SSolar-GOA v1.0: a simple, fast, and accurate Spectral SOLAR radiative transfer model for clear skies

Cachorro, Victoria Eugenia; Antuña-Sanchez, Juan Carlos; de Frutos, Ángel Máximo

doi:https://doi.org/10.5194/gmd-15-1689-2022

Articles | Volume 15, issue 4

https://doi.org/10.5194/gmd-15-1689-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-15-1689-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 15, issue 4

Model description paper

|

25 Feb 2022

Model description paper |

| 25 Feb 2022

SSolar-GOA v1.0: a simple, fast, and accurate Spectral SOLAR radiative transfer model for clear skies

Victoria Eugenia Cachorro, Juan Carlos Antuña-Sanchez, and Ángel Máximo de Frutos

Download

Final revised paper (published on 25 Feb 2022)
Preprint (discussion started on 22 Jul 2021)

Interactive discussion

Status: closed

RC1:
'Comment on gmd-2021-178', Nina Crnivec, 23 Aug 2021

The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2021-178/gmd-2021-178-RC1-supplement.pdf

Citation: https://doi.org/10.5194/gmd-2021-178-RC1
- AC1: 'Reply on RC1', Juan Carlos Antuña-Sánchez, 12 Nov 2021
  
  The authors thank to the reviewer for the effort to review the manuscript and for its fruitful and detailed comments. Attached is a PDF with the responses to the reviewer's comments.
  
  Citation: https://doi.org/10.5194/gmd-2021-178-AC1
RC2:
'Comment on gmd-2021-178', Anonymous Referee #2, 27 Sep 2021

The authors present a “simple, fast, and accurate” hyperspectral solar radiative transfer model for clear skies (SSolar-GOA v1.0). They evaluate the model against a state-of-the-art radiative transfer model (libRadtran) and observations, showing an impressive accuracy and promising applicability in a multiple of different disciplines. Although, the overall analysis, focus, and results are, to an extent, appropriate for Geoscientific Model Development, as well as novel and important, I found the writing, organization, and presentation of results severely lacking. I would recommend thorough revising before further consideration. I provide more detailed comments below, however I do think considerable revision is needed before a proper evaluation can be completed.

PRIMARY COMMENTS

As stated throughout, I found the writing rather awkward, poor, or extremely confusing in several areas. This makes it challenging to follow the rationale, results, and discussion. Please consider a careful review of the writing with extra attention paid to sections/sentences that are awkwardly written.

The presentation of results focuses are merely visual or limited to percentual differences. A lot can be learnt from linear fits, and their r2’s and RMSE’s values. See specific comments for more context.

The model seems to do a great job, but the paper would be more interesting if the authors could explore limitations of the model as well, and move faster to results and discussions. Maybe add more discussion describing how the model could be improved, what areas are lacking, what type of simulations and scientific questions cannot be explored with this model, and how other areas could potentially benefit from this. I know the authors refer to other studies, but without really giving any concrete example in the paper. This is a real breakthrough and I can indicate a few:

Yang, P., Prikaziuk, E., Verhoef, W. and Van Der Tol, C.: SCOPE 2.0: a model to simulate vegetated land surface fluxes and satellite signals, Geosci. Model Dev, 14, 4697–4712, doi:10.5194/gmd-14-4697-2021, 2021.

Braghiere, R. K., Wang, Y., Doughty, R., Sousa, D., Magney, T., Widlowski, J.-L., Longo, M., Bloom, A. A., Worden, J., Gentine, P. and Frankenberg, C.: Accounting for canopy structure improves hyperspectral radiative transfer and sun-induced chlorophyll fluorescence representations in a new generation Earth System model, Remote Sens. Environ., 261, 112497, doi:10.1016/j.rse.2021.112497, 2021.

Abstract.

The abstract is too long and contains some methodology. The abstract should concise and describe general relevance and main results. Line 12-18 could be removed. Starting the abstract with the general applicability of the study may attract interest. This section should be re-structured.

Line 10: are adapted? It looks like something is missing. It looks like it is a direct translation.

