the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 18 Jun 2025
Wind and Phytoplankton Dynamics Drive Seasonal and Short-Term Variability of Suspended Matter in a Tidal Basin
Abstract. Suspended particulate matter (SPM) is a key component of coastal ecosystems, modulating light availability, nutrient transport, and food web dynamics. Its variability is driven by a combination of physical and biological processes that interact across temporal and spatial scales. Using the Sylt-Rømø Bight as a natural laboratory and focusing on the period 2000–2019, this study integrates a long-term biogeochemical time series from the Sylt Roads monitoring program and meteorological observations with Lagrangian transport simulations and neural network modelling to disentangle and quantify the relative roles of tidal dynamics, winds, and phytoplankton mediated biological processes in shaping SPM concentrations measured at two stations near the water surface.
The findings show that wind intensity dominates short-term SPM variability, particularly at the shallow station, where SPM responds rapidly to local wind-induced resuspension. At the deep station, the wind effects appear with a delay of ~5 days, aligning with tidally induced transport timescales (~133 hours) from shallower resuspension zones, as revealed by Lagrangian simulations. Seasonal patterns are further modulated by both reduced wind intensities and the onset of biological processes, with phytoplankton blooms promoting flocculation and subsequent settling in spring and summer. Neural network experiments highlight the shifting seasonal balance between physical and biological controls: models trained on winter data overestimate summer SPM levels by up to 80 %, with only ~40 % of this discrepancy explained by weaker winds and the remainder likely reflecting biologically mediated sinking processes.
- Preprint
(4796 KB) - Metadata XML
-
Supplement
(890 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2135', Anonymous Referee #1, 10 Jul 2025
The authors introduced an interesting approach to investigate the effect of wind and phytoplankton on the seasonality in SPM concentration. Although the subject is of great interest, the study itself has many flaws and shortcomings. The manuscript is too long, with many repetitions, there is no clear difference between results and discussions. The flaws are
1) The SPM concentration time series is patchy and does not catch tidal variability. This may lead to an overestimation of wind effects as tidal variability is not included in the data.
2) The Lagrangian particle trajectories are not representative for SPM transport. The added value of the model results is limited.
3) As the correlation between wind and SPM concentration is low, the authors still ascribe the higher winter concentrations (when biological effects are small) to claim wind effects. Do you need wind to explain the seasonality? Would biology be not sufficient. What are the arguments to claim that higher SPM concentrations align with winds?
4) The effect of biological processes on SPM dynamics is described indirectly using Chl-a and temperature as a proxy of phytoplankton. This is not convincing as both parameters do not reflect the complicated biological processes.
5) Physical and biological induced flocculation is not well described in the manuscript, resulting in statements that are only partially true.
SPM, SPM level etc. is not a synonymous of SPM concentration: when you mean the latter add (replace by) ‘concentration’.
1. Introduction
The introduction need to be rewritten as it is a now a string of sentences without a common storyline. It should remind the state of the art background information (many references are from older literature >25yrs, which is ok, but there has been some progress since) and sets out the main research questions and why they are of importance.
L43: What do you mean by ‘primary production gradients’
L45-46: I would not introduce here the study site.
L49: Not clear how the import of SPM and OM is reflected by the heterotropic nature of the Wadden Sea. Import means from the North Sea?
L51: Above you describe the Wadden Sea as a tidal basin, here you switch to an intertidal area. The reference in L53-55 are not about intertidal areas. I suggest to not focus here on intertidal areas
L62: Schubel, 1974 does not seem the right reference (it is about sediments and not wind effect)
L63: Fettweis et al. (2012) is not about flocculation
L65: This is a repletion of L41
L68: Fettweis & Van den Eynde is not about biological processes
L68-69: Flocculation is a process of aggregation and break up. It is enhanced by the occurrence of marine gels (TEP), but occurs also without biological influence, due to the cohesive properties of fine-grained minerals (especially clay minerals).
L70-72: Flocs nearly always contain OM in various forms (adsorbed OM molecules, detritus and living OM).
L72-74: Add a reference
L80: Not clear how ‘excessive nutrient loads promote organic sedimentation’. Replace ‘organic’ by ‘organic matter’. How can organic matter (generally a density close to water density) settling without mineral ballast (biomineral floc)? Add a reference.
L83: How important is this (that can impact seagrass) in your study?
L85-86: It is becoming a fast evolving research topic, and many new research has been published on this subject.
L96: Two monitoring station, above you only mention one.
2. Data and Methods
L110-112: What are the 10-20% other than tidal forcings?
L119: The two references are not on tides. They are probably not needed here. Here the tidal range is 1.7m while on L109 it is 2m.
L144-145: I don’t think that higher harmonics and over-harmonics need references. Rephrase L144-149.
L160: You released tracers in the intertidal areas, correct? At the beginning of inundation?
