The effects of rain and evapotranspiration statistics on groundwater recharge estimations under semi-arid environments

Turkeltaub, Tuvia; Bel, Golan

doi:https://doi.org/10.5194/hess-2022-257

Preprints

https://doi.org/10.5194/hess-2022-257

Preprints

15 Aug 2022

The effects of rain and evapotranspiration statistics on groundwater recharge estimations under semi-arid environments

Tuvia Turkeltaub¹ and Golan Bel² Tuvia Turkeltaub and Golan Bel Tuvia Turkeltaub¹ and Golan Bel²

Show author details

¹Zuckerberg Institute for Water Research, Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus 8499000, Israel
²Department of Solar Energy and Environmental Physics, Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus 8499000, Israel

Received: 06 Jul 2022 – Discussion started: 15 Aug 2022

Abstract. Better understanding the effects of rainfall and evapotranspiration statistics on groundwater recharge requires long time series of these climate variables. However, long records of the relevant variables are scarce. To overcome this limitation, time series of the rainfall and evapotranspiration are often being synthesized using different methods. Here, we attempt to study the dependence of estimated groundwater recharge on the synthesis methods. We focus on regions with semi-arid climate conditions and soil types. For this purpose, we used long records of climate data that were measured in two semi-arid locations with similar seasonal and annual potential evapotranspiration rates but different seasonal rain distributions. Stochastic daily rain and potential evapotranspiration time series were synthesized according to the monthly empirical distributions. This suggested synthesis method preserves only certain aspects of the measured statistics. Therefore, different correction methods were considered to match the synthesized time series to the measured annual or monthly statistics. Groundwater recharge fluxes were calculated using the 1D Richards equation, for several typical semi-arid soil types, and by prescribing the synthesized rain and potential evapotranspiration as atmospheric conditions. The estimated groundwater recharge fluxes are sensitive to the synthesis method. However, the ratio between the groundwater recharge and the total rain does not show the same sensitivity. The effects of the synthesis methods are shown to be independent of the data used as observations. These findings suggest that assessment of groundwater recharge under current and future climate conditions depend on the synthesis method used for rain and evapotranspiration.

Download & links

Preprint (PDF, 1127 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (1127 KB)

Supplement (1373 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (1127 KB)
Supplement (1373 KB)
BibTeX
EndNote

Journal article(s) based on this preprint

Tuvia Turkeltaub and Golan Bel

Interactive discussion

Status: closed

CC1:
'Comment on hess-2022-257', Tongchao Nan, 20 Sep 2022

It is common to synthetically generate essential data series including rain and evaporation based on limited station record in practice. This paper aimed at the impact of how these synthetic series are generated/corrected and investigated several methods such as fixed number of rainy days, monthly averaging, annual averaging etc (five synthesis methods for rain and seven for ETref). These synthetic data are then used in predicting groundwater recharge in one-dimensional Hydrus models with different soil types, and factors influencing groundwater recharges including methods to synthesize data as well as climate and soil type, are analyzed. The results indicated that groundwater recharge statistics indeed depend on data statistics perserved in synthesis methods. It brings attention to the synthesis methods to generate rain and ETref in groundwater applications. I do believe that the conclusions in this paper are very important in a broad community.

However, I agree that the authors are onto a good idea and the manuscript is well written, clear and detailed.

My only major concern is that: this study bears a significant limitation of considering only perfectly uniform soil, which greatly constrains the potential values of this study, as is also stated by the authors. If not so, I'd recommend acceptance of this manuscript without hesitation. Hence, it is highly suggested that the authors take factors such as topography and soil structure into account in their ongoing work.
Specific comments:

Lines 28-29: Gurdak et al. 2007 apeared twice. Are these two references the same?

