Atmospheric blocking and climate extremes in Germany in present and future climate

Lohmann, Richard; Purr, Christopher; Ahrens, Bodo

doi:10.5194/egusphere-2025-3670

Preprints

https://doi.org/10.5194/egusphere-2025-3670

Preprints

19 Sep 2025

| 19 Sep 2025

Status: this preprint is open for discussion and under review for Natural Hazards and Earth System Sciences (NHESS).

Atmospheric blocking and climate extremes in Germany in present and future climate

Richard Lohmann, Christopher Purr, and Bodo Ahrens

Abstract. Atmospheric blocking is linked to extreme weather and climate events like heatwaves, heavy rainfall, and calm weather. The statistical relationship between blocking and extreme events in Germany is quantified in atmospheric reanalyses ERA5, ERA-20C, and 20CRv3, and in historical and future CMIP6 climate simulations. This targets the reliability assessment of climate projections regarding extreme events in the 21st century. The analysis of the atmospheric reanalyses in the period 1961–2010 indicates that days with blocking see heatwaves 10–11 times and heavy precipitation events or calms 1.5 to 3 times more often than days without blocking. These empirical relationships are also seen in historical CMIP6 simulations for the large-scale phenomena heatwaves and calms, but not for heavy precipitation events (with odds only 1–1.5 times higher given a day with blocking than without). In the simulated future climate, the relationship of blocking with the three extreme event types changes only moderately. Inconsistent blocking trends in the projections, particularly in summer, obstruct the robust projection of extreme events in Germany despite the stable relationship between blocking and heatwaves and calms in most of the CMIP6 simulations. Furthermore, the results confirm the need for better representation of precipitation extremes in climate models.

Received: 29 Jul 2025 – Discussion started: 19 Sep 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 6978 KB)

Supplement (76 KB)

Download & links

Richard Lohmann, Christopher Purr, and Bodo Ahrens

Status: open (until 08 Mar 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2025-3670', Anonymous Referee #1, 15 Dec 2025 reply
The study uses atmospheric reanalyses (ERA5, ERA-20C, 20CRv3) and coupled model simulations (CMIP6) with historical and future forcing (SSP5-8.5) to investigate the link between atmospheric blocking and the occurrence of different types of extreme events (heatwaves, heavy precipitation and calms) in Germany. The key finding emerging from the reanalyses is that weather extremes, and particularly heatwaves, occur more frequently when associated with blocking than in the absence of blocking. This finding is also found in CMIP6 simulations but with considerable discrepancies and spread.

The main strength of the study is the large number of datasets that are analysed (3 reanalyses, 6 CMIP6 models with historical and future simulations), which gives robustness to the methodology. However, the novel contribution of the study with respect to existing literature is not clearly indicated. Some methodological assumptions are insufficiently justified, and their uncertainties should at least be discussed. Moreover, the presentation of results in the manuscript should be reconsidered – the authors use multiple data sources, but their figures practically only show ERA5, when it would be far more interesting to show a comparison between the different reanalyses and CMIP6 models for the metrics being investigated.

While there is potential for interesting results, I believe important work is needed to bring the manuscript to standards worthy of publication, particularly in the presentation of results and writing.

Comments

The motivation and questions of the study need clearer and stronger emphasis. The authors mention that previous studies have performed similar analyses on individual CMIP models, hence presumably the novelty lies in extending the analysis to more models and reanalyses. However, the link between previous research, the proposed scientific questions, and the original contribution of the study is unclearly conveyed.

In their definition of heatwaves, the authors perform their analysis on a single grid point (51N, 10E). They point out that this point represents Germany, but this choice seems aleatoric and the authors provide no convincing explanation for this. The same applies to the analysis on heavy precipitation, where the authors consider two grid points in western and eastern Germany. The uncertainty of choosing single grid points (and why those specifically) should at least be discussed.

I believe the choice of figures should be reconsidered. Other than Fig. 5, all figures show ERA5 only. This study would be considerably improved if all reanalyses were shown in the Figures, and their differences were also discussed. The text frequently refers to the mean of the 3 reanalyses, but this is not shown in the figures. Similarly, CMIP6 models are used but not shown in figures (except F5). A much better figure format for F1-4 would be showing the reanalyses and the CMIP6 MMM with model uncertainty/spread, perhaps including individual models in the SM. This would also help reduce the lengthy tables and make the text easier to follow in the Results section.

The authors begin Results subsections (e.g., 3.1.2, 3.2, 3.3) by stating that they select regional domains based on the patterns emerging from Figures 2, 3, 4, respectively, which only show ERA5. This assumption should at least be justified. Determining the domains for the analysis based on ERA5 only, particularly when the spatial patterns of the other data products are never shown in the manuscript, seems hardly robust.

The writing should be considerably improved, particularly in the Results and Discussion section. This includes using more specific and technical language; improving the logical flow of sentences; and ensuring that all statements are clearly supported either by published evidence or by the paper’s results. Large parts of the Results section can be omitted or summarized more coherently. It would be beneficial to guide the reader more in the interpretation of new results and their implications in a broader context.

