A Collaboration Between SilverLining and Amazon Sustainability Data Initiative to accelerate climate research and promote equity
Climate modeling is among the most computing-intensive activities in all of science. Today, nearly all full production climate model simulations are run in government supercomputing facilities in the most economically developed countries, limiting capacity and access for researchers everywhere, and most significantly in the Global South. The massive datasets generated by climate models and observations also require powerful computing systems and networks, limiting where, and by who, climate research studies can be undertaken These major technology constraints slow climate science globally and prevent stakeholders in less developed parts of the world from engaging equitably in research and decision-making.
COP26: SilverLining Executive Director Kelly Wanser joins Ana Pinheiro Privette of the Amazon Sustainability Data Initiative (ASDI) on stage and Jean-Francois Lamarque of the National Center for Atmospheric Research (NCAR) and Chris Lennard of the Climate Systems Analysis Group (CSAG) at the University of Cape Town on-screen during COP26 in Glasgow, Scotland.
As part of SilverLining’s Safe Climate Research Initiative, and collaboration with Amazon Sustainable Data Initiative (ASDI), SilverLining supported an advanced study of near-term climate that includes the first-ever full production climate model simulations on the cloud. The study, led by scientists at the Global Climate Dynamics (GDC) laboratory at the National Center for Atmospheric Research (NCAR), is focused on important questions in near-term climate, but is also opening a powerful new way to accelerate climate science and expand access to tools and information for climate research.
Using ASDI-donated AWS cloud resources and technical support, including AWS scalable high performance computing (HPC), scientists at NCAR are performing an ensemble of 30 simulations of the Earth system over the period from 2035-2070 under a median scenario for warming and with simulations that include and intervention to reduce global temperature using NCAR’s CESM2 climate model. Simulation of the same same scenario will be undertaken by other modeling teams using other global climate models, including the UK Meteorological Office UKESM model.
Collaborating research teams will use the data generated through these simulations to study various impacts on Earth and human systems—including agriculture, drought, flooding, and human health—in various parts of the world. Among the first consumers of simulation data in the effort are climate scientists from countries in the Global South, through a program in partnership with The World Academy of Sciences (TWAS) and funded in part by SilverLining.
“Cloud access to climate models and datasets has huge potential to capacitate developing nation scientists to actually do good research, publish in good journals, and get into the academic playing field because we’re really not there yet compared, to say, the U.S. and European and Australian researchers.”
The datasets from the simulations will be made available on the cloud via ASDI, providing open access for researchers around the world, enabling a new level of transparency and equitable opportunity to study near-term climate impacts and potential responses. Over time, the team behind the effort plans to make many of the world's top global climate models, and related datasets, available to researchers around the world as user-friendly services on the cloud.