HPC used to develop new climate model
By Robert Roe
The US Department of Energy has launched a ten-year project, called Accelerated Climate Modelling for Energy (ACME), using the US's resources in high-performance computing (HPC) to speed up the development of fully coupled, Earth-system models for scientific and energy applications.
This announcement is the latest in a series of recent events highlighting the importance of HPC for climate research, as reported in Scientific Computing World. The 840-page US National Climate Assessment (NCA) published in May this year, warned that climate change was already adversely affecting the United States of America. The project used massive supercomputing simulations and an innovative peer-to-peer collaboration on data management software to complete the study, as information was shared through multiple universities and organisations across the world.
Following extreme cold weather last winter, another study presented by researchers using the Ohio Supercomputer Centre researched the frequency of Arctic Storms - finding they were more frequent than previously thought. Their findings where presented at the American Geophysical Union meeting in December 2013.
Berkeley Lab’s, Bill Collins is the ACME project's Chief Scientist and head of the Earth Sciences Division’s Climate Sciences Department at the US Berkeley National Laboratory. He said: ‘We need a new paradigm for how to develop and apply climate models to answer critical questions regarding the implications of our past and future energy choices for society and the environment. To address this critical need, ACME is designed to accelerate our progress towards actionable climate projections to help the nation anticipate, adapt to, and ultimately mitigate the potential risks of global climate change.’
The initial focus will be on three climate change science drivers and corresponding questions to be answered during the project’s initial phase:
- (Water Cycle) How do the hydrological cycle and water resources interact with the climate system on local to global scales? How will more realistic portrayals of features important to the water cycle (resolution, clouds, aerosols, snowpack, river routing, land use) affect river flow and associated freshwater supplies at the watershed scale?
- (Biogeochemistry) How do biogeochemical cycles interact with global climate change? How do carbon, nitrogen and phosphorus cycles regulate climate system feedbacks, and how sensitive are these feedbacks to model structural uncertainty?
- (Cryosphere Systems) How do rapid changes in cryospheric systems, or areas of the earth where water exists as ice or snow, interact with the climate system? Could a dynamical instability in the Antarctic Ice Sheet be triggered within the next 40 years?
Over the next ten years the project aims to conduct simulations and modelling on the most sophisticated HPC systems available. This will initially involve 100+ petaflop systems located at the US Department of Energy (DOE) Office of Science Leadership Computing Facilities at Oak Ridge and Argonne national laboratories – and use exascale systems as they become available in the future.
The model will also be optimised for deployment on the National Energy Research Scientific Computing Center (NERSC), which is located at Berkeley Lab.
A report entitled ‘Accelerated Climate Modeling for Energy (ACME) Project Strategy and Initial Implementation Plan’ states ‘A goal of ACME is to simulate the changes in the hydrological cycle, with a specific focus on precipitation and surface water in orographically complex regions such as the western United States and the headwaters of the Amazon.’
To address biogeochemistry, ACME researchers will examine how more complete treatments of nutrient cycles affect feedback within the carbon–climate system, with a focus on tropical systems; and investigate the influence of alternative model structures for below-ground reaction networks on global-scale biogeochemistry–climate feedbacks.
For the cryosphere, the team will examine the near-term risks of initiating the dynamic instability and onset of the collapse of the Antarctic Ice Sheet due to rapid melting by warming waters adjacent to the ice sheet grounding lines.
The experiment would be the first fully coupled global simulation to include dynamic ice shelf–ocean interactions for addressing the potential instability associated with grounding line dynamics in marine ice sheets around Antarctica.
Other Berkeley Lab researchers involved in the program leadership include Will Riley, an expert in the terrestrial carbon cycle and co-leader of the Biogeochemical Experiment Task Team. Hans Johansen, a computational fluid dynamicist, is co-leader of the Computational Performance Task Team.
Initial funding for the effort has been provided by DOE’s Office of Science. Eight Department of Energy national laboratories (Argonne, Brookhaven, Lawrence Livermore, Lawrence Berkeley, Los Alamos, Oak Ridge, Pacific Northwest and Sandia) are combining forces with the National Center for Atmospheric Research, four academic institutions and one private-sector company in the new effort.
Robert Roe is technical writer for Scientific Computing World