the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Jun 2023
Investigation of spatial and temporal variability in lower tropospheric ozone from RAL Space UV-Vis satellite products
Abstract. Ozone is a potent air pollutant in the lower troposphere and an important short-lived climate forcer (SLCF) in the upper troposphere. Studies using satellite data to investigate spatiotemporal variability of troposphere ozone (TO3) have predominantly focussed on the tropospheric column metric. This is the first study to investigate long-term spatiotemporal variability in lower tropospheric column ozone (LTCO3, surface-450 hPa sub-column) by merging multiple European Space Agency – Climate Change Initiative (ESA-CCI) products produced by the Rutherford Appleton Laboratory (RAL) Space. We find that in the LTCO3, the degrees of freedom of signal (DOFS) from these products varies with latitude range and season and is up to 0.65, indicating that the retrievals contain useful information on lower TO3. The spatial and seasonal variation of the RAL Space products are in good agreement with each other but there are systematic offsets of up to 3.0–5.0 DU between them. Comparison with ozonesondes shows that the Global Ozone Monitoring Experiment (GOME-1, 1996–2003), the SCanning Imaging Absorption spectroMeter for Atmospheric CartograpHY (SCIAMACHY, 2003–2010) and the Ozone Monitoring Instrument (OMI, 2005–2017) have stable LTCO3 records over their respective periods, which can be merged together. While GOME-2 (2008–2018) shows substantial drift in its bias with respect to ozonesondes. We have therefore constructed a robust merged dataset of LTCO3 from GOME-1, SCIAMACHY and OMI between 1996 and 2017. Comparing the LTCO3 differences between the 1996–2000 and 2013–2017 5-year averages, we find significant positive increases (3.0–5.0 DU) in the tropics/sub-tropics, while in the northern mid-latitudes, we find small scale differences in LTCO3. Therefore, we conclude that there has been a substantial increase in tropical/sub-tropical LTCO3 during the satellite-era.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(2474 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2474 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1172', Anonymous Referee #1, 17 Jul 2023
This manuscript provides a global assessment of the lower-tropospheric ozone (LTCO3) retrievals from four satellite instruments, mainly focusing on vertical sensitivities and long-term stabilities. It then goes on to create a harmonized long-term (1996-2017) time series of LTCO3, from which inter-annual variations in global LTCO3 are evaluated. As the first effort to investigate this issue, this paper is well-designed and does not exhibit obvious flaws. The topic is significant considering the stronger relevance of LTCO3 with anthropogenic activities and ozone radiative forcing than tropospheric ozone. Certain details are lack in the present form of this manuscript, which I encourage the authors to elaborate more during the revision. I support the publication of this paper on ACP if the following comments can be addressed.
1) One paragraph of discussing why LTCO3 warrants investigation (relative to tropospheric ozone) should be added to the Introduction Section. For example, LTCO3 (if detectable) can be more relevant with surface ozone pollution and anthropogenic ozone radiative forcing.
2) The satellite sensitivity to LTCO3 (DOFs) varies spatially, especially following the distribution of LTCO3 abundances. So hemispheric averaged DOFs in Table 2 cannot provide a comprehensive enough insight of this sensitivity. I suggest to additionally include a map or histogram of DOFs (associated with Figure 3) of each instrument. Another issue to further investigate is: should we also consider DOFs at each location when evaluating the significance of derived trends in LTCO3? Are DOFs too small in some "green polygons" in Figures 6 and 7?
3) Figure 2: in many cases, LTCO3 from Ozone Sondes and a priori are close to each other, while applying the AKs (satellite and sonde-AK) modifies them into different (usually opposite) directions (e.g., in the OMI global case). Should provide an explanation on this issue.
4) Figure 3: The increasing trends in LTCO3 (as discussed later in the paper) might be responsible for some differences between instruments averaged for varying periods. This fact and the expected impacts on the inter-comparison here should be discussed.
5) Line 240-243: More details should be provided to justify the choice to only drop 2003 in the merging for GOME-1, since 2002 seems also to potentially introduce artificial trends.
6) Figure 1 is unnecessary (at least in the main text) considering more clear and relevant information is already available in Table 2.
