Dam Removal Express Assessment Models (DREAM). Part 1: Model development and validation Les modèles DREAM d'évaluation rapide d'effacement de barrage. Partie 1: Développement du modèle et validation

Many dams have been removed in the recent decades in the U.S. for reasons including economics, safety, and ecological rehabilitation. More dams are under consideration for removal; some of them are medium to large-sized dams filled with millions of cubic meters of sediment. Reaching a decision to remove a dam and deciding as how the dam should be removed, however, are usually not easy, especially for medium to large-sized dams. One of the major reasons for the difficulty in decision-making is the lack of understanding of the consequences of the release of reservoir sediment downstream, or alternatively the large expense if the sediment is to be removed by dredging. This paper summarizes the Dam Removal Express Assessment Models (DREAM) developed at Stillwater Sciences, Berkeley, California for simulation of sediment transport following dam removal. There are two models in the package: DREAM-1 simulates sediment transport following the removal of a dam behind which the reservoir deposit is composed primarily of non-cohesive sand and silt, and DREAM-2 simulates sediment transport following the removal of a dam behind which the upper layer of the reservoir deposit is composed primarily of gravel. Both models are one-dimensional and simulate cross-sectionally and reach averaged sediment aggradation and degradation following dam removal. DREAM-1 is validated with a set of laboratory experiments; its reservoir erosion module is applied to the Lake Mills drawdown experiment. DREAM-2 is validated with the field data for a natural landslide. Sensitivity tests are conducted with a series of sample runs in the companion paper, Cui et al. (2006), to validate some of the assumptions in the model and to provide guidance in field data collection in actual dam removal projects.