the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Jul 2026
Quantifying and Predicting Austral Summer Mesopause Height with a Temperature-Based Upwelling Proxy
Abstract. The mesopause, the coldest region of Earth’s atmosphere near 85–100 km, is the boundary between mesosphere and thermosphere, yet its variability remains poorly quantified. Using 24 years (2002–2025) of the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) temperature measurements, we show that the high-latitude austral summer mesopause height is closely linked to the T80 index—a temperature- based proxy at 80 km for the strength of summer polar upwelling. On interannual timescales, the T80 index is almost perfectly correlated with the December mesopause height across 30° S–80° S, with a correlation coefficient (R) of 0.98 and a sensitivity of 0.36 km K−1 at 60° S. On daily timescales, this relationship weakens but remains significant (R = 0.48) at latitudes poleward of 55° S. The temperature field exhibits a vertical dipole response to T80 index: stronger upwelling (lower T80) induces adiabatic cooling below ~90 km and chemical/radiative heating above ~90 km through bottom-up coupling among dynamical, microphysical, and chemical processes. This dipole temperature response shifts the temperature profile downward, thereby lowering the mesopause height. The strong month-to-month persistence of both the T80 index and mesopause temperature provides substantial predictive skill, enabling one-month-ahead forecast of December mesopause height with correlation up to 0.90. Our results indicate that internal dynamics dominate the interannual summer mesopause variability, while signals of 11-year solar cycle and long-term CO2 cooling cannot be reliably isolated from the 24-year observations.
- Preprint
(2601 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-3749', Anonymous Referee #1, 31 Jul 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-3749/egusphere-2026-3749-RC1-supplement.pdfCitation: https://doi.org/
10.5194/egusphere-2026-3749-RC1 -
RC2: 'Comment on egusphere-2026-3749', Anonymous Referee #2, 29 Aug 2026
Review comments on Zhang et al. (2026)
This manuscript shows a strong correlation between mesopause height and the T80 index and argues that the T80 index “controls” mesopause-height variability and can be used to predict interseasonal mesopause height. I agree that the temperature near 80 km is strongly correlated with mesopause height and that both temperature and mesopause height may exhibit substantial persistence on subseasonal to seasonal timescales during summer. Therefore, it may be possible to estimate mesopause height in December/January from the temperature in November/December.
However, I have several concerns regarding the physical interpretation of the results. I am also not yet convinced that the scientific advance demonstrated in the present manuscript is sufficiently clear for publication in ACP. In particular, the manuscript needs to better distinguish correlation from causality and to clarify the scientific value of predicting monthly mean mesopause height.
I suggest that the authors address the following four major comments before publication. I therefore recommend major revision.
Major comments
- Interpretation of lower-thermospheric warming
The authors, following Zhang et al. (2025c), appear to attribute the warming above approximately 90 km mainly to chemical and radiative processes. However, I am not convinced that this is the only, or necessarily the dominant, interpretation. Dynamical processes associated with changes in the residual circulation can also produce mesospheric cooling together with lower-thermospheric warming through adiabatic cooling and heating.
For example, Miyoshi et al. (2015) demonstrated with a whole-atmosphere model that the high-latitude mesosphere cools while the lower thermosphere warms during an SSW event. They showed that this temperature response is associated with changes in gravity-wave drag and the residual circulation. The review by Baldwin et al. (2021) also discusses several studies demonstrating the impact of circulation changes on lower-thermospheric temperature during SSWs (e.g., Liu and Roble, 2002; Funke et al., 2010).
Importantly, the numerical model used by Miyoshi et al. (2015) reproduced the lower-thermospheric warming through dynamical changes, demonstrating that changes in the residual circulation alone can make a substantial contribution to this temperature response.
I therefore wonder whether the chemical and radiative effects proposed in the present manuscript are sufficiently large to dominate over the dynamical/adiabatic temperature response associated with the same circulation changes that produce mesospheric cooling. In particular, the observed correlation between O/O3-related chemistry and lower-thermospheric temperature does not by itself establish that the chemical/radiative processes cause the temperature variations. Both quantities could potentially respond to a common dynamical driver, such as changes in the residual circulation.
A quantitative separation of the dynamical and chemical/radiative contributions would probably require additional modeling and may be beyond the scope of the present manuscript. Nevertheless, I strongly suggest that the authors discuss the alternative dynamical explanation, particularly adiabatic heating associated with changes in the residual circulation, and clarify that the relative contributions of dynamical and chemical/radiative processes remain uncertain.
- Scientific significance and applicability of the prediction
I am not yet convinced that the scientific significance of the main finding is sufficiently clear. Because gravity-wave-driven residual circulation is a major control on temperatures in the mesosphere and lower thermosphere, a strong relationship between mesospheric temperature and mesopause height may be expected from the underlying dynamics.
The authors should therefore clarify what new physical understanding is provided by the relationship between T80 and mesopause height beyond the known dynamical coupling in the MLT region.
I am also uncertain about the practical or scientific value of predicting monthly mean mesopause height one or more months in advance. If this prediction has applications to other research areas—for example, atmospheric coupling, space weather, long-term climate studies, or interpretation of upper-atmospheric observations—the authors should explain these applications explicitly in the Introduction and/or Discussion. This would help establish the broader significance of the proposed prediction method.
