| dc.description.abstract | Douliu is one of the most air-polluted areas in Taiwan. Factors that caused the deterioration of air quality in this area might attribute to emissions, chemical reactions, meteorology, and geographical distribution, etc. Recent modeling studies point out that high PM2.5 concentration events occurred are usually associated with favorite weather conditions (i.e., high-pressure peripheral circulation and weak synoptic weather). To make a statement on the deterioration of PM2.5 events, we employed a new technology, Unmanned Aerial Vehicle (UAV), and an aerosol lidar to observe the high resolution of meteorological and aerosol profiles at Douliu during November 16th-20th, 2020. A series of data validation procedures had applied to ensure the UAV data quality. As a result, the UAV measurements (i.e., meteorological parameters and PM2.5 concentration) are reliable and show in good agreement with vertical structure obtained from lidar observation. In addition to observations (i.e., UAV, lidar, surface air quality, and meteorology data), we also use a WRF model to run a reanalysis for this case study.
According to our results, the roles of land-sea breeze, temperature inversion layer, and leeside vortex in aerosol transportation and dispersion over Douliu during the weak synoptic weather were discovered. The horizontal dispersion condition did not change much during the night of the observation, which was accompanied by the wind speed smaller than 1 m s-1. On November 17th and 18th when the high pollution event occurred, there were two temperature inversion layers in vertical. One was the surface radiative inversion whose depth was about 200 m, and another was the elevated subsidence inversion which was located at around 800-1000 m. The subsidence of the air also led to the dry condition in the sky and was beneficial for radiative cooling at the surface at night. Therefore, the strength of the temperature inversion near the surface increased, compared with the strength of 2.6 ℃ km-1 on the night of 16th and 4.5 ℃ km-1 on the night of 19th, the inversion strength on 17th and 18th were up to 8.0 ℃ km-1 and 12.1 ℃ km-1, respectively. Therefore, these worse vertical dispersion conditions caused the maximum of hourly PM2.5 concentration to be up to 70 μg m-3 in the evening of 17th and 99 μg m-3 in the evening of 18th. The sea breeze circulation developed at about 9 a.m., and the specific humidity and PM2.5 concentration increased at the same time at noon, because the sea breeze transported the pollutant and water vapor from the coastal area to Douliu. The structure of sea breeze front was distinct on November 18th at 1 p.m., and the leading edge of the sea breeze front reached a height greater than 1000 m which was demonstrated from the lidar NRB data. When the land breeze gradually developed, the air quality improved after 18 p.m. on 17th and 21 p.m. on 18th. The result of the numerical simulation indicated there were three leeside vortexes during this case, two were associated with the cyclonic circulation and one was associated with the anticyclonic circulation. The location, structure, as well as direction of rotation with respect to the leeside vortex showed differently is due to the variation of the synoptic winds and how they interacted with the terrain. The cyclonic vortex led northward of the pollutant near the surface, and the anticyclonic vortex brought the dry air of subsidence with the northerly wind over Douliu. Overall, this study used the high temporal resolution vertical observation data from UAV and aerosol lidar to dissect critical factors of PM2.5 deterioration in Douliu area. This work also highlights that the high temporal resolution of UAV and lidar observations can serve as a powerful tool to improve our understanding of the relationship between atmospheric vertical profile and air quality. | en_US |