Effects of Upwelling Intensity on Nitrogen and Carbon Fluxes through the Planktonic Food Web off A Coruña (Galicia, NW Spain) Assessed with Stable Isotopes

Round 1

Reviewer 1 Report

The authors' revisions generally meet my expectations, that is to say I recommend publication. This is merited by the quality of their responses, in general, and I thank them for the improvement to a generally sound and well-presented manuscript.
However, I am disappointed that the authors do not understand that their data, being collected along a gradient of time, are a time series. Time series data are not defined by the length or resolution of the time series. Time series are important to recognise because, being collected along a gradient of time, each data point is not fully independent of the next, hence each data point does not bring a full degree of freedom for statistical analysis (i.e. they are temporally autocorrelated, a form of pseudoreplication). Moreover, temporal trends in otherwise unrelated dependent and independent variables can result in Type 1 statistical error. In future work, I recommend the authors use the appropriate time series methods, especially because ecology is becoming more and more aware of such issues. Critically, when these issues are ignored, they can lead to wrong conclusions, which compromises scientific integrity. Time series data do not always need time series analysis but in those cases one must prove that there is no temporal autocorrelation in model residuals (i.e. that the non-time-series model assumptions are met).
Best wishes for the future

Author Response

see attached file: 

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors,

 

thanks for to clear response letter and modification in the manuscript.

I have no further comments.

 

Live long and prosper

 

 

 

Author Response

see attached file

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The article, Effects of upwelling intensity on nitrogen and carbon fluxes through the planktonic food web off A Coruña (Galicia, NW Spain) assessed with stable isotopes, is generally well written but suffers from two main flaws that affect the utility of the results/conclusions, and which I would like to see addressed.

(1) Many potential readers may not be from northern Spain and would be interested in general (i.e. not regional) upwelling effects on the food web via nutrients. These readers could be easily addressed by an extra paragraph in the discussion exploring similarities and differences with other upwelling systems, including Chile (e.g. Reddin et al. 2015), California (e.g. Vokhshoori et al., 2014) or South Africa (e.g. Hill & McQuaid 2008). What features are regional, which are global?

(2) The result of the regression analyses are clearly important for future work of the authors (and probably wider), but no respect is paid to the proper analytical handling of time series data, which can otherwise bias the results of standard ordinary least squares regression. I checked that the software used by the authors, PAST, does include proper time series methods (at least the recent 4.0 version does; see especially ARMA), so this should not be difficult. Ecology is rapidly becoming aware that temporal autocorrelation can bias both the significance and coefficient of standard statistical tests, so if the results of this paper are to be useful, they must be done right.
The distributions in Figure 4 (a and d) and 6 (a & b) also look to be non-normal; check and transform where appropriate. Also Fig. 7, where the high values have a huge contribution on the slope: the x axis needs log (x+1) transformation.

Also see line-by-line comments below

Figure 2. How are these y axis numbers on such a different magnitude to compared to Fig. 4?

Figure 3. State what the white points represent. What interpolation method is used?

References above.

Reddin, C. J., Docmac, F., O’Connor, N. E., Bothwell, J. H., & Harrod, C. (2015). Coastal upwelling drives intertidal assemblage structure and trophic ecology. PloS one, 10(7).

Hill, J. M., & McQuaid, C. D. (2008). δ13C and δ15N biogeographic trends in rocky intertidal communities along the coast of South Africa: evidence of strong environmental signatures. Estuarine, Coastal and Shelf Science, 80(2), 261-268.

Vokhshoori, N. L., & McCarthy, M. D. (2014). Compound-specific δ15N amino acid measurements in littoral mussels in the California upwelling ecosystem: a new approach to generating baseline δ15N isoscapes for coastal ecosystems. PLoS One, 9(6).

