the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 31 Mar 2021
Evolution of the long-term and estuary-scale phytoplankton patterns in the Scheldt estuary: the disappearance of net growth in the brackish region
Dante M. L. Horemans
Yoeri M. Dijkstra
Michèle Tackx
Patrick Meire
Tom J. S. Cox
Abstract. Estuaries often show regions in which Chlorophyll-a (Chl-a) accumulates. The location and magnitude corresponding to such accumulation result from a complex interplay between processes such as river flushing, salinity, nutrients, phytoplankton grazing, and the light climate in the water column. Of particular interest is the long-term evolution of the estuary-scale Chl-a distribution in the Scheldt estuary (Belgium/Netherlands) in spring. From 2004–2007, we observed a limited spring-bloom in the brackish region. This bloom intensified in 2008–2014 and disappeared after 2015. This long-term evolution in Chl-a has been linked to simultaneous long-term trends in the suspended particulate matter (SPM) distribution and the improvement of the water quality, which affects grazing of Chl-a by zooplankton. However, this hypothesis has not been systematically investigated. In this paper, we apply two approaches to test this hypothesis. In the first approach, we analyze long-term in situ observations covering the full estuary. These observations include the SPM concentration, zooplankton abundance, and other variables affecting the Chl-a concentration, and show a long-term estuary-scale evolution in not only the SPM distribution but also in zooplankton abundance, freshwater discharge, and maximum photosynthetic rate. In the second approach, we apply a model approach supported by these observations to determine which of the changed conditions may explain the observed change in Chl-a. Our results suggest that a change in SPM alone cannot explain the Chl-a observations. Instead, mortality rate and grazing by zooplankton mainly explains the long-term estuary-scale evolution of Chl-a in spring. Our results highlight that insight into the zooplankton dynamics is essential to understand the phytoplankton (cf. Chl-a) dynamics in the Scheldt estuary.
- Preprint
(2544 KB) - Metadata XML
-
Supplement
(95 KB) - BibTeX
- EndNote
Dante M. L. Horemans et al.
Status: closed
-
RC1: 'Comment on bg-2021-59', Huib E. de Swart, 17 Apr 2021
Overall: this is a paper about an important topic. The paper is well structured and well written. The aims are clear and results are interesting. Clearly, this work is beyond a case study, as it also provides generic and useful knowledge for understanding similar phenomena in other estuaries.
My main comments concern the applied methods (to find the light extinction coefficient), the choice of the model and the formulations/assumptions used in that mode, the presentation of the results and the discussion (see comments below). Based on these, my recommendation is major revision.
-------------------------------------------------------------------------
Comments about the scientific content
Section 2
L101: Why only two measurement points, how useful is it in that case to apply an exponential fit? One would typically expect a number of points here at different distances and use these to make a fit.
Furthermore, it is not clear why the distance was chosen to be 40 cm, please explain.L117: Why use this model instead of e.g. a state-of-the-art model like Delft3D? Alternatively, the authors could have used the model of Arndt et al. (2011), why was decided otherwise? See also later comment: a discussion about the model limitations is missing.
L125: Why is salinity considered to be static and prescribed? That does not seem close to reality.
This comment also applies to the statement in l177 ‘to correct for the large temporal variability in the discharges’.L126: Does the statement ‘Scheldt…well-mixed’ apply to the entire estuary, i.e., there no stratification anywhere? Please add a citation to support this.
L135-136: If being so precise here, then why not consider a dynamic salinity? The choices made seem a bit ad hoc.
Eq. 2b: Shouldn’t the left-hand side be (H+\zeta), rather than H?
Eq. (3)
* Where does this come from, please add a citation here.
* The meaning and dimensions of the symbols g1, g2, Z… is not clear. It takes until Table 1 to conclude that g1 and g1 have the unit s^{-1} L and so the Z’s are in L^-1. It then takes until l205 to find the unit of the Z’s, but there ind. L^{-1} is used; but what is ‘ind’?
* The text ‘extension (section 3.2.4)’ is not satisfactory: motivation should be presented here.
Eqs. (3) and (6): Please remove the ‘dots’ on the right-hand side of the second equation, they’re not used in any other equation.
L155: Why a time-averaged growth rate?
And why is this about implementation, one would expect this to be about the choice for the formulation of \mu, which originates from biological literature.L156: The argument to use Eq. (4) to save computational costs is a bit unexpected. There is no information about the time that a typical run lasts and how much does that costs, is that of the order of a week or more (as suggested by this statement)?
Note: this comment also applies to l170: please add quantitative information.
L160: It is suggested here that E has the unit PAR, but in Table 1 another unit for E is used. Please use either of them.
L162: Specify how the seasonal variation of E_00 is described in the model. Table 1 only contains one value (is that setting for the default setting?)
Eq. (5): The formulation is now such that E is only nonzero during one single day. This should probably be for all daytimes (multiple time intervals).
Eq. (6): Meaning and dimensions of the calibration parameters are not clear. In (6) \mu_01 is raised to a power with a (dimensional) temperature T (expressed in Kelvin?) as an exponent?
Table 1: Values for w_P, k_bg, k_P are expected to originate from biological literature.
Third line: there appear two parameters \mu_max and \alpha here?
Section 3
Figure 3 is very hard to read, because of the many curves with error bars and shading areas. Suggest to present Chl-a and calanoid abundance in different panels.
Consider adding ‘suggestion lines’ between the data points that correspond to a single event. That would help to ‘guide the eye’ of readers.
L237/Figure 6a: the unit seems not correct: from Eqs. (1) and (4) it follows that the unit of \mu_max is 1/s.
L240/Fig. 6b: Unit? From Eq. (4), it seems that the simplest way to write it is PAR^{-1} s^{-1}.
Fig. 7: Recommend to use a different color to plot the model results.
Section 3.2 (sensitivity analysis) is quite exhaustive and has a bit the style of a logbook. Suggest to shorten it and present less figures: not all seem to be necessary. Alternatively, move this section to an Electronic Supplement and only summarize the main findings of the sensitivity analysis in the main text.
Section 4
* L374: It is a bit unexpected that at this stage, two new state variables are introduced, viz. oxygen and fish abundance. If they are important, then why were they not considered earlier? And isn’t oxygen affecting the phytoplankton as well (through respiration)?
L353/372/379: This paper is about verifying a hypothesis (L7). It is then a bit surprising that yet two other hypotheses are formulated that are not verified. The title of Section 4.2 is also unclear in the sense of ‘reconstructions of concentrations’; what is the purpose here, to present a conceptual model?
* Section 4.2 is not really satisfactory. At the same time, important other information is missing, for example about the model limitations (what processes are neglected and how is that justified?) Recommendation: remove the current text of 4.2 and add text about model limitations and context (see next comment).
* What would add value to this study is to put the results in a broader context: in what way do these results of this study contribute to understanding phytoplankton dynamics in other estuaries? This would avoid readers to get the impression that this is a case study.
Appendices
The paper is long and it contains details that are of interest to only a small number of readers. To increase attractiveness and readability, please move all appendices to a separate electronic supplement.
-----------------------------------------------------------------------------
Minors:
Overall:
Remove commas before ‘and’.
Add commas before ‘which’ (or replace ‘which’ by ‘that’).
L3: Why of ‘particular’ interest?
Note: this comment also applies to l27: the Scheldt is definitely interesting, but it not clear what is ‘particular’ about it with respect to other estuaries.
The comment also applies to l37.L9-11: long sentence that is difficult to read. Please split.
L14: explain
L17-19: This text is almost identical to that of l1-3 of the abstract. Please rephrase.
L30: drastically, mainly
L32-35 long sentence, with several subordinate clauses. Suggest to rephrase/split it.
L38: What is meant by ‘we’ here, the authors? Clarify by means of a citation.
L42-44: a complicated sentence, suggest to split it.
L47: observations, to
L51-52: the introduction ends quite abruptly. It would help to write the last paragraph such that it provides an overview of what is in the remaining sections.
L54-55: suggest to move this to the last paragraph of the introduction (see previous comment).
L58: relatively small compared to what?
L66-67: why three citations needed?
L69-73: split.
L76: Is there an unofficial website as well? And perhaps provide the link here.
L87-89: difficult to read (too many subordinate clauses), please split/rephrase.
L134: dynamics, because
L139-148: this is one sentence. Please split.
L149-150: As a start? Further, the word ‘simple’ is rather subjective, the expression in Eq. (3) is already quite complicated.
L173: ‘sensitivity study ... supported … observations’ (? Not clear what is meant here, please rephrase).
L191-193: expected this in the last paragraph of the introduction (see earlier comment).
L242/261/330: please avoid the word ‘conclude’ while being in a section on results. Change e.g. by ‘The long-term ..’ , ‘The results show..’, The… thus shows..’.
L243: extent
L245 and hereafter: Present tense is used for activities that took place in the past. Compare with, for example, the text in Section 5.
L291: Not clear why a new symbol m_new is introduced here.
L304: k_c italic.
L309: Expected this motivation in Section 2, when designing the model experiments.
L336: If the reason is unclear then there is little to discuss. Suggest to write ‘Possible reasons for… are discussed….’
L338-338: Suggest to move this to the last paragraph of the introduction (see previous comment).
L367: surprising, because
-----------------------------------------------------------------------------------------------
Citation: https://doi.org/10.5194/bg-2021-59-RC1 -
AC2: 'Reply on RC1', Dante Horemans, 10 Jun 2021
We thank Prof. Dr. de Swart for his time to review our manuscript. His main comments concern 1. the applied methods (to find the light extinction coefficient), 2. the choice of the model and the formulations/assumptions used in that model, 3. the presentation of the results and the discussion. Below, we present a point-to-point response to his main concerns. We would be happy to also address all other minor comments along with a revised manuscript.
1. The applied methods (to find the light extinction coefficient)
We used the data available within the OMES monitoring program. We agree that measuring at different depths may be preferable in case we expect strong vertical stratification of SPM in the euphotic region. However, given the high turbidity in the Scheldt estuary, the euphotic depth is relatively small (~ dm) compared to the total water depth (~ m). We thus only expect phytoplankton growth near the water surface where we do not expect strong vertical stratification of SPM. Therefore, it is preferable to have a better estimate at the water surface instead of an estimate covering the full water column. We will clarify this in the manuscript.
2. The choice of the model and the formulations/assumptions used in that model
Why use this model instead of e.g. a state-of-the-art model like Delft3D? Alternatively, the authors could have used the model of Arndt et al. (2011), why was decided otherwise? See also later comment: a discussion about the model limitations is missing.
The main reason why we used an idealized model is that it allows us to apply an extensive sensitivity analysis consistently and efficiently over all uncertain parameters, which is a requirement given the scope of the manuscript. We could have alternatively done this within another framework such as that of Arndt et al (2011), which we would have then needed to simplify more as their model already contains too many parameters for our scope. We decided to use iFlow as we already have a track record for this and verified that sediment and flocculation are sufficiently well represented in this model.
For the revision, we suggest including an extensive motivation for choosing these specific simplifying assumptions and the framework that we chose in the introduction and method sections. Also, we will discuss model limitations and strengths better in a separate discussion section (see also comment below).
Why is salinity considered to be static and prescribed?
We motivate the choice of using a static longitudinal salinity profile by the fact that, firstly, salinity gradients are of minor importance to SPM transport in the Scheldt estuary and, secondly, that phytoplankton cells are passive and move with the tidal water flow.
This comment also applies to the statement in l177 ‘to correct for the large temporal variability in the discharges’.
The assumption of using a constant freshwater discharge is validated in our sensitivity analysis, which shows only a minor impact of variability in freshwater discharge on the accumulation of phytoplankton in the brackish region.
