the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Feb 2025
The primacy of dissolved organic matter to aquatic light variability
Abstract. Absorption and scattering by optically active constituents (OACs) modify the sunlit aquatic light environment, facilitating the derivation of biogeochemical data products at scales spanning in situ to satellite observations. Excluding solar illumination and atmospheric effects, variability in an optical parameter arises from changing OAC concentrations, wherein observed patterns in the spectral evolution of data products are associated with the connectivity and spatiotemporal dynamics of OACs. In open-ocean water masses far from terrestrial and riverine inputs, the content and mixture of OACs principally relates to the dynamics of the microbial loop—a trophic pathway describing the cycling of microbial primary producers (i.e., phytoplankton), remineralizers (e.g., bacteria and archaea), plus dissolved organic and inorganic materials (as applicable). Historical models of open-ocean optical data products, such as the normalized water-leaving radiance, [LW (λ)]N, primarily invoke chlorophyll a (Ca)—a commonly used proxy for phytoplankton biomass—as the ubiquitous independent variable governing aquatic light variability. Formulation of [LW (λ)]N as a function of Ca content assumes an idealized microbial loop wherein phytoplankton variability modifies other OACs, including the colored (or chromophoric depending on the literature) portion of the dissolved organic matter (DOM) pool, hereafter CDOM. The prescription in which Ca maximally captures oceanic light variability (hereafter primacy) is tested herein using eigen analyses on three independent bio-optical datasets to assess the shapes and associations of the principal and secondary eigenfunctions of aquatic [LW (λ)]N observations. The analyses reveal [LW (λ)]N variations to be more strongly associated with changes in CDOM than Ca—even for oligotrophic and oceanic datasets—indicating that CDOM dynamics are more variable and exhibit greater independence from Ca than formerly ascribed. Blue and green band-ratio algorithms routinely used for remote sensing of Ca are found to be maximally sensitive to CDOM—rather than Ca—variability based on partial correlation coefficients relating eigenfunction scalar amplitude functions to field or derived observations, plus validation tests of OC algorithm performance. Spectral subset eigen analyses indicate expansive spectral range observing improves the independence in retrieving CDOM absorption and Ca. The combined findings indicate expanded spectral observations supported by recent domestic and international satellite missions constitute a new and unique opportunity to optically characterize surface ocean phytoplankton stocks without relying on explicit or implied empiricisms requiring CDOM and other OACs to vary consistently with Ca. Shapes and associations of the eigen functions suggest a greater diversity of trophic pathways drive OAC dynamics—e.g., in addition to phytoplankton contributing CDOM via cellular lysis, excretion, and grazing—and are consistent with advancing knowledge of the microbial loop in the decades after bio-optical formulations based on Ca were proposed.
Competing interests: The contact author has declared that neither of the authors has any competing interests.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(2395 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2395 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-4163', J. Xavier Prochaska, 03 Mar 2025
The manuscript by Houskeeper and Hooker presents a novel perspective on the primacy of colored dissolved organic matter (CDOM) over chlorophyll a (Ca) in driving aquatic light variability. This manuscript makes a significant contribution to the field of ocean optics by challenging long-standing assumptions about the primacy of chlorophyll-a in driving aquatic light variability. Their comprehensive analysis of three independent bio-optical datasets provides compelling evidence that CDOM absorption—not chlorophyll a—represents the dominant factor influencing spectral variability in aquatic environments. This finding has profound implications for ocean color remote sensing, potentially reshaping algorithm development and improving our ability to monitor marine ecosystems from space. The authors' demonstration that expanded spectral ranges (including UV and NIR domains) improve the independent retrieval of optical constituents provides timely insights for utilizing data from new hyperspectral satellite missions like PACE. Their work effectively bridges historical perspectives on ocean optics with contemporary understanding of microbial loop dynamics, offering a more nuanced view of the biogeochemical processes influencing marine optical properties. Overall, the paper represents a valuable contribution that could significantly advance our ability to interpret and utilize ocean color observations in a changing climate.
