A Regression Model of Stream Water Quality Based on Interactions between Landscape Composition and Riparian Buffer Width in Small Catchments

Round 1

Reviewer 1 Report

This paper present an interesting approach for exploring the relationship between landscape features and water quality based on the consideration of the interaction between riparian buffer strips and land use compostion. However, as compared to the many studies on the same theme, the dataset used is very small (8 catchments only). Moreover, the index they you to describe water quality seems to reduce its variability, and one of the independent variable of the regression is not really a quantitative variable. This tends to weaken the significance of the results and of the conclusions, although they seem fairly  plausible.

my suggestions to strengthen the study:

-add more landscape pattern indices in the analysis  (e.g.  average distance of arable land to streams, % of forest in valley bottoms, etc.)

-compute regressions for individual parameters rather than for IWQ only

-enlarge the possible interpretations of the results

more comments to be found in the annotated manuscript

Comments for author File: Comments.pdf

Author Response

REVIEWER #1

This paper present an interesting approach for exploring the relationship between landscape features and water quality based on the consideration of the interaction between riparian buffer strips and land use composition. However, as compared to the many studies on the same theme, the dataset used is very small (8 catchments only). Moreover, the index they you to describe water quality seems to reduce its variability, and one of the independent variable of the regression is not really a quantitative variable. This tends to weaken the significance of the results and of the conclusions, although they seem fairly  plausible.

Dear reviewer,

We very much appreciate your will to review our paper and we did our utmost to comply with your comments and suggestions.

We agree that the sample is not large, but please see that the target area – the Environmental Protection Area  (EPA) – is a relatively small headwater catchment occupied with a limited number of perennial sub-catchments. Within this framework, the option was not to include the largest number of sub-catchments, but to span the largest number of sub-catchments with the widest range of forest to agriculture ratios, buffer strip widths and IWQ indices within the EPA. Besides, it was important to span the entire EPA longitudinally. The result were 8 catchments. The choice for the IWQ was because it represents the most commonly used measure of water quality in Brazil (https://www.ana.gov.br/panorama-das-aguas/qualidade-da-agua/indicadores-de-qualidade), and therefore is readily understood by water planners, water authorities and the general public. In that context, the use of IWQ may help these players in decision making. Nevertheless, in the revised version we extended the regression analyses to the individual parameters. Besides, we related these latter results with issues independent from our main topic (interactions), such as spatial scale or antecedent rainfall conditions.

Please find below the replies to your remarks.

my suggestions to strengthen the study:

-add more landscape pattern indices in the analysis  (e.g.  average distance of arable land to streams, % of forest in valley bottoms, etc.).

Dear reviewer, we sought to do that but please remember that the focus of the paper was to test interactions among relevant parameters controlling water quality in streams. The interaction between two terms is frequently used in many studies, while interpretations are relatively straightforward based on performance indices or graphical approaches. However, the use of a larger number of landscape metrics would imply the development of higher order interaction models. As can be read in several papers and forums (see for example https://stattrek.com/multiple-regression/interaction.aspx), analysts usually steer clear of higher-order interactions, like X1X2X3, since they can be hard to interpret. This was the reason not to expand the analysis to a larger number of metrics.  

-compute regressions for individual parameters rather than for IWQ only.

Dear reviewer, in the revised version we replicated the regression analyses to the individual parameters, as suggested. We started to announce that in the Introduction, with the sentence:

“The regression models were first applied to the IWQ and then to its formation variables, with the purpose to identify the most influencing ones.”

Then we added the following sentence to the appropriate Methods sub-section

“In a second run, the regression analysis was replicated for water temperature, pH, dissolved oxygen, turbidity and total dissolved solids, which are the formation parameters of IWQ, to evaluate their specific roles in the studied area.”

Then, we represented (please see new Table 5) and described the results as follows:

The results of multiple regression applied to the IWQ parameters are depicted in Table 5. Only the regressions with interaction term were considered in this second run. The results suggest a dominance of turbidity in the control of IWQ in the studied sub-basins. For this parameter, the coefficient of determination (R² = 0.6) is satisfactory and all regression coefficients are significant at p-level ≤ 0.05. The results for total dissolved solids are characterized by a moderate R² = 0.5, but the regression coefficients are not significant. The results for the other parameters are characterized by a low R² = 0.1 and non-significant regression coefficients.

Finally, we interpreted the results, in the Discussion section, as follows:

The regression results based on individual parameters (Table 5) exposed a significant relationship between catchment variables (landscape composition, buffer strip width), including their interactions, and water turbidity, but did not reveal identical influences of those variables or their interactions on other parameters measured in water. It should be remembered, however, that water quality parameters may respond differently to catchment variables depending on spatial scale or antecedent rainfall conditions, as noted in Uriarte [31]. We therefore clarify that our results are valid at the studied spatial scales (Figure 4) and antecedent rainfall conditions (Table 2). Transposition to other settings needs verification.

-enlarge the possible interpretations of the results

Dear reviewer, we added several sentences to the Discussion section where we consider other possible interpretations for the results. Please see the yellow-shaded sentences in the Discussion Section.

more comments to be found in the annotated manuscript

We positively replied to all comments in the PDF file, which we uploaded into the Water platform. The consequent changes in the revised text are yellow-shaded in the revised manuscript.

Reviewer 2 Report

Dear Authors!

I have checked your manuscript. I find the topic is very interesting, but I think need to improve it's international significance! I have written my remarks, questions and advices into the text (see attached file). I found that your manuscript is valuable, but it need major revsions.

Comments for author File: Comments.pdf

Author Response

REVIEWER #2

Dear reviewer,

We very much appreciate your will to review our paper and we did our utmost to comply with your comments and suggestions. We replied to all comments in the PDF file, which we uploaded into the Water platform. The consequent changes in the revised text are yellow-shaded in the revised manuscript.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Dear Authors! As I see, you modified your draft according my remarks, therefore I can accept your publication in this form. Best wishes Reviewer 1

Author Response

Comment

Dear Authors! As I see, you modified your draft according my remarks, therefore I can accept your publication in this form. Best wishes Reviewer 1

Answer: Many thanks for your consideration