Emissions estimates of volatile organic compounds (VOC) from agricultural pesticide application are critical for state agencies to meet air quality standards of ground-level ozone. State regulations in California require to reduce emissions of VOCs from fumigant and non-fumigant pesticides in non-attainment areas. As demand for food production increases with a growing population, compliance with air quality standards may become challenging related to agricultural emissions. Currently, California Department of Pesticide Regulation (CDPR) empirically calculates the VOC emissions using an emission potential (EP) value, representing the volatile fraction of a certain pesticide. However, an EP value may not represent the actual emissions related to application rate, pesticide properties and time of the application. Many studies suggested that pesticide emissions from soil applications can widely vary based on meteorological and environmental conditions. In this work, a deterministic model is demonstrated that numerically computes time dependent emissions with user inputs of hourly meteorological data and pesticide properties. The soil-physics based model captures the dynamic transport processes of heat, moisture, and pesticide at the soil-atmosphere interface. Model simulations showed good performance against field measurements for a fumigant and two surface applied herbicides. Emission estimates from the mechanistic model were also compared to the CDPR’s estimates. This method can be used by regulators to calculate VOC emission inventory at a particular geographic location. Sensitivity analyses of the model might help regulators to assess the influence of different meteorological conditions for a particular pesticide formulation.