Recent Practices and Advances in Air Quality Forecasting

From the Archive—In case you missed's a look back at the November 2013 issue of EM, which discussed some of the innovative approaches being implemented and tested to improve air quality forecasts.

by Prakash Doraiswamy

State and regional air quality agencies issue short-term air quality forecasts to provide the public with advance warning of possible air quality episodes and actions they might take to protect their health. Air quality forecasts are delivered using an air quality index that converts concentrations of multiple pollutants such as ozone and particulate matter into a simple metric that indicates the quality of the air and the associated health effects for different populations. These forecasts are typically generated based on an analysis of weather patterns, statistical models, human judgment, and more recently, through use of regional photochemical models.

Back in 2002, EM featured plans for air quality forecasting in the United States, including a new partnership between the U.S. Environmental Protection Agency (EPA) and the National Oceanic and Atmospheric Administration (NOAA) to implement a model-based forecasting system. Since then, methodologies and approaches used to forecast daily air quality have improved tremendously, including the use of single and multiple model-based air quality forecast guidance, near-real-time data postprocessing and bias correction using measured data from surface networks and possibly satellite data, and the implementation of wild fire emissions for issuing next-day air quality forecasts.

While recent budget cuts have forced these agencies to divert funding away from model-based short-term air quality forecasts, these approaches are still in practice by research groups. In this issue of EM, we present three articles that provide an insight into innovative approaches being implemented and tested to improve these air quality forecasts.