the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Aug 2026
A Compilation of Marine Photosynthesis–Irradiance Data from 14C Incubation Experiments
Abstract. Photosynthesis-irradiance (PI) experiments that measure the uptake of 14C labelled bicarbonate over a light gradient established in an incubator are foundational to computations of primary productivity and models of photosynthesis in ecosystem models. We present a compilation of PI data collected over the last 50 years with largely consistent methods, totalling 3751 PI curves from the North Atlantic (2384 curves), Arctic (1209), North Pacific (90) and Indian (68) oceans. Unlike many compilations, we present original measurements of photosynthetic rate and irradiance, together with derived photosynthesis-irradiance (PI) model parameters as well as detailed metadata including experimental design, environmental conditions, and methodological variation. Using this database and a suite of PI models, we show how parameter estimates typically vary across observations and model formulations, and demonstrate how differences in experimental design and methodology can systematically affect parameter values. We show that quality statistical conversion of parameter values from one model to another is often possible, although direct estimation of parameters from original data is always preferable.
- Preprint
(4672 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (extended)
- RC1: 'Comment on essd-2026-651', Anonymous Referee #1, 17 Oct 2026 reply
Data sets
A Compilation of Marine Photosynthesis–Irradiance Data from 14C Incubation Experiments M. M. Amirian et al. https://zenodo.org/records/21908199
Model code and software
piCurve: an R package for modeling photosynthesis-irradiance curves M. M. Amirian and A. Irwin https://github.com/Mohammad-Amirian/piCurve
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 298 | 177 | 129 | 604 | 153 | 121 |
- HTML: 298
- PDF: 177
- XML: 129
- Total: 604
- BibTeX: 153
- EndNote: 121
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1/ General comment
The manuscript is interesting for the large primary production dataset it presents and uses to show the usefulness of a new data processing package. The paper presents 3751 PI curves (with raw data) covering 50 years, with very homogeneous methods within the Bedford Institute of Oceanography. It proposes a harmonization of the parameters between 7 PI models, including the very practical double tangent. It is methodologically interesting, well conducted, and the work seems useful for integrating heterogeneous historical datasets. The methodological protocols etc. are well described in detail.
However, the introduction that seeks to justify this study is too focused on the sources of error for others versus an us who wants to solve everything. To justify the importance of the study, the introduction uses criteria for comparing models and production rate that are caricatural, which would require a more objective rewriting of the introduction. The words of the introduction can too easily lead the young generation of researchers to say that the old data are all bad, by omitting to mention the difficulties of field measurements and the natural variability of basic measurements (14C incorporation) on natural communities on board.
The results are presented with a good breakdown of the parameters acquired alongside the database, but the objectives of the study lack clarity, compared to what a biologist, for example, might expect to find in this type of work. The discussion is really too brief: it summarizes more than it discusses anything. How to use this type of work on a personal database does not seem very clear. The ideas from the last sentences of the discussion deserve more development and clarity to be applied by users of their identical data.
The paper does not sufficiently discuss the problem of different lights that have been used over time; understandable, but what about the impact on their considerations? The paper also does not discuss the methodological transition to current optical approaches. PP proxies are nowadays often estimated via variable fluorescence, which changes the nature of the parameters. Is it applicable? In the discussion (see also remarks below), the parallel with what is said in the introduction about global assessments to be specified is not discussed at all: what can this work propose potentially change regarding the uncertainties of PP global rates, regardless of the scale at which they are considered?
2/ Review of the manuscript and the data set :
1. The manuscript:
-are the data and methods presented new? The data presented are a compilation of numerous data from the photosynthesis-Irradiance (IP) relationships measured between 1972 and 2022 by the BIO, Bedford Institute of Oceanography, specifying the publications in which they were used.
-Is there any potential of the data being useful in the future? The data can be used to recalculate primary production flows in the area concerned, making comparisons/calibrations with different estimates of PP.
-Are methods and materials described in sufficient detail? The method of incorporation of 14C by microalgae is recalled. It is well specified that the incubation periods are from 0.5 to 4 hours: which is important to say because this also constitutes a strong source of variability between the data. Indeed, incubations of 30 minutes will give estimates close to raw production, whereas incubations of 4 hours in the studied area will give an intermediate estimate between raw and net PP, with possibly an accumulation of the effects of photoinhibition at high intensities (so a more developed beta parameters, etc.). There is a small lack of precision at this level.
-Are any references/citations to other data sets or articles missing or inappropriate? Generally correct, but there are some inappropriate quotes such as those from the author himself to refer to the explanation of the different phases of PI relationships: the names and the meaning of slopes and plateau have other historical references than that of the author who practices here inappropriate self-citation (L196-198).
-Is the article itself appropriate to support the publication of a data set? Yes, the article with these illustrative charts is appropriate to accompany the publication of the data, preferably with a more objective introduction (see specific comments by lines below). Finally, the question of using these data to calibrate current optical models is not mentioned (but we guess that yes).
