Development of a global 30&thinsp;m impervious surface map using multisource and multitemporal remote sensing datasets with the Google Earth Engine platform

Zhang, Xiao; Liu, Liangyun; Wu, Changshan; Chen, Xidong; Gao, Yuan; Xie, Shuai; Zhang, Bing

doi:https://doi.org/10.5194/essd-12-1625-2020

Articles | Volume 12, issue 3

https://doi.org/10.5194/essd-12-1625-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/essd-12-1625-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 12, issue 3

Data description paper

|

15 Jul 2020

Data description paper |

| 15 Jul 2020

Development of a global 30 m impervious surface map using multisource and multitemporal remote sensing datasets with the Google Earth Engine platform

Xiao Zhang, Liangyun Liu, Changshan Wu, Xidong Chen, Yuan Gao, Shuai Xie, and Bing Zhang

Download

Final revised paper (published on 15 Jul 2020)
Preprint (discussion started on 21 Jan 2020)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Reviewer comments to manuscript "essd-2019-200"', Anonymous Referee #1, 03 Mar 2020
- AC1: 'Reply to Referee #1', Liangyun Liu, 02 May 2020
SC1: 'test', zhixin qi, 25 Mar 2020
SC2: 'comments on essd-2019-200', X. P. Liu, 25 Mar 2020
- AC3: 'Reply to SC #2', Liangyun Liu, 02 May 2020
RC2: 'Review of Zhang et al.', Anonymous Referee #2, 11 Apr 2020
- AC2: 'Reply to Referee #2', Liangyun Liu, 02 May 2020

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Liangyun Liu on behalf of the Authors (02 May 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (08 May 2020) by David Carlson

RR by Anonymous Referee #1 (17 May 2020)

Suggestions for revision or reasons for rejection

This paper generates an accurate global coverage impervious surface map with multiple remote sensing datasets. The generated map was proven to have optimal accuracy measurement compared with other state-of-art impervious surface products. I appreciate the author’s response to my questions in the last round of review. I agree that this paper can be published after a minor revision.

1. Line 20-25. “First, the global impervious and non-impervious training samples were automatically derived by combining the GlobeLand30 land-cover product with VIIRS NTL and MODIS enhanced vegetation index (EVI) imagery. Then, based on global training samples and multi-source and multi-temporal imagery, a random forest classifier was trained and used to generate corresponding impervious surface maps for each 5°×5° cell of a geographical grid. Finally, a global impervious surface map, produced by mosaicking numerous 5°×5° regional maps, was validated by interpretation samples and then compared with five existing impervious products (GlobeLand30, FROM_GLC, NUACI, HBASE and GHSL). ”

At the beginning of abstract readers are informed this work will present a global impervious surface map at 30 m resolution, but why did the authors introduce the map was generated for each 5°×5° geographical grid? please double-check the content.

2. Line 59. “GlobeLand30 combined the pixel-based classification, segmentation and manual editing based on the high resolution imagery to develop the 30-m impervious surface map as an independent layer with the user’s accuracy of 0.867 (Chen et al., 2015).”

GlobeLand30 provides users with an “artificial surface layer” instead of an “impervious surface map”. In the original literature which released Globeland30, the “artificial layer” was defined as follows: “Artificial surfaces mainly consist of urban areas, roads, rural cottages and mines, which are primarily based on asphalt, concrete, sand and stone, bricks, glasses, and other materials. They can be divided into three subclasses including high reflectance, low reflectance and vegetated type.” (Chen, J., Chen, J., Liao, A., Cao, X., Chen, L., Chen, X., ... & Zhang, W. (2015). Global land cover mapping at 30 m resolution: A POK-based operational approach. ISPRS Journal of Photogrammetry and Remote Sensing, 103, 7-27.)

I agree that the major content of artificial surfaces is represented by impervious surfaces, but other small vegetation patch areas should not be simply ignored. I suggest the authors should make more clarifications towards this issue, instead of directly using “artificial surfaces” as “impervious surfaces”. Discussion about the impact of “artificial surfaces” can be made in your paper.

3. Line 158. “The Shuttle Radar Topography Mission digital elevation model (SRTM DEM), provided by the NASA JPL at a resolution of 1 arc-second (approximately 30 m) and covering the area between 60° north and 56° south (Farr et al., 2007), was a useful auxiliary dataset for impervious surface mapping over mountainous areas because impervious surfaces mainly located in the flat areas and Sentinel-1 SAR data usually reflected high backscatter similar to the impervious surfaces in mountainous areas.”

The reason for using the DEM dataset for classification is not clear. There are a lot of cities being built on mountains or having hills, e.g. Lisboa, Porto, Prague, Rome, et al. The DEM dataset may cause confusion in classification. Such negative influences may need to be discussed in your manuscript.

Hide

ED: Publish subject to minor revisions (review by editor) (29 May 2020) by David Carlson

AR by Liangyun Liu on behalf of the Authors (02 Jun 2020) Author's response Manuscript

ED: Publish subject to technical corrections (16 Jun 2020) by David Carlson

AR by Liangyun Liu on behalf of the Authors (17 Jun 2020) Author's response Manuscript

Short summary

The amount of impervious surface is an important indicator in the monitoring of the intensity of human activity and environmental change. In this study, a global 30 m impervious surface map was developed by using multisource, multitemporal remote sensing data based on the Google Earth Engine platform. The accuracy assessment indicated that the generated map had more optimal measurement accuracy compared with other state-of-art impervious surface products.