A Contamination Predictive Model for Escherichia coli in Rural Communities Dug Shallow Wells

Round 1

Reviewer 1 Report

The article, entitled "A Contamination predictive model for Escherichia coli in rural communities dug shallow wells", presented for review, concerns a very interesting and socially important aspect of the environmental impact of well surroundings on the safety of people using them.

The presentation of the article does not raise any objections, the number of self-citations is not exceeded, the cited articles are mostly new studies.

The construction of the experiments also does not raise any major objections, as does the presentation and description of the results.

During the reading of the manuscript, several inaccuracies were found and therefore a request for clarification or correction.

1. I will ask for accurate data, e.g. in the methodological part of the GPS locating of sampling.

2. Please align the type of fonts in figures/tables because there is a little chaos here.

3. The authors did not specify how the samples were collected and how they were analyzed. Are you sure the same reference methods were used in all cases? If YES, please indicate it clearly in the methodological part, if NO, I don't know if such results can be compared!

4. The article promises to be interesting, but actually already in the abstract we learn that after applying complex analyzes the results are not surprising, or even obvious. I am concerned that the impact of this study on the scientific community will be negligible. According to the reviewer, the article in this form does not contribute much to science. Maybe it would be worth remodeling it into a more methodical study, or with a comparison of the fit of different models?

Author Response

The questions marked as Can be improved  and Must be improved, were improved. Are highlighted in the text.

point 1 - answered. Maps with all 25 communities were included as supplementary material, indicating the 115 collection points

point 2 - answered

point 3 - the details of the collections were included in the methodology

point 4 - The job description was better presented, including the number of communities and households involved in the research. This is a study carried out in Brazil, where there is a lack of research in this line, being admitted by the population of several rural communities that groundwater is completely uncontaminated and can be consumed without prior treatment.

Reviewer 2 Report

Shallow and dug wells are widely used not only in Brazil, but also in other countries. Their potential dangers are well known and, mostly, in the areas where they occur, the technical preparation is not adequate for a comprehensive, analytical investigation, which is why the set goal is extremely useful: "predict the contamination of a given water source can often be less expensive and offering better efficiency", even if other researchers have already raised the importance of this.

In order to eliminate the inadequacy of the Generalized Linear Models (GLM) already used by others for this purpose, the authors aimed to set up a GLM that can predict the probability of Escherichia coli infection in dug wells depending on the predictive environmental variables.

The given methodology is understandable, transparent (please choose another colour instead of yellow) and processes a sufficient number of samples to set up the model.

Most of the findings are not surprising and not unexpected, e.g. the diameter of the well or the effect of the proximity of the pig farm, but in any case the quantified, statistically supported value represents a significant value and importance.

The conclusions drawn are correct, the developed model needs further improvement, but the results confirm the correctness of the chosen methods.

It would also be advisable to investigate the effect of the depth of the well.

Author Response

A apresentação do modelo foi melhor realizada e uma discussão foi feita com a profundidade do poço conforme solicitado pelo revisor. Um melhor detalhamento do desenho da pesquisa também foi realizado.

Round 2

Reviewer 1 Report

Thanks for the explanations and the corrected version, which looks very good in this form. I have no more comments.

Author Response

Thanks for the review