Model-Based Assessment of Preventive Drought Management Measures’ Effect on Droughts Severity

Round 1

Reviewer 1 Report

The research is very interesting and important, not only the Torola River Basin area.

The materials and methods are unfortunately difficult to read. It may not be correctly understood by the reader. One of the characteristics of a scientific article is to be readable and easy to understand so that other scientists and practitioners can use the method described.

Please take my comments into account when improving the manuscript:

 

1. There are technical errors in the manuscript, e.g. hidrological (in Fig. 1), Table 3 has no title, there are author's notes, in Figures 11, and 14, units are not described.
2. In Figure 3 place rainfall and streamflow on a single horizontal axis. In its current form, the graph is illegible and difficult to understand. 
3. All abbreviations derived from the SWAT model need to be explained, e.g. POT_FR, POT_VOLX, PND_FR, PND_PVOL, PND_PSA (Table 2).
4. Was the monthly soil moisture in the base model verified?
5. Table 2 shows the areas where changes have been made. Is there information on how many RWH ponds and check dams and ponds were introduced into the SWAT model?
6. Did the author of the model influence the location of PDMMs? Were parameters other than environmental parameters taken into account?
7. Drought severity shown as a deviation from the threshold level in mm is difficult to perceive. The note refers to Figures 5 and 6.
8. 1575 km2 is a small study area, is it possible to do the analysis for a larger area?

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

General comments

The paper uses SWAT model and threshold method for drouht identification in the Torola River basin, El Salvador.

This study is based on one hydrologic station and a non-referred number of meteo stations. Also on a DEM map, a FAO soil map of the world and also a FAO land use map.

As stressed in the conclusions a better characterization of agriculture (population,irrigation, crops, timings) is needed and longer series of meteo and hydrologic data, as well as a more dense network (or re-analysis data).

 

Some concepts need to be clarified

Drought concepts. Drough definition, as well as drought event, in the context of the sudy.

Please note that

A dry season may be known as "mid summer drought" but it is not a drought.

From a general point of view, and using the month as time scale, one or two months below the 20th percentile or another threshold level (variable soil water storage or streamflow) possibly will not be identified as a drought or drought event.

See your reference  (here the the time step is one day)

B. Heudorfer & K. Stahl | Comparison of different threshold level methods for drought propagation analysis Hydrology Research | 48.5 | 2017

 

Corrections

pg 1

Abstract

lines 11, 12 – time period of the evaluation?

 

pg 3

lines 93  - the average (period?) annual evapotranspiration is not 186 mm. Be aware of providing published values from questionable sources

See fig 7 from  WILFRIED BRUTSAERT (2020). Spatial Distribution of Global Landscape Evaporation in the Early Twenty-First Century by Means of a Generalized Complementary Approach. Journal of Hydrometeorology, vol 21,2,  2020 AMS

 

pg 3

line 105  2.2.1. Model set-up

line 125  2.2.2. Model set-up repeated  (check)

line 123

Table 1  - include the number of rainfall and temperature stations; include the source references of published or on-line data (in references)

 

Above line 157

Include the definition of the parameters modeled by the Pothole routine and its final value; also for the Pond routine

For measures allocation provide the number of RHW ponds and the number of Check dams and ponds

 

Line 200

Eqs 5 and 6     Please check, a subscrit i is lacking? The duration refers to each sub-basin i?

 

Line 224

Table 3  Title of the table

 

In section 3.2.

