Intermittent Rivers as a Challenge for Freshwater Ecosystems Quality Evaluation: A Study Case in the Ribeira de Silveirinhos, Portugal
Round 1
Reviewer 1 Report
This paper sets out to compare the physico-chemical conditions and macroinvertebrate composition of an intermittent river with environmental standards required by the European Water Framework Directive (WFD). It is clear that the macroinvertebrate metrics are well below the WFD values expected for small rivers in the north of Portugal, while the physicochemical conditions are more or less what might be expected by these standards. This is potentially a useful result and judging by the title of the paper such differences are what the authors wish to explore. However, the reasons for the biological mismatch are not well explained and the authors need to provide a better account of why macroinvertebrate composition is not what the environmental standards expect. For instance, I would have thought that intermittent rivers are likely to have a lower diversity than permanently flowing waters (or at least a different composition) and that this would be a reason why the EQS standard of 30 families (Table 3) was not met. But this is not explicitly stated. Too much of the Results and Discussion is instead taken up with details of other studies and not enough with a description of how various taxa responded to the observed gradient in stream permanence. The picture may become clearer if the authors provided an MDS ordination of their biological data. This would indicate to what extent compositional changes do indeed follow this gradient.
I am concerned about several details that lessen my confidence in the results of the study. The total number of individuals caught at each site on each sampling occasion appears to be very low. In my experience taking a sample over three, one metre strips, of stream bed is likely to provide thousands (if not tens of thousands) of individuals. The methods state the net was dragged over the substrate. Such a procedure is unlikely to be very effective at capturing specimens unless the sediment is turned over in front of the net either by hand or by foot. Perhaps this is what was done, but it has not been well expressed in the text. The authors need to give some assurance about the effectiveness of their sampling. Also no data are given about the width or the depth of the river at any of the sites. This would provide some idea of how much habitat was available. In addition, information about the nature of the river bed and the general habitat at each site should be given, e.g was the river bed composed of cobbles or gravel and were riffles sampled at all sites or were pools also present?
I am also not convinced by the explanation given for the low pH values at the upper two sites. The authors suggest low flow accounts for the lower pHs, but I do not understand the connection. Is it possible that upstream sites received more groundwater than downstream sites and that the groundwater was low in oxygen but high in carbon dioxide? Groundwater in which much microbiological metabolism occurs can have a low pH until it is exposed on the stream bed where it becomes reoxygenated. It is also suggested that the presence of high iron levels at RS-4 and RS-5 might contribute to the higher pH at these sites. Is there any data on iron concentrations at these sites?
The paper should be better focussed on the questions posed and this could be achieved by deleting sections which are not particularly relevant. For example the following lines could be deleted without loss to the overall study: lines 33-43; 92-99; 274-279; 330-335; 393-401. These sections cover either well known facts that do not need repeating or they digress from the topic at hand.
The following points should also be considered:
Line 44: the correct word here and throughout the paper is ‘type’, not ‘typology’.
Line 167: ‘anthropogenic’ rather than ‘anthropic’.
Line 186: ‘sampling’ rather than ‘the harvest’.
Line 213: delete ‘in a small ….components’.
Line 377: ‘low fee values owned by EPT taxa’ does not make sense, yet it is important to make clear your meaning at this point. The low pH values at RS_1 and RS_2 do not seem to have resulted in low numbers of EPT taxa. There are fewer EPTs at sites RS_4 and RS_5 but can this be attributed to higher pHs? As I have said above, the numbers of specimens at all sites seem to me to be unusually low for the stated sampling effort, and thus I wonder to what extent any conclusions can be drawn confidently.
Line 418: what is meant by the word ‘waterline’? Waterline refers to the edge of a water body, a concept that seems out of place here.
Lines 428-449: these conclusions should be entirely re-written concentrating perhaps on how WFD metrics should be modified for intermittent rivers. This, I think, could be one of the most useful aspects of this study.
Some sections of the manuscript need editing to make the meaning clearer or the text more concise. Examples are: lines 67-70; 138-140; 199-200, 202-203; 247-252; 279-283.
Author Response
We are grateful for all questions raised by the reviewer, which helped us to clarify some ideas in the manuscript. All changes are easily identified in track changes.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Table 2 and Fig.S1 are not included.