While their eigendecomposition analysis is thorough, I have concerns about terminology and methodology. The authors repeatedly refer to "EMA" (end-member analysis) without adequately explaining this approach or its relation to established methodologies. Have they invented it? If so, I am not sure it warrants a name nor acronym. After investigating their cited works and testing their CDOM prescription against the Loisel+2023 dataset (see: https://github.com/ocean-colour/bing/blob/ema/nb/EMA/Explore_EMA.ipynb), I found their approach exhibits sensitivity to algorithm variants that warrants further discussion. The authors appear to have developed this method across several publications, but it requires more explicit contextualization for readers unfamiliar with their previous work, particularly regarding how different implementations might affect results.
The manuscript notably omits sufficient discussion of detrital absorption, which significantly contributes to absorption at wavelengths below 500nm. The authors acknowledge overlapping spectral characteristics between Ca and CDOM but don't adequately address how detrital absorption might influence their conclusions about CDOM primacy. Given that detritus can substantially affect spectral signatures in the same regions they analyze, this omission represents a gap in their analysis framework. A more comprehensive treatment of all optically active constituents would strengthen their argument regarding the relative importance of CDOM in aquatic environments.
Throughout the manuscript, the authors consistently refer to their analysis as an "eigendecomposition" rather than using more widely recognized terms like Principal Component Analysis (PCA) or Empirical Orthogonal Functions (EOF). While technically correct, this terminological choice may unnecessarily distance their work from the broader scientific literature, potentially impeding recognition of their method's relation to established techniques. Additionally, the manuscript would benefit significantly from showing the mean spectra alongside their eigenfunctions, as well as demonstrating reconstructed spectra with varying CDOM and Ca concentrations to illustrate how these constituents individually contribute to spectral variability.
The authors should acknowledge that their eigendecomposition approach assumes linear relationships among spectral variables, which may not fully capture non-linear aspects of aquatic optical properties. Ocean color properties often exhibit complex non-linear relationships, particularly in optically complex waters. This impacts some of the conclusions drawn. In particular, on Line 503-505, the authors write: “.. the leading eigenfunction would be anticipated to capture nearly all of the variance of the dataset.”. This only holds if the relationships are linear.
While their analysis provides valuable insights into the primary modes of variability, explicitly discussing these limitations would provide readers with important context for interpreting their findings. This is especially relevant given their conclusion that CDOM, rather than Ca, drives the primary mode of variability in aquatic light fields—a finding that challenges conventional assumptions in ocean color remote sensing.
A last, potentially challenging request: I am going to insist that the authors make the RSE2021 and RSE2022 datasets public, not merely by request. These are too valuable to leave to the chance that the 2 authors become unavailable, etc.
Here are some additional, more minor comments for the authors to consider. In order
of appearance, not importance.
- Include numbers in the Abstract, i.e. be quantitative.
- I encourage the authors to reference Cael+2020 near line 95
- Line 137: maybe specify that the “optical signatures” are spatial not spectral
- Lines 139-142 :: The leading components are not always (maybe not even typically) dominated by broadband features. Those are taken out by the mean. One example I know where the first modes are *very* informative are galaxy spectra. Please reword these lines accordingly.
- Somewhere in the last paragraph of the Introduction, I encourage the authors to cite the recent papers by Z. Erickson (2020 and 2023)
- Somewhere, consider citing Siegel+2013 as a previous reference showing/asserting that CDOM dominates absorption at bluer wavelengths.
- Line 313 – Please cite a reference supporting the assertion that “scattering processes confer less spectral dependencies”
- Line 373 – I have to admit I am not familiar with the term “R2 statistics”
- Line 436 – I struggle to parse this paragraph. Found myself mainly confused reading it. Maybe provide additional context?
- Line 447 – “are stable” -> “are [un]stable” ?
- Line 460 – I was not persuaded that item c) was actual, independent evidence
- Lines 489- I encourage authors to include equations describing the processes written about here.
- Line 560 – Consider citing Prochaska & Frouin (2025) here.
- The conclusions are largely summary + self-promoting. If that is the standard for this Journal, no problem.
