Using the NASA Chemistry Climate Model for Future Climate Simulations
Overview
Earth’s climate is changing due to the accumulation of human-produced greenhouse gases (GHGs) such as carbon dioxide (CO2) in the atmosphere. There are many impacts of the increased abundance of GHGs, most notably a long-term increase in the average temperature at the Earth’s surface, known as global warming. The effects of global warming are felt in many facets of the climate system, including in the changes of global weather patterns, the retreat of glaciers and sea level rise, and changes in the locations, amounts, and duration of precipitation. These shifts are occurring alongside other changes in the composition and chemistry of Earth’s atmosphere, such as the occurrence of the Antarctic ozone hole, discovered in the 1980s.
A team of researchers at NASA’s Goddard Space Flight Center are using global Earth-system models that include interactive chemistry, called chemistry-climate models (CCMs), to investigate the evolution of the stratospheric ozone layer and its response to the reduction of emissions of ozone depleting substances following the Montreal Protocol treaty instituted in 1989. The emission changes that go into the simulations need to be accounted for in concert with other developments, such as those caused by the rise of GHGs in the atmosphere. In order for the team to have confidence in the projections generated by their CCM, they must first have confidence in its capabilities at simulating the evolving climate state, so they are systematically evaluating their CCM for its ability to simulate basic aspects of GHG-driven climate change.
Project Details
The CCM uses NASA’s Goddard Earth Observing System (GEOS) model, running GEOS as a coupled atmosphere-ocean model, building off of previous work producing coupled sub-seasonal to seasonal scale weather forecasts. The CCM evaluation includes two parts. The first is a pre-industrial simulation in which the oceanic and atmospheric state is set to conditions expected for the year 1850 and the model is run for 150 years, assuming fixed pre-industrial CO2 abundance and pre-industrial emissions of gases and aerosols to develop an approximately equilibrium state of the atmosphere and ocean. The second part of the evaluation starts from the near-equilibrium state of the pre-industrial simulation, but instantaneously quadruples the abundance of CO2. This system is then run for 150 years until it also reaches a new quasi-equilibrium state that reflects a climate system warmed by the increased CO2. These two simulations are used to calculate the equilibrium climate sensitivity (ECS) of our model, a metric that allows us to compare the climate system modeling capabilities of the GEOS CCM to other climate models.
Results and Impact
The ECS of the GEOS CCM is found to be 2.6 °C (nearly 5 °F), which is the calculated increase in the Earth’s average surface temperature under a doubling of atmospheric CO2 abundance from pre-industrial levels—a state expected to be met in the real world around the year 2060. This ECS—and other diagnostics computed from these simulations—is well within the range of previously evaluated climate models, which enhances the team’s confidence in subsequent projections with the GEOS CCM.
Why HPC Matters
The high-performance computing resources at the NASA Advanced Supercomputing (NAS) facility at Ames Research Center, particularly the Pleiades supercomputer, are essential to carrying out this work. These simulations made use of almost twenty 40-core CPU nodes apiece, which allowed the team to run a year of simulated model time in a little over a day. The 300 years of simulated model time here would have taken almost twenty years of real-world time to run on a typical desktop computer—assuming it could provide the memory and disk space needed to perform the calculation—while the team achieved results with Pleiades in a little more than six months.
What's Next
This work is foundational to future work with the GEOS CCM by establishing the basic credibility of the coupled atmosphere-ocean model for simulating future climate change. Future work will introduce detailed chemistry mechanisms and impose time varying changes in CO2 abundance and aerosol and gas emissions to more realistically produce changes in ozone and atmospheric composition in future climate models.
Presentation
Researchers
- Peter Colarco, NASA Goddard Space Flight Center Please enable Javascript to see email address.
- Sarah Strode, Morgan State University / NASA Goddard Space Flight Center Please enable Javascript to see email address.
- Feng Li, University of Maryland, Baltimore County / NASA Goddard Space Flight Center Please enable Javascript to see email address.
- Andrea Molod, NASA Goddard Space Flight Center Please enable Javascript to see email address.
