17:15 〜 19:15
[AAS11-P10] Annual Monitoring of Urban PM2.5 on a High-Rise Rooftop in Tokyo: Analysis of Vertical Profile near the Surface and Evaluation of a Numerical Model
キーワード：PM2.5、都市キャノピー層、鉛直濃度分布、化学輸送モデル
Urban-origin air pollutants are thought to act as cloud condensation nuclei (CCN) or ice nuclei (INP) in urban heavy rainfall (UHR) events. Kajikawa et al. (2023) used a numerical model to show that urban-origin air pollutants influence precipitation intensity and wet deposition in UHR while highlighting discrepancies between observed and simulated concentrations and variations of particulate matter (PM2.5) near the surface.
To investigate this issue, we analyzed year-round PM2.5 mass concentration data obtained from a light-scattering PM2.5 monitor installed on the rooftop of a building (65 m) at Waseda University’s Nishi-Waseda campus in Shinjuku, Tokyo, from April 2018 to April 2019.
After correcting for overestimations caused by relative humidity effects, no apparent diurnal variation of PM2.5 mass concentration was observed at the 65m site. Furthermore, the vertical concentration difference between the surface and 65m was slight at night, and the upper layer of the urban canopy showed relatively lower concentrations during the day. Using principal component analysis (PCA) while considering meteorological elements and precursor pollutants (NOx, Ox), we examined factors related to the vertical concentration differences. We confirmed that it follows a power-law distribution based on ground-level wind speed. Additionally, using vertical profiles of temperature and relative humidity at Tokyo Tower, cluster analysis of potential temperature gradients revealed that significant vertical concentration gradients were observed at the 65m height during atmospheric instability. Therefore, the disappearance of the diurnal variation in PM2.5 mass concentration at the 65m site is believed to result from the balancing effect of increased emissions during the day, counteracted by daytime instability and increased wind speed. Moreover, simulations using the NHM-Chem regional meteorological-chemical model revealed apparent diurnal variations in PM2.5 mass concentration at the 65m site, with excessive vertical concentration differences. The evident diurnal variations in the upper urban canopy layer can be attributed to an overestimation of ground-level wind speeds. Although surrounding factors may strongly influence the wind speed measurements at the urban canopy level, the simulation indicates stronger winds near the surface than observations, suggesting more substantial diffusion around the 65m height.
Based on these results, we consider that vertical diffusion near the surface and within the urban canopy layer is one of the key points to be improved in numerical models.
To investigate this issue, we analyzed year-round PM2.5 mass concentration data obtained from a light-scattering PM2.5 monitor installed on the rooftop of a building (65 m) at Waseda University’s Nishi-Waseda campus in Shinjuku, Tokyo, from April 2018 to April 2019.
After correcting for overestimations caused by relative humidity effects, no apparent diurnal variation of PM2.5 mass concentration was observed at the 65m site. Furthermore, the vertical concentration difference between the surface and 65m was slight at night, and the upper layer of the urban canopy showed relatively lower concentrations during the day. Using principal component analysis (PCA) while considering meteorological elements and precursor pollutants (NOx, Ox), we examined factors related to the vertical concentration differences. We confirmed that it follows a power-law distribution based on ground-level wind speed. Additionally, using vertical profiles of temperature and relative humidity at Tokyo Tower, cluster analysis of potential temperature gradients revealed that significant vertical concentration gradients were observed at the 65m height during atmospheric instability. Therefore, the disappearance of the diurnal variation in PM2.5 mass concentration at the 65m site is believed to result from the balancing effect of increased emissions during the day, counteracted by daytime instability and increased wind speed. Moreover, simulations using the NHM-Chem regional meteorological-chemical model revealed apparent diurnal variations in PM2.5 mass concentration at the 65m site, with excessive vertical concentration differences. The evident diurnal variations in the upper urban canopy layer can be attributed to an overestimation of ground-level wind speeds. Although surrounding factors may strongly influence the wind speed measurements at the urban canopy level, the simulation indicates stronger winds near the surface than observations, suggesting more substantial diffusion around the 65m height.
Based on these results, we consider that vertical diffusion near the surface and within the urban canopy layer is one of the key points to be improved in numerical models.