the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Mar 2026
Increase of water vapour above the Swiss Plateau from 1995 to 2025 observed by ground-based microwave radiometry
Abstract. For climate change research, it is important to have multiple independent measurement techniques. The ground-based microwave radiometer at Bern has been operated since 1994 and allows the independent derivation of the linear trend of the integrated water vapour column or integrated water vapour (IWV). According to the Clausius-Clapeyron equation, the water vapour saturation increases with increase of temperature. There is also a water vapour feedback since water vapour is a natural greenhouse gas and amplifies man-made global warming. In Switzerland, climate change is stronger than at many other places in the world. We analyse observations of the tropospheric water radiometer (TROWARA) which monitored IWV above Bern from 1995 to 2025. The relative IWV increase is 5.1 %/decade. Evaluation of coincident IWV data from ERA5 (reanalysis of European Centre for Medium-Range Weather Forecasts) gives a trend of 3.7 %/decade. The ERA5 surface air temperature in Bern increased by 0.47K/decade from 1995 to 2025. Thus, we get 10.9 % more IWV for a 1 K increase in case of TROWARA and 7.8 % more IWV for a 1K increase in case of ERA5. Though the IWV trends of TROWARA and ERA5 slightly differ, both datasets agree in the fact that water vapour above the Swiss Plateau significantly increased by 11 % or 16 % from 1995 to 2025. This strong increase of water vapour certainly has an impact on weather, climate, and hydrology.
- Preprint
(5670 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Review of "Increase of water vapour above the Swiss Plateau from 1995 to 2025 observed by ground-based microwave radiometry" by Hocke et al. ', Anonymous Referee #1, 31 Mar 2026
- AC2: 'Reply on RC1', Klemens Hocke, 06 Apr 2026
-
AC1: 'Comment on egusphere-2026-989', Klemens Hocke, 06 Apr 2026
Dear Reviewer 1,
Thank you for your careful review! We corrected all of the 39 points which you found and the manuscript is now much better.
Please find attached the pdf file with our point-to-point response to your review.
Best regards, Klemens Hocke
-
AC4: 'Reply on AC1', Klemens Hocke, 16 Apr 2026
Please ignore this comment. I wrongly posted the RC1 reply as a comment.
Best regards, Klemens Hocke
Citation: https://doi.org/10.5194/egusphere-2026-989-AC4
-
AC4: 'Reply on AC1', Klemens Hocke, 16 Apr 2026
-
RC2: 'Comment on egusphere-2026-989', Anonymous Referee #2, 14 Apr 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-989/egusphere-2026-989-RC2-supplement.pdf
- AC3: 'Reply on RC2', Klemens Hocke, 16 Apr 2026
-
RC3: 'Comment on egusphere-2026-989', Anonymous Referee #3, 20 Apr 2026
The manuscript presents a long-term time series (1995 to 2025) of integrated water vapour (IWV) derived from the TROWARA microwave radiometer over Bern, Switzerland, and analyses long-term trends and the scaling relationship with surface air temperature. Such long observational records are valuable and important for climate monitoring.
However, while the dataset itself is clearly of interest, the scientific analysis remains limited, and several methodological as well as conceptual issues reduce the robustness and interpretability of the results. In its current form, the manuscript does not provide substantial new insight beyond what is already well established in the literature.
The main findings, namely an increase in IWV over time and an approximate scaling with temperature, are not novel in themselves. The manuscript does not clearly demonstrate what additional physical or methodological insight is gained from the present analysis. At this stage, the study reads more like the presentation of a valuable dataset than a manuscript that advances scientific understanding.
My main concerns are summarised below.
Major concerns
Context and positioning within the literature
The introduction lacks a clear structure and reads largely as a sequence of studies rather than a synthesis that motivates the present work. A clearer framing of the research gap and the specific contribution of this study is needed.
The manuscript discusses the relationship between total column water vapour and surface temperature but does not adequately refer to key studies in this field, such as Trenberth et al. (2005) and O'Gorman and Muller (2010). Proper positioning within the existing literature would help clarify both the motivation for the study and the limitations of the chosen approach.
Trend analysis and statistical robustness
The trend analysis relies on a simple linear fit and does not sufficiently account for:
- autocorrelation in the time series,
- low-frequency variability (e.g. the seasonal cycle),
- large-scale modes of climate variability and teleconnections (e.g. ENSO, NAO),
- rigorous estimation of trend uncertainty.
