An Ensemble Processing Application Under Development and Testing at the NOAA Aviation Weather Testbed

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Thursday, 6 February 2014: 9:30 AM
Room C201 (The Georgia World Congress Center )
David Bright, NOAA/NWS/NCEP/AWC, Kansas City, MO; and J. S. Smith, B. R. J. Schwedler, G. Liu, and S. A. Lack

One of the key strategic objectives of the National Oceanic and Atmospheric Administration (NOAA) Aviation Weather Testbed (AWT) is to accelerate the realization of advanced weather forecasting capabilities for the Next Generation Air Transportation System (NextGen). Under NextGen, advanced probabilistic weather forecasting is utilized for efficient and optimal decision making, particularly under critical weather conditions. NOAA has also established a long-term Next Generation Strategic Plan that includes the goal of building a Weather-Ready Nation (WRN). A WRN ensures society is prepared for and responds to extreme weather-related events. To achieve the WRN vision, the NOAA National Weather Service (NWS) recognizes the growing demand for Impact-Based Decision Support Services (IDSS). These IDSS are realized in part through improved modeling and ensemble systems, translated into forecast confidence and reliable probabilistic information, and then effectively communicated to NWS customers.

To move toward IDSS through efficient and innovative ensemble processing, particularly for Aviation Weather Center (AWC) operations and NextGen, the AWT is developing and evaluating an initial release of the AWC's Ensemble Processor (EP). The EP is built around a MySQL database that provides an end-to-end framework for ensemble-based research, operations, and decision support. In addition to supporting basic ensemble interrogation and statistics, the system includes a builtin Java-based application program interface (API). The API allows users to build Java plug-ins for specialized, mission-specific ensemble post-processing. Queries are displayable through an HTML 5 web application with options to write GRIB2 output. Through a series of customizable XML schema, automated real-time processing populates the MySQL database with user-defined ensemble membership, produces statistics through intrinsic and stored MySQL queries, executes custom built API plug-ins, and writes GRIB2 files to disk. The GRIB2 files are available for input to existing operational software such as NAWIPS and eventually AWIPS-2.

The EP fills a gap in current ensemble post-processing capabilities within the AWT. Initial experimental deployment focuses on the NOAA/NWS Short-Range Ensemble Forecast (SREF); although, the EP is designed for flexibility in spatial resolution and membership size. This paper will describe the EP in more detail, highlight its capability for mission-specific IDSS research and application development, and demonstrate its experimental SREF deployment at the AWT.