Saturday, June 24, 2017
 

Key Questions

  • How can modeling be best integrated and improved with respect to skill, efficiency, and adaptability?

Objective 1

A framework for linking, coupling, and nesting models.

NOAA requires a framework for connecting and optimally exploiting its environmental models. This framework needs to provide standards for interoperability, the exchange and upgrade of model components, a modeling structure to address the spectrum of spatial and temporal scales, coupling across physical domains, connectivity between physical and ecosystem modeling, and effective data assimilation. Establishing an Earth System Prediction Capability (ESPC) will extend predictive capability from days to decades based on that enhanced understanding, and help identify and quantify uncertainty and risk. This objective aims to improve model nesting capabilities that optimize modeling, data assimilation, and prediction between different spatial/temporal scales and coverage, as well as enabling a robust operations-to-research (O2R) environment that facilitates research and subsequent transitions to applications and operations.

R&D Targets:

  • Develop Earth System Modeling Framework (ESMF) connectivity coupling the atmosphere, ocean, land, and ice at global and regional scales for NOAA’s operational numerical models, serving as an initial NOAA ESPC capacity

  • Initialize modeling techniques and capabilities for coupling physical domains and ecosystem domains

  • Prototype optimal nesting between NOAA’s operational global, regional, and coastal ocean models, as well as relevant operational ecological models

Objective 2

Advance Earth system modeling development, addressing underlying processes and relationships, seamless connectivity across spatial and temporal scales, and coupling across domains.

NOAA requires development, testing, and transition to applications and operations of state-of-the-art Earth system models that address fundamental processes and relationships relevant to changes in the ocean’s physical and biological state. Processes of interest include forcing, fluxes, and feedbacks across ocean, atmosphere, cryosphere, and land interfaces, extreme weather events, feedbacks in the global carbon and other biogeochemical cycles, stratospheric and tropospheric changes and interactions with climate, Arctic predictions and climate-related changes, sea-level rise, decadal predictability, and space weather prediction. A key element of this objective is moving toward robust ecosystem modeling.

 R&D Targets:

  • Extend NOAA’s radiative transfer modeling capability to additional satellite sensors while demonstrating improved surface emissivity modeling, increased accuracy, and more efficient computation

  • Demonstrate skilled modeling of sea-ice, particularly for the Arctic region, incorporating improved modeling of ice processes, e.g. ice melt, and coupling with atmospheric and ocean forcing

  • Demonstrate a data-assimilating common-core surface and subsurface transport, mixing and fate (e.g., dispersion) modeling capability for ocean, coastal, and local scales

  • Prototype data-assimilating hydrodynamic modeling capabilities that include nutrients, phytoplankton, zooplankton, and detritus (NPZD), and geochemistry, on relevant temporal and spatial scales for the oceans and coasts

  • Prototype modeling for understanding the factors affecting ocean and coastal ecosystems structure, function, and dynamics, building on initial NOAA capacity for projecting significant environmental changes over the next several decades and early warnings about threats to critical coastal and marine ecosystem services

Objective 3

Establish quantified uncertainties for NOAA’s predictions and projections.

Models introduce uncertainty into predictions/projections due to how input data are used, how conditions and processes are modeled, and how approximations are employed. Consequently, modeling uncertainties need to be determined and integrated with observation measurement uncertainties to establish overall prediction/projection uncertainty. Result differences due to model differences, as seen through ensemble prediction, are a measure of the uncertainty associated with specific predictions/projections. The integration of observation and model uncertainties is required to determine the uncertainty of predictions/projections and to provide a more useful decision-making product.

 R&D Targets:

  • Quantify model uncertainty and skill for all NOAA operational models and forecast products, including quantified understanding of the uncertainties between different climate models in their projections of sea ice, atmosphere-ocean-cryosphere interactions, and ocean heat storage

  • Develop an initial capability to produce objective uncertainty information for models and products from the global to the regional scale

  • Prototype an ensemble prediction system for evaluating probability at multiple spatial and temporal scales

  • Improve probabilistic predictions, with routine evaluations of the skill and accuracy of operational wind, solar, and moisture forecasts

  • Develop raw and post-processed probabilistic products easily accessible at full spatial and temporal resolution

Objective 4

Advance data integration and assimilation into Earth system modeling

Data assimilation is a critical element of any environmental modeling system, anchoring model results with observations to enhance representativeness and predictive skill, extracting return on NOAA’s investments in its observing system. New data assimilation techniques, new instrumentation and sources, and non-standard or intermittent data, e.g., unmanned aerial and ocean vehicles, integrated ocean observing system instruments, and instrumented marine mammals, require R&D for transitions into NOAA applications and operations. NOAA will conduct research on data assimilation for improved representation and predictive skill of: high-impact events (e.g., tornadoes, hurricanes, severe storms, floods/droughts, poor air quality, winter weather, fire weather, marine and coastal weather, short-term climate variability); economic sectors requiring significantly improved forecast services (e.g., aviation, emergency management, renewable energy); aviation-relevant issues (e.g., convection, ceiling, visibility); and fine-scale predictions of near-surface conditions.

R&D Targets:

  • Prototype data assimilation methods for: coupled modeling; two-way nested modeling; and transport and fate modeling

  • Develop hybrid and ensemble assimilation methods for standard, non-standard, and intermittent observations

  • Assimilate non-NOAA IOOS, private sector, and international GEOSS data, particularly non-satellite data, into NOAA research and operational models, addressing feasibility, data quality, skill improvement

  • Demonstrate enhanced ocean data integration and assimilation for current and emerging data types, specifically salinity, ocean color parameters, synthetic aperture radar parameters (e.g. high-resolution winds, swell spectra), HF radar, freshwater inputs (riverine), and biogeochemical data

  • Prototype integration of newly available ice thickness data and improved (automated) ice-coverage data within NOAA's operational suite of forecast models for improved ice modeling and to inform the surface energy budget

Objective 5

Produce best-quality reference data

Many R&D activities require high-quality long-duration observation datasets. Quality, in part, is determined by how well the data represents the best understanding of the observations. Improved information, understanding, and techniques for retrievals, calibration, sampling, and representation need to be applied to accumulated datasets via reprocessing and reanalysis to ensure that the data represents the best currently possible understanding of the observations.

 R&D Targets:

  • Reanalyze extended operational satellite observation records to generate calibrated and refined analysis of global and regional climate temperature, precipitation, and related ecosystem changes and trends

  • Reanalyze operational model results, examining differences for enhanced understanding of environmental processes and relationships