3. The presentation of the results and discussion
Section 3.2 (sensitivity analysis) is quite exhaustive and has a bit the style of a logbook. Section 4.2 is not really satisfactory. What would add value to this study is to put the results in a broader context.
We agree with the suggestion to move the Appendix and the results of the sensitivity analysis presented in Section 3.2.1 to an Electronic Supplement. Additionally, we propose a bigger reorganization of the manuscript in which zooplankton is introduced from the start so that no new elements are introduced in the discussion. Furthermore, section 4.2 will be removed and the discussion will focus on discussing the model context, limitations, and applicability to other estuaries.
Citation: https://doi.org/10.5194/bg-2021-59-AC2
-
AC2: 'Reply on RC1', Dante Horemans, 10 Jun 2021
-
RC2: 'Comment on bg-2021-59', Anonymous Referee #2, 21 Apr 2021
Overall Statements:
The manuscript “Evolution of the long-term and estuary-scale phytoplankton patterns in the Scheldt estuary: the disappearance of net growth in the brackish region” by Dante M. L. Horemans, Yoeri M. Dijkstra, Michèle Tackx, Patrick Meire, and Tom J. S. Cox describes the variations of Chl-a distribution in the Scheldt estuary within the time interval 2004 to 2018. The authors try to understand the high spring Chl-a concentrations in the brackish region during the years 2008-2014 which was not observed during other periods. In a first step they analyse observed data such as Chl-a concentration itself, zooplankton abundance, salinity, suspended matter concentration, and light climate related data. Except for some difficulties (see below) these investigations appear sound. In a second step the authors apply an over-simplified model which is not able to explain the phenomena observed within the data. Using parameter-variations they try to show that zooplankton grazing may be the important mechanism to reproduce the elevated spring Chl-a concentrations within the brackish region. The authors argue to use this simple model in order to have good performance. But this does not help as the model does not tackle the phytoplankton dynamics appropriately.
I suggest applying a more adequate model, which describes the investigated time interval in a transient manner and resolves different phytoplankton and zooplankton groups. Boundary conditions should be defined temporally variable. The excellent observation-derived data sets may be used for validation.
Based on these, my recommendation is reject.
Detailed remarks:
L 5: Define the brackish region according to salinity and position.
L 16 ff:
- Do you know similar studies for other rivers and estuaries? Gypens et al 2013 (http://dx.doi.org/10.1016/j.jmarsys.2012.10.006) studied the Scheldt estuary and came to more detailed and opposing results.
- Other parameters may play a role too (see McQuatters–Gollop and Vermaat (2011), doi:10.1016/j.seares.2010.12.004)
- Winter values of zooplankton may play a key role (Dudeck et al 2021, doi:10.1093/plankt/fbab011)
L20: The processes are governed by temperature variations .. salinity variations .. nutrient dynamics.
L59: Give the position of the gauge station.
L64: Silicate concentrations?
L76: Refer to English written peer reviewed articles only.
L77ff: Are the Chl-a measuring methods compatible? I know that in Rijkswaterstaat some efforts have been made to homogenize both approaches.
Figure 1: Indicate the brackish region and give km-positions for at least some of the stations.
L97: Are these methods compatible?
L109: Please introduce first the context of mu-max and alpha.
L116: Give the coordinates and resolution of the model area.
L131: ETM – please introduce this in full words.
L143: To define seaside concentration constant is questionable.
L159: The formula should have “minus alpha” not “alpha”.
L161: The formula should have “+ kc” and “+kp”.
L196: You mean surface concentrations?
L208ff: You often indicate the maximum as the mean plus standard deviation. This is not really the maximum.
L212: I do not understand the argument: “ .., as the estuary is narrow and shallow in the upstream region.”
L225: Can you really justify that SPM is in 2015 – 2018 (at 80 – 120 km) significantly larger than during the other periods, and KD is not significantly larger?
Figure 2 caption: b) is not shown
Figure 3 caption: It is not in the range km 60 – 90 but in the range km 60 – 80.
L227: You used the numbers of May?
L231: This is not always true: See March: The discharge is almost equal but not the salt intrusion.
L242: Say increasing or decreasing development.
L261: from L252-L253 I’ve learned it is decreasing.
L261- L266: I do not understand this section.
L269: You see an accumulation at km 40-60. This is not your brackish region.
L334-L335: Highlight this by a figure and discuss it seriously.
L351: Salinity: This is in opposite to Fig. 5b: The intrusion of salt is largest in 2015 – 2018 for April and later.
L353: Have you studied the planktonic community?
L398: High SPM normally induce light inhibition for phytoplankton.
L432: This formula has 2 degrees of freedom. How do you find an optimum?
L436: The time-dependence of the parameters are questionable.
L451 ff: The method to calculate SPM for the early periods is very critical, as deep-water SPM structures are governed rather by benthic-pelagic interactions than by surface variations.
L458: Neglecting background and phytoplankton induced light extinction is very critical.
Figure C1. Please give units.
Citation: https://doi.org/10.5194/bg-2021-59-RC2 -
AC1: 'Reply on RC2', Dante Horemans, 11 May 2021
We thank Anonymous Referee #2 for their time to review our manuscript. In summary, the core argument for Anonymous Referee #2 to reject our work is in the comments that the model ‘is an over-simplified model’, ‘is not able to explain the phenomena observed within the data’, and ‘does not tackle the phytoplankton dynamics appropriately’. The reviewer is additionally very specific on the elements that need to be in a model, which is a model that ‘investigates timeinterval in a transient manner and resolves different phytoplankton and zooplankton groups’ and that ‘Boundary conditions should be defined temporally variable’.
We do recognize that our conclusions may be too strongly suggested regarding the importance of grazing and are happy to revise this. Otherwise, we do believe our model is well-suited for the scope of our study. Below, we motivate this by a point-to-point response to the main concerns of Anonymous Referee #2.
Reply to reviewer main comments
Over-simplified model/does not tackle the phytoplankton dynamics appropriately
The goal of our study is to investigate the hypothesis that the observed rise and fall of a mid-estuary phytoplankton bloom could be explained by sediment concentration. We deliberately kept our model simple and focused on the effects of sediment-induced shading to be able to study exactly this without any other obscuring effects. Hence, we could do an exhaustive sensitivity study over all parameters. This yields a very robust result that rejects the hypothesis.
We do agree with the reviewer (also see several of the comments below) that it might be too premature to point to grazing as the alternative explanation. Our model might indeed be too simple to show this without doubt. We would be happy to revise our discussion in this respect, primarily concluding that phytoplankton mortality is important and stressing that there could be multiple explanations to this, of which grazing is one.
We would agree that the model is too simple to simulate the precise phytoplankton dynamics through the year. However, this is not the goal of this study and we would agree with the reviewer that there are other models and studies that are much more suitable for that.
is not able to explain the phenomena observed within the data
The phenomena that we are looking for are the approximate location and order of magnitude of a phytoplankton bloom. Within this scope, we agree that the model with a system-constant mortality rate (cf. Reference case) is not able to capture the accumulation of phytoplankton in the brackish region in spring in 2008-2014. This is exactly what we conclude in our manuscript and is used to support the main conclusion rejecting the hypothesis that observations can be explained by changes in sediment dynamics only.
However, when using a spatially varying mortality rate related to observed zooplankton concentrations, our model is at least 90 % accurate in all three periods in the region for which we have zooplankton data (beyond km 60). This shows that our model is able to explain the large-scale phenomena provided mortality is chosen in a suitable way.
Reply to reviewer suggestions
The model should investigate the timeinterval in a transient manner
We solve the equations in equilibrium state and not in a transient manner. We argue that this assumption is acceptable because, firstly, the accumulation of phytoplankton in the brackish region covers approximately two months, which is large compared to the time scale of a bloom (~ weeks). In other words, the system has enough time to evolve towards equilibrium. Secondly, we observed the accumulation of phytoplankton consistently over 7 consecutive years (2008-2014). If the system were to be sensitive to initial conditions (e.g. exact temperature and discharge in winter and early spring), and we would thus have to solve the system in a transient manner, we would expect more variability over these 7 years.
On the contrary, we believe that assuming equilibrium state strengthens the conclusions as it shows that the results do not depend on precise conditions in winter or early spring. This allows us to present a sensitivity analysis consistently and efficiently, which is a requirement given the scope of the manuscript.
The model should resolve different phytoplankton and zooplankton groups
Taking one group of phytoplankton which represent the community average is a common approach in literature. Given the scope of this study, we believe this is sufficient as interaction between different species is unlikely to affect the conclusions regarding the effect of sediment on blooms. Furthermore, in the extended model, we do parametrically include effects of two zooplankton groups, based on observations [see Eq. (3)].
Boundary conditions should be defined temporally variable
As our model allows for an extensive sensitivity analysis, we validated the assumption of using constant boundary conditions. The results show a minor impact of variability of the boundary conditions on accumulation of phytoplankton in the brackish region (see Figure 9).
Reply to the major other comments
Gypens et al 2013 (http://dx.doi.org/10.1016/j.jmarsys.2012.10.006) studied the Scheldt estuary and came to more detailed and opposing results.
Assuming that Anonymous Referee #2 means with ‘opposing results’ that Gypens et al. (2013) conclude that ‘grazing pressure plays a negligible role’, this indeed seems to contradict our results. However, we read that the corresponding results are not shown in their paper and it is not explained what experiment was done to get to this conclusion. We thus cannot determine how we should compare our results to the results of Gypens et al. (2013) regarding the sensitivity to grazing.
We do however agree that we cannot strictly draw conclusions about grazing, only about mortality and are happy to change this in our manuscript (see reply to comment 1).
On other aspects, the results presented in Gypens et al. (2013) show similar patterns compared to our results.
Other parameters may play a role too (see McQuatters–Gollop and Vermaat (2011), doi:10.1016/j.seares.2010.12.004)
We agree that there are potentially many parameters affecting phytoplankton dynamics. Here, we want to focus on the effects of sediment and accept that some other model parameters represent many other processes and parameters. Specifically, our mortality parameter describes multiple processes related to water quality as described by the referenced paper and we would be happy to discuss this better (also see reply to comment 1)
Winter values of zooplankton may play a key role (Dudeck et al 2021, doi:10.1093/plankt/fbab011)
The referenced paper does not relate to the Scheldt but to the English Channel. As mentioned above, we observed the accumulation of phytoplankton consistently over 7 consecutive years (2008-2014). If the system were to be sensitive to initial conditions (i.e., winter values), we would expect more variability over these 7 years. Especially because these years included some of the coldest and warmest winters alike in recent history.
L451 ff: The method to calculate SPM for the early periods is very critical, as deep-water SPM structures are governed rather by benthic-pelagic interactions than by surface variations.
This may be true for a transparent system but the Scheldt estuary is a turbid system in which phytoplankton growth is only possible near the water surface. The euphotic depth equals 1.15-0.45 m, much smaller than the depth.
L458: Neglecting background- and phytoplankton-induced light extinction is very critical.
We do include background- and phytoplankton-induced light extinction in the model runs, but not when estimating kc. We agree that neglecting background and phytoplankton-induced light-extinction may be critical in a transparent system. However, the Scheldt estuary is very turbid. Using the Chl-a observations and kP coefficient presented in Table 1 (which complies with values found in the literature), we obtain a kc-value of 67 m2 kg-1 instead of 72 m2 kg-1 for 2015-2018 when we do include phytoplankton-induced light extinction. When we also include background included light-extinction, we find a value of 66 m2 kg-1. As shown by our sensitivity analysis (Figure 8b), the impact of this slight difference on the model results is negligible. However, we would be happy to change this in our manuscript.
