Every day during the 8/29 - 9/6/2000 period, hourly surface ozone values in the Houston, TX area exceeded the 125 ppbv standard and approached 200 ppbv on some days. This ozone pollution episode occurred during the month-long Texas Air Quality Study (TexAQS 2000), which was aimed at identifying the chemical and meteorological processes that cause these kinds of extreme ozone events in the Houston area. The first three days of the pollution episode (8/29 - 8/31) were characterized by very light-wind conditions during midday and an off- to onshore flow reversal in the afternoon, whereas during the remaining six days (9/01 - 9/06) the wind was blowing steadily from westerly or easterly directions all day. In this paper we will examine the effect of these different flow conditions on the three-dimensional distribution of ozone, ozone peak values, and transport pathways.

To document the horizontal and vertical distribution of ozone we used primarily data gathered with the National Oceanic and Atmospheric Administration / Environmental Technology Laboratory airborne ozone and aerosol lidar. The nadir-looking lidar measured vertical profiles of ozone concentration and aerosol backscatter between the surface and about 2500 m above ground. Back trajectory calculations using data from the wind profiler network deployed in the greater Houston area provided information on the origin and transport pathways of the ozone plumes that were detected with the airborne ozone lidar. We also incorporated airborne in situ chemistry measurements to characterize the chemical makeup of these ozone plumes.

The very high ozone concentrations and widespread ozone exceedances that occurred on 8/29, 8/30, and 8/31 were linked to a combination of two meteorological factors: a) stagnant conditions during midday allowed a buildup of ozone plumes over and near the source regions and b) aged pollution plumes emitted into the morning offshore flow were recirculated and transported back over the source areas by the afternoon sea breeze. Even though the meteorological conditions on these three days were very similar we will demonstrate that subtle differences in the flow pattern and the onset and strength of the sea breeze caused significant differences in ozone distribution and ozone peak values.

During the later part of the pollution episode, when well-defined flow conditions prevailed (9/01 - 9/06), peak ozone values in the Houston vicinity were lower and exceedances registered by the Houston Regional Monitoring (HRM) network were not as widespread compared to the previous three days. However, the airborne lidar and in situ chemistry sensors detected ozone plumes downwind of Houston with peak values well above exceedance level. On 9/06 for example, the airborne ozone lidar measured ozone concentrations of more than 125 ppbv up to 150 km downwind of Houston. This suggests that on days with a steady synoptic flow the Houston ozone plume is exported to the surrounding rural regions, where HRM network coverage is sparse at best. Consequently, the HRM network measurements, concentrated in and near the Houston metro area, indicate relatively clean conditions, while exceedances are occurring in the sparsely monitored regions.