the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Aug 2026
The Hyperspectral Bio-Optical Observations Sailing on Tara (HyperBOOST) dataset
Abstract. During the Tara Europa Expedition, the HyperBOOST (Hyperspectral Bio-Optical Observations Sailing on Tara) team, part of the larger research program TREC (TRaversing European Coastlines), carried out the collection of discrete and continuous measurements to explore the bio-optical properties of water at the land-sea interface. During the 409 days of the expedition between April 2023 and August 2024, 202 HyperBOOST stations were sampled along the European coastlines, including the Baltic Sea, the North Sea and the Mediterranean Sea. Discrete biogeochemical samples were collected in triplicate at all stations using Niskin bottles, including dissolved and particulate organic carbon, suspended particulate matter, chlorophyll a and other biologically relevant pigments, as well as samples for dissolved and particulate absorption. Along-track continuous AOP measurements were carried out using a So-Rad (solar tracking radiometry platform), while IOPs were collected and processed through the automated In-line system/software. After binning and quality control, 236,567 one-minute binned particulate absorption and attenuation spectra, 26,589 five-minute binned back-scattering spectra, and 7001 five-minute binned So-Rad reflectance spectra were available for further analyses and colocation with the station data. Overall, the Tara Europa expedition allowed to systematically sample European coastal waters at unprecedented temporal and spatial coverage, crossing regions characterized by a wide range of optical and biogeochemical conditions. The combined use of different instruments or measurements approaches also provided the opportunity to assess the internal consistency of the dataset by performing several optical closure assessments. The HyperBOOST dataset provides a valuable resource for advancing and evaluating existing and devising novel bio-optical models in optically diverse waters, enhancing satellite ocean-colour applications and product validation.
- Preprint
(9659 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 08 Oct 2026)
- RC1: 'Comment on essd-2026-623', Giuseppe Zibordi, 23 Aug 2026 reply
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 108 | 41 | 12 | 161 | 19 | 18 |
- HTML: 108
- PDF: 41
- XML: 12
- Total: 161
- BibTeX: 19
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of the manuscript ‘The Hyperspectral Bio-Optical Observations Sailing on Tara (HyperBOOST) dataset’ by Costanzo et al., submitted to Earth System Science Data (Reviewed by G. Zibordi; August 23, 2026)
The manuscript presents a comprehensive dataset of bio-optical measurements performed in European Coastal waters during the Tara Europa Expedition. The work is relevant and definitively requires consideration. Still, the authors should address a number of comments before the work is published.
Comments
Line 30: While the ‘unprecedented spatial coverage’ of European coastal waters is appreciated, the ‘unprecedented temporal coverage’ may require some clarification. Are the authors referring to the short-term temporal resolution of the field measurements?
Line 51: Stating that ‘no coordinated initiative’ did not address systematic measurements of optical properties across the full extent of the European coastal domain beyond the COASTLOOK program, looks at least inaccurate. The BiOMaP program, which is cited in the manuscript, is a clear example of a measurement program that addressed both coastal and open sea bio-optical measurements in the major European Seas including the relatively little explored Black Sea. The sentence should be revised.
Lines 107-108: What high-resolution multi-year regional satellite data product was used to plan the measurement stations? The 1-km offshore mean distance of the selected stations may suggest concerns on the quality of most satellite data products. Some more details are needed.
Line 138: The acronym ESA is added aside HyperBOOST. The acknowledgments make clear that ESA supported HyperBOOST. Still, a single notation ‘ESA HyperBOOST’ or ‘HyperBOOST’ should be applied across the manuscript.
Line 146: Where applicable, Table 1 should provide details on the spectral range and resolution of data products.
Line 168: Are the CDOM absorption data included in the dataset corrected for any background signal to account for temperature and salinity perturbations? Was any quality control procedure put in place to assess the final CDOM absorption spectra?
Line 199: The IOCCG protocols ‘Volume 1’ published in 2018 is that detailing the determination of absorption coefficients of marine particles through spectrophotometric techniques using GF/F filters! The IOCCG protocols ‘Volume 6’ cited in the manuscript does not look to be the correct reference!