Line by line

Line	Comment
22	Verb missing. Perhaps “Upwelling is enriched in…”?
23	‘All plankton fractions’ seems misplaced. Precede with ‘at’?
26	Unclear: “are rapidly diluted in upper trophic levels”
27	“plankton consumers” to “consumers of plankton”
	Too much use of passive form e.g. change “Linear equations were provided…” to “We provide linear equations…”
37	“but there are also” to “but can be also”
39	“in a few days”
44	“because of the increasing in” to “because of an increase in”
45-6	“effect on different components of the food webs is largely undetermined”. This statement seems to overlook a huge body of literature e.g. the above references and more. Perhaps tone down to something like “needs further research”.
61-2	Lines 61 and 62 seem to contrast each other. Use ‘however’ to begin sentence in L62.
64	Reference needed for: “as diatoms are the dominant phytoplankton in upwelling ecosystems” (is this the case everywhere?)
68	“products were sustained” to “products are sustained”
90-2	If only diatom abundance is used, what is the need for the sentence before it? (“Phytoplankton species composition was determined…”). Link better.
115	Active voice, especially to begin the sentence: “Mesoplankton (200-5000 μm) was collected” to “We collected mesoplankton (200-5000 μm)…”
126	“Dried [how?] and finely ground [how?] aliquots…”. Methodological detail needed for reproduction lacking. Preservation used?
128-30	“Samples were not acidified to remove carbonates”. Justification is insufficient since most researchers just split the sample to avoid this issue, taking d15N from the unacidified half. Probably just go with that zooplankton are only very lightly calcified and acidification is not needed e.g. Ng JS, Wai TC, Williams GA (2007) The effects of acidification on the stable isotope signatures of marine algae and molluscs. Mar Chem 103: 97–102 Either way, delete: “may not cause substantial modification in carbon isotope results, but it”
147	I am sure it is published elsewhere, but briefly mention how geographically representative this location is: “in a cell of 1°×1° centered at 44°N, 9°W”
158	Not appropriate methods for time series (unless you can show that there is no autocorrelation in their residuals?): “correlation (Pearson coefficient) and linear regression (ordinary least squares)”
178	“All variations of the index were significantly correlated”. These are from the same data, so this statement is obvious. Would be better to use an ecological reason for scaling e.g. to match UI to temporal turnover of tissue for stable isotopes e.g. O'reilly et al 2002 Best practice would be to present results from all of these UIs to give an impression of the scaling of relationships. U15 is also smoothed artificially (averaging), forcing greater temporal autocorrelation (see methods comments) O'reilly, C. M., Hecky, R. E., Cohen, A. S., & Plisnier, P. D. (2002). Interpreting stable isotopes in food webs: recognizing the role of time averaging at different trophic levels. Limnology and oceanography, 47(1), 306-309.
226	“could be described by linear regressions, predicting the changes”. Then prediction intervals would be ideal rather than standard confidence intervals. But this is low priority compared to ensuring that analyses are appropriate for time series.
249-51	Add an error value to these enrichment means e.g. SD
257-9	Contradiction: is there a seasonal pattern or isn’t there? Use statistics to test
260	“δ15N values resulted linearly” to “δ15N values were linearly”
265 & 278	What are these correlations for?
273	“seston and trap particles were not related to upwelling index values…” This is where a scaling approach would be interesting? Which resolution of UI matched these best?
282	“with equivalent slopes”. Not significantly different does not mean equivalent.
302-3	“but not in larger size-fractions of plankton”. At the single scaling of UI you tested…
308-9	Awkward sentence structure: Perhaps add “inconsistent,”, after “during upwelling were”
322	“particles during non-upwelling periods”
325	“cause also” to “also cause” Superscript 15
355	“similitude” to “similarity” “seston and sediment trap”. Perhaps these categories actually overlap?
356, 360 & 361	“Rapid”- how rapid?
358	End sentence after “ecosystems [40]”. Then start “Meanwhile…”
363	Delete “that” in “shelf causes that the”. “upwelled water in this region to contain”
364	“variable isotopic” to “variability of isotopic”
368	“different integration times”. O’Reilly reference would be useful here. Also different feeding times.
373	“the commonly accepted value of 3.9‰ increase per trophic level”. I think the value you refer to is a mean, not a set value e.g. Vanderklift, M. A., & Ponsard, S. (2003). Sources of variation in consumer-diet δ 15 N enrichment: a meta-analysis. Oecologia, 136(2), 169-182.
375	“composed mainly by zooplankton”. Give some numbers…
381	“challenging the extended application of a single value of isotopic fractionation per trophic level when estimating trophic positions”. As above, I don’t think this is a novel challenge. Assuming a set value is poor practice. The references 45 and 46 support this. Also: Barnes C, Sweeting CJ, Jennings S, Barry JT, Polunin NV (2007) Effect of temperature and ration size on carbon and nitrogen stable isotope trophic fractionation. Funct Ecol 21: 356–362
404	“their importance” to “diatom importance”
405	“as in other upwelling ecosystems”. Needs expanding and references
407	Vague: “the relevance of diatoms”
420-425	I would delete these first two sentences. The first is questionable and a reutterance of the methods. The second, even if true, wouldn’t affect the stats if the bias is evenly spread (only the regression intercept). The third sentence finally gets into the scientific meat.
433	“values estimated from the regression line computed in this study will provide a quantitative index”. Then this needs to be done properly (or proven that it is done so)
436	This statement is far from new, and needs evaluating against the backdrop of other upwelling systems to be more widely useful
440	“leaved” to “left”