Perhaps a more minor comment but including calms in the category of “extreme events” seems a little counterintuitive. The authors place the importance of calms in the context of impacts on energy production, but the distinction from other extremes should at least be better framed. Since the motivation (i.e., the impact on the energy sector) is not sufficiently discussed in the introduction, reader is left questioning whether the analysis on calms is relevant to the paper’s objectives at all.

The Results, Discussion and Conclusions section could all be shortened. When reviewing the text, it should be kept in mind that the frequent use of acronyms and repetitive sentences make the text quite difficult to follow.

Reply
Citation: https://doi.org/10.5194/egusphere-2025-3670-RC1
RC2:
'Comment on egusphere-2025-3670', Anonymous Referee #2, 01 Mar 2026 reply
Overview
Lohman et al. aim to investigate the statistical relationship between blocking and extreme events in Germany, using reanalysis datasets and CMIP6 model outputs. They quantify the odds ratios of three extremes (heat wave, heavy precipitation, and calm) regarding atmospheric blocking. They also argue that the relationship between blocking and the three extreme types changes only moderately.
While the topic itself is societally relevant and falls within the journal's general scope, I find that the study suffers from potentially quite strong sensitivity of the results to the chosen grid point and a lack of mechanistic interpretation. Although the journal (Natural Hazards and Earth System Sciences) publishes studies on assessments of natural hazards, the present study primarily presents statistics that are likely highly dependent on the selection of grid points, without sufficient dynamical and mechanistic discussions.
Therefore, I am unable to recommend this manuscript for publication in Natural Hazards and Earth System Sciences.

Major comments
I am seriously concerned about the robustness of the results, as the authors’ choice of the single grid point appears arbitrary and may not adequately represent Germany. This study provides numerous statistics and numbers, all of which heavily rely on the validity of the choice of the data point. Figure 5, which serves as a summary plot, can therefore be quite misleading if the choice of the data point is not proven to be valid. The authors need to justify this choice more rigorously or, preferably, to base their analysis on a spatially averaged region or multiple grid points representing Germany or relevant subregions. This would substantially enhance the robustness and credibility of the results.

To be honest, this manuscript focuses too heavily on statistical relationships without sufficient dynamical and mechanistic interpretations and discussions. I am left with many “why” questions. Although the authors touch on potential mechanisms for the relationship between blocking and extremes, their arguments remain speculative, only referring to previous literature. The authors might want to add their own analyses on other variables to compare with previous studies if needed. Such mechanistic discussions would enhance the robustness of the results.

In addition, Section 4 is overly long and needs to be more structured. I recommend moving the mechanistic discussions on each extreme to Section 3.

Other comments
Please clarify which ensemble members are used for the selected models.

L133: What is the meaning of “subjectively identified based on the maxima”?

L159: Please correct the part “all reanalyses respectively CMIP6 simulations”.

How do the authors define extremes in ssp585? Do they use the same threshold as in historical? What are the changes in the mean frequency of blocking and extremes in their analysis?

Reply
Citation: https://doi.org/10.5194/egusphere-2025-3670-RC2

Richard Lohmann, Christopher Purr, and Bodo Ahrens

Supplement

https://doi.org/10.5194/egusphere-2025-3670-supplement

Richard Lohmann, Christopher Purr, and Bodo Ahrens

Viewed

Total article views: 1,485 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
1,255	198	32	1,485	45	29	31

HTML: 1,255
PDF: 198
XML: 32
Total: 1,485
Supplement: 45
BibTeX: 29
EndNote: 31

Views and downloads (calculated since 19 Sep 2025)

Month	HTML	PDF	XML	Total
Sep 2025	974	10	5	989
Oct 2025	85	16	4	105
Nov 2025	49	30	5	84
Dec 2025	65	51	13	129
Jan 2026	41	38	1	80
Feb 2026	30	46	3	79
Mar 2026	11	7	1	19

Cumulative views and downloads (calculated since 19 Sep 2025)

Month	HTML	PDF	XML	Total
Sep 2025	974	10	5	989
Oct 2025	85	16	4	105
Nov 2025	49	30	5	84
Dec 2025	65	51	13	129
Jan 2026	41	38	1	80
Feb 2026	30	46	3	79
Mar 2026	11	7	1	19

Viewed (geographical distribution)

Total article views: 1,381 (including HTML, PDF, and XML) Thereof 1,381 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 03 Mar 2026

Short summary

This study investigates the relationship between atmospheric blocking and the extreme events heatwaves, heavy rainfall and calm events in Germany in atmospheric reanalyses and CMIP6 climate simulations. In the reanalyses, the statistical relationship is more pronounced between blocking and calms than between blocking and heavy precipitation. In the simulated future climate, the frequency of the three extreme event types increases with nearly unchanged relationship of blocking with the extremes.


Total:	0
HTML:	0
PDF:	0
XML:	0