Citation: https://doi.org/10.5194/egusphere-2023-1172-RC1 -
AC1: 'Reply on RC1', Richard Pope, 23 Sep 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1172/egusphere-2023-1172-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Richard Pope, 23 Sep 2023
-
RC2: 'Comment on egusphere-2023-1172', Anonymous Referee #2, 26 Jul 2023
In this work, Pope et al. provide an “Investigation of spatial and temporal variability in lower tropospheric ozone from RAL Space UV-Vis satellite products” that is suitable for publication in the TOAR-II special issue. Given their extensive expertise with the data, however, the authors sometimes seem to be too brief/quick in passing their message and conclusions to a possibly less-informed audience. Therefore, the following clarifications/improvements are suggested:
Lines 15, 67, 135, 137, 178-180: In contrast with common practice based on the work of Rodgers (2000), the terms (vertical) sensitivity and degrees of freedom are both used without distinction. Whereas the former rather refers to the integrated weighting function (averaging kernel row sum) as a measure of the fraction of the info that comes from the retrieval, the latter refers to the kernels’ diagonal elements only as an indication of the number of independent layers that is retrieved. Ideally, the authors would address both, though clearly distinguished, in their discussion. Other retrieval performance measures like the effective vertical resolution of the LTCO3 layer and the retrieval weight offset would be most helpful in the discussion, but might not be mandatory.
Relatedly, for Figure 1: Is this for all retrieved satellite pixels within 90° latitude bands? For clarity, possibly provide the effective latitude range (and number of averaged observations) for each month given the lack of observations during polar night? Additionally, it would be helpful to specify whether the lowest red curve indeed corresponds to the LTCO3 retrieval, and to what extent this kernel agrees with or differs from ideal AK behavior (for results discussion). On the other hand, for a full and proper interpretation of the results in Figs. 6 and 7, it would be appropriate to have a view on the DFS (or integrated sensitivity) on the same spatial resolution (latitude-longitude boxes) as Figs. 6 and 7), or at least the spatiotemporal resolution of Fig. 4. The shifts in spatiotemporal resolutions between all consecutive figures (except for 6 and 7) seems to hamper a fully correlative interpretation.
Lines 53-56 and 62-64: The inconsistencies between the instruments as described in Gaudel et al. (2018) also originate from different prior information choices, and different top-level definitions for vertical integration and ozone burden calculations.
Lines 72-73: “In this study, we explore the spatiotemporal variability of lower tropospheric column ozone (LTCO3, surface to 450 hPa)…” Please explain the choice of this fixed pressure level, in terms of the fixed (?) retrieval grid and vertical sensitivity (stratospheric info permeating the tropospheric retrieval output, also see first comment).
Section 2.1 and Table 1: What is now in a footnote sounds too important not to be included in the main text. The main differences between the product versions are expected to be explained, possibly summarized as different parameter settings in Table 1 (instead of the product link for example, which is the same for all and hence superfluous as a separate column). The conclusions on the LTCO3 differences between the products are expected to relate to this then.
Lines 117-118: Are all co-locations considered in the analysis, or only the closest in space or time? Also possibly provide more information or reference(s) on “representation errors”. Something like “spatiotemporal sampling difference errors” may sound clearer.
Line 128: “sonde sub-column profile (DU) on the satellite pressure grid” The sonde data are typically not provided as sub-columns, and certainly not on the satellite pressure grid. Please provide a reference and/or some clarification for the conversion of the initial sonde data to this derived product.
Section 3.2: Do these results agree with the ozonesonde comparisons then? Please briefly discuss.
Lines 265-266: “Based on the difference between 2002 and 2005, a global scaling is applied in 2003 and 2004 for the SCIAMACHY spatial fields.” Does this mean that the scaling changes linearly in time, with yearly resolution?
Line 296: “Overall, these anomalies suggest there has been limited change in LTCO3, between 1996 and 2000” This cannot be seen from Fig. 6. Rather immediately refer to Fig. 5, as is done on the next line. Together with that, it would seem appropriate to also discuss the decrease in LTCO3 towards the end of the time series, from about 2013-2014 onwards.
Lines 320-323: The study of Wespes et al. (2018) comes with some caveats that should be mentioned for proper interpretation. More importantly, this does not straightforwardly allow concluding that all “studies using IR products tend to show significant negative trends globally, while studies using UV-Vis products show significant increasing trends in the tropics/sub-tropics.” See also comment on lines 53-56 and 62-64, which could then be specified with “other improvements” on line 339.
Technical corrections:
Lines 29-30: Providing a maximum does not say anything about the retrievals in general. This statement is too optimistic.
Line 35: Indicate GOME-2 from which platform.
Lines 40-41: Add whether this conclusion agrees with or contrasts common literature?
Line 57: “one” to “on one”
Line 60: remove “that there is”
Line 70: “so might the processes controlling variability in retrieved TO3” is not very clear. Please extend/rephrase.
Line 78: rephrase “presents are results”
Line 92: remove “ozone”
Line 112: The IASI-FORLI products are not in Table 1.
Line 176: rephrase “is remains”
Line 207: rather “corrected for” than “tolerated” ?
Figure 4: Mention latitude ranges for different drift results in caption.
Figure 5: Is there a reason for not mentioning SCIAMACHY in the caption for the 2003-2005 period?