Citation: https://doi.org/10.5194/bg-2021-59-AC1
-
CC1: 'Comment on bg-2021-59', Josette Garnier, 18 May 2021
The Scheldt estuary has been studied for a long time in a context of changes in nutrient inputs by the river after improvement of wastewater treatment, especially. The long-term data sets on observed nutrients, phytoplankton and zooplankton provide indeed an opportunity for analyzing the dynamic and trends of the interactions within the food chain.
The use of a model for () interpreting such data is however essential due to the complexity of the hydro-sedimentary processes and how they drive the as much complex biological interactions.
However, despite the simplified Iflow model seems well adapted for interpreting salinity gradient, suspended solids variations, phytoplankton is poorly represented as a unique compartment and without nutrient limitation, with the hypothesis of no limitation. It is difficult to understand why the authors, well known for their work on silica, did not take this important nutrient into account, and hence the diatom and non-diatom compartments.
In addition, it is a pity to have a rather sophisticated model and not include the major variables being discussed, i.e. two types of zooplankton. Its role is treated by magic parameters that arrive at the end to spatially and temporally constrain a phytoplankton mortality rate formulated as a first-order process in relation to algal biomass.
It is even mentioned by the authors: "reasoning still needs to be verified by a model that explicitly resolves zooplankton dynamics ". It would be recommended to better represent phytoplankton and zooplankton in their model to address their scientific questions with a convincing approach.
Such a paper would deserve a publication in Biogeosciences. However, the above remarks call for a resubmission after deep revision of the biogeochemical modelling approach.
Citation: https://doi.org/10.5194/bg-2021-59-CC1 -
AC3: 'Reply on CC1', Dante Horemans, 10 Jun 2021
We thank the Referees for their time to review the manuscript. Although the Referees see the high potential in studying the (combined model approach and) multi-annual observations in the Scheldt estuary, they point out that the reason for using the presented model approach requires further clarification. We believe that we can satisfy all Referees’ comments by (1) a thorough revision of the text, focusing much more on the reasons for our assumptions, embedding in literature, and using the discussion to reflect on the validity of our work rather than to speculate. Furthermore, (2) the Referees’ suggestions inspired us to extend the model by distinguishing between the two dominant phytoplankton groups (i.e., freshwater and marine diatoms). We hope that our response and suggestions allow us to resubmit the manuscript after deep revision. We thank you for your continued interest in our research.
1. General reply and outline for the revised manuscript
In the following, we briefly motivate our model approach, discuss the main weaknesses of its presentation in the manuscript, and propose solutions to tackle these concerns. Next, we respond to the individual remarks of Josette Garnier and Anonymous Referee #3.
1.1 Explanation of our method
The goal of our study is to describe the appearance and disappearance of Chl-a accumulation in the brackish region in spring in 2004-2018 and analyze whether this is due to changes in physical or biological characteristics. Here, the observations are the core. Our model approach is a complementary tool to interpret the data. This is our motivation to construct the model as such it is mainly data-driven and most of the parameters directly follow from the observations. We aim to minimize the number of variables and calibration parameters that we cannot validate using data.
This approach is different compared to state-of-the-art modeling studies such as Arndt et al. (2011), Gypens et al. (2013), and Naithani et al. (2016). In the latter modeling studies, the phytoplankton-zooplankton(-nutrient) dynamics are explicitly resolved over one year, assuming multiple phytoplankton and zooplankton groups. Such models require quite a few calibration parameters that are poorly constrained (e.g., maximum grazing rate, mortality rate per species). These parameters are generally calibrated by fitting to data and then assumed to be fixed in time. Although assuming fixed parameters may be acceptable when focusing on one year, we study an observed trend change, suggesting that (some of these) parameters must have changed over time.
Before our work, the observed trend change in Chl-a was poorly described and it was unclear whether this trend change is related to changes in physical characteristics (e.g., sediment, discharge, temperature) or changes in biological characteristics. In our study, we can constrain this to a change in biological characteristics related to phytoplankton mortality that seems to have some correlation with zooplankton grazing. However, we can at this moment not constrain this further as detailed data is lacking.
We see this as the rejection of the hypothesis that changes in the physics explain the observations and opening a new research question motivating to investigate the biological changes in more detail.
1.2 Suggested outline for the revised manuscript
We see that our approach suffers from chronological reasoning; we start with a light-limited model for phytoplankton growth and then add the grazing functions to the model only in the discussion and speculate about other aspects that affected grazing. Hence, grazing seems an afterthought, which is strengthened by the fact that no data can constrain the final result stating that some change to the functions g1 and g2 must have occurred.
Therefore, we propose to thoroughly restructure the manuscript:
- In the introduction, we will more explicitly state the goal as done above and set the scope of this study (comments by Referee #3). Furthermore, we will more elaborately present the state-of-the-art models and applications to the Scheldt (comments by Referee #3) and motivate our model approach.
- In the methodology section, we will start from the state-of-the-art descriptions and explicitly explain why our simplifications are justified within our scope and how the remaining parameters should be interpreted (comments by all Referees). We will present only one model and will not add new aspects later in the discussion (comments by all Referees). This model will directly include the effect of grazing and distinguish between the two dominant phytoplankton groups (i.e., freshwater and marine diatoms) (comments Josette Garnier).
- We will shorten the results, focusing on the main results and removing the lengthy sensitivity study (comments by Huib de Swart)
- We will thoroughly revise the discussion, focusing on model interpretation in the context of the literature, model limitations, implications for other estuaries, and further work.
Please, find below our response to the main concerns of Josette Garnier and Anonymous Referee #3.
Josette Garnier
Overall, we see that we should better motivate the choices and assumptions made in our model. We can support each of our assumptions based on our data, thereby resolving your concerns. Additionally, we should include the two zooplankton classes from the start, as opposed to introducing them as a model variation at the end. Both points would be fairly easily addressed. We explain this in more detail below.
It is difficult to understand why the authors did not take silica into account, and hence the diatom and non-diatom compartments
We believe that the Referee refers to Si-limitation which sometimes occurs at the downstream boundary of the Scheldt, depending on the season. This is not the focus area of our study. The assumption of a minor impact of N-P nutrient limitation in spring follows from observations (LINES 64-67). Similarly, the Si concentrations are at least one order of magnitude larger than the corresponding half-saturation constants in spring. We will add this to the manuscript.
As the phytoplankton abundance dominantly consists of diatoms (Maris and Meire, 2007; Muylaert et al., 2009; Maris and Meire, 2009, 2013, 2017), we do not distinguish between non-diatoms and diatoms. However, we do realize based on your concerns that it may be useful to distinguish between freshwater and marine diatoms [following, for example, Vanderborcht et al. (2002), Naithani et al. (2016)]. We will add this distinction to the manuscript and present and discuss the corresponding results.
Zooplankton is treated by magic parameters that arrive at the end
We agree that we should present only one model to avoid chronological reasoning (see general response above). We included two zooplankton classes using a data-driven approach. A phytoplankton mortality rate that is linear to the zooplankton abundance is an accepted model set-up (Steele and Henderson, 1992). If we were to resolve the zooplankton dynamics explicitly, we would end up with more calibration parameters besides g1 and g2, such as the mortality rate of zooplankton, which is a parameter that is notoriously hard to constrain.
Anonymous Referee #3
We agree that our approach needs clarification, motivation, and restructuring, including a discussion of state-of-the-art models (see our general response above). In the following, we present a point-to-point response to the other concerns of Anonymous Referee #3.
The title is misleading
We acknowledge the title would benefit from a more concise formulation; with ‘long-term’, we mean ‘multi-annual’, and ‘net growth’ should be replaced by ‘accumulation of Chl-a’.
The claim that decreased herbivory is responsible for the intermediate CHL-a maximum is poorly supported by any kind of evidence
We agree that Section 4.2 contains hypotheses and should be removed. A discussion on the evolution of grazing parameters is the best possible given the available data, which cannot constrain this further. Still, we think we have made significant progress constraining the observed phenomenon to some change in biological characteristics as (direct effects) of changes in physical characteristics (sediment, discharge etc) are insufficient (see also general response above).
As "deus ex machina" the authors impose a 7-fold decrease in "calanoids grazing efficiency"
We discuss the importance of mortality/grazing in the discussion Section 4.1. The required 7-fold decrease in the mortality rate/grazing parameter is indeed not explained within our model context and the data does not provide conclusive evidence here as well. Nonetheless, such variations comply with grazing experiments (LINES 364-366) and are also found in other modeling studies.
The pretty central functions Z1 and Z2 are not given at all
The functions Z1 and Z2 and the corresponding extrapolation are defined in the results section (LINES 309-314). However, we agree that it would be better to mention this already in the methodology section when we introduce the mortality rate [Eq. (3)].
The analysis is restricted to the spring bloom only
This statement by the Referee is correct and should be mentioned more explicitly in the manuscript.
Why are seasonal data displayed and discussed?
We see that this may be confusing as we focus on spring blooms. We will change this and only present the values for the relevant months in a table.
Much higher CHL in 2004-2007 for 80km<x<120km were not analyzed
This can be explained by the lower mortality rate in 2004-2007, resulting in a slower decrease of Chl-a in the downstream direction. We will mention this in the manuscript.
Lacking discussion of the literature
We agree that our model approach is lacking a clear motivation/discussion, including similar state-of-the-art modeling studied in the Scheldt estuary. We will more elaborately present the state-of-the-art models and applications to the Scheldt. We will thoroughly revise the discussion, focusing on model interpretation in the context of the literature, model limitations, implications for other estuaries, and further work (see the general response above).
Citation: https://doi.org/10.5194/bg-2021-59-AC3
-
AC3: 'Reply on CC1', Dante Horemans, 10 Jun 2021
-
RC3: 'Comment on bg-2021-59', Anonymous Referee #3, 19 May 2021
The paper of Horemans et al ("Evolution of the long-term and estuary-scale phytoplankton patterns in the Scheldt estuary: the disappearance of net growth in the brackish region") documents a joint data and modeling study for explaining time-variable accumulation of Chl-a in the Scheldt estuary. I see the merit of the study in presenting valuable observations on physiological changes in phytoplankton over the season or on the abundance of herbivorous grazers along the estuary. Also the modeling puts appropriate emphasis on SPM dynamics as a major driver of primary production in a highly turbid environment. For such systems, so far not many studies sought to combine both new observations and new modeling. As a consequence, I see a high potential value in the paper, which the authors unfortunately fail to exploit properly.
The major problem lays in the approach and its presentation. Already the title is misleading. The issue of "long-term evolution" is nowhere really addressed, nor is the "disappearance of net growth in the brackish region". Apart of the intermediate appearance of a CHL-a maximum in the brackish zone, the study does not discuss any long-term dynamics. After reading the paper two times, I could not find any hint on temporal patterns in net growth rate. It took the second read to grasp the goal of the study, which is understanding the intermediate appearance of a CHL-a maximum. The storyline is blurred by many weaknesses in the structuring and selection of material, of the language, and of graphical presentation. Most importantly, the claim that decreased herbivory is responsible for the intermediate CHL-a maximum is poorly supported by any kind of evidence. And it would take little to convince me, due to similar findings of preceding literature (see below). Observations of zooplankton abundance are shown for x (=distance from the mouth) larger than 70km, thus having little overlap with the zone of interest (40km<x<80km). How were those values extrapolated? The pretty central functions Z_1(x) and Z_2(x) are not given at all! As "deus ex machina" the authors impose a 7-fold decrease in "calanoids grazing efficiency" for the interim period (2008-2014, Table 1) without a clear justification, which process might cause such a drastic inactivation of herbivory. By the way, the notion "grazing efficiency" is simply wrong - as are many biological notions used in the paper (see below). The authors suggest that a roughly 20-25% increase in turbidity (Fig. 4) may have hampered predation rates. This explanation clearly suffers from the mismatch of numbers and most of all ignores that in the upstream Scheldt (x>80km) a lot of calanoids and non-calanoids seem to do very well at much higher turbidity. Finally, a such produced CHL-a maximum displays only a small agreement with the data (Fig. 10b). To conclude, the papers leaves us with a fully unresolved pattern, which is disappointing for a joint data-modeling study. It also remained unclear whether the analysis is restricted to the spring bloom only (Mar-Apr, e.g. Fig. 4, 7-11) or the entire season (Fig. 2,5-6). For example, in the caption of Fig.3 I can only guess whether "in spring" relates to "calanoids and non-calanoids" only - or to "Time-averaged Chl-a concentration" as well. If the focus is on the spring bloom, why are saisonal data displayed and discussed? Data on phytoplankton physiology available for 16 stations have not been shown over the entire transect: here I would expect an important hint on the role of, e.g., photo-acclimation in (non-) producing part of the CHL-a maximum.
The overall low quality of the work is in addition reflected by the lacking discussion of the literature.
Did the authors just forget to do their basic homework, or did they follow a hidden agenda when ignoring previous modeling studies for the same estuary (or the nearby Oosterschelde)? First, from the many estuarine works of Karline Soetaert and her group, only an old one is cited. The very recent and related paper of Jiang, Soetaert et al (Biogeosciences 2020) is not mentioned although it prominently discusses the chlorophyll maximum, which includes a list of further studies on this common pattern in many estuaries, often reported for US estuarine systems (see Fig. 13 in Jiang et al). Similarly, from the model studies of Pierre Regnier and Sandra Arndt only one is cited, but in a false way (as nutrient limitation does play a major role in estuarine phytoplankton dynamics during summer and autumn, at least in the downstream part).
Naithani et al (Hydrobiologia 2016), alike Jiang et al, devised an ecological model including zooplankton dynamics for the Scheldt estuary, hence already accomplished what has been envisioned by Horemans et al in their outlook. That paper already tried to assess the role of trophic cascading for the spatio-temporal estuarine distribution of phytoplankton.
Wirtz (PLOSone 2019) has shown that CHL accumulation in the turbid Wadden Sea indeed is likely the result of intense carnivory by suspension feeders and juvenile fish, which lowers top-down control by herbivores on light-limited autotrophs. A dominant role of trophic links has been backed up by Jiang et al insofar also in their simulations bivalve grazing critically controls phytoplankton distribution. However, here bivalves directly impact phytoplankton concentration in a negative way, while in the study of Wirtz they do so in a positive way by preferably removing zooplankton. These contrasting effects may well be a candidate for explaining the intermediate emergence of the CHL maximum insofar reflecting either changes in the total biomass or the community composition (and thus feeding preference) of estuarine bivalves. However, without a clear presentation of herbivorous biomass along the estuarine transect this route of thinking remains pure speculation. If the single observation within the brackish zone revealing no temporal trend in herbivorous abundance (not identical to biomass!) would be confirmed for the entire zone, the investigation would need to follow new routes. This brings me to the question why the authors have focused on a pattern where the data is incomplete while other interesting features of the spatio-temporal phytoplankton distribution such as the much higher CHL in 2004-2007 for 80km<x<120km were not analyzed. This choice would have eased the constraining of the otherwise poorly documented inverse modeling experiments.To conclude, I cannot recommend publication of this paper in BG.
From the numerous minor comments I focus on language, where three types of issues are most apparent:
* semantic errors (e.g., L16, L19, L24, L46-47, L49, ...)
* repetitive wordings (e.g., 2 x "estuary" in title, 14 x "observe" and 6 x "shows" p.10-11, 4 x "captures" L255-257, ...)
* wrong biological notions ("phytoplankton grazing" L21, "grazing efficiency" typically denotes the energetic yield of ingestion, "mu_max" usually denotes maximal growth rate but is here used for maximal photosynthesis rate, "phytoplankton-induced exponential light extinction coefficient" = self-shading coefficient, "zooplankton grazes .. primary production" L358, "volume-weighted phytoplankton concentration"?, "volume-weighted zooplankton abundance" = biomass ?, ...)Citation: https://doi.org/10.5194/bg-2021-59-RC3 -
AC4: 'Reply on RC3', Dante Horemans, 15 Jun 2021
We thank the Referees for their time to review the manuscript. Although the Referees see the high potential in studying the (combined model approach and) multi-annual observations in the Scheldt estuary, they point out that the reason for using the presented model approach requires further clarification. We believe that we can satisfy all Referees’ comments by (1) a thorough revision of the text, focusing much more on the reasons for our assumptions, embedding in literature, and using the discussion to reflect on the validity of our work rather than to speculate. Furthermore, (2) the Referees’ suggestions inspired us to extend the model by distinguishing between the two dominant phytoplankton groups (i.e., freshwater and marine diatoms). We hope that our response and suggestions allow us to resubmit the manuscript after deep revision. We thank you for your continued interest in our research.
1. General reply and outline for the revised manuscript
In the following, we briefly motivate our model approach, discuss the main weaknesses of its presentation in the manuscript, and propose solutions to tackle these concerns. Next, we respond to the individual remarks of Josette Garnier and Anonymous Referee #3.
1.1 Explanation of our method
The goal of our study is to describe the appearance and disappearance of Chl-a accumulation in the brackish region in spring in 2004-2018 and analyze whether this is due to changes in physical or biological characteristics. Here, the observations are the core. Our model approach is a complementary tool to interpret the data. This is our motivation to construct the model as such it is mainly data-driven and most of the parameters directly follow from the observations. We aim to minimize the number of variables and calibration parameters that we cannot validate using data.
This approach is different compared to state-of-the-art modeling studies such as Arndt et al. (2011), Gypens et al. (2013), and Naithani et al. (2016). In the latter modeling studies, the phytoplankton-zooplankton(-nutrient) dynamics are explicitly resolved over one year, assuming multiple phytoplankton and zooplankton groups. Such models require quite a few calibration parameters that are poorly constrained (e.g., maximum grazing rate, mortality rate per species). These parameters are generally calibrated by fitting to data and then assumed to be fixed in time. Although assuming fixed parameters may be acceptable when focusing on one year, we study an observed trend change, suggesting that (some of these) parameters must have changed over time.
Before our work, the observed trend change in Chl-a was poorly described and it was unclear whether this trend change is related to changes in physical characteristics (e.g., sediment, discharge, temperature) or changes in biological characteristics. In our study, we can constrain this to a change in biological characteristics related to phytoplankton mortality that seems to have some correlation with zooplankton grazing. However, we can at this moment not constrain this further as detailed data is lacking.
We see this as the rejection of the hypothesis that changes in the physics explain the observations and opening a new research question motivating to investigate the biological changes in more detail.
1.2 Suggested outline for the revised manuscript
We see that our approach suffers from chronological reasoning; we start with a light-limited model for phytoplankton growth and then add the grazing functions to the model only in the discussion and speculate about other aspects that affected grazing. Hence, grazing seems an afterthought, which is strengthened by the fact that no data can constrain the final result stating that some change to the functions g1 and g2 must have occurred.
Therefore, we propose to thoroughly restructure the manuscript:
- In the introduction, we will more explicitly state the goal as done above and set the scope of this study (comments by Referee #3). Furthermore, we will more elaborately present the state-of-the-art models and applications to the Scheldt (comments by Referee #3) and motivate our model approach.
- In the methodology section, we will start from the state-of-the-art descriptions and explicitly explain why our simplifications are justified within our scope and how the remaining parameters should be interpreted (comments by all Referees). We will present only one model and will not add new aspects later in the discussion (comments by all Referees). This model will directly include the effect of grazing and distinguish between the two dominant phytoplankton groups (i.e., freshwater and marine diatoms) (comments Josette Garnier).
- We will shorten the results, focusing on the main results and removing the lengthy sensitivity study (comments by Huib de Swart)
- We will thoroughly revise the discussion, focusing on model interpretation in the context of the literature, model limitations, implications for other estuaries, and further work.
Please, find below our response to the main concerns of Josette Garnier and Anonymous Referee #3.
Josette Garnier
Overall, we see that we should better motivate the choices and assumptions made in our model. We can support each of our assumptions based on our data, thereby resolving your concerns. Additionally, we should include the two zooplankton classes from the start, as opposed to introducing them as a model variation at the end. Both points would be fairly easily addressed. We explain this in more detail below.
It is difficult to understand why the authors did not take silica into account, and hence the diatom and non-diatom compartments
We believe that the Referee refers to Si-limitation which sometimes occurs at the downstream boundary of the Scheldt, depending on the season. This is not the focus area of our study. The assumption of a minor impact of N-P nutrient limitation in spring follows from observations (LINES 64-67). Similarly, the Si concentrations are at least one order of magnitude larger than the corresponding half-saturation constants in spring. We will add this to the manuscript.
As the phytoplankton abundance dominantly consists of diatoms (Maris and Meire, 2007; Muylaert et al., 2009; Maris and Meire, 2009, 2013, 2017), we do not distinguish between non-diatoms and diatoms. However, we do realize based on your concerns that it may be useful to distinguish between freshwater and marine diatoms [following, for example, Vanderborcht et al. (2002), Naithani et al. (2016)]. We will add this distinction to the manuscript and present and discuss the corresponding results.
Zooplankton is treated by magic parameters that arrive at the end
We agree that we should present only one model to avoid chronological reasoning (see general response above). We included two zooplankton classes using a data-driven approach. A phytoplankton mortality rate that is linear to the zooplankton abundance is an accepted model set-up (Steele and Henderson, 1992). If we were to resolve the zooplankton dynamics explicitly, we would end up with more calibration parameters besides g1 and g2, such as the mortality rate of zooplankton, which is a parameter that is notoriously hard to constrain.
Anonymous Referee #3
We agree that our approach needs clarification, motivation, and restructuring, including a discussion of state-of-the-art models (see our general response above). In the following, we present a point-to-point response to the other concerns of Anonymous Referee #3.
The title is misleading
We acknowledge the title would benefit from a more concise formulation; with ‘long-term’, we mean ‘multi-annual’, and ‘net growth’ should be replaced by ‘accumulation of Chl-a’.
The claim that decreased herbivory is responsible for the intermediate CHL-a maximum is poorly supported by any kind of evidence
We agree that Section 4.2 contains hypotheses and should be removed. A discussion on the evolution of grazing parameters is the best possible given the available data, which cannot constrain this further. Still, we think we have made significant progress constraining the observed phenomenon to some change in biological characteristics as (direct effects) of changes in physical characteristics (sediment, discharge etc) are insufficient (see also general response above).
As "deus ex machina" the authors impose a 7-fold decrease in "calanoids grazing efficiency"
We discuss the importance of mortality/grazing in the discussion Section 4.1. The required 7-fold decrease in the mortality rate/grazing parameter is indeed not explained within our model context and the data does not provide conclusive evidence here as well. Nonetheless, such variations comply with grazing experiments (LINES 364-366) and are also found in other modeling studies.
The pretty central functions Z1 and Z2 are not given at all
The functions Z1 and Z2 and the corresponding extrapolation are defined in the results section (LINES 309-314). However, we agree that it would be better to mention this already in the methodology section when we introduce the mortality rate [Eq. (3)].
The analysis is restricted to the spring bloom only
This statement by the Referee is correct and should be mentioned more explicitly in the manuscript.
Why are seasonal data displayed and discussed?
We see that this may be confusing as we focus on spring blooms. We will change this and only present the values for the relevant months in a table.
