Due to the lapse in federal government funding, NASA is not updating this website. We sincerely regret this inconvenience.
Visualization Theater Renamed for Longtime Collaborator Piers Sellers
July 17, 2017
To reach a safer future, we will need the resources of everybody here. The scientists, the policy makers, and the industrialists, all working together towards a common goal. And that goal is a planet that can continue to support life, including all of us.
— Piers Sellers (1955–2016)
This quote greets visitors to the Dr. Piers J. Sellers Data Visualization Theater, formally dedicated by NASA Goddard Space Flight Center officials on July 17. Renaming the NASA Center for Climate Simulation (NCCS) Data Exploration Theater for Sellers is a fitting tribute to the late scientist, astronaut, and science manager whose collaborations with the NCCS and related organizations spanned decades.
A First-of-its Kind Computer Model
As the science of global warming emerged during the 1980s, researchers took interest in applying their computer models to timeframes longer than 3-to-5-day weather forecasts. Sellers had moved to Goddard from his native England in 1982 and soon recognized that modeling the atmosphere was not enough. Realistic climate simulation meant adding the biosphere and its living vegetation.
“Piers had this idea of putting life in a model,” said research colleague Lahouari Bounoua, who is currently associate chief of Goddard’s Biospheric Sciences Laboratory. “Plants are very useful to the climate because they have many functions.” Among others, they use solar energy, pull water from the soil, capture carbon dioxide from the air, and produce oxygen and water vapor. The process of photosynthesis contributes to cooling the atmosphere through transpiration—leaves diffusing water vapor through their pores.
These processes and much more were represented by mathematical equations in the first-of-its kind Simple Biosphere Model (SiB), announced to the science world in a 1986 Journal of the Atmospheric Sciences paper (which has more than 2,400 citations). The model predicted temperature, water at the ground and above in the vegetation canopy, soil moisture, and other variables. From the start, Sellers and co-developers designed SiB for use within atmospheric general circulation models (AGCMs). Fulfilling that plan took a larger team of scientists.
Coupling Models for Climate Projections
Enter NASA’s Earth Observing System (EOS) Program. In the early 1990s, EOS offered 10-year research grants to help prepare for launching the EOS satellite fleet and effectively use the data from satellites once in orbit. Proposing to be one of these Interdisciplinary Science (IDS) teams was a natural fit for Sellers, “who had connections almost everywhere,” Bounoua said. “He had a vision of the science he wanted to do but knew the limitations of the tools that we had.” Since Hal Mooney’s group at Stanford University was doing complementary research, the Goddard and Stanford groups merged, with Sellers as principal investigator (PI) and Mooney as co-principal investigator. Their proposal secured funding, and the “Sellers-Mooney IDS” encompassed 15 groups at laboratories and universities across North America and one in France.
A centerpiece of the Sellers-Mooney IDS was running simulations to project how increased carbon dioxide in the atmosphere would affect climate in future decades. The NCCS stood ready to host the simulations, but first the researchers had to develop a new version of SiB and couple it to an AGCM—a feat no one had attempted. Joining Sellers in this endeavor were Goddard co-investigators Jim Collatz and Bounoua, who joined the team in September 1993.
As Bounoua explained, the original SiB was a model for micro-meteorology, understanding “what was going on at very local scales, like a farm or garden. When you go to an AGCM, your vision is global.” Developing SiB2 included changing the physics and aerodynamic turbulence at the surface, where land and atmosphere interface, and improving many aspects of the carbon, water, and energy cycles.
The AGCM came from a key IDS participant: Dave Randall’s group at Colorado State University (CSU). It took 2 years to couple SiB2 to the CSU AGCM and make the models work together. Bounoua recalled daily group lunch meetings with Sellers, who carried a Cinderella lunchbox given to him by his daughter, and frequent middle-of-the-night code debugging sessions. While there was pressure with such a large group relying on them, “we had a good time, and if I had it to do over again, I would do it in a minute,” he said.
Coupled models in hand, the team proceeded to run five simulation experiments at the NCCS. Inspired by his role as EOS Terra satellite project scientist, another Sellers innovation was to use a myriad of vegetation observations to initialize the simulations. The simulations ran for 10 years of spin-up to reach equilibrium and then projected the climate 30 years into the future. One control experiment left carbon dioxide at starting levels. The other four experiments doubled carbon dioxide for 1) vegetation only, 2) vegetation only with a down-regulated carbon uptake rate, 3) solar radiation only, and 4) solar radiation and vegetation with a downregulated carbon uptake rate.
These global simulations needed all the computational horsepower that the NCCS could muster. The NCCS of the mid-1990s, when it was known as the NASA Center for Computational Sciences, had a CRAY C98 supercomputer with 6 vector processors and 2 gigabytes of memory. It offered then-world-class performance of 6 gigaflops, that is 6 billion floatingpoint operations per second. “There was nothing else that could run the model at the time,” Bounoua said. “I was the biggest user of the NCCS for 2 years in a row!”
The experiments showed that, in a world with doubled carbon dioxide, transpiration will drop and accordingly air temperature will increase over the tropical continents, amplifying the warming due to solar radiation. As it shared the first results from a climate model incorporating vegetation and photosynthesis, the March 1996 Science paper about these experiments has been cited more than 600 times.
The April 1996 Journal of Climate paper on SiB2 itself has more than 1,200 citations. Bounoua and Collatz pushed Sellers to release SiB2 to the public and supported it for 2 years. Researchers from leading weather and climate groups came to Goddard to learn the model. “People took that science and implemented it in their local models, from the U.S., to Europe, to China and Japan,” Bounoua said. “Everyone is using the theory that we implemented first.” Twenty years later, SiB2 and adaptations are still being used at NASA, CSU, and worldwide.