Line 281: “similar”
Line 287: “Pirovano et”
Line 289: rephrase “There are a…”
Citation: https://doi.org/10.5194/egusphere-2023-1172-RC2 -
AC2: 'Reply on RC2', Richard Pope, 23 Sep 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1172/egusphere-2023-1172-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Richard Pope, 23 Sep 2023
-
CC1: 'Comment on egusphere-2023-1172', Owen Cooper, 05 Aug 2023
Comments by O. R. Cooper can be found in the attached pdf.
-
AC3: 'Reply on CC1', Richard Pope, 23 Sep 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1172/egusphere-2023-1172-AC3-supplement.pdf
-
AC3: 'Reply on CC1', Richard Pope, 23 Sep 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1172', Anonymous Referee #1, 17 Jul 2023
This manuscript provides a global assessment of the lower-tropospheric ozone (LTCO3) retrievals from four satellite instruments, mainly focusing on vertical sensitivities and long-term stabilities. It then goes on to create a harmonized long-term (1996-2017) time series of LTCO3, from which inter-annual variations in global LTCO3 are evaluated. As the first effort to investigate this issue, this paper is well-designed and does not exhibit obvious flaws. The topic is significant considering the stronger relevance of LTCO3 with anthropogenic activities and ozone radiative forcing than tropospheric ozone. Certain details are lack in the present form of this manuscript, which I encourage the authors to elaborate more during the revision. I support the publication of this paper on ACP if the following comments can be addressed.
1) One paragraph of discussing why LTCO3 warrants investigation (relative to tropospheric ozone) should be added to the Introduction Section. For example, LTCO3 (if detectable) can be more relevant with surface ozone pollution and anthropogenic ozone radiative forcing.
2) The satellite sensitivity to LTCO3 (DOFs) varies spatially, especially following the distribution of LTCO3 abundances. So hemispheric averaged DOFs in Table 2 cannot provide a comprehensive enough insight of this sensitivity. I suggest to additionally include a map or histogram of DOFs (associated with Figure 3) of each instrument. Another issue to further investigate is: should we also consider DOFs at each location when evaluating the significance of derived trends in LTCO3? Are DOFs too small in some "green polygons" in Figures 6 and 7?
3) Figure 2: in many cases, LTCO3 from Ozone Sondes and a priori are close to each other, while applying the AKs (satellite and sonde-AK) modifies them into different (usually opposite) directions (e.g., in the OMI global case). Should provide an explanation on this issue.
4) Figure 3: The increasing trends in LTCO3 (as discussed later in the paper) might be responsible for some differences between instruments averaged for varying periods. This fact and the expected impacts on the inter-comparison here should be discussed.
5) Line 240-243: More details should be provided to justify the choice to only drop 2003 in the merging for GOME-1, since 2002 seems also to potentially introduce artificial trends.
6) Figure 1 is unnecessary (at least in the main text) considering more clear and relevant information is already available in Table 2.
Citation: https://doi.org/10.5194/egusphere-2023-1172-RC1 -
AC1: 'Reply on RC1', Richard Pope, 23 Sep 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1172/egusphere-2023-1172-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Richard Pope, 23 Sep 2023
-
RC2: 'Comment on egusphere-2023-1172', Anonymous Referee #2, 26 Jul 2023
In this work, Pope et al. provide an “Investigation of spatial and temporal variability in lower tropospheric ozone from RAL Space UV-Vis satellite products” that is suitable for publication in the TOAR-II special issue. Given their extensive expertise with the data, however, the authors sometimes seem to be too brief/quick in passing their message and conclusions to a possibly less-informed audience. Therefore, the following clarifications/improvements are suggested:
Lines 15, 67, 135, 137, 178-180: In contrast with common practice based on the work of Rodgers (2000), the terms (vertical) sensitivity and degrees of freedom are both used without distinction. Whereas the former rather refers to the integrated weighting function (averaging kernel row sum) as a measure of the fraction of the info that comes from the retrieval, the latter refers to the kernels’ diagonal elements only as an indication of the number of independent layers that is retrieved. Ideally, the authors would address both, though clearly distinguished, in their discussion. Other retrieval performance measures like the effective vertical resolution of the LTCO3 layer and the retrieval weight offset would be most helpful in the discussion, but might not be mandatory.
Relatedly, for Figure 1: Is this for all retrieved satellite pixels within 90° latitude bands? For clarity, possibly provide the effective latitude range (and number of averaged observations) for each month given the lack of observations during polar night? Additionally, it would be helpful to specify whether the lowest red curve indeed corresponds to the LTCO3 retrieval, and to what extent this kernel agrees with or differs from ideal AK behavior (for results discussion). On the other hand, for a full and proper interpretation of the results in Figs. 6 and 7, it would be appropriate to have a view on the DFS (or integrated sensitivity) on the same spatial resolution (latitude-longitude boxes) as Figs. 6 and 7), or at least the spatiotemporal resolution of Fig. 4. The shifts in spatiotemporal resolutions between all consecutive figures (except for 6 and 7) seems to hamper a fully correlative interpretation.