It would be very useful to apply RDA or CCA analysis (CANOCO programme), plots of stations in relation to environmental factors or the association of benthic macroinvertebrates with environmental factors can be obtained.
*If the length of gradient of the theoretical variable presents a value less than three times the standard deviation of the benthic macroinvertebrates then they respond linearly to the theoretical variables and the RDA analysis is considered more suitable for the environmental gradient. If the length is longer, the CCA analysis is performed.
Comments for author File: 
Comments.pdf
Author Response
We are grateful for all questions raised by the reviewer, which helped us to clarify some ideas in the manuscript. All changes are easily identified in track changes.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
This version of the MS is much improved from the original. The authors have dealt with my suggestions well and I am satisfied that the abundances of taxa, although low, are trustworthy and can be used to draw conclusions. The photos of the sites included in the supplementary material give a good impression of their physical nature and just how small some of them are.
I have a number of suggestions, mostly for improvement of wording:
Line 21: ‘surrounded by’ is better than ‘wrapped in’.
Line 26: word missing after ‘ecological’.
Line 37: ’type’ not ‘typology’ and elsewhere in the MS e.g. line 402.
Line 51: ‘riverbed’ not ‘bed river’.
Line 56: ‘communities characteristic of ….’.
Line 59: ‘between the two periods….’.
Line 97: ‘macroinvertebrate evaluation’.
Fig.1: it is difficult to read the wording on this figure. I think a higher resolution image is needed.
Line 122: ‘overtime; one of …’.
Line 135: insert bracket i.e. ‘(Figure 1)’.
Line 144: ‘dominated’ rather than ‘predominated’.
Line 145: “At this site..’ Throughout the MS where ‘in’ is used to refer to a site ‘at’ should be substituted.
Line 152: delete: ‘artificial territories in the area such as’.
Line 157: ‘Additionally, at each …..’.
Line 164: ‘substrate; the river bed was not actively disturbed by hand or foot in front of the net. At each site ….’.
Line 190: delete ‘using a …….up’.
Line 192: delete ‘up’; delete ‘as a ….key’.
Line 200: ‘parameters, allowing us to ….’.
Line 239: delete ‘the boundaries scale’.
Line 246: delete ‘A DCA analysis ….afterward’. Start the sentence with ’A canonical …’ Were the abundance data transformed, e.g. to logs or square roots, before implementing the CCA?
Line 257: ‘at’ not ‘in’.
Lines 268-269: delete ‘dominated ……...increases’.
Line 297: delete ‘when compared to’ and insert ‘those in’.
Line 303: delete ‘an analysis of the main components’.
Line 370: ‘despite their resilience’.
Fig. 3: wording on this figure is also difficult to read and a higher resolution image is needed.
Line 405: ‘However, an increase was observed’.
Line 410: The word ‘fee’ is incorrect but I’m not sure what the correct word is because your meaning is not clear here.
Line 428: delete: ‘in relation to reference conditions when’.
Table 2: the sample at RS_5 for spring 20 seems to have very low abundances of taxa. Are data missing for this sample? For example, no chironomids were recorded, but they were present at all the other sites on this occasion. The numerous horizontal lines are confusing and no explanation is given as to why some are continuous and some are partially dotted. I’d recommend single continuous lines are used to distinguish data from each of the sites.
Table 3: it is not clear how EQR is derived from IPtIN. Mostly the values are identical but there are small differences between the two at some sites on some occasions.
Author Response
Reviewer comments #1
This version of the MS is much improved from the original. The authors have dealt with my suggestions well and I am satisfied that the abundances of taxa, although low, are trustworthy and can be used to draw conclusions. The photos of the sites included in the supplementary material give a good impression of their physical nature and just how small some of them are.
R: We are grateful for all questions raised by the reviewer, which helped us to clarify some parts in the manuscript. All changes are easily identified in track changes.
I have a number of suggestions, mostly for improvement of wording:
Line 21: ‘surrounded by’ is better than ‘wrapped in’.
R: We appreciate the comment, and the change has been made.
Line 26: word missing after ‘ecological’.
R: We appreciate the comment, and the change has been made.
Line 37: ’type’ not ‘typology’ and elsewhere in the MS e.g. line 402.
