Urban noise exposure assessment is generally performed with static sources and long-term average indicators such as Lden or LAeq. This approach is poorly suited to both emerging urban air mobility and road traffic. Indeed, drone operations are intermittent, highly route-dependent, and temporally concentrated, while road traffic noise varies strongly on short time scales, due to the dynamics of individual vehicles. This contribution introduces a methodology to model dynamic population noise exposure in urban environments which merges drone noise and road traffic noise in a unified simulation framework. In this novel methodology, road traffic is modelled using a microscopic traffic simulation, producing vehicle trajectories, speeds, and accelerations at fine temporal resolution (1 second). Using the CNOSSOS-EU noise emission and propagation models, the resulting noise levels are simulated dynamically at receiver locations. Drone operations are modelled as moving aerial sources with specified flight trajectories and characteristics (altitude, speed, operational phases), generating time-varying sound levels which are also propagated to ground receivers. The two contributions are energetically summed to produce exposure time series and derived indicators (e.g., short-term LAeq, noise events, exceedance), enabling comparisons between scenarios that include or exclude drones. The methodology is then applied to a Swedish test case, in the city of Stockholm, showing its feasibility, and revealing early results on the interactions between drone overflights and traffic noise, depending on the time of day. By enabling scenario testing, the proposed methodology will enable the optimisation of drone route and scheduling, linking drone operation to population exposure while including the contribution of road traffic. More elaborate scenarios will be studied in the future.