"Sediment dynamics in water bodies has been the subject of much study. Previously studies on moving sediments has focussed on their role in morphology. Recently, the transport and accumulation of sediment in natural rivers has become of interest to environmental scientists, as many pollutants are associated with sediments. Deterministic models still commonly used in engineering practice model sediment transport with average quantities. The latest research models use stochastic and now particle tracking approaches to better reflect nature. Random variable analysis is starting to be used but there is little data to test the new models. Grain scale measurements of transport have only recently been achieved as instrumental capabilities have increased. Increasing computing capability now has the potential to treat fluvial sediment systems with true probabilistic based simulations in which the fate of each particle can be individually tracked and modelled. Such a form of sediment modelling is still limited by the existing knowledge of the relevant grain scale physics rather than by computational limitations. The oldest but still most difficult questions to answer about river sediments is whether they do or do not move under the action of the flow and when in motion, where do they rest and for how long. The current proposal will use advanced instrumentation to measure hundreds of moving grains and local flow velocities. This data will be used to develop probabilistic approaches for transport rate prediction, further development of an existing discrete particle model will allow the time history of their grain movement including rest periods and their ultimate fate to be statistically described. This level of information will be required in the next generation of modelling tools that environmental scientists will need to predict the impact of pollutants associated with sediments that move and are then retained within river deposits for periods of time."