This introduces the potential for misleading or overly confident results (see, e.g., Wilks, 2011).
Established approaches for trend detection and uncertainty estimation (e.g. Weatherhead et al., 1998; Mieruch et al., 2008; Schröder et al., 2016; Borger et al., 2022; and related work therein) are not considered. Overall, the current methodology raises concerns regarding the robustness and statistical significance of both the reported trends and the derived IWV–temperature scaling.
Closely related to this point, the manuscript does not investigate whether the observed IWV variability is linked to large-scale modes of climate variability. Given the length of the time series, such an analysis would be highly valuable. Assessing possible teleconnections could help to separate long-term trends from natural variability, improve the interpretation of interannual and decadal fluctuations, and strengthen the physical understanding of the observed changes. At present, the absence of such an analysis limits the interpretability of the results.
Interpretation of the IWV–temperature scaling
The manuscript reports a scaling of approximately 10.9 % / K and implicitly compares this result to Clausius–Clapeyron expectations. However, Clausius–Clapeyron scaling applies to saturation vapour pressure, whereas column water vapour is also influenced by atmospheric dynamics, relative humidity, and vertical structure.
Values exceeding about 7 % / K are not unexpected at a local scale, but this requires careful discussion and interpretation. In its present form, the manuscript risks overstating the physical significance of the reported scaling, and the discussion should be revised accordingly.
Technical comments
Height correction
The applied height correction is not well justified. A correction based on a water vapour scale height would be more physically consistent (e.g. Weaver and Ramanathan, 1995). In addition, the cited reference appears to be incorrect, as the original work is by Bock et al. (2005). Moreover, for relative trends, such a correction is generally unnecessary.
Instrument intercomparison
The intercomparison between ERA5, TROWARA, radiosonde, and GNSS is based on monthly values for only one year.
For this purpose, a comparison using daily or hourly collocated data would be more appropriate. This would substantially increase the sample size, allow a more robust assessment of bias, scatter, and correlation, better reveal systematic differences between the observing systems.
Using only 12 monthly means limits the statistical value of the comparison and may obscure relevant discrepancies.
Recommendation
While the dataset is valuable, the manuscript in its current form is not suitable for publication. Substantial revisions are required with respect to methodology, statistical treatment, physical interpretation, and the overall depth of discussion.
I therefore recommend that the manuscript not be accepted in its current form, and encourage resubmission after major restructuring, a more rigorous analysis, and a more substantial scientific discussion.
Literature
Bock, O., Keil, C., Richard, E., Flamant, C. and Bouin, M.-n. (2005), Validation of precipitable water from ECMWF model analyses with GPS and radiosonde data during the MAP SOP. Q.J.R. Meteorol. Soc., 131: 3013-3036. https://doi.org/10.1256/qj.05.27
Borger, C., Beirle, S., and Wagner, T.: Analysis of global trends of total column water vapour from multiple years of OMI observations, Atmos. Chem. Phys., 22, 10603–10621, https://doi.org/10.5194/acp-22-10603-2022, 2022.
Mieruch, S., Noël, S., Bovensmann, H., and Burrows, J. P.: Analysis of global water vapour trends from satellite measurements in the visible spectral range, Atmos. Chem. Phys., 8, 491–504, https://doi.org/10.5194/acp-8-491-2008, 2008.
O'Gorman, P. A., and Muller, C. J.: How closely do changes in surface and column water vapor follow Clausius–Clapeyron scaling?, Environ. Res. Lett., 5, 025207, https://doi.org/10.1088/1748-9326/5/2/025207, 2010.
Schröder, M., Lockhoff, M., Forsythe, J. M., Cronk, H. Q., Van den Haar, T. H., and Bennartz, R.: The GEWEX Water Vapor Assessment: Results from intercomparison, trend, and homogeneity analysis of total column water vapor, J. Appl. Meteorol. Clim., 55, 1633–1649, https://doi.org/10.1175/JAMC-D-15-0304.1, 2016.
Trenberth, K. E., Fasullo, J., and Smith, L.: Trends and variability in column-integrated atmospheric water vapor, Clim. Dyn., 24, 741–758, https://doi.org/10.1007/s00382-005-0017-4, 2005.
