the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Jul 2024
Validation of ACE-FTS version 5.2 ozone data with ozonesonde measurements
Abstract. Two decades of ACE-FTS version 5.2 (v5.2) ozone data (2004–2023) are evaluated with ozonesonde data from across the globe. The biases between the ACE-FTS and ozonesonde measurements are first estimated by analysing coincident data pairs. A second approach is taken for the validation by comparing the ACE-FTS and ozonesonde monthly mean time series, with the former generated by sampling the ACE-FTS data within latitude/longitude boxes (i.e., ±5°/±30°) surrounding the stations and calculating the monthly averages. The biases, correlations, variation patterns and the mean states of the two time series are compared. The biases estimated in this way exhibit more consistent and smoother features than using the coincident pair method. The ACE-FTS and ozonesonde monthly mean time series are highly correlated and exhibit similar variation patterns in the lower stratosphere at all latitudes. The ACE-FTS instrument drifts for each station are assessed in terms of the long-term linear trends relative to ozonesondes, which, although highly stable, may have their own minor changes with time. The ACE-FTS ozone profiles exhibit in general high biases in the stratosphere, increasing with altitude up to ~10 % at around 30 km, and have local maximum differences with ozonesonde profiles at the tropopause heights. The ACE-FTS instrument drifts are generally insignificant overall in the stratosphere with high variation between the stations. Averaging the individual station instrument drifts within several latitude bands results in small insignificant drifts of within ±1 % dec-1 in the northern mid- to high latitudes, and the southern high latitudes, and a small positive insignificant drift of 0–3 % dec-1 in the tropics and southern mid-latitudes with overall uncertainties at 2–3 % dec-1 (2σ level) in the low stratosphere. In the troposphere, the average ACE-FTS instrument drifts vary with altitude and exhibit large drifts between -10 and +10 % dec-1 with uncertainties of 10 % dec-1.
-
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
(4011 KB)
-
Supplement
(4243 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4011 KB) - Metadata XML
-
Supplement
(4243 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Thorough validation paper, maybe a bit detailed and lengthy', Anonymous Referee #1, 25 Jul 2024
The paper reports on the validation of ACE-FTS version 5.2 ozone profile data using ozonesondes. The underlying analysis is thorough and correct.
The main conclusions from the paper are that ACE-FTS ozone profile data agree with sonde data around 20 km altitude, are about 10% higher than sonde data at 30 km. At the extratropical tropopause, ACE-FTS ozone data tend to be 5 to 15% higher than the sondes. Temporal ozone variations over the last 20 years are seen very similar by ACE-FTS and by the sondes. There are no significant drifts between the sonde and ACE-FTS time series. The paper is generally well written, although I find it somewhat lengthy and detailed. Nevertheless I think it is acceptable in its present form.I have only two minor suggestions for changes:
In Figure 10, I am wondering why the same data gaps appear for the ozone sondes and ACE-FTS. This is most notable at Irene where the gap goes from 2008 to 2013. Are really the same months missing in both data sets? I think this is probabably a plotting error and should be fixed.In Figure 15, I find the green area misleading. In all the previous plots, the green area denoted the standard deviation range of the individual station results, whereas error bars denoted the standard error of the mean. Different from that, the green region in Fig. 15 denotes the standard error of the mean drift. For consistency with the previous plots, I suggest to change that and use error bars, instead of the green area.
Citation: https://doi.org/10.5194/egusphere-2024-1916-RC1 -
AC1: 'Reply on RC1/RC2', Kaley Walker, 13 Sep 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1916/egusphere-2024-1916-AC1-supplement.pdf
-
AC1: 'Reply on RC1/RC2', Kaley Walker, 13 Sep 2024
-
RC2: 'Comment on egusphere-2024-1916', Anonymous Referee #2, 02 Aug 2024
This paper addresses the quality of ACE-FTS v5.2 ozone, and instrument drifts by comparing with correlative ozonesondes between upper troposphere/lower stratosphere (UTLS) and mid-stratosphere (~30 km). The overall scientific approaches and technical methods are sound. The research results are also useful to the community. There are, however, some issues/questions that need to be addressed or clarified before publication.
The following are specific comments and questions for this paper
- Line 20. “The ACE-FTS ozone profiles exhibit in general high biases in the stratosphere, increasing with altitude up to ~10% at around 30 km, …”. The ACE-FTS ozone profile does not show systematic high bias in the stratosphere. The positive biases are mainly for altitudes above ~20 km. For altitudes below 20 km and a few kilometers above tropopause, it tends to show negative biases up to ~5 -10% (Figure 7). To make it clear, I suggest modifying the original text to “The ACE-FTS ozone profiles exhibit in general high biases in the stratosphere for altitudes above 20 km, increasing with altitude up to ~10% at around 30 km, …”, or other similar sentences.
