| 15 Apr 2026
Future Changes in Severe Frontal Precipitation Events over Europe and Their Drivers
Abstract. Atmospheric fronts are closely linked to extreme precipitation across the mid-latitudes, which is projected to intensify in many regions under global warming. Understanding the physical drivers of these changes is essential to improve confidence in climate projections. Here, we analyze projected changes in seasonal heavy and extreme frontal precipitation events over Europe using the CMIP6 and EURO-CORDEX ensembles, combining event frequency analysis with frontal composite cross-sections to assess the changes of the underlying thermodynamic and dynamic processes. We find that the number of heavy frontal precipitation events increases by up to 50 % per degree of global warming, while extreme events are projected to more than double per degree. Large-scale circulation changes account for most regional reductions in frontal extremes, but contribute only weakly to the widespread increases. Thermodynamic changes, however, dominate the intensification of extremes. Increases in specific humidity are the primary driver of more intense events, while changes in the frontal circulation are minimal, likely because a more stable atmosphere counteracts potential strengthening from enhanced latent heat release.
Title: Future Changes in Severe Frontal Precipitation Events over Europe and Their Drivers
Authors: Schaffer et al.
Manuscript submitted to Weather and Climate Dynamics
This article analyses changes in frontal precipitation using CMIP6 and CORDEX data. Front detection is based on equivalent potential temperature and the 850-hPa TFP parameter, with precipitation linked to identified fronts using a 300 km distance criterion. The definitions of 'strong' and 'extreme' precipitation are based on percentiles and vary between models and periods.
The changes appear to align with previous findings and expectations. The change patterns depict the well-known projected poleward shift of the storm tracks, specifically a triple pattern of change with intensification in an elongated band across the UK, at least during some season. This band is flanked by regions with reduced changes to the north and south of it.
The frontal composites show that the changes align with the hypothesis that relative humidity changes are minimal, and that CC scaling suggests a disproportionate increase in specific humidity in the warm sector compared to the cold sector, which ultimately causes the observed increase in horizontal gradients of equivalent potential temperature. The authors conclude that the observed changes are driven by thermodynamic rather than dynamic changes and hypothesis that changes in the stability offset changes in increased vertical motion from latent heat release.
The results are solid and supported by the findings. The methods are accepted, although the numerous thresholds could have been discussed more broadly given the associated uncertainties. The debate surrounding the definition of fronts is hinted at and could be expanded upon, particularly since the dynamically more relevant potential temperature fronts appear to remain unchanged in a warmer climate.
My main concern is the distinction between thermodynamic and dynamic changes, and the hypothesis that dynamic changes are suppressed by changes in tropospheric stability.
Firstly, the definition of 'dynamic change' is somewhat open to interpretation. Does it refer to a potential change in large-scale deformation that drives frontogenesis? Or is it simply a change in frontal frequency? Or is it vertical motion obtained from inverting the QG omega equation? If the authors argue that dry dynamic changes do not underlie the observed changes, I suggest computing the leading-order tendencies of frontogenesis and compare those to the frontogenesis tendency due to latent heating to quantitatively substantiate this argument. Previous studies argue that frontogenesis due to latent heating can be in the same order of magnitude as that due to tilting and deformation (Igel and Heever, QJ, 140, 139-150, 2013) and to show that dynamic changes are indeed small (assuming that you refer to dynamics in terms of large-scale deformation and tilting) than a more detailed analysis of their changes in the frontogenesis equation would be helpful.
Second, the hypothesis that static stability changes suppress dynamic changes is interesting, but moisture seems not considered in this argument. The dry static stability matters most for the cold sector, where an increase would weaken downward motion. In the warm sector what matters is the effective stability that takes moisture into account and it has been shown that this asymmetry in vertical motion across fronts indeed increases because of a reduction in effective static stability in the warm sector, which tends to increase vertical motion in this quadrant of the cyclone (O'Gorman, JAS, 68, 75-90, 2011). (Dry) static stability alone, in particular if zonally averaged, might obscure changes in the individual sectors of the cyclone.
Â
Composites:
Fig. 4–9: Is it possible that substantial variability within each sample masks any clear circulation signal in the composite mean? Section 2.5 could benefit from an example illustrating how the cross sections are computed.
Is the increase in specific humidity resulting in increased CAPE? Is CIN reducing? Both fields could be added to the composites with a small lag relative to the precipitation event.
Results:
L265: Due to the model inconsistency in mesoscale changes (see Figures 6 and 7), I would not conclude that dynamic changes are of secondary importance. The change is much more uncertain than the increase in humidity and no conclusion seems possible. This fits into the bigger picture that circulation changes are often much more uncertain in projections than changes in temperature or humidity.
A few more thought could be added about unclear changes in precipitation efficiency in the future and moisture availability.