Changes in a brown trout (Salmo trutta L.) population result from interaction among various mechanisms which are dependent on environmental conditions and biotic processes. In reaches influenced by the presence of dams, the instream flow in the bypassed section is not the only parameter which affects the population. Flood episodes, the general connectivity of the bypassed section, and the characteristics of the substrate which define the availability and quality of spawning grounds may also have a crucial impact. The design and fine-tuning of tools which take environmental parameters into account can improve our understanding of the dynamics of such influenced populations. In this perspective, a deterministic model (MODYPOP) has been developed in an attempt to integrate all these factors and to test the effect of different long-term scenarios of influenced flow regimes on the structure of trout populations. MODYPOP was applied to three populations and three reaches (on the Roizonne, Neste d'Aure and Lignon du Forez rivers in France). For each stream, experiments were carried out on a bypassed section downstream of a hydropower station, before and after an increase in the minimum instream flow due to relicensing. These experiments allowed integrating into MODYPOP local phenomena (impact of flood episodes, impact of flushing, impact of downstream migration of juveniles and adults) affecting the populations during the study period and then calibrating them. To estimate the change in the population due to the increase in minimum instream flow, different long-term simulations were run, selecting discharge patterns at random. These scenarios help to evaluate the time required for the population to return to a range close to habitat saturation after an improvement in the hydraulic habitat or following a flood event. These applications have enabled determining the relative importance of changes in population density due to different types of events. Copyright © 2008 John Wiley & Sons, Ltd.