Line 14: “sufficient accuracy” – can you provide a r²? A RMSE in percentage? Anything that exemplifies what that means.

Line 28: Avoid wording like “obviously” in scientific writing.

Introduction.

Line 32- Earth-atmosphere System

Line 36- energy?

Etc is a vague word and should probably be used minimally.

Line 45 – what is etc? be precise. Please define the spectral wavelengths associated with UV, visible, etc.

Line 55 – do not refer other studies in this way. Just write these between brackets.

Line 60 – etc.

Line 71 – 1-10 nm is low to medium? Don’t you mean medium to high?

Line 79 – libRadtran reference?

Material and methodology.

This sections is way too long and could be substantially reduced, with some of the sections moved into a Supplementary material or appendix.

Line 139 – etc.

Line 142 – Earth

Line 159 – the BLB law.

Line 160 – which component?

Line 161 – This gives rise? What does that mean?

Line 163 – etc. Paragraph 3.1?

Line 164 – there are two verbs in this sentence.

Line 165 – you already defined RTE before.

Line 166 – to solve -> solving

Line 168 – specific problem involved? This is so general. Give examples.

Line 172 – ETR?!

Line 173 – for the diffuse component only.

Line 174 – Not only to the atmosphere, but adapted for canopies to:

Sellers, P. J.: Canopy reflectance, photosynthesis and transpiration., Int. J. Remote Sens., 6(8), 1335–1372, doi:10.1080/01431168508948283, 1985.

Line 179 – ETR?

Line 197 – BLB law.

Line 212 – period missing.

Line 224 – Again, 1-10 nm is a very resolution.

Line 226 – what is this error?

Line 231 – Thank you for giving the link to the model. How can the direct component be higher than the global one for some wavelengths?

Line 233 – You already defined BLB.

Line 289 – Use the symbol of micrometers.

Line 321 – 1 DU instead of 1 dobson.

Section 3.3. This list of items could be a Table.

Results.

Fig1. Add degrees to the numbers next to SZA. Write down Direct-horizontal instead of dir-how. Figures should be directly interpretable.

Line 444 – Before the comparison? What?

Fig 1 and 2 could be combined into one single figure, with the top row being fig 1 and bottom row fig 2. Ozone = 300 DU, not Dobson. Add units of all the other parameters too.

Fig3 should include SZA= 6 deg as well. Be consistent.

Fig 4 is repetitive and could probably be moved into supplementary material.

Line 533 - see libRadtran user Ìs guide, 2015? Please reference appropriately.

Fig 4 and 5 could be combined into a single one too. Same thing about adding degrees next to the SZA numbers throughout.

Fig 6 – what are the different colors? Please use an include color scheme suitable for colorblind people.

Fig 7 – Please add the runs from libRradtran here for comparison too.

Line 586 – How do you know the agreement is “excellent”? Visually, it looks great, but could add some statistics into your evaluations? A linear fit with observed/simulated with libRadtran versus SSolar-GOA (r², RMSE, and slope) could tell us so much about model performance.

Fig 8 - Please add the runs from libRradtran here for comparison too.

Fig 9 – This is not your work, could probably be moved into supplementary material. Please add the full citations in the figure, e.g., Kurucz, 1992.

Line 634 – add comma after ‘To this’.

Fig 10 – show linear fit with r² and RMSE.

What is the purpose of Fig 11?

Fig 12 – show linear fit with r² and RMSE.

Conclusion.

Line 708 – avoid huge and extensive.

Line 711 – avoid enormous.

Citation: https://doi.org/10.5194/gmd-2021-178-RC2
- AC2: 'Reply on RC2', Juan Carlos Antuña-Sánchez, 12 Nov 2021
  
  The authors thank to the reviewer for the effort to review the manuscript and for its fruitful and detailed comments. Attached is a PDF with the responses to the reviewer's comments.
  
  Citation: https://doi.org/10.5194/gmd-2021-178-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

AR by Juan Carlos Antuña-Sánchez on behalf of the Authors (12 Nov 2021) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (12 Dec 2021) by Sylwester Arabas

Dear Authors,

Thank you for providing the revised manuscript and the reply to the reviews.

