Response to Reviewer 1

I reviewed the manuscript of Dr. Flor Vermassen et al. and think it is an important contribution to our scientific community. The data presented in the manuscript are rather rare and provide, as written by the authors themselves, an interesting baseline for assessing the speed of change in the abundance and composition of planktonic foraminifera assemblages in the region for the coming decades.

I therefore support the acceptance of this manuscript after some rather minor revisions.

We thank the referee for a careful and useful review which points out a series of (minor) corrections. We also agree that the way in which we quantified and interpreted cytoplasm-bearing versus empty test should be clarified and now have done so (see below).

While the study is interesting and provides unique data, it mostly confirms previous findings and I believe that the authors could go a bit deeper in some of their observations and discussion, especially about the distinction of “living” and “dead” specimens.

It is unclear if the authors systematically distinguished cytoplasm-bearing and empty specimens. This, however, is important as they refer to “living foraminifera” often in the manuscript and consider that all specimens sampled were “living” rather than transported there. A distinction was made (mentioned in the conclusion as well as images of specimens with cytoplasm) but it is not numerically discussed or displayed in the manuscript and thus, leaves a lot of unknown concerning what would (or not) support their conclusions.

We agree that we had not been fully clear whether a systematic distinction was made between cytoplasm-bearing and empty specimens. The answer is that a systematic counting of cytoplasm content was only performed at one station (SO21-26-10). Our observations while picking the specimens at other stations, although not quantitative, confirmed the general pattern found in SO21-26-10, namely that the bigger/thicker shells at greater depths were indeed generally empty. Vice versa, we observed that the thinner, smaller empty shells in the shallower samples were virtually all cytoplasm bearing. We now made added a new panel to Fig. 9, showing the cytoplasm percentages.

In addition, we made clear that we assume that cytoplasm-bearing individuals were alive, i.e. ‘living’, and that the empty tests were presumed to be dead and sinking (although neither of those two assumptions are necessarily correct, strictly speaking).

Another point: The authors sampled relatively close to the coasts in or near fjords and in very shallow environments (several stations with a bathymetry <500 m depth). It has already been discussed that planktonic foraminifera do not leave/are not abundant in such environments (e.g., Schmuker 2000). I believe that this should appear somewhere in the discussion.

This is of course very true – we added a sentence in the discussion that points this out in amore straightforwatd manner:

“It is well-known that foraminiferal communities are rare in coastal and shelf environments (Schmuker 2000)".

Specific comments:

Introduction:

1. 55-57, which month were the plankton tows sampled for these studies?

Pados: Late June/early July 2011

Volkmann: July/August

Carstens: August

Darling: July

1. 60, and G. uvula, displaying similarities in their ecological preferences with T. quiqueloba.

The change is made.

1. 65, the species is not resident but it could still stay there for months and reproduce there – we do not know.

Yes, we agree.

Figure 2: a) It is difficult to read the legend, written in white on a light grey background. Please write it in black and increase the text size. b) please provide the station location.

We improved the figure/legend for legibility. The station locations are already provided in three other figures and would obscure the satellite image.

Methods:

1. 107, a multinet is generally equipped with 5 nets, why did the authors sample only 4 depths intervals, also, why sampling down to 1000 m depths and not with the classical 700 to 0 m?

The interval 200-100m was missing in the text, we have added this. We were not aware 700m was more classical than 1000m but we will sample to 700m next time.

1. 108, why stop the net between intervals?

To make sure the net closed before hauling the next section.

1. 111, for samples picked in Sweden, was the pH measured beforehand?

No, but pteropods were still plentiful and we did not observe signs of dissolution.

2.2, were all specimens manually oriented the same way on slides? In other words, are the size measurements of each foraminifera comparable? If not, what is the error linked to particles/foraminifera orientation?

Specimens were either placed on their dorsal or ventral side (but not sideways). Therefore, measurements of surface area and maximal diameter are comparable.

Overall, was a distinction made between living/cytoplasm-bearing and dead/empty specimens? If yes, please provide details.

Counting of cytoplasm content (colour) was only performed at one station (SO21-26-10).  At other stations, we were unable to count cytoplasm colour, but visual observations confirmed the general pattern observed at station SO21-26-10- (See also response to the first comment).

