In this thesis, I aim to contribute to the research towards improving the sustainability of habitat life in streams affected by anthropogenic activities. With the growing trend of altering natural water regimes due to hydropower operations, irrigation, and wastewater treatment plants, for instance, it is vital to minimize environmental effects. Here, I aim to address issues regarding human-altered streams that have received less attention in the community. In particular, I intend to provide solutions to improve the water management in the reservoir systems as well as investigating the role of macroroughness elements on the ecosystem functioning. For the water management part, the aim is to improve the global efficiency of storage systems, with the main focus being on the environmental aspect. Minimum flow releases nowadays are widely used to reduce the environmental effects associated with streams affected by anthropogenic activities. However, these static flow release rules have been found to markedly affect the stream ecosystem, mainly by reducing biodiversity. Aiming to reduce the environmental impacts caused by reservoirs, I propose a Direct Policy Search (DPS) framework based on defining dynamic flow release rules. I employ the mathematical form of the Fermi-Dirac statistical distribution to formulate dynamic non-proportional redistribution rules that partition the flow for energy production and environmental use. The energy production is calculated from technical data and the environmental indicator associated with flow releases by integrating the Weighted Usable Area (WUA) for fishes with Richter’s hydrological indicators. State-of-the art multiobjective evolutionary algorithms are used to find optimal flow release policies that have efficient economic and environmental efficiency. The results show that non-proportional flow releases can substantially improve global efficiency, specifically the ecological one, of the hydropower system when compared to minimal flow releases. This is mainly due to better water management of flood events that enables flow releases that mimic natural flow variability. In the part of my Thesis on streambed macroroughness, I investigate the contribution of macroroughness elements to stream ecosystem functioning, which generally is not considered due to their relatively small scale, using both modeling and fieldwork approaches. Macroroughness elements contribute to stream ecosystem functioning mainly by providing shelter zones for fishes as well as enhancing reach scale gas exchange. I develop an environmental indicator based on the aforementioned benefits of macroroughness elements on stream ecosystem functioning. Environmental indicators help define improved water management strategies by better characterizing the environmental impacts associated with human altered flow regimes. I define a physically based analytical model to estimate the size of wake areas downstream of macroroughness elements and make use of the derived distribution approach to calculate the statistical distribution of wake areas. To validate the concept, the model is applied to four exemplary streams having different statistical diameter size distributions of macroroughness elements. Furthermore, the proposed environmental indicator is included in the DPS framework to solve a multiobjective reservoir optimization problem. Next, empirical evidence is provided for the effects of macroroughness elements on stream ecosystem function