Citation: https://doi.org/10.5194/egusphere-2024-4163-RC1 -
AC1: 'Reply on RC1', Henry Houskeeper, 10 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2024-4163/egusphere-2024-4163-AC1-supplement.pdf
-
RC2: 'Comment on egusphere-2024-4163', Fernanda Maciel, 12 Mar 2025
The manuscript proposes a change in paradigm in ocean color remote sensing, providing evidence that CDOM absorption -rather than Ca- primarily drives water-leaving light variability. They additionally make a strong case for exploiting information in the INV range of the spectra, clearly highlighting current limitations in both in situ and satellite datasets. The work has potentially great scientific implications, as it could change the way that aquatic light variability is thought, and ocean color algorithms are designed. However, I believe that the authors should address some comments detailed below to make their point stronger.
The authors acknowledge the important role of different OACs in the variability of aquatic light, with the purpose of ultimately demonstrating the primacy of CDOM over Ca. OACs in water are typically grouped in three different categories: CDOM, phytoplankton and non-algal particles (NAP) that can covary or not. However, the work does not assess the potential contribution of non-algal particles (NAP) to light variability (apart from briefly mentioning inorganic particles in lines 410-412). How relevant are NAPs in the study’s datasets? Can their contribution be reasonably neglected? If so, this needs to be clearly justified. How does this omission in the datasets limit the findings of the work? How about the extrapolation of the findings to waters where NAPs contribution may be more relevant.
CDOM absorption coefficient (aCDOM) is an optical property itself, more directly related to Lw than Ca. Therefore, one could argue that it would be fairer to use phytoplankton absorption (aphy) rather than Ca in this study. They additionally have another important difference: while the spectral shape of CDOM absorption is smooth and can be roughly defined by two quantities (aCDOM at a reference wavelength, e.g.,440 nm, and an exponential decay exponent), aphy can present much more spectral variability in natural waters, not being as easily parameterized. Can these differences give an “advantage” to CDOM in the analyses presented in the manuscript, particularly when computing linear correlation coefficients with the eigenfunctions? Please include any pertinent discussion about this.
Following the previous comments, why did the authors decide to use only Pearson correlation? I suggest they consider including (additionally) a non-linear correlation coefficient (e.g., Spearman’s rank correlation). This could enrich the results and discussion.
I think the work could greatly benefit from exploiting the datasets a bit more, complementing the results of the principal components analysis. For example:
-For completeness and better contextualization, I suggest including a figure with the measured spectra when describing each dataset, so the reader can have a quick visualization of the spectral variability represented in the work.
-It would also be beneficial to report the correlation coefficient between aCDOM and Ca in the considered datasets, as well as a table with the median, quartiles and ranges of aCDOM and Ca measurements. Could it be the case that aCDOM can “explain” more variance because it already had greater variability in the datasets, while Ca varied in a more limited range (maybe it is necessary to consider some relation between aphy and Ca to do this comparison)?
Finally, I suggest the authors include an example of improved Ca retrieval after determining aCDOM first. I know this is a non-trivial and challenging request, but I think it would be a very convincing proof that the change of paradigm is necessary.
Other minor and more specific comments are:
The references throughout the manuscript to “end-member analysis (EMA)” are somewhat confusing. At first I thought that it was a stablished methodology (the term is actually used in other disciplines), but then I realized that it is unrelated to the methodologies used in other disciplines, and it rather seems to refer to a type of algorithm developed by the authors in previous works, referring to the use of wavebands in the extremes of the spectra to estimate CDOM absorption. I think this should be clearly stated in the manuscript to avoid any confusion.
Lines 19-20: “Spectral subset eigen analyses indicate expansive spectral range observing improves the independence in retrieving CDOM absorption and Ca.” I cannot understand this sentence, please rephrase it.
Lines 58-84: Although I find the information summarized in these paragraphs very interesting, I do not think that the level of details included here is necessary, particularly because it does not flow with the reading of the previous and following paragraphs. I would recommend shortening this part and reducing the number of citations to those only necessary to support the current work.
Line 84: The reference to Ruddick et al. (2023) is missing.
