Operations by Unmanned Aircraft Systems (UAS) or ‘drones’ pose third party risks to persons on the ground. The key question is: are these third party risks acceptably low? This thesis evaluates the level of risk posed by a concept UAS surveillance-based operation in the port of Rotterdam, by means of modeling and simulation. In a preceding MSc thesis, a modeling and simulation approach to estimate third party risk posed by drone-based parcel delivery service is researched. This modeling and simulation approach has been extended in quite a number of directions. Firstly, the details of the intended drone-based surveillance operation have been defined in collaboration with the Port of Rotterdam Authority. Secondly, the dynamic model of a quadcopter has been replaced by a dynamic model of a fixed-wing UAS. Thirdly, the influence of wind turbulence on the stochastic deviations of the drone flight have been modeled. Fourthly, for the seaport area a method has been developed for estimating local population densities, rather than using census-based data. These models have been integrated in a Monte Carlo (MC) simulation of the intended surveillance operation in the port of Rotterdam. The outcome of the MC simulations shows that the estimated third party risk level posed per drone flight hour to persons on the ground is about a factor 4 higher than a maximal allowable threshold proposed by the Joint Authorities for Rulemaking of Unmanned Systems (JARUS). The simulation results also show that the shallow glide angle of the fixed-wing UAS proposed for the surveillance operation causes a relatively large crash impact area on the ground, which leads to a relatively large expected number of persons to be hit by a ground crash. As expected, the main contributions to third party risk stem from parts of a surveillance flight that are near the city, not from parts towards the sea. Another interesting finding is that the estimated third party risk level remains the same if wind turbulence induced flight deviations are ignored. The modeling and simulation results developed in this thesis provide useful feedback to the further development of drone-based operations in the port of Rotterdam.