Lines 212-213. Is the procedure applied for the removal of phytoplankton pigments a community shared one? Is this particular methodology, with clear reference to the amount of methanol and time of interaction with particles, documented in any relevant publication? If yes, a reference should be added.
Line 220: Is there any quality control procedure applied to asses the derived absorption spectra by particles?
Line 233: The 4.67% relative uncertainty quantified for TChla is solely due to measurement repeatability. This should be specified to avoid this uncertainty value is interpreted as the overall uncertainty of the proposed pigments concentration.
Line 260: It is not clear what ‘unsmoothed’ refers to. It should be detailed.
Line 282: Does So-Rad also compensate for tilt or for azimuth direction only?
Line 304: Were all the TriOS radiometers involved in HyperBOOST fully characterized for their non-ideal performances (i.e., cosine response, stray-light, etc.). This should be clearly stated because it is a key element for HyperCP processing and also for the quantification of products uncertainties. Please also note that it is ‘stray-light’ not ‘stay-light’.
Line 309: The sentence ‘ … the NIR correction defaults to no-offset correction, …’ is not clear. Is there any missing word?
Lines 317-318: The sentence ‘… we relaxed the fraction of measurements to be removed to be 50% (from the typical 10%)‘ looks questionable. Isn’t that the number of typically removed measurements is 90% and it is here expanded (i.e., relaxed) to 50%?
Lines 337-338: The SatCon software tool only allows to convert raw data into ascii uncalibrated or calibrated values. If Satcon is mentioned, the above details should be provided. From the text it appears that some other code was applied to actually process the data.
Line 353: If n_sw is applied accounting for its spectral dependence, this should also appear in Eq. 10 (i.e., \lambda should be added to the n_sw symbol).
Line 355: The diffuse attenuation coefficient K_Lu is determined through an approximation. This should be stated. In addition, a reference should be provided and an explanation for the 0.5 factor should be added.
Line 359: The IOCCG 2018 report is not in the reference list.
Line 350: There is no mention to self-shading corrections. Are these applied? I assume the HTSRB used in HyperBOOSTS is a free-fall system with floating collar (please confirm). In such a case, details on the quality assurance procedures put in place to minimize the impact of additional shading perturbations by the free-fall system, should be summarized.
Line 369: It appears that a word is missing in ‘… not inline …’.
Line 370: Not clear what ‘TSG’ means.
Line 393-394: If the uncertainty mentioned here is simply due to measurement repeatability, it should be stated. Measurement uncertainty includes contributions from many more sources, not only the measurement repeatability.
Tables 3 and 4: The number of decimals should be harmonized across the whole tables.
Line 603: The sentence indicating that this work extends the range of values in the a_NAP axes with respect to previous studies focussing on European waters, appears very speculative. This is only justified by two points exceeding 80% when comparing the results from this work with those for the North Sea in Zibordi and Berthon (2024). For consistence, the Authors should also state that their range of values for the a_CDOM and a_ph axes are smaller than those presented in previous investigations related to European Seas (at least with respect to Zibordi and Berthon 2024).
Line 644: Were the So-Rad and HTSRB radiometers inter-calibrated? Some statement on their absolute calibration should be added.
Line 653: It should be clarified that the comparability of in-water/above-water results of this study with those of Pitarch et al. (2020) applies for the moderately optically complex waters case. Results from the clear waters case, also presented in Pitarch et al. (2020) are generally much better.
Line 664: What are the 1-sigma propagated uncertainties? Are those provided by HyperCP, but restricted to So_Rad? Are they indicating the measurement repeatability as previously considered for other data? This should be clarified for both So-Rad and HTSRB data products.
Line 691: When referring to the collection and processing of samples, the sentence ‘… they are comparable without the need of intercalibration’, does not make sense. I assume the Authors want to say that the standardization applied to sampling and laboratory analysis makes their results consistent across the whole campaign. But this does not mean they cannot benefit from inter-comparisons. Definitively, inter-comparisons would help to identify measurement issues, if any. The Authors should revise the text.