The brown trout (Salmo trutta L.) is one of the vertebrate species presenting the highest degree of intraspecific biological diversity. Human activities are threatening this biodiversity, and many endemic populations now face a medium-term risk of extinction. An individual-based model called DemGenTrout was developed to improve the management of these populations, by providing (i) a descriptive tool to study the functioning of a brown trout population at the stream scale, (ii) a predictive tool to evaluate the medium-term impacts of anthropogenic disturbances on native populations. The DemGenTrout model was first parameterized with demographic, genetic, and environmental data collected over 5 years on the Lesse River drainage (Belgium). A methodology for attributing a birth year to tagged trout of this hydrological system was developed. This information was used to determine the values of several parameters of the main four processes of the model (i.e., survival, growth, reproduction, movement). Second, the model was optimized and validated using the Pattern-oriented modelling strategy, which consists of systematically testing how well the model reproduces multiple patterns observed in the field. Third, the sensitivity of the model to its parameters was analysed. Two scenarios mimicking anthropogenic disturbances were simulated over 35 years: (i) a barrier to upstream spawning migration, (ii) stocking with hatchery-reared trout during a 10-year period. Both of them appeared to have a strong short-term impact on the demogenetic structure of the wild trout population. The migration barrier mostly impacted abundance, whereas genetic issues arose when a significant number of stocked fish survived in the wild. Stocking also appeared to act on a longer time frame if hatchery and wild trout had similar survival and spawning probabilities. ; (AGRO 3) -- UCL, 2012