We changed to: “At station SO21-26-10, the colour of cytoplasm-bearing individuals was counted, but this was the only station where time allowed for quantification of cytoplasm colour.”

Results:

1. 183 – 187, which water masses are present between 150 and 500m depth?

We had phrased it awkwardly, the sentence has now been changed to:

“Below this, a 500-800 m thick water mass occurred – the Atlantic Water – which is characterised by higher temperatures (>0°C) and relatively high salinity (>34.9 g/kg).”

Figure 3: What is Figure X?

Figure 4, we fixed this.

3.2.1. Are the concentrations presented here the ones of all foraminifera living and dead (cytoplasm bearing and empty shells)?

Yes, exactly, we now clarified this in the methods:

“The reported concentrations include all tests, regardless of cytoplasm content.”

1. 262, Living in the top 100m or found/present in the top 100m?

It is indeed more correct to state “present” and we changed this.

1. 265, same comment.

We have changed it to “present”.

1. 284, the authors found more (relative) T. quiquelobain depth, from 200 to 1000m depth. Could they discuss/comment this finding in the manuscript?

The numbers are thin, but we add a comment that this is likely due to its association with Atlantic Water.

1. 295, what does “predominant” mean? Could you give a relative percentage of cytoplasm-bearing specimens at the different depths?

Absolutely, we have now done this in Fig 9B and added the numbers to this paragraph:

“This revealed that the red-coloured individuals dominate the top 50 m (64% of cytoplasm-bearing tests were red-coloured at 0-50 m depth), while the green/yellow type were dominant at 50-100m (40% were red-coloured). At 100-200 m and 200-500m, the percentage of tests with red cytoplasm was 60 and 67% respectively, but it should be noted that these numbers are rather uncertain due to the low number of tests at these depths (at this station). Due to time constraints, quantification of cytoplasm colour was not feasible at other stations, although qualitative observations during picking confirmed that the top 50m was dominated by tests with red-coloured cytoplasm.”

Figure 8: Specimen c has a “decaying” cytoplasm suggesting the individual is probably dying. Based on the histogram (left) I do not think that there is (statistically) relatively more “red” specimens in the surface layer than deeper. Maybe the authors could comment on that and specify if it is significant or not. finally, for this figure, the cytoplasm color is not discussed in the manuscript, I am not sure it provides very relevant information.

There is a 20% difference between the 0-50 and 50-100m interval at station SO21-26, as we now stated and showed in the manuscript (see previous comment). We deliberately did not discuss the causes of the cytoplasm colours very much, as this will be the topic of a follow-up paper, which will focus on the microbiome. Still, we thought it is worth showing/mentioning in this manuscript.

Discussion:

1. 310, without a clear distinction of cytoplasm-bearing / empty test, it is difficult to be so adamant in the wording. Even if I agree, I would suggest the authors to slightly tone down.

With the new Fig. 9B and clarifications made regarding cytoplasm (see previous comments), this statement should now be better supported.

1. 327-328, which “could” suggest that… I would add the could. Sampling, picking, storage, etc., usually break the spines, even in very healthy specimens.

Yes, we definitely agree and have made the change.

1. 357, please also provide the p-value

P-value has been added.

1. 366 to 374, please provide seasons or month(s) for the number from the literature mentioned. It is indeed well-known that the species displays a highly seasonal pattern of abundances.
2. 378-379, planktonic foraminifera are also just very well-known to not inhabit such environments… Rather than food availability only, what could explain the low concentrations is just the sampling sites and the bathymetry.

Indeed, it is well known that planktonic foraminifera do not inhabit shelf/coastal environments, but there must be physical/biological reasons why they avoid this habitat – here we list some of these. Just by itself, there is no reason why planktonic foraminfera would not be able to live say 100m water, there need to be associated environmental/biological factors preventing them to live/thrive there (such as the shallow depths impeding their reproduction cycles).

1. 405, also cite Manno and Pavlov 2014

OK, done.

1. 405 – 408, the authors should here again stress the fact that no clear distinction between cytoplasm-bearing and empty shells was made. Are the bigger and/or thicker shells found more in depth systematically empty? Do they also observe thinner, smaller empty shells in the shallow samples (suggesting a reproductive event without gametogenesis…)?

Quantification of cytoplasm colour was done at one station, see the new Fig 9B. Indeed, the reviewer correctly points out that bigger and/or thicker shells found at large depths were systematically empty, see the 500-1000m interval at that station. We have a follow-up study in the pipeline which will discuss N. pachyderma reproduction (also including coiling data).

