NGD Sub-Projects
The “Next Generation Development (NGD) Activities“ are sub projects that will develop new science and computational capabilities for E3SM versions v3 and v4 with a path for integration into E3SM within 5 years.
Targeted for use in Phase 3, E3SM versions v3 and v4 will feature notable developments in advancing the fully- coupled E3SM to answer the more demanding questions in the corresponding science campaigns. Advancing better simulations requires the performance expected from the 100 PFLOP and Exascale systems planned for DOE centers. Since the developments for the v3 and v4 model are longer-term and higher risk, they require less day-to-day oversight and management than the Core activities with nearer-term deliverables. Nevertheless, the research does target inclusion of new developments into the modeling system within five years, so they are not open- ended.
To address the v3/v4 science questions, four large subprojects and two smaller research tasks will advance the E3SM Project toward its v3 and v4 objectives.
- The Software and Algorithms for Exascale Subproject will pursue major algorithmic changes in model formulation and develop a sustainable software system in preparation for the leap to
- The Non-hydrostatic Atmospheric Model Subproject combines computational and scientific advances to extend the atmosphere model to simulation at convection-permitting and convection resolving scales, with horizontal grid resolutions between 1 and 5 km, bypassing the need for cumulus
- The Atmospheric Physics Subproject will improve parameterizations in the current E3SM Atmosphere Model to correct identified biases in the simulated climate for LR and HR
- The Land and Energy Subproject will improve and extend the existing Land model to further account for natural, managed and man-made systems and their interactions, with an emphasis on the relationships between the energy system and the water
Additionally, in two smaller efforts, E3SM will begin development of a coastal ocean model and couple an existing dynamic land ice model to the modeling system. These capabilities are important for addressing the science questions related to the biogeochemistry and cryosphere systems science drivers.
Note that new capabilities in ocean and ice modeling will also be advanced through “Ecosystem projects” supported by DOE’s Scientific Discovery through Advanced Computing (SciDAC) and Earth System Modeling programs. In sum, by initiating development for v3/v4 in parallel with the more modest v2 development and the v2 simulation campaign, major developments will become ready in Phase 3 to take advantage of Exascale computing to advance Earth system science and prediction.
