the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Jun 2022
Reconstructed daily ground-level O3 in China over 2005–2021 for climatological, ecological, and health research
Abstract. Accompanied by the continuous declines of PM2.5, O3 pollution has become increasingly prominent and has been targeted by the Government of China to protect climate, ecosystem, and human health. Although satellite retrievals of column O3 have been operated for decades and nationwide monitoring of ground-level O3 has been offered since 2013 in China, climatological variability of ground-level O3 remains unknown, which impedes understanding of the long-term driver and impacts of O3 pollution in China. Here we develop an eXtreme Gradient Boosting (XGBoost) model integrating high-resolution meteorological data, satellite retrievals of trace gases, etc. to provide reconstructed daily ground-level O3 over 2005–2021 in China. Model validation confirms the robustness of this dataset, with R2 of 0.89 for sample-based cross-validation. The accuracy of the long-term variations has also been confirmed with independent historical observations covering the same period from urban, rural and background sites. Our dataset covers the long time period of 2005–2021 with 0.1°×0.1° gap-free grids, which can facilitate climatological, ecological, and health research. The dataset is freely available at Zenodo (https://zenodo.org/record/6507706#.Yo8hKujP13g; Zhou, 2022).
- Preprint
(2306 KB) - Metadata XML
-
Supplement
(1352 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on essd-2022-187', Anonymous Referee #1, 27 Jul 2022
The authors generated a long-term gap-free surface ozone concentration dataset by taking advantage of machine learning approach on the basis of satellite retrievals and meteorological variables. Both the sample-based cross validation and the independent evaluation with historical observations reveal good data accuracy, making this dataset a valuable data source for ozone pollution and health exposure risk assessment studies. Overall, the paper is well organized. The following comments and suggestions could be considered when revising the manuscript.
- Section 2.1: since the authors aim at predicting daily MDA8 rather than O3 average, the “O3” in the title should be changed to MDA8 to avoid misleading.
- More details for land use cover should be provided. Did the authors use the ratio of each specified type of land cover within a grid, or simply the principal land cover type as a categorical variable? Needs clarify
- Since satellite-based O3 and NO2 retrievals suffer from significant data gaps, how did the author deal with gaps? This is critical to the generation of daily gap-free MDA8 maps.
- Figures 3 and 5: large MDA8 values were underestimated when comparing with observations, what are possible reasons, needs to discuss in the text.
- the number 2 of R-squred should be supperposed, please correct throughout the paper.
Citation: https://doi.org/10.5194/essd-2022-187-RC1 -
CEC1: 'Comment on essd-2022-187', David Carlson, 29 Jul 2022
Authors and editors agree to stop this submission.
Citation: https://doi.org/10.5194/essd-2022-187-CEC1 -
EC1: 'Comment on essd-2022-187', Qingxiang Li, 31 Jul 2022
This manuscript is rejected.
Citation: https://doi.org/10.5194/essd-2022-187-EC1
Status: closed
-
RC1: 'Comment on essd-2022-187', Anonymous Referee #1, 27 Jul 2022
The authors generated a long-term gap-free surface ozone concentration dataset by taking advantage of machine learning approach on the basis of satellite retrievals and meteorological variables. Both the sample-based cross validation and the independent evaluation with historical observations reveal good data accuracy, making this dataset a valuable data source for ozone pollution and health exposure risk assessment studies. Overall, the paper is well organized. The following comments and suggestions could be considered when revising the manuscript.
- Section 2.1: since the authors aim at predicting daily MDA8 rather than O3 average, the “O3” in the title should be changed to MDA8 to avoid misleading.
- More details for land use cover should be provided. Did the authors use the ratio of each specified type of land cover within a grid, or simply the principal land cover type as a categorical variable? Needs clarify
- Since satellite-based O3 and NO2 retrievals suffer from significant data gaps, how did the author deal with gaps? This is critical to the generation of daily gap-free MDA8 maps.
- Figures 3 and 5: large MDA8 values were underestimated when comparing with observations, what are possible reasons, needs to discuss in the text.
- the number 2 of R-squred should be supperposed, please correct throughout the paper.
Citation: https://doi.org/10.5194/essd-2022-187-RC1 -
CEC1: 'Comment on essd-2022-187', David Carlson, 29 Jul 2022
Authors and editors agree to stop this submission.
Citation: https://doi.org/10.5194/essd-2022-187-CEC1 -
EC1: 'Comment on essd-2022-187', Qingxiang Li, 31 Jul 2022
This manuscript is rejected.
Citation: https://doi.org/10.5194/essd-2022-187-EC1
Data sets
Reconstructed daily ground-level MDA8 O3 over 2005-2021 in China Chenghong Zhou; Fan Wang; Yike Guo; Gregory R. Carmichael; Cheng Liu; Yan Wang; Meng Gao https://zenodo.org/record/6507706#.YpMbmHZBxLl
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,106 | 439 | 44 | 1,589 | 107 | 34 | 41 |
- HTML: 1,106
- PDF: 439
- XML: 44
- Total: 1,589
- Supplement: 107
- BibTeX: 34
- EndNote: 41
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1