Sugestion: compile and present the duration of each agr. and hydr. drought event; it seems that you have

agric. drought  26 months/15 years

hydrol. drought   13 months/15 years

 

Line 244

In fig 4 Agricultural droughts in event VII - check months

 

pgs 8, 9  and others

The text is difficult to read; it is necessary to go back to fig 4 to remember the numbers/names of each agricultural and hydrological event;

suggestion : refer the events in the text by the time period

 

line 241 – corresponds to period Oct2008 – Apr2009 which is also the longer with a 7 month duration

 

lines 243 – question:  what happens in August-September 2015? the hydrologic drought, with high severity, disappear suddenly!! Can you comment in the text please

 

lines 246, 247    Figure 5. Monthly agricultural drought severity in mm(base line scenario).

cut the rest it is not necessary

lines 248, 249    Figure 6.  the same

 

lines 256, 257 – indicate the total number of RWH ponds, here or before

 

line 295 Figure 7. It would be much more clear if you have in the x- axis date as month-year;

or in the caption …are applied: Oct05-Jan06; Oct2008-Apr2009    and so on…

 

Conclusions

As stressed in the conclusions a better characterization of agriculture (population, irrigation, crops, timings, observed periods of crop stress when cropping is possible) is needed and longer series of meteo and hydrologic data, as well as a more dense meteo network (or re-analysis data).

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper discusses an interesting case-study, comparing the impacts of different measures for drought mitigation on agricultural and hydrological droughts.

The list of references is very broad and the methodology of the research well detailed.

The following suggestions can improve the paper:

Lines 53-56: Revise the statement as the lack of coordination is a feature different from the lack of assessment of the long-term impacts of the preventive measures

Line 76: Check if it is opportune to cite here the drought severity indices used for evaluating baseline and PDMMs  scenarios

Line 93: Check if 186 mm is the correct value of average EPT. It seems very low

Table 2: Explain the meaning of “200-1000 mm” (is the rainfall stored in the pound?

Line 154: It is suggested to add a short statement about the measure “ check dam and pond” with some detail

Lines 257-261: The hypothesis that the agriculture drought severity is lower during the rainy season when RWH pond is applied, requires that infiltration is increasing, This occurs if the bottom of pond is permeable. It is required to discuss better the hypothesis. In particular it is necessary to explain the possible effects of irrigation in the sub-basin

Lines 273-276: The comment about  the hydrological severity reduction seems present some contradictions . It is suggested to give more details.

Fig.10 -11: Since the location of measures (presented at the bottom of Table 2) is complex, it is not easy to distinguish the effects on subbasins with the forest conservation measures and subbasins downstream. It is suggested to indicate also within Fig 10 or 11 the location of subbasins where the above cited measure is applied.

Fig. 13: The same suggestion is given for maps of subbasins where the check dam and ponds measure is applied.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 4 Report

This manuscript evaluates the effects of three potential PDMMs, RWH ponds, forest conservation, and check dams, on agricultural and hydrological drought severity in the Torola River Basin. Some of vital points are not well presented. I have several major concerns which should be addressed before the manuscript can be considered for publication.

 

Comments:

1.      The authors just validated the SWAT model by using the discharge data, without addressing the soil moisture, which would lead to a large uncertainty in evaluating agricultural drought. So the performance of simulating soil moisture should be involved in SWAT model evaluation. 

2.      In calibrating and validating the SWAT model, how about the performance of low flow simulation? It should be paid more attention since hydrological droughts are greatly related with the low flow.

3.      How to model the three PDMMs in SWAT was not clear. Only changing the model parameter values in Table 2 ? It is too simple to represent such complex PDMMs and thus may lead to less credible results. More detailed descriptions about the application of PDMMs in SWAT are needed.  

4.      Agricultural droughts are largely related to human activities, such as artificial irrigation, channel diversion. Ignoring such impacts may lead to the results of agricultural drought evaluation questioned.

5.      In setting the thresholds to identify droughts, why the twentieth percentile was determined? It should be explained more, since different percentiles may correspond to different drought event numbers. 

6.      Model outputs indicate that soil erosion decreases by up to 30% in the subbasins where RWH ponds are applied. How to get this conclusion? And how about its reliability?

7.      Table 3, the optimum parameter values should be given.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I have no comments. The work has been corrected by the guidelines.

Reviewer 4 Report

All of my comments have been addressed. There is no further comment.