L172-174: How can the passive tracers be representative for SPM, that undergoes resuspension and deposition phases?
L187-188: Can you mention the sampling frequency and the total amount of data. Was the sampling done during each weather conditions? Are the samples taken at the surface?
L190-192: Be a little more specific with the method. Do you use HPLC for Chl?
L195-196: The reference to PANGAEA is not useful, as there you have to search for the right files. I have downloaded the data in Rick et al. (2023), they contain data from to the period 2014-2019. The data before 2014 can be found in Rick et al. (2017) https://doi.org/10.1594/PANGAEA.150032
Can you check if these cover all the data you have used.
L200-203: The link can be omitted as also available in ‘Data Availibilty’
L205: Sea surface elevation versus SSH : use height instead of elevation?
L210: Do you mean; when the samples (cfr PANGAEA) were taken?
L215: Can you specify which environmental parameters were used to predict SPM concentration? How do they relate to SPM concentration?
L216: Are these environmental data continuous time series? If not, what is the effect of patchiness on the outcome?
L245-247: Chl-a concentration depends mainly on nutrients and grazing. How good are the two chosen proxies?
L247-256: References are missing here. This looks like a description of processes: put in introduction?
L252: I am not aware that detritus (is it not always organic?) contributes to flocculation actively. Detritus can be incorporated into flocs by sticky substances such as TEP.
3. Results
There is overlap between 3.1 and 3.2 (wind effect is explained in both chapters). I would advise to look only at results without interpretation of the data (keep this for the discussion).
3.1 Seasonality of SPM concentration
L267: Adding a grid for every year would facilitate the reading of the figure
L270-276: This belongs to discussion section (or intro).
L281-284: SPM concentration (and also Chl-a) has a log-normal distribution, taking the arithmetic mean and standard deviation is thus not correct (therefore that the standard deviation is in L284 larger than the mean). A better approach is to use the geometric mean and standard deviation.
L295: The colors in the figure 3 are confusing. Could you use the same color for SPM and Chl respectively.
L316, Fig 4: The large spreading in March suggest that there is a difference between the pre-bloom values (most of them) and the post-bloom ones. Can you check it and maybe adapt the Figure. Further is it useful to keep monthly correlation when the seasons can be split up into winter, spring-bloom, summer, late-summer bloom, autumn?
L299: The difference between both station in terms of correlation is small, delete this statement.
L300-303: When resuspension processes are the same, then also deposition ones. For phytoplankton and SPM to settle together means that the phytoplankton is contained in the floc in winter.
L306: R² is always < 1 per definition
L306: The sentence is not fully correct. Phytoplankton is part of the SPM, both are retained on the filter. If the correlation weakens, then it means that settling of flocs and phytoplankton differs and thus not all phytoplankton is attached to flocs as in winter. It also means that there is a difference in Chl and SPM concentration over the water column. Near the surface the Chla concentration would be higher, because free phytoplankton will settle slower than biomineral flocs.
L310: Which broader variables?
L324: Did you use all wind data? I would suggest to use only the wind data at the time of the samples and to compare this curve with the SPM concentration, as I suspect that the sampling has a biais towards good weather.
L339-343: I see that the correlation is overall very weak, although often significant. This means that the instantaneous wind speed does not explain a large portion of the variability in SPM concentration. Figure 6 seems not necessary for me.
L351: 120 hours, is this prior to the sampling date?
L353-356: Still the correlation is low, meaning that wind speed can only partially explain some of the observed variability in SPM concentration.
L357: ‘crucial role’: this phrasing is exaggerated, delete ‘crucial’
L359: What is the role of ‘indirect resuspension’, which I would consider as wave induced resuspensions. Do you have information on waves? I would expect that the role of waves in resuspending sediments is higher than of wind. Wind will result in changes of the advective transport of particles and waves in resuspension.
3.2 Resuspension and Time scales of Inner Basin Transport
3.2.1: Is not needed
L373-381 and Figure 8: I only see marginal differences between low and high tide. I would suggest to remove the Figure and the text. You mention sampling depth, see my comment for L187-188.
L383: The difference in current velocities between ebb and flood are a better indicator for SPM concentration during sampling than its (non) relation with SSH.
L400-405: In Figure 7 you used a 120h period for both stations, although 12h would have been better for the shallow station? Again, the R² is low, so that the explanatory variability of wind on SPM concentration is small. Figure 10 is not necessary.
L419-420: Seems logic to me that a storm in summer has a similar effect (increases resuspension) as a storm in winter. Delete ‘even’ in L419
L426: Although there is a link between SPM and a passive tracer, it is not the same. How representative are the results shown in Figure 11. Why not add SPM characteristics to the passive tracers (critical shear stress for erosion and settling velocity)?
L430: Can you explain what you mean by ‘stochastic fluctuations’?
L431: 80% of the velocity variability is due to tide, do you mean by ‘relatively minor (wind forcing)’ the remaining 20%? You should be more specific.