Conclusions:

1. 431-435, the authors should discuss the bathymetry and the fact that some stations (from what I understood) are rather close to the shore.

We have added a passage on the role of bathymetry and coastal environment in the discussion section:

“Common reasons to explain low abundances in (inner) shelf regions and coastal areas are the high variability in the physical and chemical environment, high turbidity and suspended sediment load, and shallow water depths impeding foraminifer reproduction cycles (Schmuker, 2000; Zamelczyk et al., 2021).”

1. 435-436, are cytoplasm individuals dominating the surface samples? No data in the manuscript show that. This is very likely true and I agree but, one should provide evidence (numerical) for that.

We agree that we had not been not clear about this. We have now added an additional figure showing that cytoplasm individuals dominate the surface level at station SO21-26-10 ­– see the new Fig. 9B and adjustments to the text, where the numbers have now been added.

Response to Reviewer 2

The manuscript of Vermassen and coauthors on ‘The distribution and abundance of planktonic foraminifera under summer sea-ice in the Arctic Ocean’ presents a new and unique data set on the understanding of Neogloboquadrina pachyderma from so far uncharted regions between Greenland and the North Pole, and confirms earlier findings on the population dynamics, and which would need to be discussed in more depth.

We thank the reviewer for their thorough review and useful suggestions, which will substantially improve the manuscript. We resolved the concern regarding the main conclusion, by being more clear about the which area our interpretations are valid for – i.e. the perennially ice-covered Arctic Ocean, not the entire Arctic Ocean.

However, I don’t concur with the major conclusion of the manuscript that ‘N. pachyderma is the only species present underneath the perennial ice cover in the region between the North Pole–Greenland, and that sub-polar species have not migrated into the central Arctic Ocean’, and which should be tuned down and changed to avoid overinterpretation.

We can see now that our text appeared to suggest that no species are migrating at all in the entire Arctic Ocean. We will now clarify more carefully that our results and interpretations are valid for the perennially ice-covered part of the central Arctic Ocean.

In the abstract, we write

“Our results would suggest that the anticipated turnover from polar to subpolar planktonic species in the perennially ice-covered part of the central Arctic Ocean has not yet occurred, in agreement with a recent meta-analysis from the Fram Strait which suggested that increased export of sea-ice is blocking the influx of Atlantic-sourced species.”

In the conclusions, we write

“N. pachyderma was the only species present underneath the perennial ice cover in the region between the North Pole–Greenland at the time of sampling, which would suggest that sub-polar species have not yet migrated into the central, perennially ice-covered Arctic Ocean. This is consistent with previous research showing that subpolar species are currently largely ‘blocked’ from entering the central Arctic basin due to increased sea-ice export through the Fram Strait (Greco et al., 2022), although another potential pathway exists via the Barents Sea.”

Plankton net samples from several sampling locations only document a snapshot of the population dynamics of a large region, and cannot conclude on missing elements (other planktic foraminifer species) and on the entire Arctic Ocean.

We agree that plankton net samples only document a snapshot of the population dynamics of a large region and species can be ’missed’. Nevertheless, our data, taken over the course of 6 weeks, were consistent in terms of species composition.  We would like to clarify that we certainly did not intend to speculate on the status of the ‘entire’ Arctic Ocean – our data and interpretation concern the perennial ice-covered part of the central Arctic Ocean. See the changed text in relation the previous comment.

Also, the rather complex current system of the Arctic Ocean would need to be analyzed and interpreted for the transport of Atlantic plankton elements to know when and where any of these may arrive. For example, in the Abstract it’s stated that the ‘… results indicate that the anticipated turnover from polar to subpolar planktonic species in the Arctic Ocean has not yet occurred, in agreement with recent studies from the Fram Strait’, and which is based on the misunderstanding of the Artic circulation pattern; the Fram Strait (as well as the sampling sites north of Greenland) are located at the ‘end’ of the Arctic circulation where Arctic waters are transported into the GIN seas and the North Atlantic. Therefore, it’s very unlikely to find Atlantic sourced plankton north of Greenland and in the Fram Strait.

