Advance Numerical Models

NCAR has a proud and unique tradition of collaboration with scientists from universities, national laboratories, and other research organizations to create, maintain, and distribute "community" climate and weather models that integrate the best available expertise across institutions. This effort extends from continuous basic theoretical and observational research, which underpins model development and evaluation, to provision of user support services. NCAR emphasizes ongoing advancement and redefinition of the atmospheric model components, integration and maintenance of complete models, and user support, while partners often lead in the development of additional model components such as sea ice and oceans. Scientists all over the world rely on these openly available tools to perform their research. In return, they contribute to model evaluation and development. During the past 30 years, community modeling has become one of the most important services provided by NCAR to university researchers and other interested users, and it will continue to be one of our top priorities. Over the next three to five years, NCAR will work with its partners to

• Improve the Community Climate System Model (CCSM) and the Weather Research and Forecasting (WRF) model
  • Develop and release CCSM version 4, with improved representation of the carbon cycle, the nitrogen cycle, and atmospheric chemistry
  • Develop and release Advanced Research WRF version 4, with improved data assimilation and microphysical and boundary-layer processes and exchanges
  • Continue development of WRF and CCSM variations, such as WRF-Chem, WRF-Fire, and the Whole Atmosphere Community Climate Model (WACCM)
• Develop and release new community modeling systems that incorporate new atmospheric components, offer state-of-the-art representation of a greater number of Earth system processes, and select and involve full testing of appropriate discretizations, grid refinement, and data assimilation approaches
• Continue research on data assimilation methods and parameterization, evaluate the impact of these new techniques on model performance, and continue enhancing the data assimilation capabilities of the NCAR community models
• Build innovative, extensible, and maintainable software design into the initial definition of Earth System Models. Determine a set of best practices for developing and modifying scientific model software, including requirement specifications, design reviews, and procedures for software testing and validation. Establish organizational incentives to follow these best practices
• Develop new software capabilities and infrastructure, including advanced software for massive parallelization, as well as new computational methods and experimental modeling platforms to advance numerical modeling and basic algorithmic and computational fluids research across the atmospheric and related sciences

Over the next decade, NCAR will move towards unified atmospheric and Earth system modeling. We will continue the integration of chemical, upper atmosphere, climate, and weather modeling, with particular attention to land surface modeling and dynamical processes work across our weather and climate groups. We plan to push ahead rapidly with use and evaluation of a first-generation combination of WRF with CCSM, the Nested Regional Climate Model (NRCM), described in more detail in the climate frontier section below. We will enhance our efforts in integrated assessment modeling (IAM), focusing both on improving capabilities and better integrating IAM approaches and experimentation with larger-scale Earth system modeling. Over the longer term, we plan to create a new, unified atmospheric modeling system capable of accurate prediction on time scales from hours to decades that is useful for both weather and climate prediction. This system will provide the basis for an advanced community Earth System Model that will include representation of oceans, sea and land ice, land cover, the upper atmosphere, and a large number of biogeochemical and ecological processes. We foresee extensive collaborations with existing and new partners in these efforts.