Spray drift is a major contributor to pesticide losses in the atmosphere leading to nontargeted ecosystem exposure. An overview of currently used ground-application spray drift models highlighted gaps in the description of the processes occurring during and after spraying. The ADDI-Spraydrift model was developed to bridge some of these gaps and provide a comprehensive, yet concise description of spray drift processes. The model is based on a random walk approach that describes droplet emission, dispersion and evaporation, ground deposition and canopy interception, and accounts for atmospheric stability regimes, and in-canopy turbulence. It predicts airborne concentration, canopy interception and deposition to the ground downwind from the sprayer. The sensitivity analysis to wind speed, active matter content, leaf angle, canopy height, leaf area index, ejection angle and velocity demonstrated the consistency of the model behaviour. The model was calibrated and evaluated against the Ganzelmeier et al. (1995) sedimentary spray drift data in vineyards. The model satisfactorily predicted droplets deposition to the ground downwind from the field boundary with a mean deviation between modelled and measured deposition of 1.3%. A discrepancy was observed at 3 and 5 m downwind from the field boundary attributed to the sensitivity to spraying conditions for which several hypotheses are discussed. Bearing in mind the need to explicitly describe the emitted droplet size and velocity, the overall predictive performance of the model appeared to be sufficient for assessing and comparing application techniques efficiency, quantifying pesticide loss and bystanders or resident exposure and evaluating the efficiency of mitigation measures.