We are rather surprised with the statement that the entire Fram Strait would only receive Atlantic Waters that have entirely recirculated around the Arctic Ocean.  The well-known West Spitsbergen Current runs through the Fram Strait and provides a direct pathway of Atlantic Waters into the central Arctic Ocean, providing a plausible pathway for Atlantic (subpolar) species to enter the central Arctic Ocean. Several studies have previously reported Atlantic-sourced species in the Fram Strait (Carstens & Wefer (1997); Pados & Spielhagen (2014); Volkmann (2000)) .

New papers are accepted for publication and in print, which show the invasion of new species in polar waters from the North Atlantic Current entering the Arctic Ocean on the eastern (Norwegian and Russian) side, which has not been sampled here. Therefore, it may be wise not to make too general statements on this topic.

We agree and will clarify in the manuscript that our data and interpretations concern the perennially ice-covered part of the central Arctic Ocean. We believe our statements are therefore compatible with subpolar species entering Norwegian and Russian shelves, which are located in the seasonal ice zone and indeed long the pathway of incoming Atlantic Water, as has also been reported previously by Volkmann (2000). If the new papers can be shared, we would be happy to add a section in our manuscript discussing our results in the light of those studies.

Overall, the paper reads good, but includes many little flaws, which need to be corrected before the paper can be accepted for publications. For example, salinity has no unit; all maps would need coordinates (N, E, and W) for orientation (e.g., Figs. 1 and 2); I guess I know what the Figure 7 should show, but the information is difficult to from the rather ‘special’ kind of design, and the authors may convey the information in a clearer kind of way. ‘Concentration’ of foraminifers is seawater needs to be changed to ‘standing stock’.

We thank the reviewer for pointing out these minor flaws have made the corrections accordingly.

Page 2, lines 49-50: Please give references for the ‘knowledge of resident pelagic communities in the remote, perennially ice-covered regions is minimal.’ There are quite some papers available on planktic forams and other sea-ice related biota. For N. pachyderma and T. quinqueloba, please also have a look at the paper of Simstich et al. (2003, ecology and isotopes).

There are only two studies on live planktonic foraminifera in the perennially ice-covered Arctic region to (Bé 1960 and Carstens and Wefer, 1992), but we are happy to include any other references or studies we might have missed. Of course, there are many more studies located in the marginal ice zone/seasonal ice zone.

Line 107: The 200-100 m water depth interval needs to be included.

Thank you for spotting this, we corrected this.

Line 108: No ‘cod ends’ but ‘sampling cups’.

The technical name for these devices is ‘cod ends’, but we adopted the change for better reading.

Line 164: Were the nutrients analyzed from filtered or unfiltered seawater?

Unfiltered.

Various lines: Change maximal to maximum

We have adopted the change.

E.g., lines 237-240: change to past tense. In general, please be careful with past and present tense, etc.

We changed the tense and checked the manuscript for consistent use of tense.

In general, numbers without units (except of salinity) under 12 should be spelled out.

We have adopted this change.

Figure 5: Please consider making all scales the same length to allow for comparability.

In this figure, we opted to use different scales in order to optimally visualise the variability down the water column (the purpose of this figure). In Figure 6, the abundances can be compared across the sites.

Lines 287-290, and line 315: Did you take the time of sampling and synodic lunar reproduction cycle into consideration, which may have affected the size distribution? See Schiebel et al. (2017). Information of the reproduction of N. pachyderma may be obtained from the test-size distribution of assemblage. Small and large tests may indicate a living population, i.e., reproduction and growth.

This is a good point which we had indeed been thinking about, but it was rather difficult to investigate this, given the limited number of sampling events in our study. We have another study in the pipeline, discussing reproduction of N. pachyderma in detail, including some remarkable coiling data. We feel it is better to thoroughly discuss the reproduction there, than as a sidenote in this paper.

Line 295: How do you assess reproduction? Did you find any gametogenic calcite?

Yes, these tests had the gametogenic calcite , giving N. pachyderma its typical “thick-shelled” look (i.e. the form mostly present in the sediment).

Lines 299-301: It would be interesting to analyze the contents of the food vacuoles to determine the different types of algae used as food source.

Absolutely, we have another study in the pipeline that focuses on their diet.

Line 379: Please see also the papers of Brunner and Biscaye (2003) and Retailleau et al. (2009, 2011, 2012) for neritic effects.

Interesting, we read the papers and cited these.