R: We appreciate the comment, and the change has been made.
Line 51: ‘riverbed’ not ‘bed river’.
R: We appreciate the comment, and the change has been made.
Line 56: ‘communities characteristic of ….’.
R: We appreciate the comment, and the change has been made.
Line 59: ‘between the two periods….’.
R: We appreciate the comment, and the change has been made.
Line 97: ‘macroinvertebrate evaluation’.
R: We appreciate the comment, and the change has been made.
Fig.1: it is difficult to read the wording on this figure. I think a higher resolution image is needed.
R: We appreciate the comment, and the change has been made.
Line 122: ‘overtime; one of …’.
R: We appreciate the comment, and the change has been made.
Line 135: insert bracket i.e. ‘(Figure 1)’.
R: We appreciate the comment, and the change has been made.
Line 144: ‘dominated’ rather than ‘predominated’.
R: We appreciate the comment, and the change has been made.
Line 145: “At this site..’ Throughout the MS where ‘in’ is used to refer to a site ‘at’ should be substituted.
R: We appreciate the comment, and all changes were made.
Line 152: delete: ‘artificial territories in the area such as’.
R: We appreciate the comment, and the change has been made.
Line 157: ‘Additionally, at each …..’.
R: We appreciate the comment, and the change has been made.
Line 164: ‘substrate; the river bed was not actively disturbed by hand or foot in front of the net. At each site ….’.
R: We appreciate the comment, and the change has been made.
Line 190: delete ‘using a …….up’.
R: We appreciate the comment, and the change has been made.
Line 192: delete ‘up’; delete ‘as a ….key’.
R: We appreciate the comment, and the change has been made.
Line 200: ‘parameters, allowing us to ….’.
R: We appreciate the comment, and the change has been made.
Line 239: delete ‘the boundaries scale’.
R: We appreciate the comment, and the change has been made.
Line 246: delete ‘A DCA analysis ….afterward’. Start the sentence with ’A canonical …’ Were the abundance data transformed, e.g. to logs or square roots, before implementing the CCA?
R: We appreciate the comment, but the part that the reviewer suggests deleting, was included in the first review, at the request of another reviewer. Therefore, we decided to keep it, to justify the analysis that was carried out. The abundance data were not transformed. We use raw data.
Line 257: ‘at’ not ‘in’.
R: We appreciate the comment, and the change has been made.
Lines 268-269: delete ‘dominated ……...increases’.
R: We appreciate the comment, and the change has been made.
Line 297: delete ‘when compared to’ and insert ‘those in’.
R: We appreciate the comment, and the change has been made.
Line 303: delete ‘an analysis of the main components’.
R: We appreciate the comment, and the change has been made.
Line 370: ‘despite their resilience’.
R: We appreciate the comment, and the change has been made.
Fig. 3: wording on this figure is also difficult to read and a higher resolution image is needed.
R: We appreciate the comment, but this figure is the best version. The high number of physicochemical parameters (Table 1) and biological data (Table 2) makes it difficult to achieve a better resolution, given that there is a high concentration of data in the same place in the figure (centered in the CCA analysis, as well as the taxa association), indicating that there is no distinction between sites according to the parameters input. In Figure 3, too many environmental parameters appear even the ones that have no meaning for the relation of the species to the environmental parameters of the sampled stations.
Line 405: ‘However, an increase was observed’.
R: We appreciate the comment, and the change has been made.
Line 410: The word ‘fee’ is incorrect but I’m not sure what the correct word is because your meaning is not clear here.
R: We appreciate the comment. The word fee was removed since it isn’t necessary.
Line 428: delete: ‘in relation to reference conditions when’.
R: We appreciate the comment, and the change has been made.
Table 2: the sample at RS_5 for spring 20 seems to have very low abundances of taxa. Are data missing for this sample? For example, no chironomids were recorded, but they were present at all the other sites on this occasion. The numerous horizontal lines are confusing and no explanation is given as to why some are continuous and some are partially dotted. I’d recommend single continuous lines are used to distinguish data from each of the sites.
