the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 29 Aug 2025
A global consistent database of plankton and detritus from in situ imaging by the Underwater Vision Profiler 5
Abstract. Plankton and detritus are essential components of the Earth’s oceans influencing biogeochemical cycles and carbon sequestration. Climate change impacts their composition and marine ecosystems as a whole. To improve our understanding of these changes, standardized observation methods and integrated global datasets are needed to enhance the accuracy of ecological and climate models. Here, we present a global dataset for plankton and detritus obtained by two versions of the Underwater Vision Profiler 5 (UVP5). This release contains the images classified in 33 homogenized categories, as well as the metadata associated with them, reaching 3,114 profiles and ca. 8 million objects acquired between 2008–2018 at global scale. The geographical distribution of the dataset is unbalanced, with the Equatorial region (30° S – 30° N) being the most represented, followed by the high latitudes in the northern hemisphere and lastly the high latitudes in the Southern Hemisphere. Detritus is the most abundant category in terms of concentration (90 %) and biovolume (95 %), although its classification in different morphotypes is still not well established. Copepoda was the most abundant taxa within the plankton, with Trichodesmium colonies being the second most abundant. The two versions of UVP5 (SD and HD) have different imagers, resulting in a different effective size range to analyse plankton and detritus from the images (HD objects >600 µm, SD objects >1 mm) and morphological properties (grey levels, etc.) presenting similar patterns, although the ranges may differ. Therefore, recommendations are provided for the appropriate use of this data when conducting studies. A large number of images of plankton and detritus will be collected in the future by the UVP5, and the public availability of this dataset will help it being utilized as a training set for machine learning and being improved by the scientific community. This will reduce uncertainty by classifying previously unclassified objects and expand the classification categories, ultimately enhancing biodiversity quantification. The dataset that constitutes this first release is available at SEANOE.
- Preprint
(1778 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 09 Jan 2026)
- RC1: 'Comment on essd-2025-522', Anonymous Referee #1, 03 Dec 2025 reply
Data sets
A global consistent database of plankton and detritus from in situ imaging by the Underwater Vision Profiler 5 Ariadna C. Nocera, Lars Stemmann, Marcel Babin, Tristan Biard, Julie Coustenoble, François Carlotti, Laurent Coppola, Lucas Courchet, Laetitia Drago, Amanda Elineau, Lionel Guidi, Helena Hauss, Laëtitia Jalabert, Lee Karp-Boss, Rainer Kiko, Manon Laget, Fabien Lombard, Andrew McDonnell, Camille Merland, Solène Motreuil, Thelma Panaïotis, Marc Picheral, Andreas Rogge, Anya Waite, Jean-Olivier Irisson https://doi.org/10.17882/107583
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,276 | 94 | 29 | 1,399 | 31 | 48 |
- HTML: 1,276
- PDF: 94
- XML: 29
- Total: 1,399
- BibTeX: 31
- EndNote: 48
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
ESSD-2025-522
A global consistent database of plankton and detritus from in situ imaging by the Underwater Vision Profiler 5
Ariadna C. Nocera1 et al.
General comments
This manuscript presents a commendable effort to assemble and standardize a comprehensive, decade-long (2008–2018) global dataset of plankton and detritus from Underwater Vision Profiler (UVP5, HD and SD). The resulting archive of approximately eight million validated images is a valuable resource that will support future analyses of plankton biogeography, enhance biogeochemical models, and aid in developing AI-driven image classification methods.
While the study is interesting and merits publication, the manuscript requires attentive revision. Each section, from Introduction through Discussion, is currently excessive in length and would benefit significantly from concision to eliminate repetition and improve overall effectiveness.
Please, find specific comments and technical notes below, which I hope will be useful in the revision process.
Specific comments
Introduction - This section is overly lengthy and burdened by repetitions and redundancies. A thorough revision to improve its organization and concision would create a clearer, more fluent succession of topics. Specifically, the final three paragraphs should be streamlined by relocating all technical details (regarding the UVP instrument, profile and image counts, classification methodology, and its limitations) to a dedicated “Materials and Methods” section. To conclude the Introduction, I recommend a brief, focused paragraph on the utility and applications of UVP, followed by a clear statement of this study’s specific scope.
