Using meteorological data and emission data gotten in the Ise Bay District on 4-5, Aug., 1977, a numerical simulation of photochemical air pollution was conducted. The wind field was represented through an interpolation of observed data. Diffusion and photochemical reaction were computed respectively with the diffusion equation and simplified chemical equations.
   The simulation could represent for these days that developing high O₃ masses drifted northeasterly by the sea breeze which arose in the eastern or southern part of Nagoya around Atsumi bay. Computational error remained below the tolerable level in most of the object area. However, the simulation overestimated distinctly at the limited part of the Chita peninsula late on the morning of 5 th. As a cause of the deviation, it is supposed that local western sea breeze continued in this period and pollutants were tossed into higher layers over the region, because topographic and thermal convection affected the wind greatly across the peninsula. These thermal plumes could not be represented in the interpolated wind field.
   The probability of the above-mentioned cause was supported by two-dimensional test computation of the western sea breeze, including thermal and topographic effect.