R: We appreciate the comment, and indeed there was an error in the line referring to the abundance of macroinvertebrates, at the RS_5 site, in the Sp20 period. The data has all been confirmed. Changes are highlighted. Continuous lines separate sampling sites. Dashed lines separate sampling periods within each site. However, we consider the reviewer's suggestion to only use continuous lines to separate the sites.
Table 3: it is not clear how EQR is derived from IPtIN. Mostly the values are identical but there are small differences between the two at some sites on some occasions.
R: We appreciate the comment. The derivation of EQR from IPtIN is explained at the end of section "2.4.2. Biological quality elements". “The final IPtIN value is the result of the sum of the weighted metrics. Two normalization steps are performed, before the multiplication of the metrics by the weighting factor and after the sum of the weighted metrics, reaching the final value expressed in the Ecological Quality Ratio (EQR) [13,28]. The normalizations are computed through the quotient between the obtained value and the EQS value for that type of river (Table 3).”
Author Response File: Author Response.docx
Reviewer 2 Report
A word file is inserted with my comments and the correction of a paragraph concerning CCA. In the pdf I made some corrections. The word file is attached to the email sent by me later since I cannot upload it here and in this one can clearly see the changes I did in the concerning paragraph. The final paragraph follows:
The CCA analysis (Figure 3) showed that RS_5 is the site with the most distin-guished macroinvertebrate community independently of the sampling season. It seems that its position is mainly related to pH, conductivity and NH4 characterizing the 1nd axis (see also Table 1). The position of RS_5 in Sp20 is influenced by temperature characterizing the 2st axis (see also Table 1) and also related to specific group of taxa (Ass – Gastropoda; Sip and Pot – Ephem-eroptera; Cul – Diptera; Table 2). Moreover RS_4 AU19 is related mainly to P characterizing the 1st axis (see also Table 1) and Cordulegastridae (See also Table 2). RS 4 and 5 Sp19 showed a strong relation with Dry (Dryopidae) organisms (taxa with a score of 5 in the IBMWP index, medium pollution tolerance) and conductivity (see Table 1). On the other hand, the remaining sites are centered in the CCA analysis, as well as the taxa asso-ciation.
Comments for author File: Comments.zip
Author Response
Reviewer comments #2
We appreciate the English and language corrections, and all the suggestions were accepted.
Pag 6
Line 250 - This part at the beginning of the coloured classes needs to be corrected.
R: We believe that is a formatting mistake, however, we reorganize the information.
Pag 7
Line 295 - and Tabe 5 ??????
R: In our manuscript version this is not occurring. We believe it was a formatting mistake. The legend of Table 1 is “Table 1. Results of physical and chemical parameters: temperature (T), conductivity (Cond), total dissolved solids (TDS), pH, dissolved oxygen (O2), nitrates (NO3), ammoniacal nitrogen (NH4), and total phosphorus (P) and biochemical oxygen demand (BOD5), for each sampling site for the three sampling periods. Environmental quality standards (EQS) values indicated by the WFD to achieve "Good ecological status" in small northern rivers were also included. Bold values show those that exceed the EQS stipulated for the respective parameter.”
Pag 8
Line 319 - higher or lower????
R: Thanks for the comment. The information in the manuscript is right, the values of DO are higher in Sp19. The sampling period of Sp19 is closer to the final trend arrow of DO showing that this period has higher levels of DO. In opposition, the Sp20 sampling period PCA distribution is a further way to the DO arrow demonstrating that this period has lower DO levels.
Line 334 (section 3.2) - It would be very useful to apply RDA or CCA analysis (CANOCO programme), plots of stations in relation to environmental factors or the association of benthic macroinvertebrates with environmental factors can be obtained. *If the length of gradient of the theoretical variable presents a value less than three times the standard deviation of the benthic macroinvertebrates then they respond linearly to the theoretical variables and the RDA analysis is considered more suitable for the environmental gradient. If the length is longer, the CCA analysis is performed.
R: We agree with the review and a DCA followed by a CCA was conducted to plot the association with the environmental variables and the macroinvertebrate communities. This information was added to the manuscript.
Line 335 - Where is Table 2????
R: It was a mistake, there are 2 tables 3 instead of 2 and 3. Now it is correct.
Pag 10
Line 375 - Where is this figure?
R: This is a figure included in supplementary material, we added this information to the manuscript.
Author Response File: Author Response.docx
