Clustering of fast gyrotactic particles in low-Reynolds-number flow.

Systems of particles in turbulent flows exhibit clustering where particles form patches in certain regions of space. Previous studies have shown that motile particles accumulate inside the vortices and in downwelling regions, while light and heavy non-motile particles accumulate inside and outside the vortices, respectively. While strong clustering is generated in regions of high vorticity, clustering of motile particles is still observed in fluid flows where vortices are short-lived. In this study, we investigate the clustering of fast swimming particles in a low-Reynolds-number turbulent flow and characterize the probability distributions of particle speed and acceleration and their influence on particle clustering. We simulate gyrotactic swimming particles in a cubic system with homogeneous and isotropic turbulent flow. Here, the swimming velocity explored is relatively faster than what has been explored in other reports. The fluid flow is produced by conducting a direct numerical simulation of the Navier-Stokes equation. In contrast with the previous results, our results show that swimming particles can accumulate outside the vortices, and clustering is dictated by the swimming number and is invariant with the stability number. We have also found that highly clustered particles are sufficiently characterized by their acceleration, where the increase in the acceleration frequency distribution of the most clustered particles suggests a direct influence of acceleration on clustering. Furthermore, the acceleration of the most clustered particles resides in acceleration values where a cross-over in the acceleration PDFs are observed, an indicator that particle acceleration generates clustering. Our findings on motile particles clustering can be applied to understanding the behavior of faster natural or artificial swimmers.

Introduction

Systems of particles in turbulent flows are widely observed in many natural phenomena such as dust/sand storms, clouds, bacterial suspensions, and in oceans, lakes, and reservoirs [1, 2]. Aside from their ubiquitous nature, these particle systems have significant impacts on public health and safety. Thus, understanding their dynamics is relevant and provides insights into these phenomena. For example, understanding the dynamics of dispersed particles in turbulent flows is essential in predicting precipitation and the occurrence of dust storms [3, 4]. Modeling the migration of motile microorganisms is helpful in the ecological risk assessments associated with harmful algae in oceans and estuaries [5, 6] and in predicting their interaction with other particles that lead to ecological harm such as disease transmission [7]. Application of the insights from these studies can be extended to issues concerning human society such as contamination of urban water systems, which is a significant source of disease outbreaks, affecting a few to thousands in a single instance of an outbreak [8].

While these works focused on the migration of systems of particles, another interesting behavior of particles in turbulent flows is that particles accumulate and form patches in certain regions of space, i.e., particles do not remain uniformly distributed [1]. For example, photosynthetic motile algae or gyrotactic phytoplanktons, which swim toward the upper part of the ocean with the brightest light, exhibit clustering wherein motile phytoplanktons are observed to be more clustered than non-motile ones [9, 10]. Mathematical models and numerical simulations have shown that this complex behavior emerges from the coupling of motility and shear in the vortical flow [9-11]. Previous studies have shown that gyrotactic cells form clusters inside the vortices and in downwelling regions or regions with downward flow [9, 11, 12]. Similar behavior is observed for light non-motile particles in which particles preferentially concentrate inside the regions of high vorticity. In contrast, heavy non-motile particles accumulate outside these vortices, which act as centrifuges ejecting the particles [13-15].

The presence of these vortices is correlated with intense and persistent centripetal particle acceleration [16]. Fluid acceleration in high-vorticity regions generates stronger multifractal cell clustering [10]. However, clusters of phytoplankton cells are still observed in turbulent flows where vortices are short-lived [9, 11]. In these low-Reynolds-number turbulent flows, non-motile cells are observed to be well-distributed, and the clustering of motile cells is dependent on the swimming parameters such as speed and stability [9, 11]. Moreover, clustering only occurs from certain combinations of swimming velocity and stability, which results in either clustering inside the vortices or downwelling regions. In this study, we investigate the clustering of fast swimming gyrotactic particles embedded in a low-Reynolds-number turbulent flow and focus on characterizing the particle speed and acceleration and its influence on particle clustering.

We simulate dimensionless swimming particles in a low-Reynolds-number turbulent flow, exploring combinations of swimming parameters (swimming speed and stability) that generates particle clustering. Here we consider swimming speeds that are relatively high compared to the previous studies [9, 11, 17]. In the next sections, we describe the methods for the numerical simulations and the numerical procedure for the data collection. Furthermore, we also discuss the clustering of particles investigated by measuring entropy [18, 19]. We also present the speed and acceleration distributions of particles for different combinations of swimming parameters.