Citation: https://doi.org/10.5194/hess-2022-257-CC1
- AC1: 'Reply on CC1', Golan Bel, 21 Sep 2022
  
  Dear Tongchao Nan,
  Thank you very much for the careful reading of the MS and for the encouraging comments.
  Indeed, the work is limited to homogeneous soil and the important effects of heterogeneity are not accounted for. However, as you mentioned, this work focus on the effects of the climate conditions synthesis methods on the simulated groundwater recharge. We felt that adding the complexity of heterogeneity will be overloading the MS. In fact, nowadays we are working on the effects of the heterogeneity in similar scenarios and in multiple semi-arid locations worldwide.
  Thank you for finding our typo with the reference in lines 28-29. We will correct that in the final version of the MS.
  
  Thanks again,
  Tuvia and Golan
  
  Citation: https://doi.org/10.5194/hess-2022-257-AC1
RC1:
'Comment on hess-2022-257', Tongchao Nan, 23 Sep 2022

It is common to synthetically generate essential data series including rain and evaporation based on limited station record in practice. This paper aimed at the impact of how these synthetic series are generated/corrected and investigated several methods such as fixed number of rainy days, monthly averaging, annual averaging etc (five synthesis methods for rain and seven for ETref). These synthetic data are then used in predicting groundwater recharge in one-dimensional Hydrus models with different soil types, and factors influencing groundwater recharges including methods to synthesize data as well as climate and soil type, are analyzed. The results indicated that groundwater recharge statistics indeed depend on data statistics perserved in synthesis methods. It brings attention to the synthesis methods to generate rain and ETref in groundwater applications. I do believe that the conclusions in this paper are very important in a broad community.

However, I agree that the authors are onto a good idea and the manuscript is well written, clear and detailed.

My only major concern is that: this study bears a significant limitation of considering only perfectly uniform soil, which greatly constrains the potential values of this study, as is also stated by the authors. If not so, I'd recommend acceptance of this manuscript without hesitation. Hence, it is highly suggested that the authors take factors such as topography and soil structure into account in their ongoing work.

Specific comments:

Lines 28-29: Gurdak et al. 2007 apeared twice. Are these two references the same?

Citation: https://doi.org/10.5194/hess-2022-257-RC1
- AC2: 'Reply on RC1', Golan Bel, 08 Dec 2022
  
  Dear Reviewer,
  Thank you for the positive assessment of our MS and for your helpful comments.
  Reply to general comment 1:
  
  Indeed, the work is limited to homogeneous soil, and the important effects of heterogeneity are not accounted for. However, as you mentioned, this work focus on the effects of the climate conditions synthesis methods on the simulated groundwater recharge. We felt that adding the complexity of heterogeneity would overload the MS. In future studies, we will address the effects of heterogeneity in similar scenarios and in multiple semi-arid locations worldwide.
  Reply to specific comment 1:
  Thank you for finding our typo with the reference in lines 28-29. We corrected it in the revised MS.
  
  Citation: https://doi.org/10.5194/hess-2022-257-AC2
RC2:
'Comment on hess-2022-257', Anonymous Referee #2, 03 Dec 2022

This study identified the important characteristics of the local rain and ETref for estimating the diffuse recharge under semi-arid climate conditions. I consider that the evaluation of the synthesis methods for rain and ETref is a great contribution since these variables have great importance on GR models. My only concern is the assumption of a homogeneous soil and only two locations considered, but they bear in mind this situation. However, some corrections are needed to improve the quality of the manuscript. Detailed corrections are presented in the pdf file.

Citation: https://doi.org/10.5194/hess-2022-257-RC2
- AC3: 'Reply on RC2', Golan Bel, 08 Dec 2022
  
