Oliver Fuhrer retweeted
Starting in May of 2025, MeteoSwiss will make its data freely and publicly available as part of the federal government's Open Government Data (OGD) policy.
OGD: bit.ly/3ScMZDP
YouTube: bit.ly/43qPHL2
#opendata #ogd #weatherdata #climatedata #meteoswiss
Oliver Fuhrer retweeted
Ab Mai 2025 stellt MeteoSchweiz seine Daten im Rahmen der Open Government Data (OGD) Strategie des Bundes frei und kostenlos zur Verfügung.
OGD Startseite: bit.ly/4krrQlq
YouTube: bit.ly/4ksHxsz
#opendata #ogd #wetterdaten #klimadaten #meteoschweiz
Oliver Fuhrer retweeted
Mauro Bianco from CSCS presents GT4Py, a DSL for #Climate simulations at SOS27. It's using SPCL's DaCe as backend :-).
"Separation of concerns is not isolation but communication."
Great talk!
Oliver Fuhrer retweeted
We are coordinating an EU-funded project that aims to use #MachineLearning in novel ways for #WeatherForecasting and related #EarthSystem modelling.
Find out more about the aims and methods of the WeatherGenerator project ➡️ ecmwf.int/en/about/media-cen…
@C2SM_ETH @cscsch
Thanks for the opportunity, dear @Wochenzeitung
Atmosphärenphysiker @ofuhrer von @meteoschweiz erklärt im Interview, wie Wettermodelle funktionieren – und wie man Wetter-Apps besser nutzt.
woz.ch/2429/das-wetter/so-al…
Extremely excited about this milestone and BIG THANKS to everybody @meteoschweiz and our partners who helped make this possible!
MeteoSchweiz hat das #Wettermodell #ICON in den operativen Betrieb eingeführt und ersetzt damit das bisherige Modell COSMO. ICON verwendet ein Dreiecksgitter, das eine höhere Auflösung ermöglicht und so die komplexe Topografie der Schweiz genauer abbildet: bit.ly/4cAD6rw
Oliver Fuhrer retweeted
Die MeteoSchweiz App startet nicht? Nicht nur bei dir: Ein Unterbruch in der Produktion macht ihr zu schaffen. Wir arbeiten daran und geben Bescheid, sobald alles wieder läuft.
Dear @swisstopo , your search is something that regularly confuses me… not finding places, results in very bizarre order, requiring exact spelling, … any plans on improving this?
Oliver Fuhrer retweeted
Here’s an early preview of ElevenLabs Music.
All of the songs in this thread were generated from a single text prompt with no edits.
Title: It Started to Sing
Style: “Pop pop-rock, country, top charts song.”
Oliver Fuhrer retweeted
More than 140 scientists met in Berlin to discuss "Earth Virtualization Engines" - a federated #HPC system to create and serve high-resolution climate data for everybody: eve4climate.org/
A technical summary:
computer.org/csdl/magazine/c…
or
arxiv.org/abs/2309.09002
🌍🌏🌎
Oliver Fuhrer retweeted
Ab #2025 können Sie unsere Wetter- und Klimadaten als Open Government Data gebührenfrei beziehen. Helfen Sie schon jetzt, den geplanten Service zu optimieren. Haben Sie Fragen und Bedürfnisse zu #OGD?
Mehr in unserem Blog: bit.ly/3rG2NoX
Oliver Fuhrer retweeted
Important paper subjecting an AI global weather model to atmospheric dynamics tests. Beautiful work by Greg Hakim at UW and Sanjit Masanam. Equation (1) was a crafty idea to tease out response to perturbation & forcing relative to smooth climatology. arxiv.org/abs/2309.10867
Oliver Fuhrer retweeted
1/ Die SRG-Gegner*innen der #Halbierungsinitiative behaupten, mehr "Markt" und weniger "Staat" würde Vielfalt steigern.
Sieht gut aus, diese Vielfalt 🧵:
Oliver Fuhrer retweeted
A Practical Deep Learning-Based Acoustic Side
Channel Attack on Keyboards
correctly identifies 93% of keystrokes via sounds over zoom calls
(reminder: scaling laws apply to ~every modality)
arxiv: arxiv.org/abs/2308.01074
Another library for interfacing PyTorch models with traditional HOC applications written in Fortran or C/C++ github.com/NVIDIA/TorchFort
1/ We do not need consensus about the „right“ way to improve local climate information. A consensus that rapid progress is essential and urgent for effectively addressing climate change is sufficient to support EVE nature.com/articles/d41586-0… 🧵
3/ Large international efforts are not uncommon to address big problems (ITER, CERN, SKA, …). Funding agencies know that effective science policy requires both a recognition of the value of varied kinds of research and a commitment to supporting a balanced scientific ecosystem.



























