联邦气象和气候局

NowPrecip is a real-time nowcasting application, operational at MeteoSwiss since 2020. A new version has introduced a novel ensemble-based approach designed to generate real-time multi-member precipitation forecasts and improve rainfall-related severe weather warnings in Switzerland. This study provides a comprehensive verification of the NowPrecip ensemble by comparing its performance with precipitation observations from five C-band dual-polarization radars operated by MeteoSwiss. In addition, the Python-based Short-Term Ensemble Prediction System (PySTEPS), a widely used open-source nowcasting framework, is employed to generate ensemble nowcasts, providing a valuable benchmark for NowPrecip performance. Both PySTEPS and NowPrecip nowcasting simulations are blended with members of the numerical weather prediction (NWP) ensemble derived from the Icosahedral Nonhydrostatic Model (ICON). To this end, we compute widely used deterministic and probabilistic verification metrics across a 3.5-year dataset over the period 2020-2023. The results show that NowPrecip systems consistently outperform PySTEPS across skill scores, while deterministic verification demonstrates that NowPrecip methods exhibit superior performance compared with the current operational NWP ensemble for lead times up to four hours.

Numerical weather prediction and climate models require continuous adaptation to take advantage of advances in high-performance computing hardware. This paper presents the port of the ICON model to GPUs using OpenACC compiler directives for numerical weather prediction applications. In the context of an end-to-end operational forecast application, we adopted a full-port strategy: the entire workflow, from physical parameterizations to data assimilation, was analyzed and ported to GPUs as needed. Performance tuning and mixed-precision optimization yield a 5.5× speed-up compared to the CPU baseline in a socket-to-socket comparison. The ported ICON model meets strict requirements for time-to-solution and meteorological quality, in order for MeteoSwiss to be the first national weather service to run ICON operationally on GPUs with its ICON-CH1-EPS and ICON-CH2-EPS ensemble forecasting systems. We discuss key performance strategies, operational challenges, and the broader implications of transitioning community models to GPU-based platforms.

The distance to the Vela Junior supernova remnant (RX J0852.0–4622 or G266.2–1.2) has long remained uncertain, limiting our understanding of its physical properties. Using Very Large Telescope/Multi Unit Spectroscopic Explorer integral field spectroscopy, we uncover chemical and kinematic connections between the nebula surrounding its central compact object (CXOU J085201.4–461753) and the nearby Herbig–Haro outflow of Ve 7–27 (Wray 16–30), indicating a shared nitrogen-rich, Fe-peak-enhanced environment. This link ties stellar birth and death, with the young star Ve 7–27 embedded in material expelled by Vela Junior’s massive progenitor, and the remnant’s blast wave is expanding through the same medium. Adopting the Gaia-based distance to Ve 7–27, we revise Vela Junior’s distance to 1.41 ± 0.14 kpc. At this distance, the remnant’s physical radius is 23.3 ± 2.3 pc, and X-ray proper motions of the northwestern rim correspond to shock speeds of (2.8 ± 0.7) × 10 ^3 to (5.6 ± 1.5) × 10 ^3 km s ^−1 . These imply an age of ∼1.6–3.3 kyr and a very low ambient density, indicating that Vela Junior is expanding within a highly rarefied wind-blown cavity carved by a massive progenitor—consistent with the nondetection of strong thermal X-ray emission. This distance update also resolves long-standing inconsistencies, with major implications for its energy budget, particle acceleration efficiency, and compact object evolution.

As climate change leads to more frequent droughts, understanding forest ecosystem health becomes increasingly critical. Monitoring forest canopy water content provides valuable insights into their resilience to these stressors. Space-based optical vegetation indices like the normalized difference water index (NDWI) are often used to monitor the water content over more extended time periods in large areas. However, the top-of-canopy view of satellites limits the sensitivity as they neglect the lower canopy and understory vegetation. This study explores the seasonal correspondence between the NDWI retrieved from Sentinel-2 satellite data and in situ measured vegetation optical depth (VOD). We use a unique time series of concurrent and complementary measurements acquired in two contrasting forest ecosystems (i.e., an evergreen coniferous and a deciduous broadleaf forest) spanning four seasons. VOD is calculated from global navigation satellite system (GNSS) data measured with one receiver antenna above and one below the canopy, assessing the full vertical extent of the forest canopy. In addition, we used the Sentinel-2 derived enhanced vegetation index (EVI), eddy flux based evapotranspiration (ET) estimates, and measurements of soil moisture, air temperature, precipitation, and shortwave incoming radiation to facilitate our interpretation. Our results showed a large seasonal variation in VOD and NDWI in the deciduous forest, a pattern that coincided with a large variation in biomass, as expected and indicated by the EVI. We saw indications that a short-term summer drought in 2022 affected VOD and ET but not the NDWI in the deciduous forest. In contrast, in the evergreen forest, we found a pronounced seasonality of canopy water content only for ET, while VOD and NDWI followed different trajectories. We conclude that the satellite-based NDWI tended to saturate at higher levels of canopy water content. VOD showed a sensitivity to changing canopy water content, as indicated by ET and soil moisture dynamics, as well as to changing canopy biomass, as suggested by varying EVI. These initial exploratory insights into the sensitivity of VOD could stimulate discussions within the community and potentially help optimize the sampling design of future VOD networks.

The long-term trend for aerosol optical properties and climate impact sensitivity in terms of radiative forcing efficiency were analyzed at a suburban station in Athens, Southeast Mediterranean, using an extensive dataset from 2008 to 2022. The study examined scattering (nsc) and absorption (nap) coefficients, scattering & Aring;ngstrom exponent (SAE), absorption & Aring;ngstrom exponent (AAE), single scattering albedo (SSA), asymmetry parameter (g), and radiative forcing efficiency (RFE). Seasonal variability was linked to meteorological conditions and human activities. Single Scattering Albedo (SSA) was lowest (0.86), and Radiative Forcing Efficiency (RFE) was highest (-61 W/m2) in winter, confirming enhanced contributions from traffic and biomass burning. Lower SAE values (1.5) in spring indicate a greater presence of coarse particles due to frequent Saharan dust events (SDEs). Daily patterns of nap and SSA reflect local emissions, with pronounced traffic-related peaks. Aerosol classification revealed that Black Carbon (BC) dominates the suburban aerosol (51 %), with mixed BrC-BC (16 %) peaking in winter and dust-pollution mixtures (13 %) increasing in spring. The presence of large particles mixed with BC (11 %) was more frequent in spring, further highlighting seasonal variability. Trend analysis showed statistically significant (ss) decreases in nsc (-0.611) and SSA (-0.003), alongside increases in nap (+0.027) and RFE (+0.270) at a 95 % confidence level, suggesting a shift toward more absorbing aerosols. The findings provide new insights and reveal a new aerosol regime, where a reduction in anthropogenic emissions is affecting the scattering rather than the absorbing aerosol component, while the impact from forest fires as a climate feedback mechanism has a significant effect in the Eastern Mediterranean. It is important for future studies and climate modelling to account for the regionally observed changes of the state of mixing of ambient aerosol leading to a shift in radiative forcing efficiency through the reduction in SSA. This is evident in the long term for the east Mediterranean region and must be accounted for in radiative forcing estimates and future climate projections.