  Dear Reviewer,
  We thank you for the careful assessment of our manuscript and for the constructive criticism. Below we address the reviewers' comments one by one. For clarity, we use regular font for the quoted reviewer comments and italicized font for our responses.
  General comments
  General comment 1: The abstract is too popular and not scientific enough to reflect the conclusion of the manuscript. Some sentences should be rewritten and supported by data.
  Reply to general comment 1: We revised the abstract and added, in multiple places, specific results and data values in relevant sentences. Specifically, the revisions appear in lines 5-7,9-10 and 14-16.
  General comment 2: In the introduction, the necessity of this study must be stated more explicitly.
  Reply to general comment 2: We added text to better clarify and emphasize the goals of the current study. In lines 59-60, we added the following text: "Specifically, we test the sensitivity of estimated GR fluxes to the daily, monthly, and annual statistics of rain and ETref."
  General comment 3: In conclusion, it is necessary to reflect what is the scientific novelty of the results of your research. On the other hand, which synthesis method performed better for both rainfall and ETref?
  Reply to general comment 3: The text in the Introduction and Conclusion sections was revised to provide the missing information (lines 19-21 and 380-382).
  General comment 4: What is the importance of understanding rainfall and evapotranspiration characteristics on groundwater recharge?
  Reply to general comment 4: Understanding the effects of rainfall and evapotranspiration characteristics on groundwater recharge is important for both fundamental understanding of the underlying physical processes and for groundwater management under future climate conditions. The text was revised to include this information (lines 19-21).
  
  Specific comments
  Comment 1: In line 1, you should eliminate the word climate in order to be clearer, since only the factor rainfall is boarded.
  Reply Comment 1: The word has been removed as suggested.
  Comment 2: In lines 37-38, this sentence should be added to the next paragraph. Since I understand that paragraph explains the Poison approach.
  Reply Comment 2: The suggestion was implemented in the revised MS (line 36).
  Comment 3: In lines 59-61, “To 60 overcome this issue, several new methods, preserving different characteristics of the measured rainfall and ETref records, are applied”, you must be more
  
  specific, which methods did you employ for rainfall and ETref?
  Reply Comment 3: The full details of all the methods implemented are provided in the Methods section and throughout the MS. It would be impossible to provide the full details in the Introduction. The text has been revised in order to refer the reader to the section with the full details (lines 59-60, 63).
  Comment 4: In lines 156-158, “the two- sample Kolmogorov-Smirnov test indicated that the synthesized and the measured distributions are statistically similar for the DS, ETDS, and ETWD
  
  methods”, which is the value of the KS test for this affirmation?
  Reply Comment 4: We now provide Table S1, in the Supplementary Information, that details the p-values of the KS tests for each method and location (line 162).
  Comment 5: In line 325, “We find that there are high correlations between the annual rain and the annual GR for sandy…”. Please specify which correlation criteria is used, it means which values
  
  correspond to high correlations. Also add these values to the sentences. Please, review this
  
  in the manuscript.
  Reply Comment 5: The text has been revised to provide the correlation coefficient values (lines 329-330). In addition, we wish to mention that these details appear in Figures S15 and S16 of the Supplementary Information.
  Comment 6: In line 361, “Here, we considered five different methods for rain synthesis and seven methods for ETref synthesis”. In total you applied 12 synthesis methods, however, in Table 1 only appear 11. Please correct this.
  Reply Comment 6: The UDDS method specifies the synthesis methods of both the rain and the ETref (both are uniformly distributed over the month). This information appears in Table 1, and we added text to emphasize this (lines 365-366).
  Comment 7: In subsection 2.2 Generation of rain and ETref time series. Consider separate rain and ETref in two subsections (e.g., 2.2. and 2.3) to facilitate the reading.
  Reply Comment 7: Since the rain and ETref time series are generated by similar synthesis methods, we would prefer to preserve the original structure of the Methods section.
  Comment 8: In line 50, you employed ETref for potential evapotranspiration. However, in line 71 you refer for reference evapotranspiration. Take this into account and make the corresponding corrections throughout the manuscript
  Reply Comment 8: Thanks for catching this. The text has been revised (lines 51-52), and the term ETref now denotes the potential evapotranspiration throughout the revised manuscript.
  Comment 9: In Figure 5, I suggest the use of letters (A,B,C,…) such as in Figure 8, to facilitate the reading and location of the graphic. For example, in lines 263-265 “Figure 5 top right panel” could be hard to follow.
  Reply Comment 9: Figure 5 has been revised to include panel labels, and the text was revised to refer to these labels.
  
