Verne Global

HPC | Research |

8 March 2018

How HPC is underpinning amazing advances in land, sea and space observation

Written by Spencer Lamb

Spencer is Verne Global's Director of Research and head's up our high performance computing work with European research and scientific organisations. He is also a member of the European Technology Platform for High Performance Computing (ETP4HPC).

In previous Verne Global blogs we’ve explored how HPC is being used throughout industry to make cars both faster and safer, to discover new materials and to advance bioinformatics. HPC has made many equally important contributions to the science of understanding our earth and the solar system around us, and it's an understanding that’s become increasingly important in the age of anthropogenic climate change.

With the assistance of HPC we can be armed with a deeper understanding of the complex, interconnected systems that surround us - and our relationship with these systems. Plus, we can mitigate the unintended effects of human behaviour, and develop new methods to take advantage of those systems. At the National Center for Supercomputing Applications located at the University of Illinois they are doing just this. Their Blue Waters supercomputer has been helping researchers better understand the natural environment for almost a decade.

Blue Waters's system works in partnership with NASA’s Terra satellite, which in its 17 years of operation has produced over 1.2 petabytes of data, yielding crucial insights about earth’s changing climate. Without the combined effort of both Terra and Blue Waters, research projects like the multi-angle imaging spectroradiometer (MISR), which measures aerosol pollution over land and water to enhance our knowledge of the lower atmosphere, would have been impossible.

The satellite, which has five instruments in all for probing the earth’s atmosphere, was the first to study the effects of pollution on human health, as well as how air pollution affects the interaction between earth’s land, oceans, and atmosphere. Another interesting HPC initiative at NASA aimed at understanding the natural environment better is DeepSat, an assembly of NASA machine learning frameworks designed to speed the analysis of large satellite images and help scientists better understand the relationship between carbon and vegetation. The DeepSat framework was used for NASA Ames Global Climate Change challenge, where it measured the effects of greenhouse gases on crop vegetation.

In Europe, similar initiatives include the Nucleus for European Modeling of the Ocean (NEMO), which is composed of three parts, the Océan Parallélisé (OPA), for studying blue ocean dynamics, Louvain-la-Neuve Sea Ice Model (LIM), for modeling sea-ice thermodynamics and dynamics, and the Tracer in the Ocean Paradigm (TOP) for the modeling of marine bio-geochemistry.

So far, the NEMO model has powered over 180 different research projects, and over 300 research papers. These cover a range of topics related to the marine ecosystem and its sustainability, from the surface salinity in the tropical pacific ocean — a key factor that influences ocean circulation — to the modeling of radiocarbons in the mediterranean sea (remnants of bomb tests in the 1950s and 1960s). Other notable projects in the field include the European Global Ocean Observing System (EuroGOOS), which while not a research institute itself facilitates cooperation among its 40 institutes from 19 countries, and helps coordinate and promote HPC-driven research efforts in the field of oceanography.

HPC doesn’t just help us understand the natural world, it can also help us to draw power from it. Carnegie Clean Energy is an Australian company developing a unique technology to harness the energy of ocean waves, and is currently the only company to operate a grid-connected wave energy project for four seasons. The solution, named CETO, has been in testing for over 10 years, and is capable of producing both zero-emission electricity and zero-emission desalinated water. The amount of potential energy in ocean waves is enormous, as high as 2 TW globally (1,012 watts). For comparisons sake, the average natural gas plant produce between 85 and 640 MW (106) of energy in a year. That means there could be the equivalent of up to 4,000 such plants in global wave energy output, if properly harnessed.

Until recently, developing devices that could withstand the corrosive environment of the ocean and efficiently capture wave energy has been a tough challenge. But using HPC and computational fluid dynamics (CFD), the team at Carnegie has been able to understand the extreme responses and loads that large waves put on wave energy devices, allowing them to validate their designs and move a step closer toward abundant, sustainable energy for coastal regions around the world.

The need for a better understanding of our natural environment is crucial to building a more sustainable future, a vision to which I am pleased to say Verne Global is wholly committed. Thankfully, being committed to sustainability doesn’t mean lowering your standards or expectations for computational power.

In locations like Iceland, abundant renewable energy gives oceanographers and other natural scientists an opportunity to utilise the latest HPC technologies, without damaging the very thing they’re hoping to study and preserve. If you’d like to find out more about how Verne Global helps its clients access sustainable HPC resources, we invite you to continue reading here.


Sign up for the Verne Global newsletter

Opinion, thought leadership and news delivered directly to your inbox once a month.