L435-439: Has been mentioned in the Method section
L439-443: There are differences, but both are also very similar.
L457: See my previous comment, I am not convinced that passive tracer give the same result as ‘active’ tracers (such as SPM).
3.3 Neural Network
L492: Salinity as a proxy for baroclinic conditions has not been explained. How do barclinic conditions affect SPM concentrations?
L497-501: The effect of temperature on flocculation in marine areas is small, and the effect of salinity on flocculation is important at the limit of the sea water intrusion in estuaries (so at low salinities).
L514: How did you obtain that wind explains 40% of the decrease?
L518-520: How did you define the seasons? As mentioned before, spring (March) may contain some typical winter conditions (pre-bloom) and autumn maybe some typical late summer conditions (summer bloom). Both may blur your results. To explain it by wind conditions is therefore not convincing.
L538-542: Is also included in tch 3.3.1
4. Discussion
L549-550: Biological processes are only introduced indirectly. There is no evidence of phytoplankton biomass or grazing in your data set.
4.1 Biological interactions
L562-564: Chl is not a good proxy for biological mediated flocculation. The latter is driven by the occurrence of exopolymers that have been excreted by the phytoplankton.
L572-575: The decoupling between SPM and Chl-a from spring onward, does not explains the formation of larger flocs with higher settling velocity. Further, the formation of larger flocs occurs already at the start of the phytoplankton bloom and not only at the end.
L575-577: Can you give a number that shows how big the influence of temperature is.
L585-588: Filter feeders will remove SPM from the system by producing fecal pellets. The POC fraction in the SPM depends on the SPM concentration (see Schartau et al., 2019).
L588-595: What is the relevance for your study? They remain speculative when you do not have data about it.
4.2 Wind & Tidal Control
This chapter is a summary of the results and not a discussion.
4.3 Neural Network
This chapter is a summary of the results and not a discussion.
5. Conclusion
L712-713: The correlation between wind speed and SPM concentration was never strong.
Citation: https://doi.org/10.5194/egusphere-2025-2135-RC1 - AC3: 'Reply on RC1', Gaziza Konyssova, 20 Aug 2025
-
RC2: 'Comment on egusphere-2025-2135', Anonymous Referee #2, 17 Jul 2025
- AC2: 'Reply on RC2', Gaziza Konyssova, 20 Aug 2025
-
RC3: 'Comment on egusphere-2025-2135', Anonymous Referee #3, 18 Jul 2025
Review of “Wind and Phytoplankton Dynamics Drive Seasonal and Short-Term Variability of Suspended Matter in a Tidal Basin” by Konyssova et al. (egusphere-2025-2135)
This is a long paper, trying to examine the seasonal and short-term variation of suspended matter in a tidal basin. Both numerical model and neural network are used in this study. However, it is very confusing to this Reviewer why the experiments are designed in this way. Do you really need both models?
In the numerical modelling part, only tide forcing is considered because of its dominate role in driving the currents in the bay. However, wind is the key factor to be studied in this work – the title of this manuscript starts with “wind”. Why not include winds in the numerical modelling? Note that the so called tidally induced transport time scale (133 hours) is also comparable to the synoptic weather time scales (several days).
Numerical model validation is only on tides using tide gauge data. However, the currents are not validated at all. The latter is relevant to the transport pathways. Note that the dynamical mechanisms are different. Accurate tides do not guarantee correct currents.
The “passive tracers” used in the numerical model experiments in this work are actually Lagrangian particles as commonly referred to in the coastal ocean modelling community. These should be called “virtual particles”, “Lagrangian particles”, or simply “particles” in Lagrangian particle tracking modelling. In contrast, “tracers” (or “passive tracers”) are often used in numerical dye experiments, which is associated with initial concentrations when they are released in the numerical model and their concentration change in space and time. It is better to change “tracers” to “particles” to avoid confusion. Otherwise, it is important to point out the tracers used in this study are different from other tracer model applications.
Linear correlation is heavily used in the analysis (Figures 6 and 7). However, it is not clear at all whether a linear relationship exists between the two variables (SPM and wind speed). Actually, the scattered data points indicate a non-linear relationship.
There are some awkward sentences and wrong word choices. For example, L20, “this study integrate …” should be changed to “in this study, we integrate …”.
Some of the statements are not rigorous at all: Lines 269-270, “… patterns reflecting the combined influence of biological activity and meteorological forcing” – how do you know that?
Citation: https://doi.org/10.5194/egusphere-2025-2135-RC3 - AC1: 'Reply on RC3', Gaziza Konyssova, 20 Aug 2025
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 345 | 37 | 18 | 400 | 32 | 10 | 21 |
- HTML: 345
- PDF: 37
- XML: 18
- Total: 400
- Supplement: 32
- BibTeX: 10
- EndNote: 21
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1