 

Author Response

see the attached file ·Esponse_to_reviewers.pdf"

Reviewer 2 Report

Review for “Effects of upwelling intensity on nitrogen and carbon fluxes through the planktonic food web off A Coruña (Galicia, NW Spain) assessed with stable isotopes” by Antonio Bode and colleagues.

 

The manuscript studies the relationships between upwelling intensity and natural abundance of stable carbon and nitrogen isotopes in seston, sedimented particles and different plankton size fractions (200 to 5000μm) sampled in two subsequent years. The data are of high quality and the authors have done a good job at interpreting and presenting their data. Overall, the manuscript is clearly written and needs only minor revisions before publication.

Minor comments:

L37: In the beginning of the introduction, the authors talk about upwelling systems in general. I would suggest to rephrase the sentence “but there are also inputs from continental origin, particularly near large rivers” into “can be influenced from near large rivers”, as this is not the case for all upwelling systems. e.g. the Peruvian Upwelling is not influenced by riverine input.

L225 I do not really get, why the authors compare the Upwelling Index with bottom waters and not surface waters. The major chlorophyll peaks, which are directly influenced by upwelling intensity are in the upper 50m. Besides, how are “bottom” water depth defined. How many meters above ground?

L325: The authors mention the potential elevation of d15N in residual water, any effects on d15N due to denitrification and/or annamox under low (<10 μmol)? Anoxic conditions are often observed in intense upwelling regions. I was wondering, if the water column in the study area is always well oxygenated or do episodes of low oxygen occur. Maybe the authors can include one sentence about oxygen conditions off A Coruña.

L376: ” ….exact taxonomic composition of plankton …was not determined..”

What a pity. The manuscript would clearly benefit from such an information. Seasonal upwelling has a strong influence on (phyto)plankton communities and therefore d13C and d15N. At the end of bloom or under low upwelling other phytoplankton than diatoms dominate the water column (e.g dinoflagellates). Additionally, the plankton community is different for different size fractions. Blooming diatoms are rather small and generally below 200 μm. Large size samples can already contain radiolaria and zooplankton. Even though I understand that it might not have been possible for this study (and I am aware of the effort), I like to emphasize that assemblage data is of great importance and should be considered in future studies.

L437: In the results/discussion part, you use the term “bottom waters”?!?

L443: Is seston necessarily “short-living”? It includes everything, also dead material (that might not be remineralized easily!

L446: The samples were not analyzed for plankton communities, therefore the authors have no proof, that diatoms are responsible for the signal.  Rephrase carefully, same for the abstract!

Author Response

see tha attached file "Response to reviewers.pdf"

Author Response File: Author Response.pdf