Much higher CHL in 2004-2007 for 80km<x<120km were not analyzed
This can be explained by the lower mortality rate in 2004-2007, resulting in a slower decrease of Chl-a in the downstream direction. We will mention this in the manuscript.
Lacking discussion of the literature
We agree that our model approach is lacking a clear motivation/discussion, including similar state-of-the-art modeling studied in the Scheldt estuary. We will more elaborately present the state-of-the-art models and applications to the Scheldt. We will thoroughly revise the discussion, focusing on model interpretation in the context of the literature, model limitations, implications for other estuaries, and further work (see the general response above).
Citation: https://doi.org/10.5194/bg-2021-59-AC4
-
AC4: 'Reply on RC3', Dante Horemans, 15 Jun 2021
Status: closed
-
RC1: 'Comment on bg-2021-59', Huib E. de Swart, 17 Apr 2021
Overall: this is a paper about an important topic. The paper is well structured and well written. The aims are clear and results are interesting. Clearly, this work is beyond a case study, as it also provides generic and useful knowledge for understanding similar phenomena in other estuaries.
My main comments concern the applied methods (to find the light extinction coefficient), the choice of the model and the formulations/assumptions used in that mode, the presentation of the results and the discussion (see comments below). Based on these, my recommendation is major revision.
-------------------------------------------------------------------------
Comments about the scientific content
Section 2
L101: Why only two measurement points, how useful is it in that case to apply an exponential fit? One would typically expect a number of points here at different distances and use these to make a fit.
Furthermore, it is not clear why the distance was chosen to be 40 cm, please explain.L117: Why use this model instead of e.g. a state-of-the-art model like Delft3D? Alternatively, the authors could have used the model of Arndt et al. (2011), why was decided otherwise? See also later comment: a discussion about the model limitations is missing.
L125: Why is salinity considered to be static and prescribed? That does not seem close to reality.
This comment also applies to the statement in l177 ‘to correct for the large temporal variability in the discharges’.L126: Does the statement ‘Scheldt…well-mixed’ apply to the entire estuary, i.e., there no stratification anywhere? Please add a citation to support this.
L135-136: If being so precise here, then why not consider a dynamic salinity? The choices made seem a bit ad hoc.
Eq. 2b: Shouldn’t the left-hand side be (H+\zeta), rather than H?
Eq. (3)
* Where does this come from, please add a citation here.
* The meaning and dimensions of the symbols g1, g2, Z… is not clear. It takes until Table 1 to conclude that g1 and g1 have the unit s^{-1} L and so the Z’s are in L^-1. It then takes until l205 to find the unit of the Z’s, but there ind. L^{-1} is used; but what is ‘ind’?
* The text ‘extension (section 3.2.4)’ is not satisfactory: motivation should be presented here.
Eqs. (3) and (6): Please remove the ‘dots’ on the right-hand side of the second equation, they’re not used in any other equation.
L155: Why a time-averaged growth rate?
And why is this about implementation, one would expect this to be about the choice for the formulation of \mu, which originates from biological literature.L156: The argument to use Eq. (4) to save computational costs is a bit unexpected. There is no information about the time that a typical run lasts and how much does that costs, is that of the order of a week or more (as suggested by this statement)?
Note: this comment also applies to l170: please add quantitative information.
L160: It is suggested here that E has the unit PAR, but in Table 1 another unit for E is used. Please use either of them.
L162: Specify how the seasonal variation of E_00 is described in the model. Table 1 only contains one value (is that setting for the default setting?)
Eq. (5): The formulation is now such that E is only nonzero during one single day. This should probably be for all daytimes (multiple time intervals).
Eq. (6): Meaning and dimensions of the calibration parameters are not clear. In (6) \mu_01 is raised to a power with a (dimensional) temperature T (expressed in Kelvin?) as an exponent?
Table 1: Values for w_P, k_bg, k_P are expected to originate from biological literature.
Third line: there appear two parameters \mu_max and \alpha here?
Section 3
Figure 3 is very hard to read, because of the many curves with error bars and shading areas. Suggest to present Chl-a and calanoid abundance in different panels.
Consider adding ‘suggestion lines’ between the data points that correspond to a single event. That would help to ‘guide the eye’ of readers.
L237/Figure 6a: the unit seems not correct: from Eqs. (1) and (4) it follows that the unit of \mu_max is 1/s.
L240/Fig. 6b: Unit? From Eq. (4), it seems that the simplest way to write it is PAR^{-1} s^{-1}.
Fig. 7: Recommend to use a different color to plot the model results.
Section 3.2 (sensitivity analysis) is quite exhaustive and has a bit the style of a logbook. Suggest to shorten it and present less figures: not all seem to be necessary. Alternatively, move this section to an Electronic Supplement and only summarize the main findings of the sensitivity analysis in the main text.
Section 4
* L374: It is a bit unexpected that at this stage, two new state variables are introduced, viz. oxygen and fish abundance. If they are important, then why were they not considered earlier? And isn’t oxygen affecting the phytoplankton as well (through respiration)?
L353/372/379: This paper is about verifying a hypothesis (L7). It is then a bit surprising that yet two other hypotheses are formulated that are not verified. The title of Section 4.2 is also unclear in the sense of ‘reconstructions of concentrations’; what is the purpose here, to present a conceptual model?
* Section 4.2 is not really satisfactory. At the same time, important other information is missing, for example about the model limitations (what processes are neglected and how is that justified?) Recommendation: remove the current text of 4.2 and add text about model limitations and context (see next comment).
* What would add value to this study is to put the results in a broader context: in what way do these results of this study contribute to understanding phytoplankton dynamics in other estuaries? This would avoid readers to get the impression that this is a case study.
Appendices
The paper is long and it contains details that are of interest to only a small number of readers. To increase attractiveness and readability, please move all appendices to a separate electronic supplement.
-----------------------------------------------------------------------------
Minors:
Overall:
Remove commas before ‘and’.
Add commas before ‘which’ (or replace ‘which’ by ‘that’).
L3: Why of ‘particular’ interest?
Note: this comment also applies to l27: the Scheldt is definitely interesting, but it not clear what is ‘particular’ about it with respect to other estuaries.
The comment also applies to l37.L9-11: long sentence that is difficult to read. Please split.
L14: explain
L17-19: This text is almost identical to that of l1-3 of the abstract. Please rephrase.
L30: drastically, mainly
L32-35 long sentence, with several subordinate clauses. Suggest to rephrase/split it.
L38: What is meant by ‘we’ here, the authors? Clarify by means of a citation.
L42-44: a complicated sentence, suggest to split it.
L47: observations, to
L51-52: the introduction ends quite abruptly. It would help to write the last paragraph such that it provides an overview of what is in the remaining sections.
L54-55: suggest to move this to the last paragraph of the introduction (see previous comment).
L58: relatively small compared to what?
L66-67: why three citations needed?
L69-73: split.
L76: Is there an unofficial website as well? And perhaps provide the link here.
L87-89: difficult to read (too many subordinate clauses), please split/rephrase.
L134: dynamics, because
L139-148: this is one sentence. Please split.
L149-150: As a start? Further, the word ‘simple’ is rather subjective, the expression in Eq. (3) is already quite complicated.
L173: ‘sensitivity study ... supported … observations’ (? Not clear what is meant here, please rephrase).
L191-193: expected this in the last paragraph of the introduction (see earlier comment).
L242/261/330: please avoid the word ‘conclude’ while being in a section on results. Change e.g. by ‘The long-term ..’ , ‘The results show..’, The… thus shows..’.
L243: extent
L245 and hereafter: Present tense is used for activities that took place in the past. Compare with, for example, the text in Section 5.
L291: Not clear why a new symbol m_new is introduced here.
L304: k_c italic.
L309: Expected this motivation in Section 2, when designing the model experiments.
L336: If the reason is unclear then there is little to discuss. Suggest to write ‘Possible reasons for… are discussed….’
L338-338: Suggest to move this to the last paragraph of the introduction (see previous comment).
L367: surprising, because
-----------------------------------------------------------------------------------------------
Citation: https://doi.org/10.5194/bg-2021-59-RC1 -
AC2: 'Reply on RC1', Dante Horemans, 10 Jun 2021
We thank Prof. Dr. de Swart for his time to review our manuscript. His main comments concern 1. the applied methods (to find the light extinction coefficient), 2. the choice of the model and the formulations/assumptions used in that model, 3. the presentation of the results and the discussion. Below, we present a point-to-point response to his main concerns. We would be happy to also address all other minor comments along with a revised manuscript.
1. The applied methods (to find the light extinction coefficient)
We used the data available within the OMES monitoring program. We agree that measuring at different depths may be preferable in case we expect strong vertical stratification of SPM in the euphotic region. However, given the high turbidity in the Scheldt estuary, the euphotic depth is relatively small (~ dm) compared to the total water depth (~ m). We thus only expect phytoplankton growth near the water surface where we do not expect strong vertical stratification of SPM. Therefore, it is preferable to have a better estimate at the water surface instead of an estimate covering the full water column. We will clarify this in the manuscript.
2. The choice of the model and the formulations/assumptions used in that model
Why use this model instead of e.g. a state-of-the-art model like Delft3D? Alternatively, the authors could have used the model of Arndt et al. (2011), why was decided otherwise? See also later comment: a discussion about the model limitations is missing.
The main reason why we used an idealized model is that it allows us to apply an extensive sensitivity analysis consistently and efficiently over all uncertain parameters, which is a requirement given the scope of the manuscript. We could have alternatively done this within another framework such as that of Arndt et al (2011), which we would have then needed to simplify more as their model already contains too many parameters for our scope. We decided to use iFlow as we already have a track record for this and verified that sediment and flocculation are sufficiently well represented in this model.
For the revision, we suggest including an extensive motivation for choosing these specific simplifying assumptions and the framework that we chose in the introduction and method sections. Also, we will discuss model limitations and strengths better in a separate discussion section (see also comment below).
Why is salinity considered to be static and prescribed?
We motivate the choice of using a static longitudinal salinity profile by the fact that, firstly, salinity gradients are of minor importance to SPM transport in the Scheldt estuary and, secondly, that phytoplankton cells are passive and move with the tidal water flow.
This comment also applies to the statement in l177 ‘to correct for the large temporal variability in the discharges’.
The assumption of using a constant freshwater discharge is validated in our sensitivity analysis, which shows only a minor impact of variability in freshwater discharge on the accumulation of phytoplankton in the brackish region.
3. The presentation of the results and discussion
Section 3.2 (sensitivity analysis) is quite exhaustive and has a bit the style of a logbook. Section 4.2 is not really satisfactory. What would add value to this study is to put the results in a broader context.
We agree with the suggestion to move the Appendix and the results of the sensitivity analysis presented in Section 3.2.1 to an Electronic Supplement. Additionally, we propose a bigger reorganization of the manuscript in which zooplankton is introduced from the start so that no new elements are introduced in the discussion. Furthermore, section 4.2 will be removed and the discussion will focus on discussing the model context, limitations, and applicability to other estuaries.
Citation: https://doi.org/10.5194/bg-2021-59-AC2
-
AC2: 'Reply on RC1', Dante Horemans, 10 Jun 2021
-
RC2: 'Comment on bg-2021-59', Anonymous Referee #2, 21 Apr 2021
Overall Statements:
The manuscript “Evolution of the long-term and estuary-scale phytoplankton patterns in the Scheldt estuary: the disappearance of net growth in the brackish region” by Dante M. L. Horemans, Yoeri M. Dijkstra, Michèle Tackx, Patrick Meire, and Tom J. S. Cox describes the variations of Chl-a distribution in the Scheldt estuary within the time interval 2004 to 2018. The authors try to understand the high spring Chl-a concentrations in the brackish region during the years 2008-2014 which was not observed during other periods. In a first step they analyse observed data such as Chl-a concentration itself, zooplankton abundance, salinity, suspended matter concentration, and light climate related data. Except for some difficulties (see below) these investigations appear sound. In a second step the authors apply an over-simplified model which is not able to explain the phenomena observed within the data. Using parameter-variations they try to show that zooplankton grazing may be the important mechanism to reproduce the elevated spring Chl-a concentrations within the brackish region. The authors argue to use this simple model in order to have good performance. But this does not help as the model does not tackle the phytoplankton dynamics appropriately.