First of all, let me point out that the GMD code availability policy extends to all code needed to reproduce results presented in the paper including scripts that automate model runs and visualization. In the case of the present paper, this should cover scripts automating the SSolar_GOA and libRadtran runs for the comparative analysis. Please provide the code required to reproduce the plots presented in the paper (e.g. as an electronic supplement to the paper or at a persistent repositry as Zenodo).

Below, I am providing notes taken while reading the revised manuscript and the exchange with the reviewers. Before proceeding, please address these points.

Sincerely Yours,
Sylwester Arabas

* Page 1 (abstract)
- line 7: physical-based -> physically-based
- line 17: sentence "Besides, ..." unclear
- line 18: remove "extensive"
- line 21: "shows a high performance ... and it underestimate" - please shorten, perhaps clarifying that the presented validation quantified the relative differences...
- line 27: remove "Obviously"
- given the journal and paper scope, I suggest to include in the abstract and introduction a mention of implementation in Python and open-source licensing

* Page 2:
- line 48: should "Research Centres" be capitalised?
- line 51: crossed out "in a"
- line 52: Hodges 1993 cited as Hodges 1990

* Page 3:
- line 93: "allows" -> "allow"

* Page 4:
- line 116: "has the drawback" -> "comes with a tradeoff" ?

* Page 5:
- line 131: the "or" in units parenthesis is unclear
- line 134: "widest used" -> "most widely used"
- line 145: rephrase "disappeared"
- line 153: "law" missing after Bouguer
- line 153: remove "easy"

* Page 6:
- line 163: "solver" -> "solvers"
- line 166/167: "but contains" -> "and comes in" ?
- line 178: "which has undergone..." - what is the purpose of this statement?
- line 181: package -> source code
- line 184: remove "really"
- line 186: two sentences in a row begin with "Therefore, "
- line 187: what does "the Mie program" refer to?

* Page 7:
- line 201: "high" -> "height" ?
- line 205: "though" -> "through"
- line 206: give example references to these models
- lines 207-209: "This is a physical, fast, ..." suggest removing this sentence
- line 209: "As already mentioned, the core of the model is the simplicity of the" -> "The crux of the model is the simple"
- line 251: rephrase "working very well" (aimed at operation within the ... range ?)
- line 220-221: "The model, in some way, ..." - unclear sentence
- line 221: "The model as described may be easily replicated by the readers, or it may be download Windows version"
-> "The SSolar_GOA v1.0 is released as free and open-source software.
It is implemented in Python offering portability across architectures and operating systems.
For download instructions, see the Code Availability section."

* Page 8:
- line 235: "and various" -> unneeded "and"
- line 140: "where implicitly it is assumed the non-interaction between these processes" -> "where the non-interaction between these processes is implicitly assumed"
- equation 7: looking at the code, the exponent at the last lambda in the denominator should be "-4" not "4"
- line 270: "and sea level" -> "and the sea-level pressure"

* Page 9:
- line 296: "model includes a file": this mixes implementation with formulation, would better sounds as "model uses tabulated coefficients" ?

* Page 10:
- line 304: remove "very"

* Page 11:
- line 338: "However, inverse to" -> "In contrast to"
- line 342: suggest removing "so care must be taken with the units of both quantities in the previous expressions"

* Page 12:
- line 373: please rephrase "as necessary approaches for developing a simple model under the consideration of non-interaction"

* Page 13:
- line 380-381: remove "but a spectral file for ρ is very easy to implement in the model"
- line 393: "are usually taken from the bibliography." - please rephrase and be precise (what "usually" and "from the bibliography" mean?)
- line 393/394: remove "Finally, we call attention to the total number of expressions/formulas..."
(or rephrase and elaborate being precise which other models you refer to and what "number of formulas" implies: implementation challenges, computational cost, ..)
- line 395: suggest adding a screenshot of the graphical user interface here

* Page 14:
- line 415: remove "bear in mind that"
- line 433: remove "Before the two comparisons,"

* Page 18:
- line 489: rephrase "interval in nm of the model"
- line 501: "it can observe" -> "one can observe" ?