Methods

The system consists of gyrotactic swimming particles embedded in a cube with low-Reynolds-number homogeneous isotropic turbulent flow. In this section, we first discuss the background of gyrotactic swimming particles, then describe the numerical simulation of the turbulent flow embedded with swimming particles. Lastly, we discuss the numerical procedure for data collection.

Gyrotaxis model

We investigate the clustering of particles and characterize the probability distributions of their speed and acceleration by allowing the particles that are seeded uniformly throughout the cube to disperse until their mass distribution reaches a statistically steady state. Each particle moves with a net velocity that is the sum of its intrinsic velocity and the velocity of the fluid affecting it. The fluid velocity is taken from the direct numerical simulation (DNS) of the Navier-Stokes (NS) equation using a dealiased pseudo-spectral code [20, 21], while the intrinsic swimming velocity is based on the gyrotaxis model of swimming [9, 10]. In particular, the swimming particle follows the equations of motion given by where v is the net velocity of the particle, is the fluid velocity, is the position, t is time, v is the swimming speed, is the swimming orientation, B is the reorientation time, is the unit vector in the z-direction, and = ∇ × u is the vorticity [9, 10]. The first term of Eq 1 describes the tendency of the particle to remain aligned along the vertical direction, while the second term captures the tendency of vorticity to overturn the particle by imposing a viscous torque on it [9, 10]. Without swimming, the particles are considered as tracer particles as the second term of Eq 1 becomes zero, and the particle becomes affected only by the fluid velocity.

Numerical simulation

To simulate the system, we first create a three-dimensional (3D) cube with a turbulent fluid flow. Then, we embed tracer particles and incorporate the swimming mechanism to the particles as described in Eqs 1 and 2. The numerical implementation is written in a parallelized Python code, where the solver for the DNS of the NS equation from references [20, 21] is used. DNS is a valuable tool in fluid dynamics that uses methods of high order and accuracy and has been used to simulate homogeneous isotropic turbulence [22]. Here we use the Python NS solver as it is easy to modify or extend and, at the same time, still provides accurate and reliable data.

The simulation process is divided into two parts, (a) the fluid part calculation and (b) the particle part calculation (as shown in Fig 1). The fluid part includes the computation of fluid velocity and fluid vorticity using DNS, while the particle part includes the computation of the position, velocity, and swimming orientation of the particles. The particle positions are initialized and allocated among each CPU for the parallel implementation. The fluid velocity and vorticity at the locations of particles are calculated using the tricubic interpolation method [23]. Eq 2, on the other hand, is numerically solved using the fourth-order Runge-Kutta method [24]. As for the time-stepping scheme for the particle calculation part, the Adams-Bashforth 3/ Moulton 4 step predictor/corrector (ABM) method is used [25].

Fig 1

Numerical simulation workflow.

The red box corresponds to the fluid calculation while the blue boxes correspond to the particle calculations.

Numerical procedure

We used a periodic cube with a length of L = [2π, 2π, 2π] and a resolution of 963 grid points, which ensures accurate resolution at small scales. The turbulent fluid flow is at Reynolds number Reλ = 59, a turbulent fluid flow where tracer particles are observed to be uniformly distributed [9]. This guarantees that no clustering of particles is induced by the fluid vorticity alone. The time step used is Δt = 0.002, which satisfies the Courant-Friedrichs-Lewy criterion to resolve temporal scales in the DNS of NS equation. When a statistically steady state is achieved (∼ 10 − 15 eddy turnover times [22]), each particle is tracked, and its position, velocity, and acceleration per time step are recorded for the next 500 time steps. The numerical simulations are performed on a 40-CPU Supermicro X11DAi-N supercomputer, and each run takes ∼ 2.8 hours for tracer particles and ∼ 4.1 to 5.2 hours for swimming particles.

Swimming parameter values explored are v ∈ [0 : 30] and B ∈ [1 : 50]. Following the standard quantification of parameters used in previous studies, we de-dimensionalize the parameters with the Kolmogorov scales so that Φ = v/u and Ψ = B/τ, where u = (ϵν)1/4 and τ = (ϵ/ν)1/2. Parameters Φ and Ψ are the swimming number and the stability number, respectively. Here, both ranges of dimensionless parameters selected are greater relative to the previous studies [9, 11], which have not been explored. This is also to investigate whether particle acceleration alone can produce clustering. We extend our simulations using other Reynolds numbers Reλ ∈ [21, 36], where clustering of gyrotactic cells was previously observed [9, 11, 17], to confirm whether clustering of fast swimming particles will also be observed in such flows. The parameters used in this study are summarized in S1 Table.