Line 104, 154: I am not familiar with the expression “spectrally expansive”, if this is well-known terminology, please disregard this comment, otherwise, I suggest to briefly define what the authors mean by this.
Lines 151, 199, 204, 210, 221, etc. What are contemporaneous measurements? Do you mean simultaneous?
Line 268: “the square root transformation improves normality” Why? Is there a reference to support this assertion?
Lines 270-276: I must admit that I am not very familiar with PCA applied to spectral datasets, and it would have greatly helped if the authors included matrices dimensions in equations (3) and (4).
Line 283-284: ”after adjusting for covariance with the biogeochemical quantity y.” How was this done? Please describe or add a reference.
Lines 335: “or fluorescence properties” instead of “of fluorescence properties”?
Lines 438-440: I do not understand the sentence that begins with: “The variability may, perhaps, correspond in part to…” I do not follow the connection with the previous sentence. Please rephrase it.
Line 580: “aCDOM and Ca absorption” instead of “aCDOM and CDOM absorption”?
Citation: https://doi.org/10.5194/egusphere-2024-4163-RC2 -
AC2: 'Reply on RC2', Henry Houskeeper, 10 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2024-4163/egusphere-2024-4163-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Henry Houskeeper, 10 May 2025
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-4163', J. Xavier Prochaska, 03 Mar 2025
The manuscript by Houskeeper and Hooker presents a novel perspective on the primacy of colored dissolved organic matter (CDOM) over chlorophyll a (Ca) in driving aquatic light variability. This manuscript makes a significant contribution to the field of ocean optics by challenging long-standing assumptions about the primacy of chlorophyll-a in driving aquatic light variability. Their comprehensive analysis of three independent bio-optical datasets provides compelling evidence that CDOM absorption—not chlorophyll a—represents the dominant factor influencing spectral variability in aquatic environments. This finding has profound implications for ocean color remote sensing, potentially reshaping algorithm development and improving our ability to monitor marine ecosystems from space. The authors' demonstration that expanded spectral ranges (including UV and NIR domains) improve the independent retrieval of optical constituents provides timely insights for utilizing data from new hyperspectral satellite missions like PACE. Their work effectively bridges historical perspectives on ocean optics with contemporary understanding of microbial loop dynamics, offering a more nuanced view of the biogeochemical processes influencing marine optical properties. Overall, the paper represents a valuable contribution that could significantly advance our ability to interpret and utilize ocean color observations in a changing climate.
While their eigendecomposition analysis is thorough, I have concerns about terminology and methodology. The authors repeatedly refer to "EMA" (end-member analysis) without adequately explaining this approach or its relation to established methodologies. Have they invented it? If so, I am not sure it warrants a name nor acronym. After investigating their cited works and testing their CDOM prescription against the Loisel+2023 dataset (see: https://github.com/ocean-colour/bing/blob/ema/nb/EMA/Explore_EMA.ipynb), I found their approach exhibits sensitivity to algorithm variants that warrants further discussion. The authors appear to have developed this method across several publications, but it requires more explicit contextualization for readers unfamiliar with their previous work, particularly regarding how different implementations might affect results.
The manuscript notably omits sufficient discussion of detrital absorption, which significantly contributes to absorption at wavelengths below 500nm. The authors acknowledge overlapping spectral characteristics between Ca and CDOM but don't adequately address how detrital absorption might influence their conclusions about CDOM primacy. Given that detritus can substantially affect spectral signatures in the same regions they analyze, this omission represents a gap in their analysis framework. A more comprehensive treatment of all optically active constituents would strengthen their argument regarding the relative importance of CDOM in aquatic environments.
Throughout the manuscript, the authors consistently refer to their analysis as an "eigendecomposition" rather than using more widely recognized terms like Principal Component Analysis (PCA) or Empirical Orthogonal Functions (EOF). While technically correct, this terminological choice may unnecessarily distance their work from the broader scientific literature, potentially impeding recognition of their method's relation to established techniques. Additionally, the manuscript would benefit significantly from showing the mean spectra alongside their eigenfunctions, as well as demonstrating reconstructed spectra with varying CDOM and Ca concentrations to illustrate how these constituents individually contribute to spectral variability.