Weaver, C. P., and Ramanathan, V.: Deductions from a simple climate model: Factors governing surface temperature and atmospheric thermal structure, J. Geophys. Res., 100(D6), 11585–11591, https://doi.org/10.1029/95JD00770, 1995.
Weatherhead, E. C., Reinsel, G. C., Tiao, G. C., Meng, X.-L., Choi, D., Cheang, W.-K., Keller, T., DeLuisi, J., Wuebbles, D. J., Kerr, J. B., Miller, A. J., Oltmans, S. J., and Frederick, J. E.: Factors affecting the detection of trends: Statistical considerations and applications to environmental data, J. Geophys. Res. Atmos., 103, 17149–17161, https://doi.org/10.1029/98JD00995, 1998.
Wilks, D. S.: Statistical Methods in the Atmospheric Sciences, vol. 100 of International Geophysics, Elsevier Academic Press, Amsterdam, 3rd Edn., 2011.
Citation: https://doi.org/10.5194/egusphere-2026-989-RC3 -
AC5: 'Reply on RC3', Klemens Hocke, 22 Apr 2026
Thank you for your careful review! Please find our point-to-point response in the attached pdf file.
-
RC4: 'Reply on AC5', Anonymous Referee #3, 27 Apr 2026
I thank the authors for their detailed response to my report. However, my main methodological concerns remain insufficiently addressed. In particular, I remain unconvinced that the current statistical treatment is adequate for the conclusions drawn in the manuscript. In addition, the manuscript still provides limited new scientific insight beyond what is already well established in the literature.
Trend analysis
A central issue concerns the assumptions underlying the trend analysis. The authors emphasize that linear regression is mathematically correct and refer to the work of Carl Friedrich Gauss. While this is correct, it does not address the core concern raised in the review. The issue is not the use of linear regression itself, but whether its assumptions are satisfied, in particular the assumption of uncorrelated residuals and the validity of the associated uncertainty estimates.
In this context, I do not find the argument convincing that annual means can be treated as independent because the water vapour residence time is much shorter than one year. Serial autocorrelation in climate time series is governed by large-scale variability and long-term forcing rather than local residence times. The importance of autocorrelation has been recognised since early work by George Udny Yule (1927) and by Gilbert Thomas Walker, in particular in his series on "Correlation in seasonal variations of weather" (1920s) and subsequent work on periodicity (1931). It is standard practice to assess autocorrelation explicitly rather than assume independence. Without such an assessment, the reported confidence intervals and significance levels of the trends remain uncertain.
Temporal aggregation
Closely related to this is the decision to base the analysis solely on annual means. While annual averaging removes the explicit seasonal cycle, it also discards substantial information and limits the statistical analysis. A framework based on monthly data would allow a more appropriate treatment of seasonality, autocorrelation, and variability.
This becomes particularly important in light of the authors’ discussion of a possible bias in the early part of the TROWARA record. In such a situation, it would be natural to consider models that allow for inhomogeneities, for example through step terms or breakpoint detection methods. Established approaches exist in the atmospheric sciences, including step-function approaches (e.g. Mieruch et al., 2008) and homogenisation and breakpoint analyses, such as penalized maximal F tests (e.g. Schröder et al., 2016). These aspects cannot be assessed within the current framework based on annual means.
Teleconnections
The manuscript states that no correlation with ENSO or NAO is found. However, given the use of annual means, this result is of limited significance. Variability associated with such modes typically operates on sub-annual timescales and with time lags of several months. A meaningful assessment would therefore require an analysis based on monthly anomalies. In the present framework, the absence of a detected signal cannot be interpreted as evidence of absence.
Consistency with previous work
The study is presented as an extension of Bernet et al. (2020), where the analysis was performed at finer temporal resolution and included a more detailed treatment of variability. In comparison, the methodological simplification adopted here is not fully justified, especially given that similar issues of variability and potential inhomogeneity are relevant in the present dataset.
Interpretation of the IWV–temperature scaling
The revised discussion appropriately acknowledges possible biases and shows that the ERA5-based estimate is more consistent with Clausius–Clapeyron expectations. However, this also weakens the robustness of the TROWARA-based scaling. If the trend is sensitive to potential inhomogeneities, the interpretation of the derived sensitivity should be further moderated.
