the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Sep 2025
Improved land mask for satellite remote sensing of oceans and inland waters
Abstract. We present an improved medium (250 m) spatial resolution land mask based on augmenting earlier results of Mikelsons et al. (2021) (https://doi.org/10.1016/j.rse.2021.112356) to reflect recent changes in global water surface coverage. This land mask update is critical for remote sensing of coastal oceans and inland waters as this is the first step to properly identify water pixels from land pixels for satellite data processing. We show that clear sky false color imagery derived for monthly and yearly time periods can be effectively used to identify changes to the surface water coverage. In addition, we also use Sentinel-2 satellite imagery to derive more accurate boundaries of new water bodies with complex geometries. We demonstrate improved coverage from satellite ocean color and inland water property retrievals with the improved land mask, including a range of new inland water bodies, as well as changes to the extent of the existing water bodies. The improved land mask can also be used for remote sensing of land, atmospheric, and cryosphere products.
- Preprint
(1298 KB) - Metadata XML
-
Supplement
(4372 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on essd-2025-543', Anonymous Referee #1, 13 Oct 2025
-
RC2: 'Comment on essd-2025-543', Anonymous Referee #2, 10 Nov 2025
SUMMARY
This manuscript described an updated global land-ocean mask representative of the period 2020 to 2025 and contrasted it with the previous mask to identify secular changes due to water management practices. The new mask refined the mask by incorporating additional satellite information, fixed artifacts in the previous mask, and captured changes of water extent. On the latter subject, the authors detailed the contraction and expansion of several lakes due to water management practices or shifts in rainfall pattern, the emergence of lakes due to new dams, and changes in coastlines due to activities such as land reclamation.
This is a clear, well-written manuscript, not only describing the improvement to an existing product, but also catalogued how human activities are impacting the coverage of water surfaces and the importance of capturing these impacts. The data is readily accessible and appears to be of high quality. There are a few minor points that can help refine the manuscript, but this is a study worthy of publication in this journal.
COMMENTS
1) Title/abstract: Both the title and abstract convey the impression that this manuscript describes a new product, but a large part of it in fact assesses changes in the water extent due to human activities. I consider this a plus, since it goes beyond the simplistic "here's a new product" to connect the product to the real world. But this should be mentioned in the abstract at the minimum, even though I recognize that this journal focuses on the data rather than results.
2) L52-55: I may have missed this elsewhere in the manuscript, but is the product on an "equal-area" grid at 230 m or a typical "equal-angle" grid (i.e., the standard rectangular latitude-longitude grid) that has a resolution of 230 m at the equator (but not at other latitudes)? I suspect it is the latter, in which case please mention the shape of the global array (e.g., "360 × 180" for a 1° grid).
3) L100-101: For context, please provide the approximate period that the existing mask represents.
4) L129-131: Looking at the website, the tool is labeled as "experimental", suggesting that it may not be permanent. Therefore, the authors should provide a description of OCView to a degree such that a researcher could reasonably replicate its functions should it no longer exist in the future.
5) L295-298: These are critical points: the entire manuscript underscores the importance of capturing changes in water extent but also the challenges in doing so. I would like to see a speculative discussion, especially from the perspectives of the authors who have done the actual hard work, on the feasibility of automation (e.g., can the clear sky imagery approach be automated?), periodic updates (e.g., how often is reasonable?), and the use of commercial satellite imagery. Such information would be highly beneficial for anyone in the community considering such an endeavor, though I would also understand if the authors prefer to exercise some restrain over such speculations.
Citation: https://doi.org/10.5194/essd-2025-543-RC2
Data sets
Improved land mask for satellite remote sensing of oceans and inland waters Karlis Mikelsons and Menghua Wang https://doi.org/10.17632/9r93m9s7cw.2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 853 | 67 | 21 | 941 | 45 | 19 | 33 |
- HTML: 853
- PDF: 67
- XML: 21
- Total: 941
- Supplement: 45
- BibTeX: 19
- EndNote: 33
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Overall, this kind of data which updated to recent state of water area is useful for the processing especially for algorithms dedicated to the ocean area. The one-file and one-byte format is convenient to use in processing data from the low-middle resolution global observation sensors. However the seasonal cycle may also affect the OC processing in the coast of the lake or big rivers and need to be considered. This data is indeed valuable in the current state, however it may have more value if it includes information of seasonal cycle orr probability (or coverage) of water cover within the pixels which may be obtained by authors already.
I think this datasets has the publication quality but it may be better to consider the following comments.
Specific comments:
Line-100 "In this study, we use a combination of yearly and monthly clear sky false color imageries from the recent years (2020–2025) ..";Â
 Do you have confirmed that the water areas have not changed except for areas listed in tables in this paper?
Line-123 " we imposed the requirement that more than 90% of high resolution (10 m) imagery pixels":
 Can you clarify that the 90% was applied to only spatial coverage or both spatial and temporal (considering seasonal change) coverage?
Line-295 "Overall, periodic updates are essential to maintain accuracy.":
 Can you propose the update period recommended for the natural and artificial changes?