The post-processing of data includes the measurement of the clustering of particles and the characterization of the velocity and acceleration of the particles. The two ways with which the clustering is measured are by calculating (a) particle density and (b) entropy. To measure the particle density, the cube is divided into smaller cubes or bins of length r. The number of particles contained in each bin is counted, then the particle density is calculated using the equation ρ = nL3/(Nr3), where n is the number of particles per bin and N is the total number of particles normalized by r3. Using the bins created, we calculate the entropy H = −∑ p log (p/r3) − ∑ p log r3, where p is the probability per bin [18, 19]. The maximum entropy H = ln(N) = 11.53 corresponds to uniformly distributed particles. Thus, high particle clustering corresponds to a low entropy value. The characterization of the velocity and acceleration, on the other hand, is done by taking their probability distributions (PDF). We also explore the distribution of the longitudinal and centripetal acceleration, and , where is the particle velocity unit vector [16].

Results and discussions

In this section, we present the results when Reλ = 59, a turbulent fluid flow where tracer particles are observed to be uniformly distributed. The same experiments have been done for other values of Reλ where the results are included in the latter part of this section as Supporting information for a more organized presentation.

Cluster formation

Particles seeded uniformly in the cube, with homogeneous and isotropic incompressible turbulent fluid flow, are allowed to disperse until their mass distribution reaches a statistically steady state. Fig 2 shows snapshots of the spatial distribution of particles in 2D and 3D for different swimming speeds. The 2D plot shows a slice across the yz-plane flattened along the x-axis with thickness L/42 where L is the cube length along the x-axis. The colormap represents the swimming speed (from Eq 1) of the particles. Here we see clustering at swimming number Φ = 17 and Φ = 33 with the highest particle clustering observed at Φ = 33 where relatively big voids in the 2D plot can be observed. However, at Φ = 66, the particles become less accumulated again.

Fig 2

Cluster formation of swimming particles.

Snapshots of the distribution of particles when Reλ = 59 for swimming numbers Φ = [0, 17, 33, 66], where the swimming number Φ = 0 corresponds to tracer particles. Colors correspond to the swimming speed of the particles.

Fig 3A shows that the PDF of the particle density follows an exponential fit. The exponent λ of the fit increases as more bins become populated with more particles. Particles with swimming number Φ = 33, where we observed the highest clustering in Fig 2, have the highest λ. The trend for Φ = 66, where we observed a decrease in clustering (in Fig 2), is comparable to Φ = 17.

Fig 3

Particle density PDF and entropy.

(A) Particle density PDF when Reλ = 59 for different swimming numbers Φ = [0, 17, 33, 66]. (B) Changes in entropy H with swimming number at Ψ = 91 (vertical dashed line for Φ = 33). (C) Entropy H as a function of stability number at Φ = 33. Error bars taken across 250-time steps.

In our simulations, the lowest entropy H is measured at Φ = 33 while the highest H is measured at Φ = 0 (represented as a dashed line in Fig 3B). Correspondingly, swimming numbers Φ = 33 and Φ = 0 have the lowest and highest λ, respectively. The behavior of H in our results has a similar trend with the previous results where clustering increases and reaches a maximum followed by a decreasing trend [9, 17]. However, in our results, a significant change in the clustering is observed when Φ is changed, while no significant difference in H is observed when Ψ is changed (see Fig 3C), which is in contrast with the previous studies where cluster formation is dominated by the stability number [9, 17]; cluster formation is only dictated by the swimming number. For other values of Reynolds number, the behavior of H is similar to when Reλ = 59 except the graphs shift towards higher Φ values (see S1 Fig). Consistent with the result when Reλ = 59, clustering in the lower Reλ values is still invariant with the reorientation time (stability number) as shown in S1 Fig.

Our results show a cluster formation mechanism different from what has been previously reported. We deem that the difference in the observed clustering is attributed to the velocity and acceleration characteristics of the particles. Thus in the next sections, we explore the probability distributions of particle velocity and particle acceleration and compare them for different values of Φ.

Velocity and acceleration characteristics of swimming particles

To gain insights into the characteristics of particles as they are dispersed or clustered throughout the space, we look into the distribution of their velocity and acceleration. We focus our investigation on the effect of swimming number Φ on clustering.

Velocity probability distribution

In Fig 4A, we show the PDFs of the velocity magnitude v = |v| of the particles (from Eq 1) for different values of Φ. Our result is consistent with previous results where PDF of velocity magnitude of Lagrangian(tracer) particles are observed to be nearly Gaussian [26, 27]. Here we observe a broadening in their distribution as Φ is increased. To quantify the broadening of the distribution, we take the Gaussian fit of the PDFs and compare the parameters σ and μ, the standard deviation, and the mean, respectively. In Fig 4B, we see a non-linear increase in σ as Φ is increased, while in Fig 4C, we observe that increasing the swimming number Φ also increases the mean speed μ. We expect these results since the particle velocity is just the superposition of both the swimming and fluid velocity as shown in Eq 1.

Fig 4

PDF of microswimmer speed when Reλ = 59.

All values are subtracted with the mean such that the distributions are centered at zero. (B) and (C) are plots of the standard deviation σ and mean μ versus the swimming speed, respectively.

Acceleration probability distribution

Fig 5 shows the plots of the PDFs of the x-components of the three types of particle acceleration, namely, the acceleration a, centripetal acceleration a, and the longitudinal acceleration a, which are de-dimensionalized with the root-mean-squared acceleration a. Our results shown in Fig 5A is in agreement with the previous studies [27, 28] where the tails of the distribution follow the exponential fit (represented as black dashed line) given by P(a) = C exp(−a2/(1 + |aβ/γ|)γ2), where β = 0.49, γ = 0.42 and δ = 1.56. Likewise, the plots for a(x) and a(x) have qualitatively comparable results with [29] where the tails of the longitudinal acceleration PDF drop more sharply compared to the centripetal acceleration PDF. Both PDFs for a(x) and a(x) have a wider range of values compared to the PDF of a, which ranges only up to a(x)/a < ±3. The inset in Fig 5A, on the other hand, shows the plot of a vs.Φ. The narrowing of the PDFs is a consequence of the increase in the a values as Φ is increased.

Fig 5

PDF of the x-components of acceleration.

Shown are the PDFs of (A) acceleration a(x), (B) centripetal a(x), and (C) longitudinal acceleration a(x) when Reλ = 59 for Φ = 33. Black dashed line in (A) corresponds to the exponential fit [27, 28] for Φ = 0. Different colors correspond to swimming numbers Φ = [0, 17, 33, 66].

The PDFs of the magnitudes of a, a, and a are shown in Fig 6. We also observe a narrowing of the distributions as Φ is increased, which is more pronounced at the high-acceleration tails. At low acceleration values, the PDFs follow a power-law fit (shown in Fig 6A–6C), which is represented as the solid black line. Except for a small difference in shifts, the PDFs follow the same trend with an average exponent of 1.93 ± 0.01. We observe a cross-over where the PDFs of swimming particles become greater compared to the PDFs of tracer particles (Φ = 0). In Fig 6E and 6F, we show the log-linear representation of the plots where the high-acceleration tails are observed to be nearly exponential. The estimated exponential fit is shown as the solid black lines. Here we observe a second cross-over where tracer particles PDF becomes highest.

Fig 6

PDF of acceleration magnitude when Reλ = 59.

(A-C) Log-log plots of the acceleration magnitude PDF where the vertical dashed line corresponds to the location of the first cross-over. (D-F) Log-linear plots of the the PDFs where the vertical dotted lines corresponds to the location of the second cross-over. Black solid lines are the estimated fits.

We observe a narrowing in the PDFs, which is attributed to the increase in a. We also observed two kinds of deviations from the distributions of tracer particles (Φ = 0), i.e., (a) the first cross-over where the PDFs of swimming particles become higher and (b) the second cross-over where the PDF of tracer particles becomes the highest. The two cross-overs observed in the PDFs may have a contribution to the clustering of swimming particles. To investigate further, we look into the acceleration characteristics of the most clustered particles or particles in bins with density ρ ≥ 0.00015. We take the average acceleration of the particles contained in each bin and plot the frequency distribution as shown in Fig 7. For all types of accelerations, we observe the highest bin counts at Φ = 33 where clustering is maximum, and the lowest counts for Φ = 0 where no clustering is observed. Peaks in the frequency distributions (shown in Fig 7A–7C) are observed at acceleration values where the first cross-over in Fig 6 is observed, while the peaks in the plots for accelerations along the x-axis (shown in Fig 7D–7F) are at 0. We can see that the most clustered particles reside within acceleration values near the first cross-over location, and the number of bins becomes zero as it approaches the second cross-over location. For all the other values of Reλ, we also observed two cross-overs in their corresponding acceleration PDFs with a slight change in locations. Consistent with the results when Reλ = 59, the peaks in the frequency distributions of the most clustered particles are also observed at the first cross-over location when Reλ = [21, 36] as shown in S2 and S3 Figs. This means that the presence of the first cross-over is an indicator that acceleration drives the clustering of swimming particles described here. On the other hand, there is no significant difference in the frequency distribution plots except for the increased peak in the plot for a(x) (Fig 7F). Thus, the cluster formation is not dictated by vortex trapping. Our results on the mechanism of cluster formation can be applied to further understand the dynamics of natural/artificial fast swimmers. While clustering is advantageous during reproduction as it enhances the encounter rates between cells [9], it can also be detrimental as it increases the transmission of diseases. With our results, we may be able to gain insights into the balance between clustering and transmission of diseases.

Fig 7

Histogram of the magnitudes and x-components of acceleration when Reλ = 59.

The number of bins (cube) per acceleration value for the magnitudes of (A) acceleration, (B) centripetal acceleration, and (C) longitudinal acceleration. (D-F) corresponds to the x-components of the three accelerations.

Conclusions

Previous studies have shown that clustering of microswimmers is greatly affected by both Φ and Ψ, which results in particles accumulating inside vortices and in downwelling (downward flow velocity) regions [9, 11, 12]. Clustering in the downwelling regions occurs specifically for particles in the limit of Ψ ≪ 1. In our simulations, the clustering is only significantly affected by swimming speed and remains invariant for any Ψ. Here, the swimming numbers explored in our simulations are much higher than the previous reports. Our results show that clustering can be achieved even without the influence of the stabilizing torque as long as the particles swim at high speeds. The highest particle clustering was achieved at moderate values of swimming number. As Φ increases, the particles move faster and become more dispersed resulting in a decrease in clustering. Furthermore, clusters emerged from balancing the effects of motility and vorticity, which has been previously discussed in the references [9-11]. The turbulent flow in our system is in a low Reynolds number, Reλ = [21, 36, 59], such that shear from vortices cannot form clusters, but by adding motility to the particles, clusters can start to form. However, adding a very high swimming velocity to the particles breaks the balance, which then results in a decrease in H. The mechanism of cluster formation presented in this paper can be applied to understand many natural/artificial swimmers with faster swimming speeds [1, 2] where clustering can be advantageous for reproduction as it enhances the encounter rates between cells, but it can also be detrimental as it increases the transmission of diseases.

Our results may also provide useful insights into water system management since most urban water contamination is caused by motile bacteria. We have previously noted that the presence of intense and persistent centripetal acceleration is correlated with intense vorticity, which is the evidence of particles trapped inside vortices [16]. In our simulations, the majority of the particles in high-density bins have small magnitudes of acceleration indicating that clustered particles are located outside regions of high vorticity. The observed clustering is similar to that of clustering from vortex ejection, which to our knowledge has not yet been observed in any work on clustering of motile microorganisms. In addition, the acceleration of most of the clustered particles resides within the cross-over location in the acceleration PDFs, where the PDFs of swimming particles become higher compared to that of tracer particles. Our results reveal that particle acceleration drives clustering.

Simulation parameters used in different Reynolds number flow.

Swimming parameters and number of particles used in each simulation for different values of Reynolds number.

(TIF)

Click here for additional data file.

Entropy H versus swimming number Φ and reorientation time B.

Changes in the entropy H with (A) swimming number Φ, and (B) reorientation time B for different values of Reynolds number Reλ at Φ = 33.

(TIF)

Click here for additional data file.

The number of bins (cube) per acceleration value (magnitude and x-components) for Reλ = 21.

Histogram for (A) acceleration, (B) centripetal acceleration, and (C) longitudinal acceleration while (D-F) corresponds to the histograms for the x-components of the three accelerations.

(TIF)

Click here for additional data file.

The number of bins (cube) per acceleration value (magnitude and x-components) for Reλ = 36.

Histogram for (A) acceleration, (B) centripetal acceleration, and (C) longitudinal acceleration while (D-F) corresponds to the histograms for the x-components of the three accelerations.

(TIF)

Click here for additional data file.

10 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Fang-Bao Tian

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf.

2. Thank you for stating the following in the Acknowledgments Section of your manuscript:

[We would like to thank the Department of Science and Technology (DOST)-SEI 252

Accelerated Science and Technology Human Resource Development Program (ASTHRDP) and University of San Carlos for supporting this research. M. Liponhay would also like to acknowledge the funding from DOST-CRADLE Program, Project No. 8419. Finally, we thank Professors Danilo M. Yanga, Christopher P. Monterola, and Christian M. Alis for the fruitful discussions about this work.]

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

[The author(s) received no specific funding for this work.]

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

3. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript is on the clustering of fast gyro-tactic particles in low-Reynolds-number flow, which is beneficial to understand various ecological phenomena, for example, harmful algal blooms in lakes. The authors' efforts are appreciated, and the authors should consider the following points to improve further the quality of this manuscript.

(1) The primary method used in this paper was reported in previous work by Durham et al. (Durham WM, Climent E, Barry M, De Lillo F, Boffetta G, Cencini M, et al. Turbulence drives microscale patches of motile phytoplankton. Nature communications. 2013). This work presents swimming particles' velocity and acceleration characteristics for different parameters. However, the numerical analysis in this work has not been verified by some experiments and relevant literature. At least, more cases of numerical simulation should be performed and analyzed for flows of various Reynolds numbers to make the conclusion powerful.

(2) The conclusion is given for fast gyrotactic particles in low-Reynolds-number flow. More limitations are required to specify the scope of the conclusion, for example, the range of low-Reynolds-number flow.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

27 Jan 2022

RESPONSE TO THE EDITOR

COMMENT 1: Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf.

RESPONSE 1: Thank you for the resources. We have

Added corresponding author’s initials beside email address.

Removed funding information from the Acknowledgements section.

Improved the citations of equations and figures in Methods and Results and discussions sections.

COMMENT 2: Thank you for stating the following in the Acknowledgments Section of your manuscript:

[We would like to thank the Department of Science and Technology (DOST)-SEI 252

Accelerated Science and Technology Human Resource Development Program (ASTHRDP) and University of San Carlos for supporting this research. M. Liponhay would also like to acknowledge the funding from DOST-CRADLE Program, Project No. 8419. Finally, we thank Professors Danilo M. Yanga, Christopher P. Monterola, and Christian M. Alis for the fruitful discussions about this work.]

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

[The author(s) received no specific funding for this work.]

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

RESPONSE 2: Concerning the Acknowledgments Section of the manuscript, we have removed the funding-related statement and incorporated this correction in the revised manuscript. We propose the following amendment to the funding statement:

“JA and ML acknowledge the Department of Science and Technology (DOST)-SEI Accelerated Science and Technology Human Resource Development Program (ASTHRDP) and the University of San Carlos for supporting this research. ML acknowledges the Department of Science and Technology (DOST) of the Philippines with Project No. 8419, 2020 under the Collaborative R&D to Leverage the Economy (CRADLE) Program (https://s4cp.dost.gov.ph/programs/cradle/). The funders had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript.”

COMMENT 3: We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

RESPONSE 3: Thank you for the information. We are changing our Data Availability statement to

“All preprocessed data files are available from 10.5281/zenodo.5904617, 10.5281/zenodo.5905360, and 10.5281/zenodo.5904923.”

We have also mentioned this in our cover letter.

RESPONSE TO THE REVIEWER

Reviewer #1: This manuscript is on the clustering of fast gyro-tactic particles in low-Reynolds-number flow, which is beneficial to understand various ecological phenomena, for example, harmful algal blooms in lakes. The authors' efforts are appreciated, and the authors should consider the following points to improve further the quality of this manuscript.

COMMENT 1: The primary method used in this paper was reported in previous work by Durham et al. (Durham WM, Climent E, Barry M, De Lillo F, Boffetta G, Cencini M, et al. Turbulence drives microscale patches of motile phytoplankton. Nature communications. 2013). This work presents swimming particles' velocity and acceleration characteristics for different parameters. However, the numerical analysis in this work has not been verified by some experiments and relevant literature. At least, more cases of numerical simulation should be performed and analyzed for flows of various Reynolds numbers to make the conclusion powerful.