Figure 9: Some of the stations are the same in the left and right panel. Better combine the two panels and use different markers / colours for the different types of stations?

We adopted the suggested changes.

Figure 11: the legend is too small to be easily read, and should be presented in a different kind of way.

We now explain the box plots in the caption.

Table 1: Change ‘Depth net’ to ‘Sampling depth interval (m)’

We adopted the change.

Plates: Are all scale bars the same length?

Yes, we clarified this in the caption.

Response to Reviewer 3

The authors provide quantitative data on the abundance and size of planktonic foraminifera in the upper water layer in the high Arctic, including at stations sampled underneath perennial sea ice. The sampling was carried out by net with fine mesh sizes and followed a depth stratified pattern and the resulting observations are therefore the first ones of its kind. As the authors note, the dataset will represent an important benchmark for the monitoring of the ongoing Atlantification of the high Arctic and the observed distributional patterns will be important for paleoceanographers analysing sedimentary records with polar planktonic foraminifera. Although there are some minor issues with the methodology and interpretation, which I outline below, the study as a whole is well conceived and appropriately documented and I recommend publication with minor revisions.

We thank Prof. Michal Kucera for his useful review, in which he points out some issues which we realise we had not been clear about and/or were previously a bit unsure about on how to handle. Due to time limitations (Vermassen only had two days between formulating a response to this review and embarking on a 6-week expedition),  we were not able to make all of the adjustments the manuscript yet, but we do intend to make the changes according to the below suggestions (i.e., after Vermassen returns from the expedition).

Comparing foraminifera from samples taken with different mesh size

Planktonic foraminifera increase the size of their shell by a factor of almost 100 throughout life. Similarly, small species and large species reach adult sizes that differ by more than a factor of 10. Considering this, there will always be two factors that hamper comparisons of absolute concentrations across studies and regions. First, the results will be affected by the mesh size that is used for sampling. This was noticed already by Berger (1969, 1971), who introduced quantitative correction factors to convert abundances among different mesh sizes. For comparison of foraminifera concentrations with other studies (e.g., Line 125), considering the different mesh sizes that they used, would it not be worth exploring what the effect of the conversion factors from Berger would be? This would allow a more direct comparison with the only other study from permanently ice covered region by Be (1960).

We absolutely agree with the tremendous importance of mesh size and implications for comparability (which is also why we pointed this out in our introduction). It is indeed interesting to the experiment with the Berger conversion factor, and we are including this now in the manuscript.

Out of curiousity, we also tested how well the Berger conversion performs – since we have the morphometric data, we can calculate different standardised abundances as if caught with different mesh sizes. For station 26, we found that the ‘mock’ 100 and 200 micron Berger-based standardised concentrations overestimated the the real/observed standardised concentration by one magnitude. So, while interesting, this does raise some questions whether the assumptions behind Bergers conversion factors are always valid.

Second, if the present population of planktonic foraminifera in any sample contains man specimens that belong to the same reproductive cohort (being the result of the same reproductive event), then the observed size distribution also reflects the temporal distance from the reproductive event. I.e., shortly after the reproductive event, the foraminifera would be smaller. Why not using the observations from the multiple sites to see if the size distribution changes through time? This will help us to constrain if there was a cohort-like reproduction affecting the entire region or if it was either very local or not present at all, like in the study by Meilland et al. (2021).

Yes, this is an interesting suggestion that is easy to include and we will do so. Regarding reproduction, please note that we are preparing a follow-up manuscript, where will discuss reproduction more comprehensively, so we are limiting the amount of discussion on reproduction in this manuscript.

Distinguishing between empty and cytoplasm-filled shells

From the methodology section of the paper (Line 112) it is not clear if empty and filled shells were consistently counted separately? This makes it difficult to interpret the observed patterns of abundance and size change with depth. It appears that the authors only counted empty and filled shells separately at one station (Fig. 8). If this is the case then one would expect at least a discussion of the effect of this simplification on the results. For exampling, knowing where large empty shells occur in the water column would provide the necessary constrain on speculations that the authors make about ontogenetic vertical migration in section 4.3.

