The Department of Energy’s (DOE) Biological and Environmental Research Program supports climate change research on a global scale, and seeks to generate the scientific knowledge base needed to inform public discussion, evaluate energy policy options, and provide tools for planning, adapting, and mitigating climate change. The ultra-scale visualization climate data analysis tools (UV-CDAT) were established to help address these needs.

UV-CDAT accomplishes a task that has never before been attempted for or by the climate community: the integration of more than 70 disparate scientific software packages and libraries for large-scale data analysis and visualization. Team members worked across institutional boundaries to develop and integrate software packages under a single framework that facilitates climate research and enables scientists to use them with little or no effort. The result is a powerful toolset that aids climate researchers with solving their most complex data analysis and visualization challenges.

The integrated, cross-institutional partnership responsible for developing and refining UVCDAT is unique in its breadth and depth of expertise. Four DOE national laboratories— Lawrence Berkeley (LBNL), Lawrence Livermore (LLNL), Los Alamos (LANL), and Oak Ridge (ORNL)—focused on the development of large-scale parallel analytics and the diagnosis of climate model simulation and observational post-processing. Two universities—Polytechnic Institute of New York University and the University of Utah—handled provenance and workflow, and streaming visualization. From the private sector, Kitware worked on software processes, cross-platform build and test suites, back-end visualization capabilities, and development of a spatiotemporal parallel pipeline; and Tech-X supplied reprojection libraries. The National Aeronautics and Space Administration’s (NASA) Goddard Space Flight Center provided three-dimensional (3D) data visualization and UV-CDAT tutorials, while the National Oceanic and Atmospheric Administration (NOAA) provided temporal and spatial regridding tools.

UV-CDAT version 1.0 was officially released in early 2013. UV-CDAT version 2.0.0, released in October 2014, is in production use for DOE and NASA projects, most notably DOE’s Accelerated Climate Modeling for Energy (ACME) project and NASA’s Climate Model Data Services. The UV-CDAT software source code is open source under the Berkeley Software Distribution license agreement to allow for community development and partnership.

Today, UV-CDAT provides users with access to more analysis and visualization products than any other single source. The tool offers unparalleled capabilities for climate scientists to address big data analytics, sensitivity analyses, heterogeneous data sources, and multiple disciplinary domains; and it incorporates existing software components in combinations that were previously difficult or even impossible.