Dear Authors,

Thank you for revising the manuscript. Let me ask for another round of revision addressing the following Reviewer's comments (line numbers refer to revised manuscript with track changes):

> My first feeling after reading the whole Intro is that the authors wanted to build a model that can overcome the limitations mentioned in paragraphs
> from line 22 to line 53. But to prove that, the new model must have better performance than previous models. However, the model performance is
> actually worse than simple Eulerian models. This is totally okay, but the Intro should be modified to better express the authors true focus, that is to serve
> as a tool to better study phytoplankton diversity and productivity in a more realistic turbulent ocean.
>
> ## Specific Comments
>
> Line 24: This is not true. To my understanding, the change of phytoplankton concentration in Eulerian models actually means the cells are moving.
>
> Line 59: increasingly what? popular?
>
> Eq3. I still don't think respiration cost Chla. Actually, I read Geider et al. 1997 again to double check if I'm wrong. Geider used a Eulerian framework, so the
> degradation of Chla is actually caused by mortality, but Chla doesn't degrade within the cell. That's the difference.
>
> Line 656: DCM is actually not captured. The patterns are quite different(Fig.4, middle panels). DCM is better simulated in the other two models (Fig.13). It
> would be interesting if the authors can explain why this happens. This is actually more meaningful.

Best regards,
Sylwester Arabas

Dear Authors,

I'm pleased to accept the paper pending technical corrections:

- Figures 2, 4, 6, 8 and 13 are composed of raster graphics, please use vector graphics as in all other plots

- avoid using italics for subscripts and superscripts which do not denote variables: "C", "N", Chl", "div", "opt", "phy", "dn", "ref", "max", "min", "m" (?), "PSII", etc

- avoid using italics for units (and use LaTeX \upmu instead of $\mu$)

- the Github and Zenodo links lead to code version labelled "second release", while the title mentions PIBM 1.0 (which is an older release not accessible via Zenodo) - please update the paper title to match the relevant code release

- it is somewhat misleading that the software name PIBM is almost nowhere to be found in the code archive (where IBM is used) - please update the code archive with a project name and version matching the paper

- spell openMPI as "Open MPI" (2 instances: p34 & p15)

- note: OpenMPI is just a particular implementation of the MPI standard, statements such as "use OpenMPI parallel computing" sound misleading - it is MPI API that the code relies on, and it can be compiled against different MPI implementations, Open MPI being just one example

- I see that the Gabriel et al. 2004 citation is suggested on the Open MPI website as a general-purpose citation for Open MPI, but it is a 20-year old conference paper about a draft of one particular MPI-2 implementation, while we are now at MPI-4.1!

- spell openMP as OpenMP (p34/l660)

- since you mention considering integration with PlanktonIndividuals.jl and point to the Julia language, please clarify in the paper that the IBM codebase is FORTAN/Matlab/shell/R

- fix URL in Ledbetter 1979 reference (htps://doi.org/https://doi.org/...)

Best regards,
Sylwester Arabas