Methods - While this section provides useful detail on the dataset's preparation, its length can be reduced to improve clarity. Suggested revisions include: integrating the first sentence of section 2.1 into the UVP description (2.2.1), moving the remainder to the dataset section (2.3), and removing the now-empty 2.1. Additionally, adding examples (L188-191) would effectively illustrate how contextual information aids in classifying objects with weak morphological attributes.
Further technical notes on methods are provided below.
Results- This section requires revision to reduce repetition and improve concision.
As an example, the text from lines 284–291 can be condensed as follows:
“Detritus was the most abundant category (90.5% of total images), indicating a substantial presence of non-living particulate organic material (Fig. 3). Copepoda dominated zooplankton (2.2%), followed by Trichodesmium (1.3%), Trichodesmium-contextual (1.1%), the plankton-like category (1%), artefacts (0.9%), Phaeodaria (0.8%), and Bacillariophyta-contextual (0.6%). Each of the remaining groups represent < 0.16% of the total number of objects.”
Section 3.6 also needs to be streamlined and made more concise.
The verbs in Results should be in the past tense.
Discussion – As with previous sections, this text would benefit from concision and clarification (see Technical notes). Key revisions should address three main points: a reinforced support to the separation of 'artifacts' from 'detritus'; the removal of the speculative link between the 'plankton-like' category and "new findings" (L430); and more realistic and practical ideas about how UVP data can advance our understanding of the pelagic ecosystem.
Technical notes (modifications/corrections)
Abstract -
L31: …their composition and fluxes, and marine ecosystems as a whole.
L34: …by to versions (SD and HD) of the…… [removed from L41]
L40: Copepoda was the most abundant planktonic taxon, followed by Trichodesmium colonies.
L43: …“presenting similar patterns, although the ranges may differ”. This part of the sentence should be clarified.
L47: ….by identifying previously unclassified objects…..
L48: …available on SEANOE (link) [include the link]
Introduction
L53: Landry [not Laundry]
L62-63: the sampling and observation methods should be listed/grouped separately and not mixed as presented now.
L64-66: the sentence “Many planktonic organisms….2024).” is misplaced because interrupts the flow focused on instruments. It should be moved before the sentence “Furthermore….”.
L67 O’Brien [not O’brien].
L70-71: In this position, the sentence “For plankton….2019)” is weak and interrupts the flow. It should be better moved to the paragraph L82-101.
The paragraph L61-72 is better closed by the sentence “Addressing these challenges….change”.
L74-76: the sentence “Conventional sampling….across disciplines.” Should be removed because useless repetition.
L86: Suggested modification: “Deep collection by net tows followed by taxonomic identification with microscopes provides accurate biodiversity data, though this sampling method damages fragile organisms and requires…..2022).”
L134-138: Suggested modification: “the Material and Methods section provides details…… The Results section presents maps..…. The Discussion section provides recommendations…..
Methods
L152,153: It should be reported also how these characteristics (volume, frequency, depth) are in the HD version.
L162 and 165: avoid repetition and combine the info on 42 features in a single sentence.
L173: clarify the meaning of “dynamics of their imagers”.
L211: …with plankton net data…..
L219: clarify the meaning of “common but not evenly distributed”.
L228: clarify how the images were “independently reviewed”.
L234: clarify what is the meaning of “their ecology” here: biogeography? vertical distribution?, seasonal distribution?
L239:…in different depth layers [or, at different depths].
L251, 252:….(down to 200 m depth)… [same for 500 m and 1000 m].
Results
L268 and 277: the proper “seasonal variation” is not visible in Figure 2; I would rather write “…remarkable annual and interannual variation”.
L269: 30°S-30°N is not the equatorial region but the tropical zone.
L310: remove “the most abundant category” because it’s a repetition.
L312: …..UVP5 HD (Figure. 5).