We realise that we had not been fully clear whether a systematic distinction was made between cytoplasm-bearing and empty specimens (see also response to reviewer #1). Indeed, a systematic counting of cytoplasm content was only performed at one station (SO21-26-10). Our observations while picking the specimens at other stations, although not quantitative, confirmed the general pattern found in SO21-26-10, namely that the bigger /thicker shells at greater depths were indeed generally empty. Vice versa, we observed that the thinner, smaller empty shells in the shallower samples were virtually all cytoplasm bearing. We have adjusted the methods section to be more clear about this, as well as pointing out in the discussion that our (tentative) interpretations are only based on one station. We also added a new panel to Fig. 9, showing the cytoplasm percentages.

Analysing the controlling factors on population size

I am afraid that a much more formal analysis then what the authors provide (Line 356) is needed to make use of the rich environmental data that the authors collected. First, if the authors assume, and this seems to be supported by Fig. 8, that the foraminifera inhabit the top 100 m of the water column, then they should use population numbers from at least two depth intervals per site, considering that they also have all environmental data for those depths available. Then, they should carry out an analysis with appropriate statistical tools (for example a generalised linear model) on data transformed in a way that they do not violate the assumptions of the method (for example log-transformation of all abundances). In this way, they could test the effect of all variables they collected, also include variable like depth and distance from ice margin, and formally show which variables have no effect. This will be extremely helpful, even for all the negative results. I am afraid that as it stands, the analysis in Fig. 9 is inappropriate and meaningless. There are too few cases and a selection as made by the authors is dangerous and misleading – one can always find a way by which another group of stations can be distinguished from the other stations and then there may even be a negative trend between abundance and chlorophyll. All of this, by the way, requires a consideration of the reproductive dynamics as described above. We must exclude the possibility that all that we see is a cohort that is getting bigger with time (and thus along the cruise) and thus more abundant in the analysed mesh size interval.

We agree that experimenting with GLMs can be a useful additional analysis ­– we previously considered this but had assessed the risk of overfitting too high, and therefore decided to explore the data with individual correlations. We would just like to point out that, for GLMs to be robust, a common rule of thumb is to have at least 10 observations per predictor, meaning that at least 70 observations  would be needed if we include all possible predictors shown in Fig. 5 (compared to the 25 observations available from multinets). Nevertheless, it will be an interesting exercise and we will experiment with a GLM.

Food source for foraminifera living below permanent sea ice

This is a very important question and the most obvious unexplained pattern is the large concentration below 50 m, i.e. below the productive zone as indicated by chlorophyll profiles. Either all of the foraminifera in the sampling interval 100-50 m resided just below 50 m and fed on phytoplankton corresponding to the bottom end of the chlorophyll distribution interval, or they fed on something else. In this context, it would seem relevant for the hypotheses that the authors make to consider the study by Greco et al. (2021), which indicates that N. pachyderma likely feeds on aggregates rather than on fresh diatoms.

We fully agree with this comment and had actually been thinking along the same lines. We will now explicitly add a sentence about the possibility of aggregate-feeding in the 50-100m interval. 

Minor comments

Line 53: here the reference to Darling and Wade (2008) is not appropriate because unlike the other studies quoted, the review paper presents no original observations. I recommend to quote primary literature by the author at this place.

Ok.

Line 64-66: the sentences should be reformulated, the first one has incorrect grammatical structure and the second one likely confused “were” for “where”.

We will make the adjustment.

Line 299: the data discussed here are not given in Fig 7 (should likely be Fig 8). The wording “dominate” is subjective and for many readers would not be consistent with the data. I recommend to use the more neutral “more abundant”, here and elsewhere throughout the paper.

Agreed, we will make the adjustment.

Line 403: rather than gametogenesis it is safer to refer to “reproduction”.

That is true, we will make the change.

Line 380: There is abundant anecdotal and less abundant but solid quantitative evidence that planktonic foraminifera do not inhabit waters shallower than about 100 m. They avoid (or are unable to live in) these waters irrespective of temperature (Puerto Rico shelf, North Sea) or food availability, and this pattern also applies to silled fjords, despite the greater water depth further inland beyond the sill. Thus, the conclusion at this place that the observed low abundances in the RYDER19 region are due to lack of food is unlikely. Darling et al. (2007) observed a completely foraminifera-free Bering Sea, despite high productivity in that region.

This is of course true, and Reviewer 1 made a similar comment. We will adjust the manuscript accordingly, removing the suggestion about food control and pointing out that foraminifera generally don’t inhabit such shallow water depths.