A comparative study of Medicane Daniel (September 2023) is performed using two high-resolution models, WRF and ICON, both configured at ∼2 km spatial resolution with comparable domains, timesteps, boundary forcing and settings. A custom optimizer harmonizes the vertical levels discratization and sensitivity experiments test different cumulus parametrizations: fully explicit, shallow-convection, deep-cumulus parameterized and ICON’s gray-zone option. Diagnostics include an objective tracker combining mean sea-level pressure and lower-tropospheric geopotential, alongside intensity metrics (central pressure and 10 m wind), precipitation patterns and point validation at Benina (HLLB) for pressure, wind and rainfall. Structural evolution is assessed through Hart’s Cyclone Phase Space (CPS) and a novel Temporal Annular Symmetric Mean (TASM), describing the three-dimensional storm structure during its warm-core phase. Both models reproduce Daniel’s track, lifecycle and tropical-like features. Explicit convection deepens the cyclone and sharpens wind maxima, but enhances small-scale variability that complicates tracking. Deep-cumulus schemes weaken extremes and broaden rainfall, while shallow-convection options provide a balance, improving precipitation placement and core thermodynamics. Model internal differences also influence results: ICON shows lower efficiency in transferring diabatic heating upward, producing a shallower warm core, whereas WRF tends to generate a stronger vortex to better retain tropical-like characteristics. CPS and TASM consistently indicate a shallow-to-deep warm-core transition and a compact, symmetric structure at peak intensity. Overall, the study highlights the importance of harmonized configurations and suggests that, at gray-zone resolutions, shallow-convection treatments often offer a good compromise for simulating Mediterranean tropical-like cyclones.

We analyze 17 Mediterranean cyclones with tropical characteristics using ERA5
reanalysis data to investigate the role of upper-tropospheric processes and their
seasonal variations in cyclone development. Our results show that cyclones
occurring in September exhibit the strongest moist convection, with Ianos, one
of the strongest medicanes on record, representing an extreme case. The devel-
opment of the warm core is anticorrelated with the value of upper-tropospheric
potential vorticity (PV), and correlated with lower tropospheric PV associated
with latent heat release. Using back-trajectories analysis, we then investigate the
presence of dry intrusions in the early and mature cyclone phases. We show that
the Lagrangian definition of dry intrusion, characterized by a descent of 400 hPa
in 48 hours, is not satisfied in most of the cyclones considered here. Instead,
weaker PV streamers with shallower descent are often observed before cycloge-
nesis. While a dry intrusion may reduce static stability and intensify convection,
leading to an accelerated warm core formation, the eventual warm core intensity
appears to depend critically on pre-existing convection and diabatic heating in
a close vicinity of the cyclone center. This is the case for Medicane Ianos. In
fact, this cyclone presents two marginally descending flows of about 200 hPa
in 60 hours associated with shallow PV streamers, one in the early stage and
one before the main tropical-like phase. Despite the modest intensity of the
PV streamer, Ianos develops the strongest warm core among all cyclones. This
motivates us to examine Ianos in detail using a WRF simulation with a grid
spacing of 3 km. The simulation confirms the need for a high-resolution model
setup to correctly represent the cyclone intensity and capture key mesoscale
mechanisms and the dominant contribution of diabatic heating to the cyclone
deepening.

Wet-snow events are a major cause of severe winter outages in the Italian high- and medium-voltage power networks, due to the accumulation of ice and snow on overhead conductors. It is estimated that, in Italy alone, the annual economic impact of these events exceeds 200 million euros.

To address this issue, RSE initiated a research program more than a decade ago that led to the development of an operational alert system for snow accumulation on overhead lines. The system, known as WOLF (Wet-snow Overload aLert and Forecasting), is designed to forecast wet-snow loads on overhead lines during snowfall events and to provide timely warnings to Italian TSOs and DSOs, enabling them to implement appropriate measures to ensure the reliability and continuity of electricity transmission and distribution. WOLF integrates precipitation and temperature fields from the WRF model with the Makkonen accretion model, which estimates the growth of snow load on a reference conductor in each domain grid cell as a function of the prevailing meteorological conditions.

Over the past ten years, observations collected at the WILD (Wet-snow Ice Laboratory Detection) monitoring station, located in the Cuneo Alps, have supported the refinement of both meteorological forecasting models and snow-sleeve accretion models. Previous case studies have shown that the primary sources of uncertainty in snow-load forecasts stem from the intrinsic limitations of the meteorological fields simulated by the NWP models used to drive the accretion model. Sensitivity analyses performed using different model configurations and global drivers revealed variable performance, without identifying a single optimal setup across the analyzed snowfall events.

The recent availability of NWP model outputs from multiple providers through the Italian open-data hub Mistral (https://meteohub.mistralportal.it/app/datasets) offers new opportunities to address these limitations. In this work, the potential benefits of a multi-model approach are assessed by combining snow-load forecasts derived from different meteorological simulations and by exploring probabilistic post-processing techniques for wet-snow prediction. The results, validated against observations from recent snowfall events in the Alpine region, indicate that this approach is promising. Further research is planned, including evaluation over a larger set of case studies, with the aim of reducing the forecast uncertainty of the WOLF system starting from upcoming winter seasons.