Evaluation of atmospheric stability as a predictive index for air pollutant dispersion in urban areas

This study aims to establish a new environmental assessment method for air pollution based on atmospheric stability evaluations under real urban conditions. Wind speed from Doppler lidar, temperature data from the Tokyo Tower, and ground surface temperatures estimated from Weather Research and Forecasting simulations were used to calculate the gradient (R_g) and bulk (R_b) Richardson numbers as quantitative indicators of atmospheric stability. The study investigates the range, frequency of occurrence, and seasonal/hourly variations of R_g and R_b and the correlations between the two indices. When the evaluation height for R_g was set to approximately 100m, the seasonal and diurnal trends of R_g and R_b were generally consistent. However, R_g exhibited significantly greater variability than R_b, likely because R_g is inversely proportional to the square of the wind speed gradient. R_b was also compared with the Pasquill classification to assess their correlation and seasonal/hourly variations. While R_b and the Pasquill classification were broadly consistent, R_b provided a more detailed and quantitatively continuous representation of atmospheric stability. In contrast, the Pasquill classification significantly overestimated the frequency of neutral conditions “ClassD”, particularly under overcast skies or during transition periods, and often misclassified both weakly and strongly stable/unstable states as neutral. Furthermore, given the critical role of the vertical wind speed profile in pollutant dispersion within urban areas, wind speed profiles were analyzed based on both R_b and the Pasquill classification. Both classifications showed that as atmospheric stability decreased, the power-law exponent of the vertical profile of mean wind speed decreased, while turbulence intensity increased.