Our earliest understanding of carbon dioxide (CO2) probably came from classes during elementary school. As we might recall, we learned that the CO2 humans breathe out is actually an important element which is essential to plant life; you might also have learnt that the bubbles coming out of soft drink also is CO2. However, this gas, which is odorless, tasteless, and colorless under atmospheric pressure is now understood to be a key factor of global warming.
The Innocent Culprit of Global Warming
As global temperatures keep rising, it remains uncertain if human-induced carbon emission is the sole reason. However, in general, scientists believe that the increase in temperatures across the earth over the past 50 years is significantly related to carbon emissions resulting from human activities. In 2007, UN’s IPCC (Intergovernmental Panel on Climate Change) revealed in its report that the evidence of global warming was unequivocal, and that human activities were very likely the driving force behind the climate change over the past 50 years. It was also revealed in Al Gore’s book, “An Inconvenient Truth,” that the increase in CO2 concentration on Earth over the past 6.5 hundred thousand years is positively correlated to the rise in global temperature. Since the Industrial Revolution, the concentration of carbon in the air has multiplied year over year, and the global temperature has also seen a corresponding gradual increase. Because of the century-long lifetime of CO2 molecules, the CO2 from the past 100 years still affects us today, and the CO2 we generate now will also likely continue to impact Earth a century later.
To avoid the catastrophic climate change caused by global warming, the IPCC pointed out in its report that countries of the world must immediately keep the increase of temperature within 2˘J and keep the concentration of CO2 under 450ppm, so that people could prevent further climate change and the associated storming rains, scorching summers and other climate-related catastrophes such as rises in sea levels from happening. Presently, in efforts to reduce carbon emissions, governments are developing renewable energies such as wind power, hydropower, and solar power. Another viable option being researched is the technology-assisted storage of CO2.
Carbon Storage – the New Savior in “Clean Coal”
According to the Energy Technology Perspective 2010, newly published by the International Energy Agency (IEA), in order to keep the rise in global temperature under 2˘J, the utilization of carbon capture and storage (CCS) technology accounts for 19% of likely approaches for emission reduction. It indicated the significance of CCS in controlling global warming. The concept behind carbon storage is the placement of CO2 into a “container” in such a way that CO2 will remain stored and not released to the atmosphere. The so-called “container” primarily refers to the sea and underground reservoirs.. Regarding the comparison of storage methods, in 2003, the journal Science presented an article discussing the advantages and disadvantages of injecting CO2 into the sea and/or underground reservoirs and pointed out that underground storage is the most feasible approach.
In fact, storing carbon underground is not a new concept or technology. In the 1970s, oil and gas exploration companies had employed a similar storage technique, injecting CO2 into underground oil wells to reduce the viscosity of petroleum and thus increase the yield. Nowadays, geological storage refers to liquefying CO2 after capturing CO2 emitted from fixed sources (e.g. the end of pipelines for a thermal power plant) and injecting the liquid CO2 into a high-porosity stratum at depths deeper than 800m below the ground surface via transporting pipelines and injection wells. Research showed that at a depth of greater than 800m, the density of CO2 is the greatest and the volume is the smallest, so that it can more stably remain trapped in pores for a long time and possibly even dissolve in deep underground water or undergo mineralization reactions with the minerals in the stratum to enhance the effect of carbon reduction.
Site Selection by Supercomputer for Geological Storage of CO2
Prior to carrying out CO2 storage, selecting a good storage site is very important. Usually, in terms of stratum composition, a good site must contain capping rock formations and a viable gas reservoir. The capping rock functions as “lid” capable of effectively trapping the injected CO2 rock to avoid polluting water resources in shallow strata. Additionally, the gas reservoir must possess appropriate porosity and permeability to be able to provide sufficient space and transitivity for the easy injection of CO2 and to reduce the burden on the injection well’s pump and motor. If geological exploration, field investigation, and field testing were entirely depended upon, the process would be prohibitively time-consuming with additional difficulty in predicting the future impact on the regional environment.
