Pier Luigi Vidale is Professor of Climate System Science and co-leads the joint NCAS-Met Office global High Resolution Climate Modelling programme. Since a major UK-Japan collaboration in 2004-2007, Pier Luigi has played a world-leading role in high resolution climate modelling and understanding the role of meso-scale processes in the global climate system. He was the Scientific Coordinator of the EU’s H2020 PRIMAVERA (defining HighResMIP for CMIP6), currently co-leads the EU’s Horizon Europe EERIE (defining HighResMIP2 for CMIP7). Pier Luigi is actively involved in the development of km-scale global models, e.g. as investigator in the EU’s H2020 NextGEMS project, and in collaborations with the Met Office. Pier-Luigi is co-Chair of the World Climate Research Programme’s Digital Earths Lighthouse Activity, aiming to define the nature and purpose of Digital Twins for climate, and is the Director of Science Collaboration in the University of Reading’s £30M AFESP programme.

Keynote talk: Bridging CMIP-class and km-scale simulation of the climate system

P.L. Vidale, A. Gettelman, M. Roberts, A. Baker, K. Hodges, T. Auerswald, A. Sommer

Physics-based weather and climate simulation is usually defined by decisions around initialisation, forcing, resolution and sub-grid parameterisations. A more practical – and complementary – challenge is balancing computational resources and desired physical realism, where effective use of top-tier supercomputers requires mastering an ever-growing range of programming paradigms, which risks restricting cutting-edge simulation to a small community possessing the necessary expertise.

Global atmospheric models currently used for climate projections began to represent cyclones in the 1970s, but global ocean models only started to resolve eddies and boundary currents in the last decade. The next advances in mid-latitude weather prediction (driven e.g. by tropical wave trains), extending to seasonal and climate prediction (driven e.g. by air-sea interactions), aim directly for global, coupled km-scale (horizontal grid spacing < 10 km) simulation. Testing how well this new class of models represents underpinning weather and climate processes crucially depends on suitable observations. Yet, at these scales, global observational data sets are often inadequate for assessment, especially at the process level. Significant progress can be made, however, by involving scientists at operational and academic centres worldwide and exploring synergies between global and regional capabilities in simulation and analysis. With this aim, some of the CMIP7 protocols, e.g. HighResMIP2, have proposed complementary experiments that are suitable for the km-scale community, while km-scale protocols, e.g. DYAMOND-3, have started to extend weather-type experiments to span climate-relevant (annual to multi-annual) time scales. 

In May 2025, the WCRP global km-scale hackathon brought the CMIP and km-scale communities together: nearly 700 participants from around the world gathered at 10 regional nodes, and several satellite nodes, for one week of coordinated, process-level analyses. The hackathon focused on the DYAMOND-3 experiments, performed with km-scale – regional and global – Earth System Models. It deployed and demonstrated a range of advanced technologies that make these exascale data sets rapidly accessible to anyone, enabling the community to work in new and complementary ways, to democratise access to global information at a local scale, and to explore how local processes affect the rest of the globe. Exciting science has emerged, e.g. on improved representation of precipitation (frequency/intensity, diurnal cycle) and its attribution to phenomena such as MCSs, tropical cyclones and atmospheric rivers, and on more realistic coupling between the land surface and the atmosphere. These advanced analyses are increasingly integrating new satellite data and emulation with machine learning. A second global hackathon will be held in June 2027, with more nodes and an increased focus on multi-scale interactions, extremes and their impacts.

Fundamentally, these techniques and technologies allow the community to move towards more participative models for the co-production and delivery of diverse sources of environmental information, for domain scientists and citizens alike.

Dr. Kelly Núñez Ocasio is an Assistant Professor in the Department of Atmospheric Sciences at Texas A&M University. Before joining Texas A&M, she held an Advanced Study Program (ASP) Postdoctoral Fellowship at the NSF National Center for Atmospheric Research (NCAR). She received her Ph.D. in Meteorology and Atmospheric Science from Penn State University in 2021 and her B.S. in Physics, along with a Curricular Sequence in Atmospheric Sciences and Meteorology, from the University of Puerto Rico, Mayagüez Campus, in 2016. Dr. Kelly Núñez Ocasio is currently a co-lead of the international Mesoamerica and Caribbean Affinity Group (MAAG) and a member of the AMS Science and Technological Activities Commission (STAC) Committee on Tropical Meteorology and Tropical Cyclones. She also serves as an Associate Editor for Monthly Weather Review and as a Rapporteur for the Eleventh International Workshop on Tropical Cyclones (IWTC-11). Her research is currently supported by the National Science Foundation (NSF AGS-2508342), investigating how tropical weather systems respond to climate change using convection-permitting regional simulations.

Keynote talk: Tropical Weather Systems Response to a Warming Climate: Single-Event and Multi-Year Simulations with Convection-Permitting MPAS-A 

Kelly Núñez Ocasio

Mesoscale convective systems (MCSs) over tropical Africa and the Atlantic provide critical rainfall while posing flood risk, and, when coupled to African easterly waves (AEWs), often serve as precursors to Atlantic tropical cyclones. Understanding how these systems respond to climate change requires kilometer-scale resolution to capture the interplay between synoptic forcing (AEW dynamics, monsoon circulation) and local thermodynamics (CAPE, CIN, moisture gradients). This talk presents a multi-scale perspective on tropical weather system evolution under warming, using convection-permitting regional simulations with the Model for Prediction Across Scales–Atmosphere (MPAS-A). Recent convection-permitting simulations of September 2006 under pseudo-global-warming conditions reveal that future MCSs evolve distinctly from thermodynamic expectations alone.

Despite increased CAPE, MCS numbers decrease due to substantial CIN increases, while rainfall intensifies over water (linked to increasing relative humidity) but not over land. Future MCSs shift southward, become less tied to the African easterly jet, and couple more strongly to the West African monsoon and AEWs. The most extreme future MCSs show stronger updrafts/downdrafts and drier mid-to-lower troposphere. These results demonstrate that synoptically forced MCSs—such as those coupled to AEW dynamics—respond fundamentally differently from the general MCS population, an outcome only capturable at kilometer scale.

Building on this motivation, I will present preliminary results from ongoing multi-year MPAS-A simulations funded by the National Science Foundation (NSF AGS-2508342), extending this single-event framework to historical and future JJAS seasons targeting different ENSO phases. This work addresses how the mean tropical climate, MCSs, and AEWs over Africa and the Atlantic—and their ENSO-phase-dependent behavior—respond to warming by end of century.

I will also discuss work done as co-lead with the Mesoamerica and Caribbean Affinity Group (MAAG), an NSF NCAR and community platform focused on fostering research collaborations, engagement, data access, and international collaboration. I will highlight research outcomes, leadership perspectives, and working-groups progress as we build regional datasets and foster cross-sector research across the region.

Prof. Masaki Satoh is a Professor at the Atmosphere and Ocean Research Institute (AORI), The University of Tokyo, and a cross-appointed Professor and Vice Director of the Typhoon Science and Technology Research Center at Yokohama National University. He received his Ph.D. in Geophysics from The University of Tokyo in 1993. Before joining AORI in 2005, he worked at the Saitama Institute of Technology and was a Senior Visiting Scholar at the University of Cambridge. He has also been affiliated with the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) and has served as a Senior Researcher at the Japan Aerospace Exploration Agency (JAXA) since 2011. His research focuses on atmospheric dynamics and numerical modeling, particularly developing and applying global high-resolution atmospheric models and studying tropical cyclones and mesoscale convective systems. He led the development of the Nonhydrostatic Icosahedral Atmospheric Model (NICAM) and has played a leading role in the international coordination of global storm-resolving modeling, including the DYAMOND project. He has also promoted synergistic research integrating high-resolution modeling with satellite observations, including those from the EarthCARE mission.

Keynote talk: Convective mass flux, vertical air motions, cloud microphysics in Mesoscale Convective Systems Revealed by a Synergistic Approach Using EarthCARE and Kilometer-Scale Models

Masaki Satoh and Woosub Roh

The vertical velocity structure in deep convection remains one of the most uncertain aspects of mesoscale convective systems (MCSs). In numerical models, vertical velocity spectra are not adequately represented at kilometer-scale (km-scale) resolution, whereas in observations, vertical velocities are not precisely measured or globally available. The EarthCARE satellite, launched in May 2024, provides the first global observations of vertical velocity through its Cloud Profiling Radar (CPR), which measures radar reflectivity and Doppler velocities worldwide. These unique observations offer a new opportunity to evaluate the dynamical structure of deep convection simulated in models.

In this study, we analyze EarthCARE Doppler radar observations by comparing them with global km-scale simulations conducted using the Nonhydrostatic Icosahedral Atmospheric Model (NICAM) together with a satellite simulator. We also examine results from other km-scale models participating in ECOMIP. We select representative cases of mesoscale convective systems to investigate how vertical motions are retrieved from the relationship between radar reflectivity and ice-hydrometeor fall speeds, and to identify characteristic features of vertical velocity structures.

Our results show that the NICAM simulations successfully reproduce the overall vertical structures observed by EarthCARE when averaged over convective systems. Analysis of the multi-model ECOMIP results will further clarify the relationships between circulation structure and cloud microphysics in deep convection, as well as reveal model deficiencies. Through this study, we demonstrate the strong potential of EarthCARE data to constrain and improve the representation of cloud microphysics and dynamics in global weather and climate models.

Dr Negin Nazarian is a Scientia Associate Professor at the University of New South Wales (UNSW Sydney). As an urban climatologist, Negin studies the complex interactions between the built environment and climate, with a keen interest in how these dynamics impact people, particularly through heat and wind. She leads the Climate-Resilient Cities lab, dedicated to advancing high-resolution urban climate modelling, refining urban canopy parameterisations (through physical or AI modelling) for regional and global climate models, and employing urban climate informatics to enhance our understanding of human exposure to urban climate challenges. She is Deputy Director at the ARC Centre of Excellence for 21st Century Weather, chairs the Board on Urban Environment at the American Meteorological Society, and is a Lead Author of the IPCC Special Report on Climate Change and Cities in the 7th Assessment Report (AR7).

Neelesh Rampal is a Climate Data Scientist at Earth Sciences New Zealand (formerly NIWA), where he has worked since 2020. His research uses generative artificial intelligence (AI) to emulate regional climate models and rapidly produce large ensembles of downscaled climate projections. His interests also include physics-informed AI and uncertainty quantification in climate projections. Neelesh completed his PhD in Climate Science at the Climate Change Research Centre at UNSW Sydney. He has co-led several research grants focused on developing AI approaches for downscaling, with particular emphasis on capturing rare extreme events and how they change under warming. His current work extends these approaches to convection-permitting scales and sub-daily timescales. As a co-coordinator of the CORDEX Task Force on Machine Learning, Neelesh coordinates international efforts in AI downscaling, including model intercomparisons, benchmarking and guidance on evaluating these approaches for use in climate projections. He is also interested in developing accessible training datasets and tools for the global community.

Check back to see the speakers list as we finalise the program.