The authors should acknowledge that their eigendecomposition approach assumes linear relationships among spectral variables, which may not fully capture non-linear aspects of aquatic optical properties. Ocean color properties often exhibit complex non-linear relationships, particularly in optically complex waters. This impacts some of the conclusions drawn. In particular, on Line 503-505, the authors write: “.. the leading eigenfunction would be anticipated to capture nearly all of the variance of the dataset.”. This only holds if the relationships are linear.
While their analysis provides valuable insights into the primary modes of variability, explicitly discussing these limitations would provide readers with important context for interpreting their findings. This is especially relevant given their conclusion that CDOM, rather than Ca, drives the primary mode of variability in aquatic light fields—a finding that challenges conventional assumptions in ocean color remote sensing.
A last, potentially challenging request: I am going to insist that the authors make the RSE2021 and RSE2022 datasets public, not merely by request. These are too valuable to leave to the chance that the 2 authors become unavailable, etc.
Here are some additional, more minor comments for the authors to consider. In order
of appearance, not importance.
- Include numbers in the Abstract, i.e. be quantitative.
- I encourage the authors to reference Cael+2020 near line 95
- Line 137: maybe specify that the “optical signatures” are spatial not spectral
- Lines 139-142 :: The leading components are not always (maybe not even typically) dominated by broadband features. Those are taken out by the mean. One example I know where the first modes are *very* informative are galaxy spectra. Please reword these lines accordingly.
- Somewhere in the last paragraph of the Introduction, I encourage the authors to cite the recent papers by Z. Erickson (2020 and 2023)
- Somewhere, consider citing Siegel+2013 as a previous reference showing/asserting that CDOM dominates absorption at bluer wavelengths.
- Line 313 – Please cite a reference supporting the assertion that “scattering processes confer less spectral dependencies”
- Line 373 – I have to admit I am not familiar with the term “R2 statistics”
- Line 436 – I struggle to parse this paragraph. Found myself mainly confused reading it. Maybe provide additional context?
- Line 447 – “are stable” -> “are [un]stable” ?
- Line 460 – I was not persuaded that item c) was actual, independent evidence
- Lines 489- I encourage authors to include equations describing the processes written about here.
- Line 560 – Consider citing Prochaska & Frouin (2025) here.
- The conclusions are largely summary + self-promoting. If that is the standard for this Journal, no problem.
Citation: https://doi.org/10.5194/egusphere-2024-4163-RC1 -
AC1: 'Reply on RC1', Henry Houskeeper, 10 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2024-4163/egusphere-2024-4163-AC1-supplement.pdf
-
RC2: 'Comment on egusphere-2024-4163', Fernanda Maciel, 12 Mar 2025
The manuscript proposes a change in paradigm in ocean color remote sensing, providing evidence that CDOM absorption -rather than Ca- primarily drives water-leaving light variability. They additionally make a strong case for exploiting information in the INV range of the spectra, clearly highlighting current limitations in both in situ and satellite datasets. The work has potentially great scientific implications, as it could change the way that aquatic light variability is thought, and ocean color algorithms are designed. However, I believe that the authors should address some comments detailed below to make their point stronger.
The authors acknowledge the important role of different OACs in the variability of aquatic light, with the purpose of ultimately demonstrating the primacy of CDOM over Ca. OACs in water are typically grouped in three different categories: CDOM, phytoplankton and non-algal particles (NAP) that can covary or not. However, the work does not assess the potential contribution of non-algal particles (NAP) to light variability (apart from briefly mentioning inorganic particles in lines 410-412). How relevant are NAPs in the study’s datasets? Can their contribution be reasonably neglected? If so, this needs to be clearly justified. How does this omission in the datasets limit the findings of the work? How about the extrapolation of the findings to waters where NAPs contribution may be more relevant.
