldi and Knutti 2007, Thober and Samaniego, 2014), from the RCP8.5 and distribution mapping-adjusted ensemble, one global-regional climate model was selected.”

This sentence is a too long and it is confusing. Could you please explain further? Please also specify why you didn’t use climate change projections from RCP 4.5.

Comment 6

Page 4, Lines 107-108:

“This selection was conducted according to the model democracy approach, which treats all climate models equally and the median model of the model ensemble is selected (IPCC, 2013).”

What is the model democracy approach? In the IPCC 2013 there is no mention about it. Moreover, please fix the references according to the journal’s requirements and in respect to the report’s recommendations (see below).

IPCC, 2013: Summary for Policymakers. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA

Comment 7

Page 4, Lines 108-110: “The G-RCM CSIRO-SMHI was chosen since it represented the median model of the major hydrological components (Mahmoodi et al., 2020b).”

A model and its collection of runs is referred to as an ensemble. Please specify what do you mean as median model. If this ensemble selection influences the hydrological components in a major way, authors should add additional discussion in case of a different selection.

Comment 8

Page 4, Lines 117-119: “The population growth rate suggested by presidency of I.R.I, Plan and Budget Organization (2019) was applied on the 2017 population data to estimate the population of the basin for the years 2045 and 2085, representative for the near and far future periods respectively (Table 1).”

If your analysis was made based on 2017 population data, why data relative to 2011 population are presented in tables (table 1 etc.)? If you didn’t use these data please remove them.

Comment 9

Page 4, Lines 123-124: “(i): To meet the future domestic, agricultural and industrial water demand, increases in the number of wells and qanats are linearly extrapolated with the estimated increases in the population of Halilrood Basin as follows:”

Linear extrapolation should be discussed in the discussion and conclusion sections. Why authors decided to use this method?

Comment 10

Page 7, Lines 186-187:

“The direction of change is shown by positive RVA, where the indicator becomes more stable within the RVA targets and negative RVA, where the indicator is moving towards an upper or lower alternative state.”

Please explain what do you mean “the indicator becomes more stable”.

Comment 11

Page 1, Line 42-42; Page 7, Line 203-204; Page 8, Line 225-226; Page 11, Line 287-288:

• “In Iran, the scarcity of rainfall, combined with climate change and population growth over the last decades, has resulted in higher groundwater extraction rates (Izady et al., 2015; Rafiei Emam et al., 2015; Mahmoudpour et al., 2016).”
• “As shown in Figure 2c and e, the GWW/GWR ratio is higher in the near and far future if the two stressors climate change and population growth are considered simultaneously (Projected-WUS).”
• “This might be due to the higher reduction in projected winter precipitation (Mahmoodi et al., 2020b).
• “This could lead to a higher groundwater withdrawal in summer season when the surface water does not meet the rising demand.”

Higher than what?

Comment 12

Page 5, Lines 135-136 “To disentangle the impacts of climate change and population growth and its combined effects on future aquifer condition and hydrologic regime, five scenarios were developed (Table 4)”

I only see 4 scenarios in table 4. Please fix the table and also explain the  symbol *.

Comment 13

Page 9, Lines 236-237

“Although annual extreme flows mainly experience a lower degree of change in the near and far future (Figure 3 and 4),..”

Should there be figure 4?

Comment 14

Table 8.

Please write the relative years under each of the two future periods.

Comment 15

Figure 3.

Please improve coloring. It is very hard to distinguish as is.