Field Tests Unite Weather And Climate Models

Although still under development, many organizations and researchinstitutions are starting to adopt ESMF. Under a partnership, groupsfrom NASA, the National Science Foundation (NSF), the National Oceanicand Atmospheric Administration (NOAA), the Department of Energy (DOE),the Department of Defense and research universities are using ESMF asthe standard for coupling their weather and climate models to achieve arealistic representation of the Earth as a system of interacting parts,unifying much of the modeling community. ESMF makes it easier to shareand compare alternative scientific approaches from multiple sources,uses remote sensing data more efficiently and eliminates the need forindividual agencies to develop their own coupling software.

“The development of large Earth system applications often spansinitiatives, institutions and agencies, and involves the geoscience,physics, mathematics and computer science communities. With ESMF, thesediverse groups can leverage common software to simplify modeldevelopment,” said NASA ESMF principal investigator Arlindo da Silva, ascientist in GSFC’s Global Modeling and Assimilation Office.

NASA’s Earth-Sun System Technology Office/Computational TechnologiesProject funds the field tests and overall ESMF development. Thepartners on the field tests are DOE’s Los Alamos National Laboratory(LANL), the Massachusetts Institute of Technology (MIT), NASA’s JetPropulsion Laboratory, NOAA’s Geophysical Fluid Dynamics Laboratory(GFDL) and National Centers for Environmental Prediction (NCEP), NSF’sNational Center for Atmospheric Research (NCAR) and the University ofCalifornia, Los Angeles (UCLA).

The newly completed field tests, known as interoperabilityexperiments, show that this new approach in coupling models can besuccessful. For instance, temperature and wind outputs are similararound the globe when ingesting data into an NSF-NASA atmosphere modelas they are for the atmosphere model used operationally by NOAA.Although most of the experiments would require exhaustive tuning andvalidation to be scientifically sound, they already show that ESMF canbe used to assemble coupled applications quickly, easily and withtechnical accuracy.

“These interoperability experiments illustrate the role ESMF canplay in integrating the national Earth science resources,” da Silvasaid. “Using existing data assimilation technology from NCEP, thefinite-volume Community Atmosphere Model, or fvCAM, was able to ingestconventional and satellite observations, a capability that could openthe door to using the fvCAM for weather as well as climate prediction.”The fvCAM, which includes land surface capabilities, was developed byNCAR, with key components from GSFC.

The second experiment combines NCEP’s data assimilation technologywith the Aries atmosphere model originally developed by the NASASeasonal-to-Interannual Prediction Project. Aries is typically coupledwith an ocean model to run experimental forecasts of phenomena such asEl Niño and its effects on precipitation. Among additional advantages,the two field tests enable the intercomparison of systems for satellitedata assimilation.

The coupled experiments have many other potential applications. Thethird experiment, combining a GFDL atmosphere-land-ice model with anMIT ocean-sea ice model (known as MITgcm), may ultimately bring newinsights into ocean uptake of carbon dioxide and other importantatmospheric gases and how this process affects the climate.

In an early independent adoption of ESMF technology, UCLAresearchers have successfully coupled their Atmospheric GeneralCirculation Model to the MITgcm for the first time and updated aprevious coupling to the LANL Parallel Ocean Program model. They madeexperimental predictions of the El Niño/Southern Oscillations with thecoupled models using initial states provided by JPL’s Estimating theCirculation and Climate of the Ocean (ECCO) project. These preliminaryresults validate ESMF performance in terms of scientific fidelity andsupport the importance of ECCO products for improving short-termclimate forecasts.

Demonstrations of the software and the field tests are taking placeat the 4th ESMF Community Meeting at MIT in Cambridge, Mass., July20–22. Reaching beyond the ESMF partnership, the research team releasesthe software to the scientific community via the Internet. NCAR, homeof the core implementation team, is scheduled to release ESMF Version2.2.0 this month.