* Page 24:
- line 579: "consistence" -> "consistency"
- line 581: "component are" -> "components are"
- line 583: "range stand out" -> "range stands out"

* Page 27:
- first paragraph of 4.2 is all bold

* Page 39:
- line 785: remove "it is true that"
- line 795: rephrase "easy-to-understand"

* Page 40:
- line 807: "and the idea is to implement this new" - rephrase, perhaps like "will be considered in further model development" (to clarify it is not available as of now)
- line 813: rephrase around "two more influent" (and remove parenthesis)
- line 815: "is easy to understand and evaluate" -> "results in concise and computationally undemanding formulation" ?
- line 815: "but more important is that" -> ". Importantly, it was shown that the assumption of"
- line 818: rephrase "one layer or multiple layer do not added"
- line 822-823: correct "for many applications in solar energy for different application agriculture"
- line 824: "Depending of the required accuracy" -> "Depending on the required level of accuracy"

* Page 41:
- line 831: correct "which is covers"
- line 836: rephrase "but it must be in mind", "easy to use, which cannot to compete"
- line 837: rephrase "It is not easy to find in" to "To the authors' knowledge"...
- line 839: correct reference year (1919)

* Code availability:
- Please elaborate what you mean by "portable". Given the code is developed in Python, it is portable across platforms and operating systems by design.

* Zenodo archive:
- the readme file mentions installation, but it's more a "installation of dependencies" rather than package installation
- the archive contains numerous cache files (__pycache__ dirs with *.pyc files)
- the archive is a tarbomb, please include a top directory in it
- the 1.7MB size of the .ico file is intriguing
- after launching the program for the first time, the parameter values do not match defaults (clicking "Load default inputs" on a freshly opened window changes many parameters)
- the software license is not specified within any of the files in the code archive, please at least include a COPYING file

* Figures:
- all figures need to be supplied in a vector graphics format (not raster graphics/screenshots)
(similarly, in the SSolar_GOA user interface, the "Save image" feature would best allow to save in publication-ready vector format (svg, pdf, ...)

* References (see https://www.geoscientific-model-development.net/submission.html#references):
- some entries have journal names abbreviated, some not (see the above URL for suggested abbreviation database),
- some abbreviations include dots, some not,
- some entries use all-words-capitalised, some first-word-capitalised format,
- Fouquart & Bonnel: missing capital letter (Earth's)
- libRadtran user's guide: multiple years unclear, missing authors (Bernhard Mayer, Arve Kylling, Claudia Emde, Robert Buras,Ulrich Hamann, Josef Gasteiger, and Bettina Richter), missing version number, please consider asking the authors to post the pdf at a permanent location (arxiv, zenodo, ...)
- ASD Full Range: wouldn't "Malvern Panalytical, 2021" better serve as label?
- Hodges 1993: unclear what the 1993 year refers to, the website pointed by the DOI does not list Hodges name, "Unitet State" - typo
- Kurucz 1992: is it this paper: https://doi.org/10.1017/S0074180900124805 (if so, the year is invalid)
- Sukhodolov 2014: bogus doi url, missing capital letter in "evaluation"
- Utrillas and Vergaz entries are coalesced

Hide

AR by Juan Carlos Antuña-Sánchez on behalf of the Authors (22 Dec 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (17 Jan 2022) by Sylwester Arabas

AR by Juan Carlos Antuña-Sánchez on behalf of the Authors (22 Jan 2022) Manuscript

Download

The requested paper has a corresponding corrigendum published. Please read the corrigendum first before downloading the article.

Article (2970 KB)
Full-text XML

Short summary

This work describes the features of a simple, fast, accurate, and physically based spectral radiative transfer model (SSolar-GOA) in the solar wavelength range under clear skies. The model is intended for a wide community of users for many different applications, was designed to be easily replicated, and has sufficient accuracy. The validation of the model was carried out through extensive comparison with simulated spectra from the LibRadtran and with direct and global spectral measurements.