HR: 08:00h
AN: A21F-01 INVITED [Abstracts]
TI: NOAA-EPA's New National Air Quality Forecast Capability: Initial Steps
AU: * Davidson, P
EM: paula.davidson@noaa.gov
AF: NOAA National Weather Service, SSMC2-15342
13254 East-West Highway, Silver Spring, MD 20910
United States
AB:
In partnership with the US EPA, NOAA has developed, tested and implemented the first two stages of a national air quality
forecast capability into the National Weather Service (NWS) operational suite. The initial capability was implemented in
September, 2004 and provided ground-level ozone predictions over Northeastern United States. In a program of phased
development and testing to expand this capability, the domain has been extended over the entire Eastern United states as of
August 31, 2005. Predictions are made with the NOAA-EPA Community Model for Air Quality (CMAQ) driven by NOAA's operational
mesoscale weather prediction model (Eta-12). The capability is an end-to-end forecast guidance system providing twice daily
predictions of hour-by-hour ground-level ozone concentrations on a 12km grid, disseminated over operational NWS and EPA
dataservers. Forecast guidance products are hosted on operational dataservers: fully backed up, with archiving and
near-real-time verification in place to monitor forecast accuracy.
In order to demonstrate readiness for operational implementation, required accuracy of 90% and reliability of 95% on-time
delivery have been demonstrated in the pre-deployment testing. During the Summers of 2004 and 2005, pre-deployment testing
of forecast domains over Northeastern US and Eastern US, respectively, have led to operational implementation of the first
two stages of the capability. Prior to pre-deployment testing, developmental testing was conducted to demonstrate
feasibility of the prototype operational configuration using forecast components for air quality (CMAQ and pollutant
emissions pre-processing) adapted from research and assessment simulations. Developmental testing identified priorities for
system enhancements needed to improve guidance accuracy; for example: improved model linkage, updated emissions information,
improved treatments of solar radiation for photolysis rate estimation, and improved treatments of vertical mixing and
transport within clouds. The current phase of developmental testing, over an expanded coast-to-coast domain, was begun
during Summer, 2005. The impact of much deeper boundary layers over western US on predicted ground-level ozone has
emphasized the need to improve vertical mixing relative to simplistic approaches that have been demonstrated to be effective
in Eastern US. We are therefore now testing more advanced vertical mixing schemes and evaluating the optimal use of
real-time satellite data on upper-atmosphere ozone in the developmental forecast capability. Within the next year, we will
convert from Eta-12 to WRF as the driving meteorological model when it becomes NOAA's new operational mesoscale numerical
prediction model. Targeted deployment of nationwide ozone forecasts within 5 years will be followed by the addition of
particulate matter forecasts and an extended forecast period- out to day 2 and beyond.
DE: 0300 ATMOSPHERIC COMPOSITION AND STRUCTURE
SC: Atmospheric Sciences [A]
MN: Fall Meeting 2005