\n\n
\n DTSTART;TZID=Europe/Berlin:20260706T110000 DTEND;TZID=Europe/Berlin:20260706T120000 LOCATION:KIT Campus Nord, IMKAAF, Gebäude 326, Raum 150, END:VEVENT BEGIN:VEVENT UID: 61888 DTSTAMP:20260703T110921Z SUMMARY:Artificial Intelligence Pathways from Weather to Climate X-ALT-DESC;FMTTYPE=text/html:
Deep learning emulates atmospheric reanalyses with high fidelity, enabling increasingly well-calibrated ensemble weather forecasts at progressively longer lead times. To extend these gains to climate-relevant horizons, AI prediction systems must produce credible forced responses to drivers of interest (e.g., greenhouse gases, land-use change). We propose a minimal, testable framework for AI climate modeling: (i) represent external forcings explicitly and restrict them to physically appropriate state tendencies; and (ii) stress-test robustness in out-of-distribution regimes, including extremes and counterfactual trajectories. Using leading climate emulators and hybrid physics-AI models, we identify coupling and development challenges and compare scaling with resolution and effective complexity. AI models do not appear intrinsically more efficient than GPU-ported dynamical models once complexity is accounted for, yet they can directly predict target variables at the desired grid without integrating the full high-frequency, multivariate state. Diverse ML downscaling strategies can partially substitute for explicit fine-scale resolution when observations are available, paving the way towards inexpensive, local risk assessment across prediction horizons
Deep learning emulates atmospheric reanalyses with high fidelity, enabling increasingly well-calibrated ensemble weather forecasts at progressively longer lead times. To extend these gains to climate-relevant horizons, AI prediction systems must produce credible forced responses to drivers of interest (e.g., greenhouse gases, land-use change). We propose a minimal, testable framework for AI climate modeling: (i) represent external forcings explicitly and restrict them to physically appropriate state tendencies; and (ii) stress-test robustness in out-of-distribution regimes, including extremes and counterfactual trajectories. Using leading climate emulators and hybrid physics-AI models, we identify coupling and development challenges and compare scaling with resolution and effective complexity. AI models do not appear intrinsically more efficient than GPU-ported dynamical models once complexity is accounted for, yet they can directly predict target variables at the desired grid without integrating the full high-frequency, multivariate state. Diverse ML downscaling strategies can partially substitute for explicit fine-scale resolution when observations are available, paving the way towards inexpensive, local risk assessment across prediction horizons
\n DTSTART;TZID=Europe/Berlin:20260707T154500 DTEND;TZID=Europe/Berlin:20260707T164500 LOCATION:KIT, Campus Süd, Gebäude 30.22, Otto-Lehmann-Hörsaal END:VEVENT BEGIN:VEVENT UID: 61500 DTSTAMP:20260703T110921Z SUMMARY:TRO-Seminar X-ALT-DESC;FMTTYPE=text/html:(1) tbd (2) tbd (3) Renewable Energy Systems in a Changing Climate(4) tbd
(1) tbd (2) tbd (3) Renewable Energy Systems in a Changing Climate(4) tbd
\n DTSTART;TZID=Europe/Berlin:20260709T091500 DTEND;TZID=Europe/Berlin:20260709T114500 LOCATION:KIT, Campus Nord, Gebäude 435, Seminarraum 2.05 END:VEVENT BEGIN:VEVENT UID: 62113 DTSTAMP:20260703T110921Z SUMMARY:First spatially continuous fog maps for the central Namib reveal seasonally shifting fog belt X-ALT-DESC;FMTTYPE=text/html:
\n\n
\n DTSTART;TZID=Europe/Berlin:20260713T110000 DTEND;TZID=Europe/Berlin:20260713T120000 LOCATION:KIT Campus Nord, IMKAAF, Gebäude 326, Raum 150, END:VEVENT BEGIN:VEVENT UID: 61560 DTSTAMP:20260703T110921Z SUMMARY:Towards a Community Modell ICON X-ALT-DESC;FMTTYPE=text/html:
TBD
TBD
\n DTSTART;TZID=Europe/Berlin:20260714T154500 DTEND;TZID=Europe/Berlin:20260714T164500 LOCATION:KIT, Campus Süd, Gebäude 30.22, Otto-Lehmann-Hörsaal END:VEVENT BEGIN:VEVENT UID: 61537 DTSTAMP:20260703T110921Z SUMMARY:TRO-Seminar X-ALT-DESC;FMTTYPE=text/html:(1) tbd (2) tbd (3) tbd (4) tbd
(1) tbd (2) tbd (3) tbd (4) tbd
\n DTSTART;TZID=Europe/Berlin:20260716T091500 DTEND;TZID=Europe/Berlin:20260716T114500 LOCATION:KIT, Campus Nord, Gebäude 435, Seminarraum 2.05 END:VEVENT BEGIN:VEVENT UID: 62114 DTSTAMP:20260703T110921Z SUMMARY:Weather Sensing with Weather dependent Backscattering of Photons in Optical Fibers integrated in High-Voltage Lines X-ALT-DESC;FMTTYPE=text/html:Ice nucleating particles (INPs) strongly influence cloud phase, precipitation, and radiative properties, yet observations of their vertical distributions remain sparse. This presentation reviews recent advances in airborne INP measurements using drones, launched balloons, and tethered balloon systems, focusing on the development of the Profiling Upper altitudes For Ice Nucleation (PUFIN) sampler for routine altitude-resolved observations. Results from multiple deployments demonstrate that INP concentrations can vary substantially with height, season, aerosol source, and boundary layer structure, highlighting the limitations of relying solely on surface measurements. These vertically resolved observations provide critical constraints for understanding aerosol-cloud interactions and improving the representation of ice formation processes in atmospheric models.