Lines 53-56 and 62-64: The inconsistencies between the instruments as described in Gaudel et al. (2018) also originate from different prior information choices, and different top-level definitions for vertical integration and ozone burden calculations.
Lines 72-73: “In this study, we explore the spatiotemporal variability of lower tropospheric column ozone (LTCO3, surface to 450 hPa)…” Please explain the choice of this fixed pressure level, in terms of the fixed (?) retrieval grid and vertical sensitivity (stratospheric info permeating the tropospheric retrieval output, also see first comment).
Section 2.1 and Table 1: What is now in a footnote sounds too important not to be included in the main text. The main differences between the product versions are expected to be explained, possibly summarized as different parameter settings in Table 1 (instead of the product link for example, which is the same for all and hence superfluous as a separate column). The conclusions on the LTCO3 differences between the products are expected to relate to this then.
Lines 117-118: Are all co-locations considered in the analysis, or only the closest in space or time? Also possibly provide more information or reference(s) on “representation errors”. Something like “spatiotemporal sampling difference errors” may sound clearer.
Line 128: “sonde sub-column profile (DU) on the satellite pressure grid” The sonde data are typically not provided as sub-columns, and certainly not on the satellite pressure grid. Please provide a reference and/or some clarification for the conversion of the initial sonde data to this derived product.
Section 3.2: Do these results agree with the ozonesonde comparisons then? Please briefly discuss.
Lines 265-266: “Based on the difference between 2002 and 2005, a global scaling is applied in 2003 and 2004 for the SCIAMACHY spatial fields.” Does this mean that the scaling changes linearly in time, with yearly resolution?
Line 296: “Overall, these anomalies suggest there has been limited change in LTCO3, between 1996 and 2000” This cannot be seen from Fig. 6. Rather immediately refer to Fig. 5, as is done on the next line. Together with that, it would seem appropriate to also discuss the decrease in LTCO3 towards the end of the time series, from about 2013-2014 onwards.
Lines 320-323: The study of Wespes et al. (2018) comes with some caveats that should be mentioned for proper interpretation. More importantly, this does not straightforwardly allow concluding that all “studies using IR products tend to show significant negative trends globally, while studies using UV-Vis products show significant increasing trends in the tropics/sub-tropics.” See also comment on lines 53-56 and 62-64, which could then be specified with “other improvements” on line 339.
Technical corrections:
Lines 29-30: Providing a maximum does not say anything about the retrievals in general. This statement is too optimistic.
Line 35: Indicate GOME-2 from which platform.
Lines 40-41: Add whether this conclusion agrees with or contrasts common literature?
Line 57: “one” to “on one”
Line 60: remove “that there is”
Line 70: “so might the processes controlling variability in retrieved TO3” is not very clear. Please extend/rephrase.
Line 78: rephrase “presents are results”
Line 92: remove “ozone”
Line 112: The IASI-FORLI products are not in Table 1.
Line 176: rephrase “is remains”
Line 207: rather “corrected for” than “tolerated” ?
Figure 4: Mention latitude ranges for different drift results in caption.
Figure 5: Is there a reason for not mentioning SCIAMACHY in the caption for the 2003-2005 period?
Line 281: “similar”
Line 287: “Pirovano et”
Line 289: rephrase “There are a…”
Citation: https://doi.org/10.5194/egusphere-2023-1172-RC2 -
AC2: 'Reply on RC2', Richard Pope, 23 Sep 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1172/egusphere-2023-1172-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Richard Pope, 23 Sep 2023
-
CC1: 'Comment on egusphere-2023-1172', Owen Cooper, 05 Aug 2023
Comments by O. R. Cooper can be found in the attached pdf.
-
AC3: 'Reply on CC1', Richard Pope, 23 Sep 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1172/egusphere-2023-1172-AC3-supplement.pdf
-
AC3: 'Reply on CC1', Richard Pope, 23 Sep 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 386 | 140 | 22 | 548 | 8 | 8 |
- HTML: 386
- PDF: 140
- XML: 22
- Total: 548
- BibTeX: 8
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
Richard J. Pope
Brian J. Kerridge
Richard Siddans
Barry G. Latter
Martyn P. Chipperfield
Wuhu Feng
Matilda A. Pimlott
Sandip S. Dhomse
Christian Retscher
Richard Rigby
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2474 KB) - Metadata XML