- Line 24, “…., and a small positive insignificant drift of 0 - 3 % dec-1 in the tropics and southern mid-latitudes with overall uncertainties at 2 – 3 % dec-1 (2σ level) in the low stratosphere”. The above sentence needs to be clarified. In Figure 15, it shows positive drifts of ~5% dec-1 in the tropic and southern mid-latitude lower stratosphere (i.e. ~16 -19 km in the tropics, ~10-14 km in southern mid-latitude), with uncertainties of ~5-10% dec-1. The results in Figure 15 are different from the above texts in line 24.
- Line 121, “… and concluded that v4.1/4.2 showed slightly larger biases than v3.5/3.6 in the middle to upper stratosphere”. The ACE-FTS v4.1 and v3.6 ozone show the largest differences in the middle stratosphere (~30km), with positive biases of ~3% and 9% for v3.6 and v4.1 ozone, respectively (see Figure 2 in Sheese et al (2022)). The v4.1 ozone is not slightly larger (i.e. 6% larger) than v3.6 in the middle stratosphere. The above sentence needs to be revised or indicates up to ~6% differences between v4.1 and v3.6 ozone in the middle stratosphere.
- Line 162, There is a typo in the Gaussian function, in the denominator square root of (2*pi*s) should be square root of (2*pi) *s (i.e. “s” is outside of the square root). Another question is why s is set to be 1. The s value should correspond to the vertical resolution of ACE-FTS vertical resolution. Since the vertical resolution of ACE-FTS is ~3 – 4 km, the "s" value would be larger than 1 (see equation (2) in Sheese et al, 2017 (https://doi.org/10.1016/j.jqsrt.2016.06.026)). Using smaller s value would result in less smoothing ozonesond profiles (e.g. not comparable to ACE-FTS vertical resolution) and create larger differences (or artifacts) when comparing ACE-FTS ozone against ozonesondes in the regions with strong vertical gradient.
- Line 199, “Laeng et al. (2021)”. The published date is 2022 (i.e. Laeng et al. (2022)). Please double check the doi number of this paper (reference section in line 822). The doi number is “https://doi.org/10.5194/amt-15-2407-2022”
- Line 454, “This is an interesting feature showing the differences appear to depend on the atmospheric temperature profiles”. Ozone in the UTLS is mainly controlled by dynamics. The larger differences between ACE-FTS and ozonesondes in the UTLS (or tropopause height) could result from larger dynamic variability and sampling biases such as using more relaxed coincident criteria. This can be seen in Figure 7, where differences between ACE-FTS and ozonesondes in the UTLS (and tropopause) are smaller than those in Figure 9 due to stringent coincident criteria. The ACE-FTS ozone retrieval in the UTLS regions could also be affected by aerosol or cloud. The above sentence (line 454) needs more clarification.
- Lines 601, “the aggregated mean instrument drifts show small insignificant drifts within ±1% dec-1”. The altitude regions for ±1% dec-1 drift needs to be specified. For example a few kilometer above the tropopause.
Citation: https://doi.org/10.5194/egusphere-2024-1916-RC2 -
AC1: 'Reply on RC1/RC2', Kaley Walker, 13 Sep 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1916/egusphere-2024-1916-AC1-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Thorough validation paper, maybe a bit detailed and lengthy', Anonymous Referee #1, 25 Jul 2024
The paper reports on the validation of ACE-FTS version 5.2 ozone profile data using ozonesondes. The underlying analysis is thorough and correct.
The main conclusions from the paper are that ACE-FTS ozone profile data agree with sonde data around 20 km altitude, are about 10% higher than sonde data at 30 km. At the extratropical tropopause, ACE-FTS ozone data tend to be 5 to 15% higher than the sondes. Temporal ozone variations over the last 20 years are seen very similar by ACE-FTS and by the sondes. There are no significant drifts between the sonde and ACE-FTS time series. The paper is generally well written, although I find it somewhat lengthy and detailed. Nevertheless I think it is acceptable in its present form.I have only two minor suggestions for changes:
In Figure 10, I am wondering why the same data gaps appear for the ozone sondes and ACE-FTS. This is most notable at Irene where the gap goes from 2008 to 2013. Are really the same months missing in both data sets? I think this is probabably a plotting error and should be fixed.In Figure 15, I find the green area misleading. In all the previous plots, the green area denoted the standard deviation range of the individual station results, whereas error bars denoted the standard error of the mean. Different from that, the green region in Fig. 15 denotes the standard error of the mean drift. For consistency with the previous plots, I suggest to change that and use error bars, instead of the green area.
Citation: https://doi.org/10.5194/egusphere-2024-1916-RC1 -
AC1: 'Reply on RC1/RC2', Kaley Walker, 13 Sep 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1916/egusphere-2024-1916-AC1-supplement.pdf
-
AC1: 'Reply on RC1/RC2', Kaley Walker, 13 Sep 2024
-
RC2: 'Comment on egusphere-2024-1916', Anonymous Referee #2, 02 Aug 2024
This paper addresses the quality of ACE-FTS v5.2 ozone, and instrument drifts by comparing with correlative ozonesondes between upper troposphere/lower stratosphere (UTLS) and mid-stratosphere (~30 km). The overall scientific approaches and technical methods are sound. The research results are also useful to the community. There are, however, some issues/questions that need to be addressed or clarified before publication.