- Use of the term “control”
Line 260 and several other parts of the manuscript state that “the T80 index controls the mesopause height variability.” I do not think that the word “controls” is physically appropriate here.
Temperature at 80 km does not itself drive or control mesopause height. Rather, T80 and mesopause height may both respond to common dynamical processes, particularly gravity-wave forcing and the associated residual circulation. Indeed, the authors themselves discuss such a mechanism in Lines 268–270.
I therefore suggest replacing “controls” with a more neutral expression such as “is strongly associated with,” or “serves as an indicator of” mesopause-height variability throughout the manuscript. The distinction between correlation and causality is particularly important for the interpretation of the results presented here.
- Section 4.2: CO2 and solar-cycle effects
I strongly suggest reconsidering, or possibly removing, Section 4.2 in its present form. Numerous previous studies have demonstrated detectable effects of increasing CO2 and solar variability on temperatures and/or the thermal structure of the mesosphere and lower thermosphere using SABER and other observations (e.g., Ramesh et al., 2015; Tang et al., 2016; Zhao et al., 2020, 2021; Mlynczak et al., 2022; Liu et al., 2024).
Therefore, the absence of statistically significant relationships in the present analysis should not be interpreted as evidence that CO2 or solar variability has little influence on mesopause variability. Rather, the lack of statistical significance may result from the methodology used here, particularly because solar-cycle, long-term CO2-related, dynamical, and other sources of variability are not explicitly separated.
If the authors wish to retain this section, I suggest applying an appropriate multiple-regression framework that separates at least the long-term trend and solar-cycle contribution, together with other major sources of variability where possible. Otherwise, I recommend removing this section or substantially revising the interpretation.
Minor comments
Figure 1: If possible, please add standard errors or another measure of uncertainty.
Lines 170–171: The wording appears to imply that T80 drives the upwelling. T80 should instead be regarded as an indicator or proxy of the circulation/upwelling. The upwelling is dynamically driven, primarily through wave forcing and the resulting residual circulation. Please revise the wording accordingly.
Line 215: I am not convinced that the method demonstrated here predicts “seasonal” mesopause-height variability. It appears to predict a monthly or subseasonal mean state based on the persistence/correlation between preceding and subsequent months. Please clarify the terminology and the actual prediction timescale.
Section 3.4: Please explain more clearly the motivation for examining daily variability and how this analysis contributes to the main objectives of the manuscript.
Reference list
Baldwin, M. P., et al. (2021): Sudden Stratospheric Warmings, Reviews of Geophysics, 59, e2020RG000708. DOI: 10.1029/2020RG000708. This is a strong review citation for SSW-driven changes extending into the thermosphere; it explicitly summarizes lower-thermospheric warming found by Liu & Roble and Funke et al.Miyoshi, Y., Fujiwara, H., Jin, H., and Shinagawa, H. (2015): Impacts of sudden stratospheric warming on general circulation of the thermosphere, JGR: Space Physics, 120. DOI: 10.1002/2015JA021894. Particularly useful for Major Comment 1 because it explicitly shows high-latitude mesospheric cooling, lower-thermospheric warming, and a reversal/change of circulation around 90–125 km.
Zhao et al. (2020): Long-Term Trends and Solar Responses of the Mesopause Temperatures Observed by SABER During the 2002–2019 Period, JGR: Atmospheres. DOI: 10.1029/2020JD032418. Directly relevant to separating long-term and solar-cycle effects in SABER mesopause temperatures.
Tang, C., et al. (2016): The response of the temperature of cold-point mesopause to solar activity based on SABER data set, JGR: Space Physics, 121, 7245–7255. DOI: 10.1002/2016JA022538. Direct evidence that solar activity effects on mesopause temperature can be detected using SABER.
Ramesh, K., Sridharan, S., and Vijaya Bhaskara Rao, S. (2015): Influence of solar cycle and chemistry on tropical (10°N–15°N) mesopause variabilities, JGR: Space Physics, 120, 4038–4051. DOI: 10.1002/2014JA020930.
Mlynczak, M. G., et al. (2022): Cooling and Contraction of the Mesosphere and Lower Thermosphere From 2002 to 2021, JGR: Atmospheres, 127, e2022JD036767. DOI: 10.1029/2022JD036767. Particularly strong for your argument because it explicitly separates contributions from declining solar activity and increasing CO₂ and quantifies MLT contraction.
Liu, X., Xu, J., Yue, J., Liu, Y., and Andrioli, V. F. (2024): Trends in the high-latitude mesosphere temperature and mesopause revealed by SABER, Atmospheric Chemistry and Physics, 24, 10143–10157. DOI: 10.5194/acp-24-10143-2024. This is especially relevant because it specifically examines high-latitude mesosphere/mesopause trends.
Citation: https://doi.org/10.5194/egusphere-2026-3749-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 185 | 44 | 21 | 250 | 19 | 19 |
- HTML: 185
- PDF: 44
- XML: 21
- Total: 250
- BibTeX: 19
- EndNote: 19
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