CDOM absorption coefficient (aCDOM) is an optical property itself, more directly related to Lw than Ca. Therefore, one could argue that it would be fairer to use phytoplankton absorption (aphy) rather than Ca in this study. They additionally have another important difference: while the spectral shape of CDOM absorption is smooth and can be roughly defined by two quantities (aCDOM at a reference wavelength, e.g.,440 nm, and an exponential decay exponent), aphy can present much more spectral variability in natural waters, not being as easily parameterized. Can these differences give an “advantage” to CDOM in the analyses presented in the manuscript, particularly when computing linear correlation coefficients with the eigenfunctions? Please include any pertinent discussion about this.
Following the previous comments, why did the authors decide to use only Pearson correlation? I suggest they consider including (additionally) a non-linear correlation coefficient (e.g., Spearman’s rank correlation). This could enrich the results and discussion.
I think the work could greatly benefit from exploiting the datasets a bit more, complementing the results of the principal components analysis. For example:
-For completeness and better contextualization, I suggest including a figure with the measured spectra when describing each dataset, so the reader can have a quick visualization of the spectral variability represented in the work.
-It would also be beneficial to report the correlation coefficient between aCDOM and Ca in the considered datasets, as well as a table with the median, quartiles and ranges of aCDOM and Ca measurements. Could it be the case that aCDOM can “explain” more variance because it already had greater variability in the datasets, while Ca varied in a more limited range (maybe it is necessary to consider some relation between aphy and Ca to do this comparison)?
Finally, I suggest the authors include an example of improved Ca retrieval after determining aCDOM first. I know this is a non-trivial and challenging request, but I think it would be a very convincing proof that the change of paradigm is necessary.
Other minor and more specific comments are:
The references throughout the manuscript to “end-member analysis (EMA)” are somewhat confusing. At first I thought that it was a stablished methodology (the term is actually used in other disciplines), but then I realized that it is unrelated to the methodologies used in other disciplines, and it rather seems to refer to a type of algorithm developed by the authors in previous works, referring to the use of wavebands in the extremes of the spectra to estimate CDOM absorption. I think this should be clearly stated in the manuscript to avoid any confusion.
Lines 19-20: “Spectral subset eigen analyses indicate expansive spectral range observing improves the independence in retrieving CDOM absorption and Ca.” I cannot understand this sentence, please rephrase it.
Lines 58-84: Although I find the information summarized in these paragraphs very interesting, I do not think that the level of details included here is necessary, particularly because it does not flow with the reading of the previous and following paragraphs. I would recommend shortening this part and reducing the number of citations to those only necessary to support the current work.
Line 84: The reference to Ruddick et al. (2023) is missing.
Line 104, 154: I am not familiar with the expression “spectrally expansive”, if this is well-known terminology, please disregard this comment, otherwise, I suggest to briefly define what the authors mean by this.
Lines 151, 199, 204, 210, 221, etc. What are contemporaneous measurements? Do you mean simultaneous?
Line 268: “the square root transformation improves normality” Why? Is there a reference to support this assertion?
Lines 270-276: I must admit that I am not very familiar with PCA applied to spectral datasets, and it would have greatly helped if the authors included matrices dimensions in equations (3) and (4).
Line 283-284: ”after adjusting for covariance with the biogeochemical quantity y.” How was this done? Please describe or add a reference.
Lines 335: “or fluorescence properties” instead of “of fluorescence properties”?
Lines 438-440: I do not understand the sentence that begins with: “The variability may, perhaps, correspond in part to…” I do not follow the connection with the previous sentence. Please rephrase it.
Line 580: “aCDOM and Ca absorption” instead of “aCDOM and CDOM absorption”?
Citation: https://doi.org/10.5194/egusphere-2024-4163-RC2 -
AC2: 'Reply on RC2', Henry Houskeeper, 10 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2024-4163/egusphere-2024-4163-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Henry Houskeeper, 10 May 2025
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 263 | 67 | 17 | 347 | 20 | 39 |
- HTML: 263
- PDF: 67
- XML: 17
- Total: 347
- BibTeX: 20
- EndNote: 39
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Henry F. Houskeeper
Stanford B. Hooker
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2395 KB) - Metadata XML