Instrument intercomparison
My comment regarding the one-year intercomparison appears to have been misunderstood, as it referred to the analysis shown in Figure 2. The issue is not the equivalence of monthly and annual means, but the limited sample size for assessing bias, scatter, and correlation. Using only 12 monthly means does not allow a robust evaluation. Higher temporal resolution data would be more appropriate. For example, time series of daily or hourly values, together with corresponding scatter plots, would provide a more informative comparison between TROWARA and the reference datasets.
Novelty and overall contribution
More generally, while the dataset itself is highly valuable, the manuscript provides limited new scientific insight. The study essentially extends the time series presented in Bernet et al. (2020) by a relatively short period, while relying on a simplified analysis framework. It is not clearly demonstrated that this extension leads to additional physical or methodological insight beyond the existing literature. In this sense, the manuscript primarily focuses on the presentation of a valuable dataset, while the added scientific insight beyond existing studies remains limited.
Conclusion
In summary, while the revisions improve some aspects of the manuscript, the core methodological concerns and the limited novelty remain. In particular, the assumptions underlying the trend analysis are not demonstrated, the treatment of variability is limited, and the chosen temporal aggregation restricts the interpretability of the results.
I therefore maintain my recommendation that the manuscript is not suitable for publication in its current form and would require substantial further revision, including a more rigorous statistical treatment and a clearer demonstration of added scientific value.
References:
Mieruch, S., Noël, S., Bovensmann, H., and Burrows, J. P.: Analysis of global water vapour trends from satellite measurements in the visible spectral range, Atmos. Chem. Phys., 8, 491–504, https://doi.org/10.5194/acp-8-491-2008, 2008.
Schröder, M., Lockhoff, M., Forsythe, J. M., Cronk, H. Q., Vonder Haar, T. H., and Bennartz, R.: The GEWEX Water Vapor Assessment: Results from intercomparison, trend, and homogeneity analysis of total column water vapor, J. Appl. Meteorol. Clim., 55, 1633–1649, https://doi.org/10.1175/JAMC-D-15-0304.1, 2016.
Walker, G. T.: On periodicity in series of related terms, Proc. R. Soc. A, 131, 518–532, https://doi.org/10.1098/rspa.1931.0069, 1931.
Yule, G. U.: On a method of investigating periodicities in disturbed series, with special reference to Wolfer’s sunspot numbers, Philos. Trans. R. Soc. A, 226, 267–298, https://doi.org/10.1098/rsta.1927.0007, 1927.
Citation: https://doi.org/10.5194/egusphere-2026-989-RC4 - AC6: 'Reply on RC4', Klemens Hocke, 07 May 2026
-
RC4: 'Reply on AC5', Anonymous Referee #3, 27 Apr 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 570 | 168 | 54 | 792 | 28 | 39 |
- HTML: 570
- PDF: 168
- XML: 54
- Total: 792
- BibTeX: 28
- EndNote: 39
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
A unique long record of atmospheric moisture is exploited as a ground truth measurement which is of importance in evaluating climate models and datasets, in particular in verifying the underlying phsyics of water vapour-climate amplifying feedback processes as well as determining the source of precipitation including extremes. The analysis is rigorous and of wider significance to the climate science community and my assessment is that the work merits publication subject to addressing some minor comments listed below.
1) Abstract - some of the introductory material could be removed to focus on the importance, key results and wider implications e.g. the importance of water vapour for climate, why independent measurements are important, the originality of this record, key results and wider implications. Improving the abstract could enhance the impact of this work.
2) L5 - suggest "man-made" --> "human-caused" (though I recognise that men are probably more guilty!). "many places around the world" is a bit vague so could state that "Surface air temperature in Switzerland has increased at about twice the global average rate since 1995."
3) L12 - suggest removing "certainly" since this is not directly demonstrated (e.g. "has implications for")
4) L15 - "nearly twice" would be more precise
5) L17 - a more up to date IPCC report could be cited. In fact the IPCC assesses multiple lines of evidence, not just climate models, to show that human caused emissions have driven warming of climate.
6) L17-20 - the 7%/K rate is applicable for temperatures at low altitudes away from, the poles since the Clausius Clapeyron rate depands upon the absolute temperature such that the rate is larger for cooler temepratures (e.g. Allan 2012 Surv. Geophys. 10.1007/s10712-011-9157-8). An altitude difference between the observing site and the ERA5 grid point could also cause a discrepancy in mean temperature and water vapour though I presume this is quite small (I see this correction is made on L123).
