<p>Air pollution modeling as a problem of continuum mechanics provides a rigorously predictive approach invaluable for industrial control and engineering. To address the limitations of classical atmospheric Eulerian transport models, such as CHIMERE, this study introduces a new numerical method: The Unsteady Mass Conservation Approach. Rooted in the principle of mass conservation defined by the continuity equation, UMCA incorporates dynamic correction terms both instantaneous and advective to improve the simulation under non-stationary conditions. Pollutant mass concentration is modeled through a system of coupled differential equations that account for key physical processes, including advection, turbulence, atmospheric reactions, emissions, and deposition. Several important variables are analyzed to understand and evaluate the model’s behavior, such as the Reynolds number, temperature, pressure, wind speed, momentum, and the zonal and meridional wind components. The accuracy, stability, and adaptability of the UMCA method are assessed using multiple performance metrics: Pearson correlation coefficient, root mean square error, logarithmic residual gain, and residual norm. A series of validation tests is conducted, including wind-driven dispersion, short-term forecasting, spatial consistency, solver stability, and comparison with standard data assimilation techniques. The results show that UMCA is both practical and precise, with benefits in simplicity, portability, and performance. It’s also flexible enough to be applied to other complex flow problems, including ongoing work in turbulence and nonlinear atmospheric modeling.</p>