Description:
This subtopic directly addresses the Department of Commerce’s strategic objective to “Reduce Extreme Weather Impacts - develop and deploy next–generation observation, data assimilation/processing, and modeling for the environment in order to make informed planning, resources management and investment decisions."
NOAA and other federal agencies have been maintaining extensive observation networks and developing a large number of integrated earth prediction system models. Computer models developed for weather prediction, coastal ocean circulation, waves, and ice, as well as satellite remote sensing data are all computationally intensive, requiring high performance computing (HPC) to perform simulations, analysis and forecasts. An attractive alternative is to apply machine learning (ML), deep learning (DL), artificial intelligence (AI), pattern- recognition or data analytic approaches to improve the accuracy of earth prediction systems and the efficiency of automated data processing, pattern recognition and feature extraction from large volumes of datasets. Using these new techniques to improve efficiency and accuracy of the forecast products will potentially lead to earlier warnings for extreme weather and water events which have the potential to save more lives and reduce property damage.
This call invites small, high-tech firms specializing in developing novel machine learning, artificial intelligence and pattern recognition algorithms to analyze and process large volumes of computer model results and satellite imagery in order to improve the efficiency and accuracy of integrated earth prediction systems (numerical weather prediction, ocean circulation, hydrological, waves, and ice modeling systems).
The ultimate goal is to develop the next-generation of commercial applications, products and services in Information Technology (IT), autonomous vehicles, medical, insurance industries by applying machine learning or artificial intelligence technologies. Federal government agencies such as NOAA, USGS, DoD and DoE will no doubt benefit from such innovative technology, the true commercialization applications has much broader potential opportunities in multiple industries and market place.
