Sandtank-ML: An Educational Tool at the Interface of Hydrology and Machine Learning

Round 1

Reviewer 1 Report

The manuscript presents the new tool of sandtank and use of machine learning used as computer simulation of a physical aquifer model. The manuscript is well written and easy to read. The topic is more related for edutors than for scientists. However, the tool can be useful for the young researchers. Therefore, I recommend to published the manuscript with minor changes. Please, consider the following points: 

• Use of abbreviations in the introduction part, when it is used onlu once, e.g., line 61 "GW" or name of institute, which should not be clear to all
• Rename chapter Discussion to Discussion and Conclusions
• Replace some of your references or add some other, because there is almost 35% of literature published by one of the co-authors.

Author Response

We thank the reviewer for their constructive comments and for thoughtfully highlighting the utility of this work for educators and young researchers.

Point 1: Use of abbreviations in the introduction part, when it is used only once, e.g., line 61 "GW" or name of institute, which should not be clear to all

Response 1: We appreciate this reviewer’s comment and have corrected the abbreviation “GW” to “groundwater.” After reviewing the document, we have also defined “ENSO” (El Niño-Southern Oscillation, line 72); “AI” (artificial intelligence, line 56); “NSF” (National Science Foundation, line 98) to ensure readers are informed of the meaning of acronyms at their first use.

Point 2: Rename chapter Discussion to Discussion and Conclusions

Response 2: The authors have renamed the Discussion chapter to Discussion and Conclusions.

Point 3: Replace some of your references or add some other, because there is almost 35% of literature published by one of the co-authors.

Response 3: We thank the reviewer for their perspective on this. Unfortunately, because this is a methods paper and the approach we are using was developed by the authors, a number of references describing the underlying simulations approach (ParFlow) are self-citing. Excessive self-citation was not our intent, and therefore we tried to be judicious and removed any unnecessary self-citations where possible.

Reviewer 2 Report

It is very important the use of educational tools in order to improve teaching activities but, additionally to prepare our students and researchers for a better research. In this case, the review of the article entitled “Sandtank-ML: An educational tool at the interface of hydrology and machine learning” has the aim to have a look from this point of view, the utility for research and education. However, the review of the article focuses more on the value of the tool for research, rather than for teaching, given the purpose of the journal.

First of all, two things for discussion, the concept of anthropocene and the use of machine learning (ML), both basic in the article. I agree with authors about the concept and the promising use of ML, although I am sure that, especially anthropocene, is going to still on the scientific community discussion for years. Considering ML, it is still a tool basically used by researchers and it is becoming an educational tool in the recent years. The Sandtank-ML, in this sense, is an interesting tool for researchers to know basic concepts and fit wells with the aims and scope of the journal, but it will be improved if some mathematical details will be added as a new step, not only to “gain confidence in the process and technology”. It is good to know that authors are preparing new versions: “Multiple supplementary versions of the application are under development that allow users to explore more advanced topics”.

The use machine learning to know how extreme events act and how to manage possible scenarios, is becoming critical and necessary for all. This is a strength of this article, even considering the climate change.

As suggestions, to improve the paper if authors consider them, are the following:

A short introduction/concept about what the anthropocene is and a (some) reference(s) may be given, as well as a reputed definition of machine learning.

Please take care about the use of hyphens to facilitate the reading, for instance in line 59, hydrolog-ical, it is easy to read if the hyphen at the end of the line is like “hydrologi-cal or hydrolo-gical”. Some similar situations are found in lines 67, 68, 80, 92, 93, 119, 178, 262, 296, 407.

Check this in line 134, “and” because I do not know the reason to write it in italics.

Author Response

We are grateful for the reviewer’s comments and in-depth discussion on future improvements to the application. We acknowledge that the addition of “some mathematical details” as additional options in versions moving forward would serve users well and will consider this as we continue application development.

Point 1: A short introduction/concept about what the anthropocene is and a (some) reference(s) may be given, as well as a reputed definition of machine learning.

Response 1: We thank the reviewer for their comment and realize this term took away from our main messaging. While there is a rich debate on the Anthropocene, this is not central to the message we are conveying in this manuscript and have therefore removed this term throughout the document. Additionally, we have edited the definition of machine learning used to introduce the topic; this definition is referenced in the manuscript (reference 6: Jordan and Mitchell 2015).

Point 2: Please take care about the use of hyphens to facilitate the reading, for instance in line 59, hydrolog-ical, it is easy to read if the hyphen at the end of the line is like “hydrologi-cal or hydrolo-gical”. Some similar situations are found in lines 67, 68, 80, 92, 93, 119, 178, 262, 296, 407.

Response 2: We understand and appreciate this comment and will bring this up with MDPI during typesetting. It may be a result of the MDPI template and we agree that it is frustrating.

Point 3: Check this in line 134, “and” because I do not know the reason to write it in italics.

Response 3: We apologize, there was no reason for the use of italics in this case. We have removed italics from “and” in line 134.

Reviewer 3 Report

Este artículo bien trabajado no presenta la forma protocolaria de un artículo científico, si bien contiene una información, descriptiva científicamente, de un programa de aplicación práctica relevante en el ámbito actual.

Dada sus novedades o aspectos pragmáticos para la educación, resulta de interés su publicación, por el valor de la difusión científica que contiene.

Sería deseable pedir a los autores que cumplimentaron, en un futuro, la información con investigaciones que puedan demostrar su utilidad con algo más de los dos estudios de caso que se presentan en el trabajo actual.

This well-worked article does not present the formal form of a scientific article, although it does contain scientifically descriptive information on a relevant practical application program in the current field.

Given its novelties or pragmatic aspects for education, its publication is of interest, due to the value of the scientific dissemination it contains.

It would be desirable to ask the authors who, in the future, completed the information with research that can demonstrate its usefulness with something more than the two case studies that are presented in the current work.

Author Response

We thank you for the thoughtful comments and appreciate that this manuscript may not fit the typical form of a scientific article, but agree that it would be a publication of value to users in education and research.

 Point 1: It would be desirable to ask the authors who, in the future, completed the information with research that can demonstrate its usefulness with something more than the two case studies that are presented in the current work.

Response 1: Thank you for your comment, to the extent that we understand. We started with the two case studies in this manuscript, then offer open access to this application for users to interact with the interface, add to the collection of case studies, and further development. We see this manuscript as an opportunity to let people know that they can use and develop this open-source educational and research tool, therefore feel its publication is warranted.