I suggest applying a more adequate model, which describes the investigated time interval in a transient manner and resolves different phytoplankton and zooplankton groups. Boundary conditions should be defined temporally variable. The excellent observation-derived data sets may be used for validation.
Based on these, my recommendation is reject.
Detailed remarks:
L 5: Define the brackish region according to salinity and position.
L 16 ff:
- Do you know similar studies for other rivers and estuaries? Gypens et al 2013 (http://dx.doi.org/10.1016/j.jmarsys.2012.10.006) studied the Scheldt estuary and came to more detailed and opposing results.
- Other parameters may play a role too (see McQuatters–Gollop and Vermaat (2011), doi:10.1016/j.seares.2010.12.004)
- Winter values of zooplankton may play a key role (Dudeck et al 2021, doi:10.1093/plankt/fbab011)
L20: The processes are governed by temperature variations .. salinity variations .. nutrient dynamics.
L59: Give the position of the gauge station.
L64: Silicate concentrations?
L76: Refer to English written peer reviewed articles only.
L77ff: Are the Chl-a measuring methods compatible? I know that in Rijkswaterstaat some efforts have been made to homogenize both approaches.
Figure 1: Indicate the brackish region and give km-positions for at least some of the stations.
L97: Are these methods compatible?
L109: Please introduce first the context of mu-max and alpha.
L116: Give the coordinates and resolution of the model area.
L131: ETM – please introduce this in full words.
L143: To define seaside concentration constant is questionable.
L159: The formula should have “minus alpha” not “alpha”.
L161: The formula should have “+ kc” and “+kp”.
L196: You mean surface concentrations?
L208ff: You often indicate the maximum as the mean plus standard deviation. This is not really the maximum.
L212: I do not understand the argument: “ .., as the estuary is narrow and shallow in the upstream region.”
L225: Can you really justify that SPM is in 2015 – 2018 (at 80 – 120 km) significantly larger than during the other periods, and KD is not significantly larger?
Figure 2 caption: b) is not shown
Figure 3 caption: It is not in the range km 60 – 90 but in the range km 60 – 80.
L227: You used the numbers of May?
L231: This is not always true: See March: The discharge is almost equal but not the salt intrusion.
L242: Say increasing or decreasing development.
L261: from L252-L253 I’ve learned it is decreasing.
L261- L266: I do not understand this section.
L269: You see an accumulation at km 40-60. This is not your brackish region.
L334-L335: Highlight this by a figure and discuss it seriously.
L351: Salinity: This is in opposite to Fig. 5b: The intrusion of salt is largest in 2015 – 2018 for April and later.
L353: Have you studied the planktonic community?
L398: High SPM normally induce light inhibition for phytoplankton.
L432: This formula has 2 degrees of freedom. How do you find an optimum?
L436: The time-dependence of the parameters are questionable.
L451 ff: The method to calculate SPM for the early periods is very critical, as deep-water SPM structures are governed rather by benthic-pelagic interactions than by surface variations.
L458: Neglecting background and phytoplankton induced light extinction is very critical.
Figure C1. Please give units.
Citation: https://doi.org/10.5194/bg-2021-59-RC2 -
AC1: 'Reply on RC2', Dante Horemans, 11 May 2021
We thank Anonymous Referee #2 for their time to review our manuscript. In summary, the core argument for Anonymous Referee #2 to reject our work is in the comments that the model ‘is an over-simplified model’, ‘is not able to explain the phenomena observed within the data’, and ‘does not tackle the phytoplankton dynamics appropriately’. The reviewer is additionally very specific on the elements that need to be in a model, which is a model that ‘investigates timeinterval in a transient manner and resolves different phytoplankton and zooplankton groups’ and that ‘Boundary conditions should be defined temporally variable’.
We do recognize that our conclusions may be too strongly suggested regarding the importance of grazing and are happy to revise this. Otherwise, we do believe our model is well-suited for the scope of our study. Below, we motivate this by a point-to-point response to the main concerns of Anonymous Referee #2.
Reply to reviewer main comments
Over-simplified model/does not tackle the phytoplankton dynamics appropriately
The goal of our study is to investigate the hypothesis that the observed rise and fall of a mid-estuary phytoplankton bloom could be explained by sediment concentration. We deliberately kept our model simple and focused on the effects of sediment-induced shading to be able to study exactly this without any other obscuring effects. Hence, we could do an exhaustive sensitivity study over all parameters. This yields a very robust result that rejects the hypothesis.
We do agree with the reviewer (also see several of the comments below) that it might be too premature to point to grazing as the alternative explanation. Our model might indeed be too simple to show this without doubt. We would be happy to revise our discussion in this respect, primarily concluding that phytoplankton mortality is important and stressing that there could be multiple explanations to this, of which grazing is one.
We would agree that the model is too simple to simulate the precise phytoplankton dynamics through the year. However, this is not the goal of this study and we would agree with the reviewer that there are other models and studies that are much more suitable for that.
is not able to explain the phenomena observed within the data
The phenomena that we are looking for are the approximate location and order of magnitude of a phytoplankton bloom. Within this scope, we agree that the model with a system-constant mortality rate (cf. Reference case) is not able to capture the accumulation of phytoplankton in the brackish region in spring in 2008-2014. This is exactly what we conclude in our manuscript and is used to support the main conclusion rejecting the hypothesis that observations can be explained by changes in sediment dynamics only.
However, when using a spatially varying mortality rate related to observed zooplankton concentrations, our model is at least 90 % accurate in all three periods in the region for which we have zooplankton data (beyond km 60). This shows that our model is able to explain the large-scale phenomena provided mortality is chosen in a suitable way.
Reply to reviewer suggestions
The model should investigate the timeinterval in a transient manner
We solve the equations in equilibrium state and not in a transient manner. We argue that this assumption is acceptable because, firstly, the accumulation of phytoplankton in the brackish region covers approximately two months, which is large compared to the time scale of a bloom (~ weeks). In other words, the system has enough time to evolve towards equilibrium. Secondly, we observed the accumulation of phytoplankton consistently over 7 consecutive years (2008-2014). If the system were to be sensitive to initial conditions (e.g. exact temperature and discharge in winter and early spring), and we would thus have to solve the system in a transient manner, we would expect more variability over these 7 years.
On the contrary, we believe that assuming equilibrium state strengthens the conclusions as it shows that the results do not depend on precise conditions in winter or early spring. This allows us to present a sensitivity analysis consistently and efficiently, which is a requirement given the scope of the manuscript.
The model should resolve different phytoplankton and zooplankton groups
Taking one group of phytoplankton which represent the community average is a common approach in literature. Given the scope of this study, we believe this is sufficient as interaction between different species is unlikely to affect the conclusions regarding the effect of sediment on blooms. Furthermore, in the extended model, we do parametrically include effects of two zooplankton groups, based on observations [see Eq. (3)].
Boundary conditions should be defined temporally variable
As our model allows for an extensive sensitivity analysis, we validated the assumption of using constant boundary conditions. The results show a minor impact of variability of the boundary conditions on accumulation of phytoplankton in the brackish region (see Figure 9).
Reply to the major other comments
Gypens et al 2013 (http://dx.doi.org/10.1016/j.jmarsys.2012.10.006) studied the Scheldt estuary and came to more detailed and opposing results.
Assuming that Anonymous Referee #2 means with ‘opposing results’ that Gypens et al. (2013) conclude that ‘grazing pressure plays a negligible role’, this indeed seems to contradict our results. However, we read that the corresponding results are not shown in their paper and it is not explained what experiment was done to get to this conclusion. We thus cannot determine how we should compare our results to the results of Gypens et al. (2013) regarding the sensitivity to grazing.
We do however agree that we cannot strictly draw conclusions about grazing, only about mortality and are happy to change this in our manuscript (see reply to comment 1).
On other aspects, the results presented in Gypens et al. (2013) show similar patterns compared to our results.
Other parameters may play a role too (see McQuatters–Gollop and Vermaat (2011), doi:10.1016/j.seares.2010.12.004)
We agree that there are potentially many parameters affecting phytoplankton dynamics. Here, we want to focus on the effects of sediment and accept that some other model parameters represent many other processes and parameters. Specifically, our mortality parameter describes multiple processes related to water quality as described by the referenced paper and we would be happy to discuss this better (also see reply to comment 1)
Winter values of zooplankton may play a key role (Dudeck et al 2021, doi:10.1093/plankt/fbab011)
The referenced paper does not relate to the Scheldt but to the English Channel. As mentioned above, we observed the accumulation of phytoplankton consistently over 7 consecutive years (2008-2014). If the system were to be sensitive to initial conditions (i.e., winter values), we would expect more variability over these 7 years. Especially because these years included some of the coldest and warmest winters alike in recent history.
L451 ff: The method to calculate SPM for the early periods is very critical, as deep-water SPM structures are governed rather by benthic-pelagic interactions than by surface variations.
This may be true for a transparent system but the Scheldt estuary is a turbid system in which phytoplankton growth is only possible near the water surface. The euphotic depth equals 1.15-0.45 m, much smaller than the depth.
L458: Neglecting background- and phytoplankton-induced light extinction is very critical.
We do include background- and phytoplankton-induced light extinction in the model runs, but not when estimating kc. We agree that neglecting background and phytoplankton-induced light-extinction may be critical in a transparent system. However, the Scheldt estuary is very turbid. Using the Chl-a observations and kP coefficient presented in Table 1 (which complies with values found in the literature), we obtain a kc-value of 67 m2 kg-1 instead of 72 m2 kg-1 for 2015-2018 when we do include phytoplankton-induced light extinction. When we also include background included light-extinction, we find a value of 66 m2 kg-1. As shown by our sensitivity analysis (Figure 8b), the impact of this slight difference on the model results is negligible. However, we would be happy to change this in our manuscript.
Citation: https://doi.org/10.5194/bg-2021-59-AC1
-
CC1: 'Comment on bg-2021-59', Josette Garnier, 18 May 2021
The Scheldt estuary has been studied for a long time in a context of changes in nutrient inputs by the river after improvement of wastewater treatment, especially. The long-term data sets on observed nutrients, phytoplankton and zooplankton provide indeed an opportunity for analyzing the dynamic and trends of the interactions within the food chain.
The use of a model for () interpreting such data is however essential due to the complexity of the hydro-sedimentary processes and how they drive the as much complex biological interactions.
However, despite the simplified Iflow model seems well adapted for interpreting salinity gradient, suspended solids variations, phytoplankton is poorly represented as a unique compartment and without nutrient limitation, with the hypothesis of no limitation. It is difficult to understand why the authors, well known for their work on silica, did not take this important nutrient into account, and hence the diatom and non-diatom compartments.
In addition, it is a pity to have a rather sophisticated model and not include the major variables being discussed, i.e. two types of zooplankton. Its role is treated by magic parameters that arrive at the end to spatially and temporally constrain a phytoplankton mortality rate formulated as a first-order process in relation to algal biomass.
It is even mentioned by the authors: "reasoning still needs to be verified by a model that explicitly resolves zooplankton dynamics ". It would be recommended to better represent phytoplankton and zooplankton in their model to address their scientific questions with a convincing approach.
