Particulate flows are ubiquitous in Nature and industrial processes where they exhibit a rich set of fascinating and complex dynamics. This thesis develops a coupled fluid-particle simulation method to study such systems at the microscale by fully resolving all relevant physical effects. It combines the lattice Boltzmann method with the discrete element method to represent the fluid flow dynamics and to model frictional collisions between the rigid particles. Focusing on the characteristics of the applied coupling approach, several techniques for improving the simulation results are compared. An extensive calibration and validation pipeline ensures the method's accuracy. Key aspects are the explicit modeling of short-range hydrodynamic interactions between particles and an adequate temporal resolution of single collision events. This verification concept is universally applicable for fully resolved simulation approaches to derive guidelines for model parameterization. All algorithmic components are designed for efficient massively parallel execution on supercomputers. This feature is essential to tackle the vast computational costs of simulative microscale studies. Performance improvements are achieved by novel techniques for dynamic load balancing that account for the varying workload of all parts. The method is applied to simulate the erosion and transport of sediment beds consisting of several ten thousand particles. The detailed simulation data allows for deriving improved rheological descriptions of the macroscale system dynamics. This thesis thus contributes a valuable building block in the multi-scale modeling framework of particulate flows.weiterlesen