  Citation: https://doi.org/10.5194/hess-2022-257-AC3

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Publish subject to minor revisions (further review by editor) (09 Dec 2022) by Gerrit H. de Rooij

AR by Golan Bel on behalf of the Authors (09 Dec 2022) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (09 Dec 2022) by Gerrit H. de Rooij

Interactive discussion

Status: closed

CC1:
'Comment on hess-2022-257', Tongchao Nan, 20 Sep 2022

It is common to synthetically generate essential data series including rain and evaporation based on limited station record in practice. This paper aimed at the impact of how these synthetic series are generated/corrected and investigated several methods such as fixed number of rainy days, monthly averaging, annual averaging etc (five synthesis methods for rain and seven for ETref). These synthetic data are then used in predicting groundwater recharge in one-dimensional Hydrus models with different soil types, and factors influencing groundwater recharges including methods to synthesize data as well as climate and soil type, are analyzed. The results indicated that groundwater recharge statistics indeed depend on data statistics perserved in synthesis methods. It brings attention to the synthesis methods to generate rain and ETref in groundwater applications. I do believe that the conclusions in this paper are very important in a broad community.

However, I agree that the authors are onto a good idea and the manuscript is well written, clear and detailed.

My only major concern is that: this study bears a significant limitation of considering only perfectly uniform soil, which greatly constrains the potential values of this study, as is also stated by the authors. If not so, I'd recommend acceptance of this manuscript without hesitation. Hence, it is highly suggested that the authors take factors such as topography and soil structure into account in their ongoing work.
Specific comments:

Lines 28-29: Gurdak et al. 2007 apeared twice. Are these two references the same?

Citation: https://doi.org/10.5194/hess-2022-257-CC1
- AC1: 'Reply on CC1', Golan Bel, 21 Sep 2022
  
  Dear Tongchao Nan,
  Thank you very much for the careful reading of the MS and for the encouraging comments.
  Indeed, the work is limited to homogeneous soil and the important effects of heterogeneity are not accounted for. However, as you mentioned, this work focus on the effects of the climate conditions synthesis methods on the simulated groundwater recharge. We felt that adding the complexity of heterogeneity will be overloading the MS. In fact, nowadays we are working on the effects of the heterogeneity in similar scenarios and in multiple semi-arid locations worldwide.
  Thank you for finding our typo with the reference in lines 28-29. We will correct that in the final version of the MS.
  
  Thanks again,
  Tuvia and Golan
  
  Citation: https://doi.org/10.5194/hess-2022-257-AC1
RC1:
'Comment on hess-2022-257', Tongchao Nan, 23 Sep 2022

It is common to synthetically generate essential data series including rain and evaporation based on limited station record in practice. This paper aimed at the impact of how these synthetic series are generated/corrected and investigated several methods such as fixed number of rainy days, monthly averaging, annual averaging etc (five synthesis methods for rain and seven for ETref). These synthetic data are then used in predicting groundwater recharge in one-dimensional Hydrus models with different soil types, and factors influencing groundwater recharges including methods to synthesize data as well as climate and soil type, are analyzed. The results indicated that groundwater recharge statistics indeed depend on data statistics perserved in synthesis methods. It brings attention to the synthesis methods to generate rain and ETref in groundwater applications. I do believe that the conclusions in this paper are very important in a broad community.

However, I agree that the authors are onto a good idea and the manuscript is well written, clear and detailed.

My only major concern is that: this study bears a significant limitation of considering only perfectly uniform soil, which greatly constrains the potential values of this study, as is also stated by the authors. If not so, I'd recommend acceptance of this manuscript without hesitation. Hence, it is highly suggested that the authors take factors such as topography and soil structure into account in their ongoing work.

Specific comments:

Lines 28-29: Gurdak et al. 2007 apeared twice. Are these two references the same?