7) Fig.1 - the first and second box on the left are showing the same thing and the feedback does not look correct as the warming is feedbing back to greenhouse gas increases. I suggest: Emissions lead to increased CO2/CH4 --> warming of climate ---> increased water vapour ---> feedback to warmimg of climate
8) L24 - the feedback loop is fully realised once the Earth's radiative balance is attained which will also involve and equilibrium between evaporation and precipitation
9) L29 - Suggest new sentence at "For example"
10) L31 - Multiple aspects of the changing observing system could affect ERA5 including new and changing satellite instruments and their artificial drifts, changes from manual to automated ground-based humidity sensors, etc. There is evidence of an artificial drying over the global land in ERA5, particularly in the early 2000s. Although this is more pronounced in tropical regions, this effect could contribute to the smaller increases in water vapour compared to the measurements in Europe (Wang et al. 2026 Sci. Bull. https://doi.org/10.1016/j.scib.2026.01.045). In any case, changes in the observing system could indeed lead to spurious water vapour changes in ERA5.
11) L36 - See also surface water vapour changes in Switzerland from Philipona et al. (2007) GRL doi: 10.1029/2008GL036350 of around 3%/decade 1981-2005, which leads to a large increase in downward longwave radiation.
12) L42 - see also Wan et al. (2023) HESS doi:10.5194/hess-28-2123-2024
13) L46 - the larger increases in the northern hemisphere seems somewhat consistent with Allan et al. 2022 Fig.2. (changes in the Arctic are largest)
14) L65 - presumably there are other original aspects of the analysis that are introduced? Otherwise this can appear rather incremental.
15) L75 - a line intrododucing how the work aims to advance the knowledge already discussed may improve the flow and the impact.
16) L89 - if only non-rain conditions are observed, will this bias the actual water vapour content, which will tend to be higher in rainfall events? Changes in rainfall frequency, as well as natural shifts in the larger-scacle circulation, can also affect the record, in addition to the overall greenhouse gas induced warming
17) L98/L188 - "an indoor"
18) L102 - what is "on it's random"?
19) L123 - quote magnitude (e.g. 0.05 mm and also in %)
20) L141 - around 5-10%?
21) L147 - the value as a percentage could also be useful
22) L153 - please quantify
23) L155 - can the radiosonde or other independent data offer a clue?
24) L158 - mostly correct is vague
25) L159 - January looked consistent in Fig. 2?
26) Fig.6 - the trends look consistent to within uncertainty
27) L171 - the difference to Bernet et al. 2020 is negligible considering interannual variability
28) L174 - the latest IPCC report suggests an increase in contrasts between wet and dry events in a warming world (Douville et al. 2021 IPCC Chapter 8) which is consistent, though the record is short.
29) L179 - and presumably also surface station humidity which also displays a change from manual to automated sensors
30) L182 - quantify?
31) L184 - I do not understand the link to the radiosode from the microwave radiometer here? Is it meant that the radiosonde's affect ERA5?
32) L188 - could the drop in TROWARA IWV in 2002 that is larger in magnitude than ERA5 relate to the instrument site change?
33) L194 - what is wet random?
34) L196 - the increase in IWV in both makes the result more robust but does not on its own signify that it is caused by temperature changes. Saying both values are "reasonable" is vague.
35) L207 - better to quote as a % per decade as done later so this could be removed
36) L210 - the Clausius Clapeyron equation provides the physical basis for expecting increases but does not show the cause without other lines of evidence
37) L212 - trend studies over Europe?
38) L220 - are there independent coincident measurements of surface specific humidity that could compliment the IWV record?
39) Discussion/Conclusions - decreases in aerosol in the 1990s may also have impacted Europe's climate, directly through their diminishing effect on reflecting sunlight by the atmosphere and cloud (e.g. Schwarz et al. 2020 Nature Geosci. doi: 10.1038/s41561-019-0528-y) but also through altering wind patterns (e.g. Dong et al. 2023 Clim. Dyn. https://doi.org/10.1007/s00382-022-06438-3). To what extend can some of the changes relate to changes in atmospheric circulation that are related to or unrelated to greenhouse gas foreced climate change?