Reconstruction of Sentinel-2 Image Time Series Using Google Earth Engine

Round 1

Reviewer 1 Report (Previous Reviewer 3)

This latest version of manuscript is much better than its previous version. I consider it meets the criteria of publication in this journal.

Reviewer 2 Report (Previous Reviewer 1)

First of all, I would like to thank the Editorial Board of Remote Sensing for giving me the opportunity to participate in this new review process of this work, and especially the authors for the improvements incorporated to it.

In this sense, already in my previous review I considered that the article had a sufficient level of quality for its publication in Remote Sensing, which I now even consider that it has been improved with the revision made by the authors. Especially in the introduction and bibliographic review. Likewise, it is to be appreciated that the authors provide the GEE code, making it easier for the readers of this paper to apply the methodology in their own fields.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

 

First of all, I would like to thank the Editorial Board of Remote Sensing for the possibility of participating in the revision of this paper.

The paper deals with a topic that is of special interest, such as the reconstruction of satellite images affected, in this case, by cloudiness problems. Without a doubt, it is a subject of great interest and that fits directly into the topics of the Remote Sensing journal. I would also like to highlight the good quality of this work, which includes an adequate bibliographic review supported by updated bibliographical references, a well-argued methodological proposal -based on previous methods applied to other types of images such as MODIS- and, finally, a clear presentation two study areas with several examples that provide interesting results.

Likewise, it is important the fact that a tool that is available to anyone, such as Google Earth Engine, has been used for its execution, and the authors also share their own script for applying the methodology. All these elements provide added value to this work: interest, methodology, application examples, and sources.

In this sense, I consider that the article is likely to be published in Remote Sensing according to the current version, although authors are recommended to make a final review of some typographical aspects of the work, for example, the Sentinel platform sometimes appears in capital letters and others with lower case, as well as the spaces between the text and the bracket of the bibliographical references that sometimes do not appear.

Congratulations on the good work done, it will undoubtedly be an interesting contribution to the journal.

Reviewer 2 Report

The aim of this paper is to generate cloudless Sentinel-2 surface reflectance data in order to provide high-quality satellite data to use as stable products in time-series analysis. Unfortunately, this concept is not clearly explained in the manuscript. I think that the subject is of interest for a wide readership but it is not explained fluently. Please, starting from the abstract, explain which is the aim of your paper, how to reach it, which data to use, and finally describe the selcted case studies. 

By the reading I get the feeling that reflectances are confused with Vegetation Indices or with time series. The text is confused. Moreover, where is the figure to show  your study area? At lines 111-113 authors declare that cloudless Sentinel-2 NDVI and surface reflectance time series data (here the core of the confusion of the whole paper) provide high-quality remote sensing basic products (???) for agriculture, ecology, hydrology, climate and other researches. Which products? Which sector do authors select  in this paper to show the improvement due to their processing?

Generally, I suggest:

-        to rephrase the whole paper. In particular, I suggest to re-write the introduction by considering that the core of the manuscript is to apply a method to obtain cloudless Sentinel-2 data (TOA or BOA, radiance or reflectance?);

-       frame your study area. Why this choice?

-       Include both Result and Discussion sections;

-       make the conclusions consistent with the evidence presented;

-       Insert more recent references. 

Reviewer 3 Report

This is a very interesting manuscript that elaborates on using the adjusted DCT-PLS method to reconstruct the Sentinel-2 image in the case of cloud contamination.  A minor to moderate revision is required to enhance the quality of this manuscript.
My concerns are listed as follows,
(1) The abstract needs to be refined to highlight the novelty and contribution of this study. Besides, some spellings such as  Sentinel-2 should be consistent throughout this section and the full text.
(2) Description of section 2.1.2. Large Scale Study Area should be emphasized and answer why choose these two study areas?
(3) Tables 2 and 3 should be merged and better designed.
(4) What does the Xi denote in equation (6)?
(5) Discussion section containing the new findings and limitations of this study must be elaborated.
(6) Overall citations are somewhat out of date and the latest references should be added.