Science for Society: Extending Earth Science Research
Results into Decision Support Tools

(Continued)

Benchmarking Earth Model Predictions

For 40 years, NASA has been developing the capability to use the perspective of space to provide global observations of the key parameters of our Earth system. NASA uses a systems approach to enable the measurements to be explored rigorously and scientifically. The NASA science community uses the measurements to advance global and regional models and improve predictive capability. Advanced computer technologies are used to visualize the resulting global observations and predictions to generate excitement about, and understanding of, our home planet.

NASA has successfully developed and deployed the flagship Earth Observation System (EOS). As of April 2003, the US space agency has 18 Earth observation satellites on orbit carrying 80 sensors. These remote sensing satellites are observing the Earth’s atmosphere, land surfaces, ice covers, ecosystems, gravity fields, and surface deformations. About 3 terabytes of data are collected daily and transmitted to Earth receiving stations. These measurements are filtered, assimilated, and digested into the Earth Observations System Data Information System (EOSDIS) through intelligent data assimilation and modeling processes. The resulting calibrated and validated measurements of key parameters of Earth processes are assimilated into Earth science models that are responsible for producing predictions – and related knowledge useful in decision support systems. Since the Earth science research community cannot perform large-scale experiments on our environment, the Earth system model is the most optimal tool for Earth system scientists to use to perform simulations and produce “what if” scenarios of the evolution of our environment.

Earth science models represent a consolidation of scientific understanding of the range of physical processes associated with the Earth system. It is estimated that physical process field studies and routine observations must be conducted for at least 10 years before a given physical process can be represented in an integrated Earth science model. In a special issue of the EOM journal (http://www.eomonline.com), NASA presents articles on eight of the Earth science models. These include a global and regional weather prediction model, a near-term coupled atmosphere-ocean model, a seasonal-to-interannual prediction model, a long-term climate change model, a coastal biological-physical oceanography model, a land surface hydrology model, and a bio-geochemistry model. These models represent years of Earth science modeling advancement and the consolidation of scientific findings from many of the Earth science disciplines. These state-of-the-science research and operational models are used in many decision support systems today.

NASA's High-End Atmospheric Model for Climate and Weather Predictions describes an advanced high-end atmospheric model designed to produce weather and climate predictions is introduced. Although it is still viewed as an experimental model, this has been the highest resolution global weather and climate model running routinely.

Seasonal Climate Prediction and Predicting Seasonal to Interannual Climate Variations describe a class of climate models designed to predict seasonal to interannual climate variations. Since the seasonal variation of the climate system depends heavily on the ocean – which carries the memory of the past and forces the future evolution of the climate system – coupled models are used for this type of prediction. These seasonal to interannual climate models have been used to guide water management, energy consumption prediction, and fishery production predictions.

Using Earth Science Tools to Improve Seasonal Climate Prediction for Agriculture addresses spatial resolution in global climate change research. The current climate change models do not have sufficient spatial resolution for regional assessments. Due to the lack of computational resources, techniques have been developed to downscale the global model outputs to a regional spatial scale. The downscaled information can then be used in agriculture decision-making.

In Using Regional Atmospheric Models for Commercial Applications, a regional weather forecast model is introduced. The model is used commercially and operationally at many installations, including the Kennedy Space Flight center.

Modeling the Coastal Ocean Processes Within the U.S. Continental Margins, Incorporating the Land Data Assimilation System into Water Resource Management and Decision Support Systems, and Biospheric Monitoring and Ecological Forecasting represent specific purpose models on coastal ocean modeling and regional water monitoring to agriculture production prediction. These NASA-supported models are producing predictions for specialized purposes.

It is easy to see from the broad range of applications that use Earth science model products that it will be helpful to have an integrated framework for Earth system modeling. The benefits of a common framework are the driver for the next phase of the consolidation of Earth system science research results. As stated in the ESE Research Strategy: “The ultimate challenge of Earth System Science is to consolidate the scientific findings in the different disciplines into an integrated representation of the coupled atmosphere, ocean, ice, land and biosphere system.” The integrated Earth system model – currently consisting of coupled atmosphere, ocean, and land components – is used to contribute to science-based assessments of potential future changes. In the next 10 years, additional components will be added into the integrated Earth system model. These components will include ice, biosphere, solid earth, and chemistry transport models. NASA has already started projects to build a national Earth System Modeling Framework (ESMF). Through this modeling framework, various Earth science models will be able to communicate and exchange information inputs from Earth observation systems and outputs to decision support systems. The ESMF provides a unified external interface to serve a range of applications for Earth science model predictions. We are on a path towards creating a comprehensive description of the whole Earth model and an end-to-end system ranging from satellite data acquisition, data analysis, and modeling, all the way to the decision support systems.

Next Steps

As the Earth science community continues to increase our understanding of Earth system processes, there will be continued opportunities to assimilate observations and predictions into decision support tools and thereby increase the capacity to protect the home planet.

Your contributions in Earth science research and technology development are greatly valued and appreciated as the “stock-in-trade” to serve our society through applications of national priority. Our world needs the best available information and information infrastructure to enable decisions on policy and management of natural resources. To quote Dr. Jack Kaye, Director of our Earth Science Research Division,

“Science allows it,
Technology enables it,
Society requires it,
And the time is now!”

Pages 1 | 2 | 3