Such a paper would deserve a publication in Biogeosciences. However, the above remarks call for a resubmission after deep revision of the biogeochemical modelling approach.
Citation: https://doi.org/10.5194/bg-2021-59-CC1 -
AC3: 'Reply on CC1', Dante Horemans, 10 Jun 2021
We thank the Referees for their time to review the manuscript. Although the Referees see the high potential in studying the (combined model approach and) multi-annual observations in the Scheldt estuary, they point out that the reason for using the presented model approach requires further clarification. We believe that we can satisfy all Referees’ comments by (1) a thorough revision of the text, focusing much more on the reasons for our assumptions, embedding in literature, and using the discussion to reflect on the validity of our work rather than to speculate. Furthermore, (2) the Referees’ suggestions inspired us to extend the model by distinguishing between the two dominant phytoplankton groups (i.e., freshwater and marine diatoms). We hope that our response and suggestions allow us to resubmit the manuscript after deep revision. We thank you for your continued interest in our research.
1. General reply and outline for the revised manuscript
In the following, we briefly motivate our model approach, discuss the main weaknesses of its presentation in the manuscript, and propose solutions to tackle these concerns. Next, we respond to the individual remarks of Josette Garnier and Anonymous Referee #3.
1.1 Explanation of our method
The goal of our study is to describe the appearance and disappearance of Chl-a accumulation in the brackish region in spring in 2004-2018 and analyze whether this is due to changes in physical or biological characteristics. Here, the observations are the core. Our model approach is a complementary tool to interpret the data. This is our motivation to construct the model as such it is mainly data-driven and most of the parameters directly follow from the observations. We aim to minimize the number of variables and calibration parameters that we cannot validate using data.
This approach is different compared to state-of-the-art modeling studies such as Arndt et al. (2011), Gypens et al. (2013), and Naithani et al. (2016). In the latter modeling studies, the phytoplankton-zooplankton(-nutrient) dynamics are explicitly resolved over one year, assuming multiple phytoplankton and zooplankton groups. Such models require quite a few calibration parameters that are poorly constrained (e.g., maximum grazing rate, mortality rate per species). These parameters are generally calibrated by fitting to data and then assumed to be fixed in time. Although assuming fixed parameters may be acceptable when focusing on one year, we study an observed trend change, suggesting that (some of these) parameters must have changed over time.
Before our work, the observed trend change in Chl-a was poorly described and it was unclear whether this trend change is related to changes in physical characteristics (e.g., sediment, discharge, temperature) or changes in biological characteristics. In our study, we can constrain this to a change in biological characteristics related to phytoplankton mortality that seems to have some correlation with zooplankton grazing. However, we can at this moment not constrain this further as detailed data is lacking.
We see this as the rejection of the hypothesis that changes in the physics explain the observations and opening a new research question motivating to investigate the biological changes in more detail.
1.2 Suggested outline for the revised manuscript
We see that our approach suffers from chronological reasoning; we start with a light-limited model for phytoplankton growth and then add the grazing functions to the model only in the discussion and speculate about other aspects that affected grazing. Hence, grazing seems an afterthought, which is strengthened by the fact that no data can constrain the final result stating that some change to the functions g1 and g2 must have occurred.
Therefore, we propose to thoroughly restructure the manuscript:
- In the introduction, we will more explicitly state the goal as done above and set the scope of this study (comments by Referee #3). Furthermore, we will more elaborately present the state-of-the-art models and applications to the Scheldt (comments by Referee #3) and motivate our model approach.
- In the methodology section, we will start from the state-of-the-art descriptions and explicitly explain why our simplifications are justified within our scope and how the remaining parameters should be interpreted (comments by all Referees). We will present only one model and will not add new aspects later in the discussion (comments by all Referees). This model will directly include the effect of grazing and distinguish between the two dominant phytoplankton groups (i.e., freshwater and marine diatoms) (comments Josette Garnier).
- We will shorten the results, focusing on the main results and removing the lengthy sensitivity study (comments by Huib de Swart)
- We will thoroughly revise the discussion, focusing on model interpretation in the context of the literature, model limitations, implications for other estuaries, and further work.
Please, find below our response to the main concerns of Josette Garnier and Anonymous Referee #3.
Josette Garnier
Overall, we see that we should better motivate the choices and assumptions made in our model. We can support each of our assumptions based on our data, thereby resolving your concerns. Additionally, we should include the two zooplankton classes from the start, as opposed to introducing them as a model variation at the end. Both points would be fairly easily addressed. We explain this in more detail below.
It is difficult to understand why the authors did not take silica into account, and hence the diatom and non-diatom compartments
We believe that the Referee refers to Si-limitation which sometimes occurs at the downstream boundary of the Scheldt, depending on the season. This is not the focus area of our study. The assumption of a minor impact of N-P nutrient limitation in spring follows from observations (LINES 64-67). Similarly, the Si concentrations are at least one order of magnitude larger than the corresponding half-saturation constants in spring. We will add this to the manuscript.
As the phytoplankton abundance dominantly consists of diatoms (Maris and Meire, 2007; Muylaert et al., 2009; Maris and Meire, 2009, 2013, 2017), we do not distinguish between non-diatoms and diatoms. However, we do realize based on your concerns that it may be useful to distinguish between freshwater and marine diatoms [following, for example, Vanderborcht et al. (2002), Naithani et al. (2016)]. We will add this distinction to the manuscript and present and discuss the corresponding results.
Zooplankton is treated by magic parameters that arrive at the end
We agree that we should present only one model to avoid chronological reasoning (see general response above). We included two zooplankton classes using a data-driven approach. A phytoplankton mortality rate that is linear to the zooplankton abundance is an accepted model set-up (Steele and Henderson, 1992). If we were to resolve the zooplankton dynamics explicitly, we would end up with more calibration parameters besides g1 and g2, such as the mortality rate of zooplankton, which is a parameter that is notoriously hard to constrain.
Anonymous Referee #3
We agree that our approach needs clarification, motivation, and restructuring, including a discussion of state-of-the-art models (see our general response above). In the following, we present a point-to-point response to the other concerns of Anonymous Referee #3.
The title is misleading
We acknowledge the title would benefit from a more concise formulation; with ‘long-term’, we mean ‘multi-annual’, and ‘net growth’ should be replaced by ‘accumulation of Chl-a’.
The claim that decreased herbivory is responsible for the intermediate CHL-a maximum is poorly supported by any kind of evidence
We agree that Section 4.2 contains hypotheses and should be removed. A discussion on the evolution of grazing parameters is the best possible given the available data, which cannot constrain this further. Still, we think we have made significant progress constraining the observed phenomenon to some change in biological characteristics as (direct effects) of changes in physical characteristics (sediment, discharge etc) are insufficient (see also general response above).
As "deus ex machina" the authors impose a 7-fold decrease in "calanoids grazing efficiency"
We discuss the importance of mortality/grazing in the discussion Section 4.1. The required 7-fold decrease in the mortality rate/grazing parameter is indeed not explained within our model context and the data does not provide conclusive evidence here as well. Nonetheless, such variations comply with grazing experiments (LINES 364-366) and are also found in other modeling studies.
The pretty central functions Z1 and Z2 are not given at all
The functions Z1 and Z2 and the corresponding extrapolation are defined in the results section (LINES 309-314). However, we agree that it would be better to mention this already in the methodology section when we introduce the mortality rate [Eq. (3)].
The analysis is restricted to the spring bloom only
This statement by the Referee is correct and should be mentioned more explicitly in the manuscript.
Why are seasonal data displayed and discussed?
We see that this may be confusing as we focus on spring blooms. We will change this and only present the values for the relevant months in a table.
Much higher CHL in 2004-2007 for 80km<x<120km were not analyzed
This can be explained by the lower mortality rate in 2004-2007, resulting in a slower decrease of Chl-a in the downstream direction. We will mention this in the manuscript.
Lacking discussion of the literature
We agree that our model approach is lacking a clear motivation/discussion, including similar state-of-the-art modeling studied in the Scheldt estuary. We will more elaborately present the state-of-the-art models and applications to the Scheldt. We will thoroughly revise the discussion, focusing on model interpretation in the context of the literature, model limitations, implications for other estuaries, and further work (see the general response above).
Citation: https://doi.org/10.5194/bg-2021-59-AC3
-
AC3: 'Reply on CC1', Dante Horemans, 10 Jun 2021
-
RC3: 'Comment on bg-2021-59', Anonymous Referee #3, 19 May 2021
The paper of Horemans et al ("Evolution of the long-term and estuary-scale phytoplankton patterns in the Scheldt estuary: the disappearance of net growth in the brackish region") documents a joint data and modeling study for explaining time-variable accumulation of Chl-a in the Scheldt estuary. I see the merit of the study in presenting valuable observations on physiological changes in phytoplankton over the season or on the abundance of herbivorous grazers along the estuary. Also the modeling puts appropriate emphasis on SPM dynamics as a major driver of primary production in a highly turbid environment. For such systems, so far not many studies sought to combine both new observations and new modeling. As a consequence, I see a high potential value in the paper, which the authors unfortunately fail to exploit properly.
The major problem lays in the approach and its presentation. Already the title is misleading. The issue of "long-term evolution" is nowhere really addressed, nor is the "disappearance of net growth in the brackish region". Apart of the intermediate appearance of a CHL-a maximum in the brackish zone, the study does not discuss any long-term dynamics. After reading the paper two times, I could not find any hint on temporal patterns in net growth rate. It took the second read to grasp the goal of the study, which is understanding the intermediate appearance of a CHL-a maximum. The storyline is blurred by many weaknesses in the structuring and selection of material, of the language, and of graphical presentation. Most importantly, the claim that decreased herbivory is responsible for the intermediate CHL-a maximum is poorly supported by any kind of evidence. And it would take little to convince me, due to similar findings of preceding literature (see below). Observations of zooplankton abundance are shown for x (=distance from the mouth) larger than 70km, thus having little overlap with the zone of interest (40km<x<80km). How were those values extrapolated? The pretty central functions Z_1(x) and Z_2(x) are not given at all! As "deus ex machina" the authors impose a 7-fold decrease in "calanoids grazing efficiency" for the interim period (2008-2014, Table 1) without a clear justification, which process might cause such a drastic inactivation of herbivory. By the way, the notion "grazing efficiency" is simply wrong - as are many biological notions used in the paper (see below). The authors suggest that a roughly 20-25% increase in turbidity (Fig. 4) may have hampered predation rates. This explanation clearly suffers from the mismatch of numbers and most of all ignores that in the upstream Scheldt (x>80km) a lot of calanoids and non-calanoids seem to do very well at much higher turbidity. Finally, a such produced CHL-a maximum displays only a small agreement with the data (Fig. 10b). To conclude, the papers leaves us with a fully unresolved pattern, which is disappointing for a joint data-modeling study. It also remained unclear whether the analysis is restricted to the spring bloom only (Mar-Apr, e.g. Fig. 4, 7-11) or the entire season (Fig. 2,5-6). For example, in the caption of Fig.3 I can only guess whether "in spring" relates to "calanoids and non-calanoids" only - or to "Time-averaged Chl-a concentration" as well. If the focus is on the spring bloom, why are saisonal data displayed and discussed? Data on phytoplankton physiology available for 16 stations have not been shown over the entire transect: here I would expect an important hint on the role of, e.g., photo-acclimation in (non-) producing part of the CHL-a maximum.
The overall low quality of the work is in addition reflected by the lacking discussion of the literature.