2. The data quality:
-is the data set accessible via the given identifier? Yes, the data is easily accessible through the links given at the end of the article; different csv files correspond to different types of variables and are easy to identify. Methodological information (metedata) is also given very correctly and clearly.
-Is the data set complete? Are error estimates and sources of errors given (and discussed in the article)? Data from PI curves must be accompanied by environmental hydrological descriptors in order to be exploited from an ecological point of view: this is the case here for temperature, a little less so for other elements such as nutrients, reported at 70%; which remains correct and interesting. Explanations are given in detail by type of parameters. Statistical analyses of the influence of the incubation temperature are proposed in a relevant and very interesting way, because rare analyses in the bibliography. The demonstration of the superiority of the double tangent model, which is used as a central reference, could be more explicit and clear, especially for biologist users.
-Are the accuracy, calibration, processing, etc. state of the art? Are common standards used for comparison? Regarding the mathematical treatment developed in this study with the process of transforming different models into model parameters equivalent to a double tangent equation (?), I am not enough a mathematician to judge this operation, but it is not explained clearly enough, it seems to me. As a biologist, we could also expect a parallel comparison of the photosynthetic parameters themselves. This is probably not the objective of the authors because this subject has already been extensively discussed. However, a little more explanation or clarity at that level would be welcome.
-Is the data set significant – unique, useful, and complete? Yes, the dataset is significant, original, and interesting to publish.
3. Article and data set:
-are there any inconsistencies within these, implausible assertions or data, or noticeable problems which would suggest the data are erroneous (or worse). No specific problem at this level.
-Is the data set itself of high quality? Yes, the data set with explanatory variables for the processes studied as a priority is of high quality and deserves to be published.
4. Presentation quality:
-is the data set usable in its current format and size? The data set can be used without any problem in its current form; moreover, it is very complete with environmental variables, for example, and metadata.
-Are the formal metadata appropriate? Yes, no problem.
-Check the publication: is the length of the article appropriate? The article is quite short: the introduction could be more objective compared to the work proposed here (see notes below). The results are in line with the authors' objective, but the discussion is really too brief. It is a kind of synthesis that does not discuss anything, not even how to apply the new tools proposed by the community of marine biologists.
-Is the overall structure of the article well structured and clear? See previous remarks on the introductions and discussion.
-Is the language consistent and precise? No problem.
-Are mathematical formulae, symbols, abbreviations, and units correctly defined and used? Without any problem.
-Are figures and tables correct and of high quality? The tables and figures are of high quality, accurate and in correct colours.
-Is the data set publication, as submitted, of high quality? The data set is of very good quality, but the introduction and discussion deserve some modifications.
-Finally: By reading the article and downloading the data set, would you be able to understand and (re-)use the data set in the future? Why not, but for other questions. The authors' comments regarding the new tool used here, referenced in a recent publication from 2025, are not clear at the end of the discussion.
3/ Specific comments :
-The beginning of the introduction (up to line 33) talks about overall net primary production, mixing absolutely everything: observations and modelling, in order to justify the importance of continuing this type of study. We cannot calculate a range of variation in NPP estimates by mixing observation (without specifying the incubation times, moreover) and modeling approaches. This gives a fallacious justification, especially by saying that it complicates the detection of trends. This part is really too globalizing, the authors can do a simpler way to justify themselves than to sweep so broadly.
-the sentence from lines 44 to 46 has a curious justification: If incubation estimates are not made continuously at all scales, they can be done via the satellite tool. It is not the PI curve that allows it. The sentence has no meaning or is incomprehensible.
-Lines 50 to 53: a sentence that is also too general : nothing is specified or discussed in parallel about the frequency of PI relationship measurements to calculate PP rates, nor is there anything about the incubation times necessary to measure correct carbon fluxes ; another source of variability for calculating PP rates.
-Line 54: the methodological considerations outlined at this level should be presented before the simplistic sources of error that are given above.
-Line 65: the Peeters and Eilers 1978 reference chosen by the authors is surprising/unexplained in light of its use: it may be mathematically interesting, but it should be justified by comparing it to the other model of Peteers and Eilers from 1988, which was widely used by the community and is still used a lot : It is indeed integrated by default in the current fluorometers software, known as variable fluorescence, to provide an estimate of the parameters of the ETR vs Light relationships. Do the authors have a recommendation to make on this subject? For these study cases, it could be an interesting question/added value.
-around Lines 70: there is talk of different models and photoinhibition (beta), without without defining for novices (via bibliographic references, for example) the difference/link between photodamage and photoinhibition ; that some authors also consider this as a methodological artifact related to the incubation times of the methods, not occurring in the natural environment for deep samples, e.g.
-Lines 77 to 79: it’s quite caricatural to say that : "inconsistent fitting procedure". Do the authors take issue with the variability, reproducibility, of a measure of PI relationship with 14C? No.
-For results, see part 2. of this report.
-For the conclusion, see general comment at the beginning of this report §1.