Citation: https://doi.org/10.5194/hess-2022-257-RC1
- AC2: 'Reply on RC1', Golan Bel, 08 Dec 2022
  
  Dear Reviewer,
  Thank you for the positive assessment of our MS and for your helpful comments.
  Reply to general comment 1:
  
  Indeed, the work is limited to homogeneous soil, and the important effects of heterogeneity are not accounted for. However, as you mentioned, this work focus on the effects of the climate conditions synthesis methods on the simulated groundwater recharge. We felt that adding the complexity of heterogeneity would overload the MS. In future studies, we will address the effects of heterogeneity in similar scenarios and in multiple semi-arid locations worldwide.
  Reply to specific comment 1:
  Thank you for finding our typo with the reference in lines 28-29. We corrected it in the revised MS.
  
  Citation: https://doi.org/10.5194/hess-2022-257-AC2
RC2:
'Comment on hess-2022-257', Anonymous Referee #2, 03 Dec 2022

This study identified the important characteristics of the local rain and ETref for estimating the diffuse recharge under semi-arid climate conditions. I consider that the evaluation of the synthesis methods for rain and ETref is a great contribution since these variables have great importance on GR models. My only concern is the assumption of a homogeneous soil and only two locations considered, but they bear in mind this situation. However, some corrections are needed to improve the quality of the manuscript. Detailed corrections are presented in the pdf file.

Citation: https://doi.org/10.5194/hess-2022-257-RC2
- AC3: 'Reply on RC2', Golan Bel, 08 Dec 2022
  
  Dear Reviewer,
  We thank you for the careful assessment of our manuscript and for the constructive criticism. Below we address the reviewers' comments one by one. For clarity, we use regular font for the quoted reviewer comments and italicized font for our responses.
  General comments
  General comment 1: The abstract is too popular and not scientific enough to reflect the conclusion of the manuscript. Some sentences should be rewritten and supported by data.
  Reply to general comment 1: We revised the abstract and added, in multiple places, specific results and data values in relevant sentences. Specifically, the revisions appear in lines 5-7,9-10 and 14-16.
  General comment 2: In the introduction, the necessity of this study must be stated more explicitly.
  Reply to general comment 2: We added text to better clarify and emphasize the goals of the current study. In lines 59-60, we added the following text: "Specifically, we test the sensitivity of estimated GR fluxes to the daily, monthly, and annual statistics of rain and ETref."
  General comment 3: In conclusion, it is necessary to reflect what is the scientific novelty of the results of your research. On the other hand, which synthesis method performed better for both rainfall and ETref?
  Reply to general comment 3: The text in the Introduction and Conclusion sections was revised to provide the missing information (lines 19-21 and 380-382).
  General comment 4: What is the importance of understanding rainfall and evapotranspiration characteristics on groundwater recharge?
  Reply to general comment 4: Understanding the effects of rainfall and evapotranspiration characteristics on groundwater recharge is important for both fundamental understanding of the underlying physical processes and for groundwater management under future climate conditions. The text was revised to include this information (lines 19-21).
  
  Specific comments
  Comment 1: In line 1, you should eliminate the word climate in order to be clearer, since only the factor rainfall is boarded.
  Reply Comment 1: The word has been removed as suggested.
  Comment 2: In lines 37-38, this sentence should be added to the next paragraph. Since I understand that paragraph explains the Poison approach.
  Reply Comment 2: The suggestion was implemented in the revised MS (line 36).
  Comment 3: In lines 59-61, “To 60 overcome this issue, several new methods, preserving different characteristics of the measured rainfall and ETref records, are applied”, you must be more
  
  specific, which methods did you employ for rainfall and ETref?
  Reply Comment 3: The full details of all the methods implemented are provided in the Methods section and throughout the MS. It would be impossible to provide the full details in the Introduction. The text has been revised in order to refer the reader to the section with the full details (lines 59-60, 63).
  Comment 4: In lines 156-158, “the two- sample Kolmogorov-Smirnov test indicated that the synthesized and the measured distributions are statistically similar for the DS, ETDS, and ETWD
  