L312: plankton categories [not “taxonomic”].
L313: remove “compared to detritus”
L315: HD [not “hd”].
L316: define the limits of “small size range”. Remove “(Figure 5)” from here because it must be cited at the beginning of this section (see my note at L312).
L327: The units must be the same in the text and in Figure 6 (mm ESD or mm3)
L338: The genus names must be in italis
L343: …both the filter-feeders (…) and carnivores (….) had very low concentrations…
L356: …the five plankton categories resented…
L357: …(SD and HD) (Figure 8). [move here the fig citation from L358]
Discussion
L368: …the UVP5 images…
L383, 384: Move “Altogether….version.” to the beginning of section 4.1.
L395-397: Revise the English
L408-410: The info about validation should be moved to Methods, indicating more clearly how it was independently done.
L413: The factors that create artifact images are not clearly explained.
L414: “air intrusion” where? Clarify.
L435: Trichodesmium
L445: “ecological dynamics” is vague, rephrase.
L454-455: I suggest to avoid generalist sentences like this one
L458: …understanding their distribution in different environmental conditions is crucial….
L490: …integrating the UVP5 profiles with those acquired with the new instrument’s versio (UVP6 LP and HD) into the observation system…
L496: The new UVP, which can be attached or associated with different platforms, is now widespread distributed and has allowed the acquisition of innumerable profiles and 87 millions of images…
Table 1: I recommend to shorten and simplify the column titles: “Broad categories”, “Children categories” and report their explicative details in the table caption.
Moreover, I suggest to leave in the first column only the category names with the exponent numbers and move the number of vignettes to a second column.
The names: Flota, Poebious, Pseudocalansu, Munnopsis, Dolichospermum , Nodularia, Aulacantha, Aulographis, Coelodendrum, Aulatractus must be written in italics.
In many other names that are not genera or species, the italics must be removed, e.g., Chaetognatha, solytaryblack, ovigerous, Crustacea, etc.
Check the name “Euphausiace”, it should be Euphausiacea.
The reference “Christiansen et al., 2022” is not in the reference list
Figure 1: It seems to me that some vignettes lack the scale bar (10, 30, 33), while in some others the bar and/or the units are barely visible (14, 19, 20, 21, 25).
Figure 2, caption: a) Sampling effort….2018 in the global ocean. b) Latitudinal distribution…profiles. c) Maximum vertical extent of the profiles.
Figure 3, caption: …….for the 33 consistent categories ordered by descending counts…..colors.
Remove the note because it’s a repetition of info already clearly provided in Methods.
Figures 4, 5, 6, 7: Replace “[0, 100]” “[100, 500]” “[500, 1000]” with, respectively “0-100 m”, “100-500 m”, “5000-1000 m”
Figure 4: Map of the global distribution of detritus and………..
Figure 5: Move “Detritus” to the first column on the left, to leave together all planktonic categories.
Figure 6: The units must be the same in the text (L327, mm) and here (mm3).
Figure 7, caption: …and the five plankton categories. [not “functional”].
Table A1: Caption: List of the 62 projects included in the present study, with the corresponding identifying number (pprojid), ….
The title of the fourth column (pprojid) is missing.
The data-owner column should be moved after the project title and before the license.
Table A2: Caption: Morphological features computed for each imaged object…
I suggest to group the features of the same type one after the other, for example all those that refer to the grey level.
“objid” is not a morphological feature.
Area: say how it is calculated.
Median: it’s a statistical parameter that should be listed close to mean, stdev and mode.
Min and max: I think it’s more logic the opposite, i.e., 0 grey= while, 255 grey= black.
Circ: sorry, but I do not understand this formula.
Fractal: Berube and Jebrak, 1999 should be reported at the end of the table.
Fcons: Amadasun and King, 1989 should be reported at the end of the table.
Elongation: “major/minor” what? Complete the definition
What “cv” and “sr” stand for?
Table A3: The names must be carefully checked (e.g., Donoso, not Denoso; Lopez Ruben, not Lopes Rubens).