Impact of high-resolution sea surface temperature, emission spikes and wind on simulated surface ozone in Houston, Texas during a high ozone episode. (March 2017)

• Record Type: Journal Article
• Authors: Pan, Shuai; Choi, Yunsoo; Jeon, Wonbae; Roy, Anirban; Westenbarger, David A.; Kim, Hyun Cheol
• Is Part Of: Atmospheric environment. Volume 152(2017)
• Page Start: 362
• Page End: 376
• Publication Date: 2017-03
• Subjects: SST -- Emission -- Wind direction -- CMAQ -- Ozone nonattainment -- Houston; Air -- Pollution -- Periodicals; Air -- Pollution -- Meteorological aspects -- Periodicals; 551.51
• DOI: 10.1016/j.atmosenv.2016.12.030
• Languages: English
• ISSNs: 1352-2310

Abstract

Abstract: Model-measurement comparisons for surface ozone often show significant error, which could be attributed to problems in meteorology and emissions fields. A WRF-SMOKE-CMAQ air quality modeling system was used to investigate the contributions of these inputs. In this space, a base WRF run (BASE) and a WRF run initializing with NOAA GOES satellite sea surface temperature (SST) (SENS) were performed to clarify the impact of high-resolution SST on simulated surface ozone (O3 ) over the Greater Houston area during 25 September 2013, corresponding to the high O3 episode during the NASA DISCOVER-AQ Texas campaign. The SENS case showed reduced land-sea thermal contrast during early morning hours due to 1–2 °C lower SST over water bodies. The lowered SST reduced the model wind speed and slowed the dilution rate. These changes led to a simulated downwind O3 change of ∼5 ppb near the area over land with peak simulated afternoon O3 . However, the SENS case still under-predicted surface O3 in urban and industrial areas. Episodic flare emissions, dry sunny postfrontal stagnated conditions, and land-bay/sea breeze transitions could be the potential causes of the high O3 . In order to investigate the additional sources of error, three sensitivity simulations were performed for the high ozone time period. These involved adjusted emissions, adjusted wind fields, and both adjusted emissions and winds. These scenarios were superimposed on the updated SST (SENS) case. Adjusting NOx and VOC emissions using simulated/observed ratios improved correlation and index of agreement (IOA) for NOx from 0.48 and 0.55 to 0.81 and 0.88 respectively, but still reported spatial misalignment of afternoon O3 hotspots. Adjusting wind fields to represent morning weak westerly winds and afternoon converging zone significantly mitigated under-estimation of the observed O3 peak. For example, simulations with adjusted wind fields and adjusted (emissions + wind fields) reduced under-estimation of the peak magnitude of 100 ppb from 50 ppb to 7 and 9 ppb. Additionally, these sensitivity cases captured the timing and location of the observed O3 hotspots. The simulation case with both adjusted emissions and wind fields showed the best statistics for NOx (correlation: 0.84; IOA: 0.90) and O3 (correlation: 0.87; IOA: 0.92). These comparisons suggest that emissions and wind fields are important in determining the magnitude of high peaks, and wind direction is more critical in determining their timing and location. Particularly, improving the model capability to reproduce small-scale meteorological conditions favoring O3 production, such as stagnation and wind reversal, is crucial for accurate placement of locations of peak O3 and its precursors. Highlights: Study examined causes of peak surface ozone in a coastal urban environment. SST variations have large impact on distributions of pollutants over downwind areas. Wind direction is critical to spatially determine surface NOx and ozone hotspots. Emissions and wind speed are important in determining peak NOx and ozone magnitude.