  methods”, which is the value of the KS test for this affirmation?
  Reply Comment 4: We now provide Table S1, in the Supplementary Information, that details the p-values of the KS tests for each method and location (line 162).
  Comment 5: In line 325, “We find that there are high correlations between the annual rain and the annual GR for sandy…”. Please specify which correlation criteria is used, it means which values
  
  correspond to high correlations. Also add these values to the sentences. Please, review this
  
  in the manuscript.
  Reply Comment 5: The text has been revised to provide the correlation coefficient values (lines 329-330). In addition, we wish to mention that these details appear in Figures S15 and S16 of the Supplementary Information.
  Comment 6: In line 361, “Here, we considered five different methods for rain synthesis and seven methods for ETref synthesis”. In total you applied 12 synthesis methods, however, in Table 1 only appear 11. Please correct this.
  Reply Comment 6: The UDDS method specifies the synthesis methods of both the rain and the ETref (both are uniformly distributed over the month). This information appears in Table 1, and we added text to emphasize this (lines 365-366).
  Comment 7: In subsection 2.2 Generation of rain and ETref time series. Consider separate rain and ETref in two subsections (e.g., 2.2. and 2.3) to facilitate the reading.
  Reply Comment 7: Since the rain and ETref time series are generated by similar synthesis methods, we would prefer to preserve the original structure of the Methods section.
  Comment 8: In line 50, you employed ETref for potential evapotranspiration. However, in line 71 you refer for reference evapotranspiration. Take this into account and make the corresponding corrections throughout the manuscript
  Reply Comment 8: Thanks for catching this. The text has been revised (lines 51-52), and the term ETref now denotes the potential evapotranspiration throughout the revised manuscript.
  Comment 9: In Figure 5, I suggest the use of letters (A,B,C,…) such as in Figure 8, to facilitate the reading and location of the graphic. For example, in lines 263-265 “Figure 5 top right panel” could be hard to follow.
  Reply Comment 9: Figure 5 has been revised to include panel labels, and the text was revised to refer to these labels.
  
  Citation: https://doi.org/10.5194/hess-2022-257-AC3

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Publish subject to minor revisions (further review by editor) (09 Dec 2022) by Gerrit H. de Rooij

AR by Golan Bel on behalf of the Authors (09 Dec 2022) Author's response Author's tracked changes Manuscript

ED: Publish subject to technical corrections (09 Dec 2022) by Gerrit H. de Rooij

Journal article(s) based on this preprint

Tuvia Turkeltaub and Golan Bel

Supplement

https://doi.org/10.5194/hess-2022-257-supplement

Tuvia Turkeltaub and Golan Bel

Viewed

Total article views: 622 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
489	117	16	622	33	4	4

HTML: 489
PDF: 117
XML: 16
Total: 622
Supplement: 33
BibTeX: 4
EndNote: 4

Views and downloads (calculated since 15 Aug 2022)

Month	HTML	PDF	XML	Total
Aug 2022	249	59	4	312
Sep 2022	71	23	6	100
Oct 2022	70	6	0	76
Nov 2022	39	9	0	48
Dec 2022	59	19	6	84
Jan 2023	1	1	0	2

Cumulative views and downloads (calculated since 15 Aug 2022)

Month	HTML	PDF	XML	Total
Aug 2022	249	59	4	312
Sep 2022	71	23	6	100
Oct 2022	70	6	0	76
Nov 2022	39	9	0	48
Dec 2022	59	19	6	84
Jan 2023	1	1	0	2

Viewed (geographical distribution)

Total article views: 539 (including HTML, PDF, and XML) Thereof 539 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 12 Jan 2023

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (1127 KB)
Metadata XML

Short summary

Groundwater is an essential resource affected by climate conditions and anthropogenic activities. Estimations of groundwater recharge under current and future climate conditions require long-term climate records that are scarce. Different methods to synthesize climate data, based on observations, are used to estimate groundwater recharge. In terms of groundwater recharge estimation the best synthesis method is based on the daily statistics corrected to match the observed monthly statistics.


Total:	0
HTML:	0
PDF:	0
XML:	0