While the horizontal grid is fixed, the model’s vertical levels depend on atmospheric pressure and their height can vary both by location and over time. “RenderMan requires that those volumes be uniform rectangular volumes, but the GEOS model is not structured that way.” “You are creating little volumes in space and assigning them an optical density and color,” said SVS director Horace Mitchell, who collaborated with Shirah on the visualization. Central to the carbon dioxide project was Pixar RenderMan, the leading tool for a 3D computer graphics technique called volume rendering. In translating such data into compelling visualizations, the SVS uses several of the same software tools as feature film special effects companies. Running GEOS at that resolution, combined with the OCO-2 observations, yielded about 1 terabyte of total output. Weir set GEOS to use grid boxes 50 kilometers wide throughout the atmosphere. The model run consumed 336 cores of the NCCS Discover supercomputer for 22 days. This study repeated those steps for September 2014 to September 2015-capturing a 3D documentary of the ebbs and flows of carbon dioxide over a full year. As in its weather forecasting research, the GMAO assimilates the OCO-2 data, runs the model forward until the next assimilation point, assimilates data again, and so on. The GEOS modeling suite includes a constituent data assimilation system designed for ingesting carbon and aerosol observations. “We add the vertical information” by assimilating the OCO-2 data into the Goddard Earth Observing System (GEOS) computer model of the atmosphere. While providing a unique global perspective, OCO-2 concentration data shows only a top-down view of “how much carbon dioxide a column of air holds,” explained Brad Weir, a GMAO data assimilation expert affiliated with GESTAR/Universities Space Research Association. JPL processes the raw radiance data into a concentration product for researchers to use. OCO-2 takes more than 100,000 carbon dioxide observations daily. The satellite data comes from NASA’s Orbiting Carbon Observatory-2 (OCO-2), built and operated by the Jet Propulsion Laboratory (JPL). For the first time, we have satellite data, a model, and visualization come together to show that science to the public.” “Some of this is science we have known for a long time, including that carbon dioxide can be carried by winds and transported over long distances. “It is really exciting to see,” said Lesley Ott, GMAO carbon cycle scientist. “Seasonal Changes in Carbon Dioxide” belongs to a group of visualizations developed by the SVS in a collaboration with Goddard’s Global Modeling and Assimilation Office (GMAO). Visualization by Greg Shirah and Horace Mitchell, SVS Research by Lesley Ott, Steven Pawson, and Brad Weir, GMAO. This visualization represents a combination of OCO-2 satellite and GEOS model data yielding the first 3D picture of carbon dioxide throughout the Earth’s atmosphere. This honor by their computer graphics peers caps off a year that saw the visualization receive the Professional Winner award in Science Magazine's Data Stories contest. Shirah and colleagues are thrilled that their work is the sole data visualization being shown on a massive screen alongside artistic videos and feature film special effects sequences. The annual conference draws more than 14,000 attendees who line up early for seats at the Electronic Theater. ![]() “SIGGRAPH is one of the top venues for computer graphics in the world,” said Greg Shirah, lead visualizer on the project from NASA Goddard Space Flight Center’s Scientific Visualization Studio (SVS). The first screening is tonight at the Los Angeles Convention Center. From hundreds of submissions the visualization is among only 25 videos selected for the prestigious Electronic Theater. 3D Carbon Dioxide Visualization Appears in Prestigious FestivalĪ scientific visualization of 3D global carbon dioxide data computed at the NASA Center for Climate Simulation (NCCS) is appearing at this week’s SIGGRAPH 2017 Computer Animation Festival in Los Angeles, California.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |