The objective of this project is to develop indexes that relate variations in air quality to observed climate elements for the purposes of monitoring U.S. air quality conditions. Two important pollutants are ozone and particulate matter smaller than 2.5 µm in diameter (PM2.5). Some of the meteorological conditions that can affect the occurrence of harmful concentrations of these pollutants include temperature, mixing height, wind speed, wind direction, and solar radiation. Mixing height is important because this determines in part the dilution of pollutants; as the well-mixed boundary layer grows during the day, the air entrained from above is usually characterized by lower pollutant concentrations than the air within the boundary layer. Maximum daytime mixing height can be estimated from 0000GMT radiosonde data, near the time of maximum surface temperatures. However, the spatial density of radiosonde sites is quite low and there are a number of large urban areas with no nearby site. One purpose of this study was to evaluate the possible use of grid point vertical profiles from reanalysis data for estimation of mixing height. Seventeen radiosonde sites were chosen for an analysis covering the period 1986-1996. Daily values of the maximum mixing height were estimated from radiosonde data and concurrent reanalysis profiles for the gridpoints nearest to each radiosonde site. Differences between radiosonde and reanalysis estimates were acceptably small for most sites east of the western mountain chains. For sites in the west, the results were more variable. At some sites, large differences were found. Reasons for the large differences will be discussed. It appears that the reanalysis can be used to estimate mixing height in urban areas in parts of the eastern U.S., but more work is needed in the western U.S.