Besides the structural considerations mentioned above, injection of CO2 into deep strata implicates the underground migration of the massive amount of CO2 over a large area and an extended period of time. The migration time possibly is longer than a thousand years and the area is large – possibly extending 10 to 100 kilometers upward. Meanwhile, the reactions and changes that CO2 might undergo with underground water and geological strata, and mechanisms such as hydrodynamic effects, residuals, and solubility and mineral trappings are very complicated problems of multiphase flow and transport. A possible means of understanding the mechanics of CO2 storage is through the use of supercomputing technology. If a supercomputer were used to assist in drawing up many likely scenarios, conducting simulations and predicting the impact of prolonged storage on the environment, it would be conducive to speeding up the search for storage sites.
Regarding the computational requirements in searching for storage sites, we take Tokyo Bay, Japan as an example. In 2008, by taking the stratum underneath Tokyo Bay as the supposed storage site, Taisei Corporation of Japan cooperated with Japan’s Agency for Marine-Earth Science and Technology (JAMSTEC) and Lawrence Berkeley National Laboratory (LBNL) to set up 10 injection wells in a block of 70x60x2.5km3, planning to continuously pump down 1 million tons of CO2 per well per year for 100 years and observing for changes occurring over the following 1000 years. This particular geo-sequestration scenario involved nearly 10 million computational cells using the Earth Simulator (Japan’s high performance supercomputer ranked No.1 in the world between 2002 and 2004). The calculation employed 1,024 processors requiring two days at nearly 8 TFlops per second.
HPC Opening the Prolog to Carbon Reduction in Taiwan
Since 2007, NCHC has committed to the research of high performance computing for CO2 geological storage. In 2009, NCHC collaborated with LBNL to organize education programs on TOUGH2-MP (software for storage simulation) and conduct trend forums, gaining enthusiastic participation of Central Geological Survey, MOEA, Taiwan Power Company (TPC), CPC Corporation, academia, and research institutes. Furthermore, in 2010, NCHC formally worked with TaiPower Research Institute and Sinotech Engineering Consultants, Inc. to initiate Taiwan’s practice of CO2 sequestration using HPC (High Performance Computing).
In the collaboration with TPC, NCHC primarily provides assistance in the HPC area, including the enhancement in computational efficiency of TOUGH2-MP and the establishment of our country’s own parallel computing platform for CO2 geological storage. TOUGH2-MP, a simulation program developed by LBNL, is used by more than 300 laboratories worldwide to pursue research on multiphase flow in porous medium and CO2 geo-sequestration. However, this software fails to provide a user-friendly operating interface. In dealing with complex conditions, such as the irregular borders of the actual storage site and geological heterogeneities, it is unable to support local grid refinement (LGR) or to solve the problem of incompatibility with post-processing tools (e.g. Tecplot and Sufer).
Therefore, NCHChas invested time into mesh generation research to develop techniques for improving grid density at specific positions (e.g. the CO2 injection well area) to increase the accuracy of simulation. Simultaneously, NCHC is also developing interfacing tools for graphing concentration fields and flow fields as required based on the computational results of TOUGH2-MP. Now, significant improvements have been made on the dimensions of simulation (including the size of the site, the number of computational cells and the simulation time), the complexity of hydrologic and geological conditions, and the efficiency of data analysis and processing.
Based on the experience gained from collaborations with TPC and others, NCHC has also established a prototype parallel computing platform for CO2 geo-sequestration. In the future, NCHC expects to use supercomputers as the vehicle to link pre- and post-processing software including parallelized numerical simulators and mesh generators, and to develop a public HPC platform for industrial, academic and research sectors in related research and industrial applications. Moreover, to enhance the simulation efficiency, we are also surveying the viability of improving TOUGH2-MP by applying newly emerged hardware/software acceleration methods (e.g. GPU acceleration).
In Al Gore’s Footsteps
Carbon reduction has become an important global issue, but still requires more effort in gathering support from all sectors of society in accelerating the progress in slowing down global warming. In An Inconvenient Truth, former U.S. Vice President Al Gore repeatedly presented scientific arguments to persuade governments and consortiums to pay more attention to global warming. Walking in his footsteps, we will do our utmost to allow people to effectively utilize supercomputing and HPC for the exploration of CO2 geological storage. To achieve this goal, we will require computing scales of the order of Teraflops, if not Petaflops, hence the need for even more extensive research into increasing the efficiency of such HPC research.
Hopefully in the near future, detailed demonstrations of Taiwan’s CO2 geo-sequestration scenarios, integrating field investigation and test data, will be presented by the supercomputers of the NCHC, enabling group discussions among experts, scholars, and environmental citizens to explore new ways for carbon reduction.