National Earth System Prediction Capability: Metrics, Post-processing, and Products for Seasonal to Subseasonal Workshop
The National Earth System Prediction Capability (ESPC) is an interagency effort between NOAA, DOD, NASA, DOE, and NSF designed to increase collaboration, efficiency, and effectiveness in the computational modeling community. The National ESPC conducted a workshop from 28 February to 2 March 2018 aimed at a broad enterprise discussion of user needs, agency capabilities, products, gaps between needs and capabilities. Potential operational and technological solutions to address those gaps, especially potential post-processing solutions, developer metrics, and reliability metrics were explored. Metrics for this purpose can be categorized in the following groups:
- Input: observations, their quality, quantity, distribution etc.
- Process: measuring characterization of physical processes. Can range from detailed (moisture, interface fluxes) to collective (el Nino, TC genesis)
- Output: specific forecast variables, to include thresholds such as drought, flood, frost, monsoon, TC genesis)
- Outcome: improved forecasts, increased reliability
- Impact: user/customer benefit from decisions based on forecasts (safety, economy, preparedness).
This workshop emphasized (2) through (4).
This workshop provided an initial opportunity for community input to the draft S2S prediction report that NOAA is preparing in response to the Weather Act of 2017, which directs and authorizes NOAA to improve temperature and rainfall prediction and impacts such as drought, sea-ice extent etc. at two weeks to two years. The act also directs NOAA to define operational goals and objectives for these improvements and to reach out to other Agencies, academia and the private sector to help in defining the forecast, observing, monitoring, and research needs to meet these goals and objectives. National ESPC worked to gain consensus on the types of metrics to be considered in these time scales and facilitate conversation between the operational and R&D communities on how to measure performance of systems. Workshop attendance included representation from various Federal agencies participating in Federal Committee for Meteorological Services and Supporting Research (FCMSSR) and representatives from the scientific and commercial communities.
The workshop focused on the 2016 NAS S2S report, user needs, and metrics to support earth system development. Further discussions highlighted user needs for product understanding, reliability, and application to decision support. The ESPC staff will present key workshop findings.
The National Earth System Prediction Capability (ESPC) is a US interagency partnership to improve the common science and earth system modeling capability, to better support each agency’s mission from 0 to 30 years. ESPC conducted a workshop in February for a broad enterprise discussion of S2S user needs, agency capabilities, gaps between needs and capabilities, and products to support decisions. Discussions also centered around verification metrics to better support earth system development.
Ocean Model Coupling (Air-Ocean, Ice-Ocean, Wave-Ocean) on Subseasonal Through Interannual Timescales to Support the National Earth System Prediction Capability
Important decisions in sectors ranging from food security and public health, emergency management and national security rely on forecast information globally and at time scales beyond traditional weather limits. Prediction at sub-seasonal to seasonal time scales and beyond require full coupling between the components of the physical earth system. National Earth System Prediction Capability (National ESPC) is a partnership of five Federal agencies collaborating to address research and operational issues, especially coordinated transitions or research to operational or application use, across time scales ranging from synoptic to decadal. The partnership’s focus is on the subseasonal-to-seasonal (S2S) and intraseasonal to interannual (ISI) time range for which both initial conditions and boundary forcings drive the state of the coupled air-ocean-land-ice environment. This session is looking for papers describing improvements to ocean coupling (air-ocean, ice-ocean, wave-ocean) technologies and effective data assimilation for coupled systems, both for weather prediction to support improved S2S/ISI prediction as well as internally consistent ocean and atmosphere modeling. Technologies should improve representation of important coupled phenomena such as MJO, PDO, ENSO, IOD and others. (AGU Ocean Sciences Oral Session, Poster Session)Applications of an Unstructured Grid Surface Wave Model (FVCOM-SWAVE) to the Arctic Ocean: the 'ice retreat-wave growth' positive feedback mechanism by Yang Zhang et al.
Verification and Validation on the Subseasonal to Seasonal Time Scale to support the National Earth System Prediction Capability
Important decisions in sectors ranging from food security and public health, emergency management, and national security rely on communicated forecast and uncertainty information globally and at time scales beyond traditional weather limits. National Earth System Prediction Capability (National ESPC) is a partnership of five federal agencies collaborating to address research and operational issues across time scales ranging from synoptic to decadal, especially improving the two-way connection between research and operational prediction.
The partnership’s greatest focus is on the subseasonal-to-seasonal (S2S) time range for which both initial conditions and boundary forcings drive the state of the coupled air−ocean−land−ice environment. With the goal of improving assessments of S2S model prediction skill, prediction tools or technologies, and providing process-based feedback on model performance, this session is accepting papers describing verification and validation techniques and tools on the S2S time scale, for user needs with fundamental time scales extending across this time frame such as watershed prediction reliability, planting season precipitation, potential frost days, Arctic or Great Lakes ice breakup dates, ice extent, etc. (AMS Session)
Session Talks (all links external)
U.S Air Force Weather Subseasonal to Seasonal (S2S) Prediction: Pushing the Limits of Predictability to Optimize National Defense by Ralph O. Stoffler
Operational Environmental Modeling Across Scales by Hendrik Tolman
NOAA’s Strategy to Improve Operational Subseasonal Prediction for Weather Outlooks by Timothy L. Schneider
NCEP's Next Generation Coupled Forecast System for Subseasonal to Seasonal Scales by Arun Chawla
Examining the Impact of Sea Surface Temperature Coupling on Subseasonal Prediction Skill using a Prototype Unified Global Coupled System at NCEP by Suranjana Saha
Real-Time and Retrospective 3−4-Week Forecasting for the NOAA Subseasonal Experiment with the FIM-HYCOM Coupled Model by Stan Benjamin
Spatial Approaches to Evaluation of Climate Variability Projections and Predictions by Barbara Brown
Session: Stochastic and Coupled Modeling for Seamless Earth System Prediction Capabilities
Five Federal Agencies (DOD – Navy & Air Force, DOE, NASA, NSF, and NOAA – NWS & OAR) signed a charter April 2016 creating the National Earth System Prediction Capability. This effort envisions an operational National capability for a coupled earth (air, ocean, wave, ice, land, space) prediction system (or system of systems) to support National needs from days to decades, with highest priority the two week to inter-annual time frame. This session will introduce the sponsoring agencies, their scientific goals, and supporting efforts. Relevant papers to be considered for the session include sub-seasonal to seasonal prediction systems (coupling, coupled data assimilation, multimodel ensembles, verification and validation, etc.), especially systems coordinated between participating agency partners. Additional topics include predictability related to slow mode earth system processes (ENSO, MJO, QBO, PDO, etc.) and societal needs for longer-range predictions.
Ocean Ensemble Forecasting in the Navy Earth System Prediction Capability by Dr. Clark Rowley
Subseasonal to Seasonal Forecasting at NASA in support of the National Earth System Prediction Capability by Dr. David Consadine
The DoD's High Performance Computing Modernization Program by Dr. William Burnett
Monitoring the Performance of the CFS v3 Development by Dr. Malaquias Peña
NOAA Climate Program Office Contributions to the National ESPC by Dr. Wayne Higgins