A few months after those seminal papers, Sellers left Goddard to gain a space-based view of the world as an astronaut. He continued collaborating with the IDS team (with CSU’s Randall as PI) from Houston for a few years. Later, during three Space Shuttle missions he helped to build the International Space Station (ISS). Yet, studying the Earth was never far from Sellers’ mind, and he pushed to get science instruments on the ISS.
Goddard Take Two
Sellers came back down to Earth and Earth science with a return to Goddard in 2011. With his seemingly boundless energy, he took on dual roles as deputy director for the Sciences and Exploration Directorate (SED) and director of the Earth Sciences Division. Sellers quickly impacted the NCCS and its parent organization, the Computational and Information Sciences and Technology Office (CISTO).
Then SED Director Nick White wanted to elevate information technology (IT) and supercomputing. Instead of reporting to the assistant director, White had Phil Webster, CISTO chief and NCCS project manager, report to Sellers. “The first time I met him, Piers came by my office, sat down, and said, ‘Tell me what you do,’” Webster said “So, I showed him around for a whole morning.”
Having been a major NCCS user, Sellers the science manager was keen on growing Goddard’s Earth system modeling capability, although not for its own sake but as a tool needed by NASA missions. He also began to see IT more broadly as integral to success. “Piers was a big advocate of a federated model, where decisions are made collectivity,” Webster said. It takes the Flight Projects Directorate (400), Applied Engineering and Technology Directorate (500), and SED (600) working together to make a mission happen. Sellers pushed Webster to extend this federated model to IT decisionmaking and bring in the Information Technology and Communications Directorate (700) as a partner.
Reflecting as he nears retirement, Webster said, “The things I was able to accomplish within IT governance and advancing the state of the architecture at Goddard were due to his support, enthusiasm, and kicking me in the butt now and again.” Today, “the federated model is working quite well. This wasn’t his area of expertise at all, but Piers saw how it would improve processes on Center to do more science with the same amount of money,” Webster added. “He was the best boss and most influential boss I ever had.”
Sharing NASA Science
At Goddard and beyond, Sellers became a prolific presenter of NASA Earth science. His main presentation collaborators were Steve Graham of the NASA Science Communications Support Office and Lori Perkins, a visualizer in CISTO’s Scientific Visualization Studio (SVS). Perkins first worked with Sellers the astronaut on two presentations for NASA’s National Air and Space Museum events. She was impressed by his flexibility to use alternative visuals because they made the points better. “He was willing to listen and believed that good ideas could come from anywhere,” Perkins said.
Sellers the Goddard manager gave dozens of presentations a year on the Data Visualization Theater’s 20- by 7-foot hyperwall, a joint effort between the SVS and NCCS. Audiences were as diverse as foreign dignitaries, members of Congress, fellow scientists, interns, and school children. Perkins noted that he would always take time to meet interns and not only learn about what they were doing but also ask if they were having fun.
“The talks were not just entertainment,” Perkins stressed. “He would use them as a first step to a collaborative effort.” On one visit, he was pushing the Vietnamese to do citizen science, and they signed up for the GLOBE Project. A few months after meeting, the German Aerospace Center (DLR) signed a Space Act Agreement for the GRACE satellite follow-on.
While covering the gamut of NASA Earth science research, Sellers’ “talks always had a modeling component,” Perkins said. “The last thing he wanted people to think, you launch the spacecraft, collect the data, and you’re done. You have to understand the data when our planet changes.” As Webster pointed out, he never failed to mention that “the only way you can see into the future of the climate is through these models.”
Our climate future became vital for Sellers when he received a diagnosis of Stage IV pancreatic cancer in October 2015. In his final year of life, he shared a grounded optimism in media interviews. Most memorable for Perkins was a visit by actor Leonardo DiCaprio for his documentary Before the Flood.
Although quite ill at the time, Sellers spent 5 hours with DiCaprio at the hyperwall. Perkins remembers DiCaprio saying, “I’ve been studying this for 2 years and didn’t realize how complicated the Earth as a system was, all the connections.” Sellers replied, “I’ve been doing this for 30 years, and we learn something new every day. Our planet is a delicate balance, and understanding that balance takes time.” “He knew that even if he lived another 20 years, he needed to inspire the next generation to do that work,” Perkins said.
Jarrett Cohen, NASA Goddard Space Flight Center
More Information
NCCS Dr. Piers J. Sellers Data Visualization Theater
Goddard Dedicates Hyperwall to Late Climate Scientist, Astronaut
Research Papers
Sellers, P.J., Y. Mintz; Y.C. Sud, A. Dalcher (1986), “A Simple Biosphere Model (SiB) For Use Within General-Circulation Models,” Journal of the Atmospheric Sciences, 43, 505–531.
Sellers, P.J., L. Bounoua, G.J. Collatz; D.A. Randall, D.A. Dazlich; S.O. Los, J.A. Berry, I. Fung, C.J. Tucker, C.B. Field, T.G. Jensen (1986), “Comparison of Radiative and Physiological Effects of Doubled Atmospheric CO2 on Climate,” Science, 271, 1402–1406.
Sellers, P.J., D.A. Randall, G.J. Collatz, J.A. Berry, C.B. Field, D.A. Dazlich, C. Zhang, G.D. Collelo, and L. Bounoua (1996), “A Revised Land Surface Parameterization (SiB2) For Atmospheric GCMs .1. Model Formulation,” Journal of Climate, 9, 676–705.