RESPONSE 1: We thank the reviewer for this comment. We have performed additional numerical simulations using different Reynolds numbers, where clustering of gyrotatic cells is observed inside the vortices in the previous studies. We use similar values to confirm whether the mechanism of particle clustering discussed in our study can be observed in such flows.

We found the same behavior for Re_λ=[21,36] – the presence of clustering with the same characteristics with respect to velocity and acceleration distribution as when Re_λ=59, which supports our conclusions. Shown below are the corresponding snapshots of the distribution of particles for Re_λ=[21,36] for different swimming numbers.

We have also provided additional discussions in the Results and discussions section (in the third paragraph of the Cluster formation subsection and in the last paragraph of the Velocity and acceleration characteristics of swimming particles subsection) to support our conclusions and have incorporated the results in the manuscript as supporting information.

COMMENT 2: The conclusion is given for fast gyrotactic particles in low-Reynolds-number flow. More limitations are required to specify the scope of the conclusion, for example, the range of low-Reynolds-number flow.

RESPONSE 2: We thank the reviewer for the suggestion. We agree with the reviewer, and we have included the range of low-Reynolds-number flow and the basis of selection of these values in Paragraph 2 of the Numerical procedure subsection of Methods.

Our findings that motile particles cluster outside vortices when also happen for low Re numbers, i.e., when the particle speed is high enough. Given these observations, our data and findings now better support the conclusions with its scope for fluid flow at low Reynolds number and high particle speed. The results of the numerical experiments using these values have been discussed in the Results and Discussions and are presented in the Supporting information section.

In addition to the revisions made to the manuscript above, we have also made corrections in the values used to de-dimensionalize the swimming parameters, changing the values of the swimming parameters. These changes only affect the swimming parameter values reported in the figures (axis values in Figs 3 and 4 and legends of Figs 2-7) and do not affect the trend in our plots or the analyses of the data. We have incorporated these corrections in the Methodology and Results and Discussions sections of the manuscript.

We also added a new paragraph at the beginning of the Results and discussion section to add clarity in the organization of the presentation of results.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

8 Mar 2022

16 Mar 2022

RESPONSE TO THE EDITOR

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Response:

We thank the editor for this comment. We have checked our reference list and found no articles that have been retracted. We made no major changes in the reference list except for a few additional information added in reference numbers 8, 21, 24, and 25.

RESPONSE TO THE REVIEWER

Reviewer #1:

(1) Please check Eq.2; it is not written correctly. p->dp/dt

Thank you for spotting this error. We have made corrections to the expression of Eq. 2 and have incorporated it in the revised manuscript.

(2) The references for Eqs. 1 and 2 should be given.

We thank the reviewer for this comment. We have now cited the reference for Eqs. 1 and 2 in line number 66. We have also checked if there are parts in the manuscript where we may have failed to properly cite the references and made the necessary corrections in the revised manuscript. We cited the references in line numbers 29-30 and line numbers 120-122.

(3) Figures are unclear, and authors should make sure all figures are clear enough for reading.

Thank you for pointing out this concern. We have improved our figures and increased their resolutions making sure that they are all clear enough for reading. We have also revised Fig 7, S3 Fig and S4 Fig so that the axis titles are shown.

(4) Texts and languages should be improved further. For example, Navier-Stokes NS equation -> Navier-Stokes equation.

Regarding this concern, we have now improved the texts and languages used in the manuscript.

We have changed the Navier-Stokes NS equation to Navier-Stokes (NS) equation, as NS equation is repeatedly used in the manuscript. Similar necessary changes have also been made throughout the manuscript to remove inconsistencies in the usage of other terms and abbreviations.

We have also significantly improved the clarity of the discussions in our manuscript by revisiting our choices of words and by polishing our use of the English language.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

24 Mar 2022

Clustering of fast gyrotactic particles in low-Reynolds-number flow

PONE-D-21-37882R2

Dear Dr. Almerol,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Fang-Bao Tian

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

30 Mar 2022

PONE-D-21-37882R2

Clustering of fast gyrotactic particles in low-Reynolds-number flow

Dear Dr. Almerol:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Fang-Bao Tian

Academic Editor

PLOS ONE