Ice nucleating particles (INPs) strongly influence cloud phase, precipitation, and radiative properties, yet observations of their vertical distributions remain sparse. This presentation reviews recent advances in airborne INP measurements using drones, launched balloons, and tethered balloon systems, focusing on the development of the Profiling Upper altitudes For Ice Nucleation (PUFIN) sampler for routine altitude-resolved observations. Results from multiple deployments demonstrate that INP concentrations can vary substantially with height, season, aerosol source, and boundary layer structure, highlighting the limitations of relying solely on surface measurements. These vertically resolved observations provide critical constraints for understanding aerosol-cloud interactions and improving the representation of ice formation processes in atmospheric models.
\n DTSTART;TZID=Europe/Berlin:20260721T151500 DTEND;TZID=Europe/Berlin:20260721T161500 LOCATION:KIT Campus Nord, IMKAAF, Gebäude 435, Raum 205, END:VEVENT BEGIN:VEVENT UID: 61501 DTSTAMP:20260703T110921Z SUMMARY:TRO-Seminar X-ALT-DESC;FMTTYPE=text/html:(1) Predictability of the 2014 Pentecost storm (2) tbd (3) Added-Value of the Coming Decade ICONXPP Decadal Climate Predictions Ensemble for User-relevant Variables in Europe (4) tbd
(1) Predictability of the 2014 Pentecost storm (2) tbd (3) Added-Value of the Coming Decade ICONXPP Decadal Climate Predictions Ensemble for User-relevant Variables in Europe (4) tbd
\n DTSTART;TZID=Europe/Berlin:20260723T091500 DTEND;TZID=Europe/Berlin:20260723T114500 LOCATION:KIT, Campus Nord, Gebäude 3435, Seminarraum 2.05 END:VEVENT BEGIN:VEVENT UID: 62117 DTSTAMP:20260703T110921Z SUMMARY:Master Thesis: How the electron precipitation characteristics corresponds to the auroral peak height emission / Master Thesis: tbd X-ALT-DESC;FMTTYPE=text/html:Influences of aerosols on climate are qualitatively known, but the magnitude of their effects is uncertain with important implication for understanding past and future climate change. In this colloquium, I will explain some reasons for uncertainties in our understanding of aerosol effects on climate, and outline how we can make progress despite persistent multi-model spread in aerosol radiative forcing. We will see model-to-model differences for aerosol effects from global climate model simulations, and new CMIP aerosol data for use in climate simulations that inform the next IPCC assessment report of climate change. Looking ahead, I will introduce some of the plans for advancing the research field, e.g., through leading the new experiment protocol of the Aerosol and Chemistry Model Intercomparison Project, to which the most complex Earth system models currently available worldwide will contribute simulations.
Influences of aerosols on climate are qualitatively known, but the magnitude of their effects is uncertain with important implication for understanding past and future climate change. In this colloquium, I will explain some reasons for uncertainties in our understanding of aerosol effects on climate, and outline how we can make progress despite persistent multi-model spread in aerosol radiative forcing. We will see model-to-model differences for aerosol effects from global climate model simulations, and new CMIP aerosol data for use in climate simulations that inform the next IPCC assessment report of climate change. Looking ahead, I will introduce some of the plans for advancing the research field, e.g., through leading the new experiment protocol of the Aerosol and Chemistry Model Intercomparison Project, to which the most complex Earth system models currently available worldwide will contribute simulations.
\n DTSTART;TZID=Europe/Berlin:20260728T151500 DTEND;TZID=Europe/Berlin:20260728T161500 LOCATION:KIT Campus Nord, IMKAAF, Gebäude 435, Raum 205, END:VEVENT END:VCALENDAR