The following are specific comments and questions for this paper
- Line 20. “The ACE-FTS ozone profiles exhibit in general high biases in the stratosphere, increasing with altitude up to ~10% at around 30 km, …”. The ACE-FTS ozone profile does not show systematic high bias in the stratosphere. The positive biases are mainly for altitudes above ~20 km. For altitudes below 20 km and a few kilometers above tropopause, it tends to show negative biases up to ~5 -10% (Figure 7). To make it clear, I suggest modifying the original text to “The ACE-FTS ozone profiles exhibit in general high biases in the stratosphere for altitudes above 20 km, increasing with altitude up to ~10% at around 30 km, …”, or other similar sentences.
- Line 24, “…., and a small positive insignificant drift of 0 - 3 % dec-1 in the tropics and southern mid-latitudes with overall uncertainties at 2 – 3 % dec-1 (2σ level) in the low stratosphere”. The above sentence needs to be clarified. In Figure 15, it shows positive drifts of ~5% dec-1 in the tropic and southern mid-latitude lower stratosphere (i.e. ~16 -19 km in the tropics, ~10-14 km in southern mid-latitude), with uncertainties of ~5-10% dec-1. The results in Figure 15 are different from the above texts in line 24.
- Line 121, “… and concluded that v4.1/4.2 showed slightly larger biases than v3.5/3.6 in the middle to upper stratosphere”. The ACE-FTS v4.1 and v3.6 ozone show the largest differences in the middle stratosphere (~30km), with positive biases of ~3% and 9% for v3.6 and v4.1 ozone, respectively (see Figure 2 in Sheese et al (2022)). The v4.1 ozone is not slightly larger (i.e. 6% larger) than v3.6 in the middle stratosphere. The above sentence needs to be revised or indicates up to ~6% differences between v4.1 and v3.6 ozone in the middle stratosphere.
- Line 162, There is a typo in the Gaussian function, in the denominator square root of (2*pi*s) should be square root of (2*pi) *s (i.e. “s” is outside of the square root). Another question is why s is set to be 1. The s value should correspond to the vertical resolution of ACE-FTS vertical resolution. Since the vertical resolution of ACE-FTS is ~3 – 4 km, the "s" value would be larger than 1 (see equation (2) in Sheese et al, 2017 (https://doi.org/10.1016/j.jqsrt.2016.06.026)). Using smaller s value would result in less smoothing ozonesond profiles (e.g. not comparable to ACE-FTS vertical resolution) and create larger differences (or artifacts) when comparing ACE-FTS ozone against ozonesondes in the regions with strong vertical gradient.
- Line 199, “Laeng et al. (2021)”. The published date is 2022 (i.e. Laeng et al. (2022)). Please double check the doi number of this paper (reference section in line 822). The doi number is “https://doi.org/10.5194/amt-15-2407-2022”
- Line 454, “This is an interesting feature showing the differences appear to depend on the atmospheric temperature profiles”. Ozone in the UTLS is mainly controlled by dynamics. The larger differences between ACE-FTS and ozonesondes in the UTLS (or tropopause height) could result from larger dynamic variability and sampling biases such as using more relaxed coincident criteria. This can be seen in Figure 7, where differences between ACE-FTS and ozonesondes in the UTLS (and tropopause) are smaller than those in Figure 9 due to stringent coincident criteria. The ACE-FTS ozone retrieval in the UTLS regions could also be affected by aerosol or cloud. The above sentence (line 454) needs more clarification.
- Lines 601, “the aggregated mean instrument drifts show small insignificant drifts within ±1% dec-1”. The altitude regions for ±1% dec-1 drift needs to be specified. For example a few kilometer above the tropopause.
Citation: https://doi.org/10.5194/egusphere-2024-1916-RC2 -
AC1: 'Reply on RC1/RC2', Kaley Walker, 13 Sep 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1916/egusphere-2024-1916-AC1-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 268 | 101 | 20 | 389 | 52 | 9 | 11 |
- HTML: 268
- PDF: 101
- XML: 20
- Total: 389
- Supplement: 52
- BibTeX: 9
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 190 | 47 |
| China | 2 | 34 | 8 |
| Germany | 3 | 30 | 7 |
| Canada | 4 | 27 | 6 |
| France | 5 | 15 | 3 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 190
Jiansheng Zou
Patrick E. Sheese
Chris D. Boone
Ryan M. Stauffer
Anne M. Thompson
David W. Tarasick
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4011 KB) - Metadata XML
-
Supplement
(4243 KB) - BibTeX
- EndNote
- Final revised paper