Did the authors just forget to do their basic homework, or did they follow a hidden agenda when ignoring previous modeling studies for the same estuary (or the nearby Oosterschelde)? First, from the many estuarine works of Karline Soetaert and her group, only an old one is cited. The very recent and related paper of Jiang, Soetaert et al (Biogeosciences 2020) is not mentioned although it prominently discusses the chlorophyll maximum, which includes a list of further studies on this common pattern in many estuaries, often reported for US estuarine systems (see Fig. 13 in Jiang et al). Similarly, from the model studies of Pierre Regnier and Sandra Arndt only one is cited, but in a false way (as nutrient limitation does play a major role in estuarine phytoplankton dynamics during summer and autumn, at least in the downstream part).
Naithani et al (Hydrobiologia 2016), alike Jiang et al, devised an ecological model including zooplankton dynamics for the Scheldt estuary, hence already accomplished what has been envisioned by Horemans et al in their outlook. That paper already tried to assess the role of trophic cascading for the spatio-temporal estuarine distribution of phytoplankton.
Wirtz (PLOSone 2019) has shown that CHL accumulation in the turbid Wadden Sea indeed is likely the result of intense carnivory by suspension feeders and juvenile fish, which lowers top-down control by herbivores on light-limited autotrophs. A dominant role of trophic links has been backed up by Jiang et al insofar also in their simulations bivalve grazing critically controls phytoplankton distribution. However, here bivalves directly impact phytoplankton concentration in a negative way, while in the study of Wirtz they do so in a positive way by preferably removing zooplankton. These contrasting effects may well be a candidate for explaining the intermediate emergence of the CHL maximum insofar reflecting either changes in the total biomass or the community composition (and thus feeding preference) of estuarine bivalves. However, without a clear presentation of herbivorous biomass along the estuarine transect this route of thinking remains pure speculation. If the single observation within the brackish zone revealing no temporal trend in herbivorous abundance (not identical to biomass!) would be confirmed for the entire zone, the investigation would need to follow new routes. This brings me to the question why the authors have focused on a pattern where the data is incomplete while other interesting features of the spatio-temporal phytoplankton distribution such as the much higher CHL in 2004-2007 for 80km<x<120km were not analyzed. This choice would have eased the constraining of the otherwise poorly documented inverse modeling experiments.To conclude, I cannot recommend publication of this paper in BG.
From the numerous minor comments I focus on language, where three types of issues are most apparent:
* semantic errors (e.g., L16, L19, L24, L46-47, L49, ...)
* repetitive wordings (e.g., 2 x "estuary" in title, 14 x "observe" and 6 x "shows" p.10-11, 4 x "captures" L255-257, ...)
* wrong biological notions ("phytoplankton grazing" L21, "grazing efficiency" typically denotes the energetic yield of ingestion, "mu_max" usually denotes maximal growth rate but is here used for maximal photosynthesis rate, "phytoplankton-induced exponential light extinction coefficient" = self-shading coefficient, "zooplankton grazes .. primary production" L358, "volume-weighted phytoplankton concentration"?, "volume-weighted zooplankton abundance" = biomass ?, ...)Citation: https://doi.org/10.5194/bg-2021-59-RC3 -
AC4: 'Reply on RC3', Dante Horemans, 15 Jun 2021
We thank the Referees for their time to review the manuscript. Although the Referees see the high potential in studying the (combined model approach and) multi-annual observations in the Scheldt estuary, they point out that the reason for using the presented model approach requires further clarification. We believe that we can satisfy all Referees’ comments by (1) a thorough revision of the text, focusing much more on the reasons for our assumptions, embedding in literature, and using the discussion to reflect on the validity of our work rather than to speculate. Furthermore, (2) the Referees’ suggestions inspired us to extend the model by distinguishing between the two dominant phytoplankton groups (i.e., freshwater and marine diatoms). We hope that our response and suggestions allow us to resubmit the manuscript after deep revision. We thank you for your continued interest in our research.
1. General reply and outline for the revised manuscript
In the following, we briefly motivate our model approach, discuss the main weaknesses of its presentation in the manuscript, and propose solutions to tackle these concerns. Next, we respond to the individual remarks of Josette Garnier and Anonymous Referee #3.
1.1 Explanation of our method
The goal of our study is to describe the appearance and disappearance of Chl-a accumulation in the brackish region in spring in 2004-2018 and analyze whether this is due to changes in physical or biological characteristics. Here, the observations are the core. Our model approach is a complementary tool to interpret the data. This is our motivation to construct the model as such it is mainly data-driven and most of the parameters directly follow from the observations. We aim to minimize the number of variables and calibration parameters that we cannot validate using data.
This approach is different compared to state-of-the-art modeling studies such as Arndt et al. (2011), Gypens et al. (2013), and Naithani et al. (2016). In the latter modeling studies, the phytoplankton-zooplankton(-nutrient) dynamics are explicitly resolved over one year, assuming multiple phytoplankton and zooplankton groups. Such models require quite a few calibration parameters that are poorly constrained (e.g., maximum grazing rate, mortality rate per species). These parameters are generally calibrated by fitting to data and then assumed to be fixed in time. Although assuming fixed parameters may be acceptable when focusing on one year, we study an observed trend change, suggesting that (some of these) parameters must have changed over time.
Before our work, the observed trend change in Chl-a was poorly described and it was unclear whether this trend change is related to changes in physical characteristics (e.g., sediment, discharge, temperature) or changes in biological characteristics. In our study, we can constrain this to a change in biological characteristics related to phytoplankton mortality that seems to have some correlation with zooplankton grazing. However, we can at this moment not constrain this further as detailed data is lacking.
We see this as the rejection of the hypothesis that changes in the physics explain the observations and opening a new research question motivating to investigate the biological changes in more detail.
1.2 Suggested outline for the revised manuscript
We see that our approach suffers from chronological reasoning; we start with a light-limited model for phytoplankton growth and then add the grazing functions to the model only in the discussion and speculate about other aspects that affected grazing. Hence, grazing seems an afterthought, which is strengthened by the fact that no data can constrain the final result stating that some change to the functions g1 and g2 must have occurred.
Therefore, we propose to thoroughly restructure the manuscript:
- In the introduction, we will more explicitly state the goal as done above and set the scope of this study (comments by Referee #3). Furthermore, we will more elaborately present the state-of-the-art models and applications to the Scheldt (comments by Referee #3) and motivate our model approach.
- In the methodology section, we will start from the state-of-the-art descriptions and explicitly explain why our simplifications are justified within our scope and how the remaining parameters should be interpreted (comments by all Referees). We will present only one model and will not add new aspects later in the discussion (comments by all Referees). This model will directly include the effect of grazing and distinguish between the two dominant phytoplankton groups (i.e., freshwater and marine diatoms) (comments Josette Garnier).
- We will shorten the results, focusing on the main results and removing the lengthy sensitivity study (comments by Huib de Swart)
- We will thoroughly revise the discussion, focusing on model interpretation in the context of the literature, model limitations, implications for other estuaries, and further work.
Please, find below our response to the main concerns of Josette Garnier and Anonymous Referee #3.
Josette Garnier
Overall, we see that we should better motivate the choices and assumptions made in our model. We can support each of our assumptions based on our data, thereby resolving your concerns. Additionally, we should include the two zooplankton classes from the start, as opposed to introducing them as a model variation at the end. Both points would be fairly easily addressed. We explain this in more detail below.
It is difficult to understand why the authors did not take silica into account, and hence the diatom and non-diatom compartments
We believe that the Referee refers to Si-limitation which sometimes occurs at the downstream boundary of the Scheldt, depending on the season. This is not the focus area of our study. The assumption of a minor impact of N-P nutrient limitation in spring follows from observations (LINES 64-67). Similarly, the Si concentrations are at least one order of magnitude larger than the corresponding half-saturation constants in spring. We will add this to the manuscript.
As the phytoplankton abundance dominantly consists of diatoms (Maris and Meire, 2007; Muylaert et al., 2009; Maris and Meire, 2009, 2013, 2017), we do not distinguish between non-diatoms and diatoms. However, we do realize based on your concerns that it may be useful to distinguish between freshwater and marine diatoms [following, for example, Vanderborcht et al. (2002), Naithani et al. (2016)]. We will add this distinction to the manuscript and present and discuss the corresponding results.
Zooplankton is treated by magic parameters that arrive at the end
We agree that we should present only one model to avoid chronological reasoning (see general response above). We included two zooplankton classes using a data-driven approach. A phytoplankton mortality rate that is linear to the zooplankton abundance is an accepted model set-up (Steele and Henderson, 1992). If we were to resolve the zooplankton dynamics explicitly, we would end up with more calibration parameters besides g1 and g2, such as the mortality rate of zooplankton, which is a parameter that is notoriously hard to constrain.
Anonymous Referee #3
We agree that our approach needs clarification, motivation, and restructuring, including a discussion of state-of-the-art models (see our general response above). In the following, we present a point-to-point response to the other concerns of Anonymous Referee #3.
The title is misleading
We acknowledge the title would benefit from a more concise formulation; with ‘long-term’, we mean ‘multi-annual’, and ‘net growth’ should be replaced by ‘accumulation of Chl-a’.
The claim that decreased herbivory is responsible for the intermediate CHL-a maximum is poorly supported by any kind of evidence
We agree that Section 4.2 contains hypotheses and should be removed. A discussion on the evolution of grazing parameters is the best possible given the available data, which cannot constrain this further. Still, we think we have made significant progress constraining the observed phenomenon to some change in biological characteristics as (direct effects) of changes in physical characteristics (sediment, discharge etc) are insufficient (see also general response above).
As "deus ex machina" the authors impose a 7-fold decrease in "calanoids grazing efficiency"
We discuss the importance of mortality/grazing in the discussion Section 4.1. The required 7-fold decrease in the mortality rate/grazing parameter is indeed not explained within our model context and the data does not provide conclusive evidence here as well. Nonetheless, such variations comply with grazing experiments (LINES 364-366) and are also found in other modeling studies.
The pretty central functions Z1 and Z2 are not given at all
The functions Z1 and Z2 and the corresponding extrapolation are defined in the results section (LINES 309-314). However, we agree that it would be better to mention this already in the methodology section when we introduce the mortality rate [Eq. (3)].
The analysis is restricted to the spring bloom only
This statement by the Referee is correct and should be mentioned more explicitly in the manuscript.
Why are seasonal data displayed and discussed?
We see that this may be confusing as we focus on spring blooms. We will change this and only present the values for the relevant months in a table.
Much higher CHL in 2004-2007 for 80km<x<120km were not analyzed
This can be explained by the lower mortality rate in 2004-2007, resulting in a slower decrease of Chl-a in the downstream direction. We will mention this in the manuscript.
Lacking discussion of the literature
We agree that our model approach is lacking a clear motivation/discussion, including similar state-of-the-art modeling studied in the Scheldt estuary. We will more elaborately present the state-of-the-art models and applications to the Scheldt. We will thoroughly revise the discussion, focusing on model interpretation in the context of the literature, model limitations, implications for other estuaries, and further work (see the general response above).
Citation: https://doi.org/10.5194/bg-2021-59-AC4
-
AC4: 'Reply on RC3', Dante Horemans, 15 Jun 2021
Dante M. L. Horemans et al.
Data sets
Scheldt data observed in the Netherlands Rijkswaterstaat https://waterinfo.rws.nl
Scheldt data observed in Belgium De Vlaamse Waterweg http://www.omes-monitoring.be/en/data
Model code and software
iFlow version 2.9 Yoeri M. Dijkstra, Ronald L. Brouwer, and Dante M. L. Horemans https://doi.org/10.5281/zenodo.4560637
Dante M. L. Horemans et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 843 | 498 | 37 | 1,378 | 92 | 23 | 21 |
- HTML: 843
- PDF: 498
- XML: 37
- Total: 1,378
- Supplement: 92
- BibTeX: 23
- EndNote: 21
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1