Water quality in lakes and river systems has deteriorated worldwide due to intensification in land use and associated nutrient loading or changes in natural flow regimes. The most obvious impacts are increase in the frequency of harmful algal blooms caused by potentially toxic cyanobacteria and fish kills due to hypoxia. Other problems are not immediately visible or have indirect impacts like contamination by metals and pathogens, or vector borne diseases depending on wetting and increased temperature. To reduce health and economic risks posed by such water quality issues, there is an increased need for early warning systems. While Earth observation of inland aquatic systems can give an account of historic conditions and current state, integrating hydrodynamic and hydrologic modelling tools with predictive capabilities allow for timely intervention and optimised management options. On a local scale Earth observation can be used to drive hydrodynamic simulations for short term prediction of harmful algal blooms in specific water bodies allowing for early warning and providing operating strategies for risk minimisation for, e.g., water treatment plants or reservoirs (case studies shown here). Combined with local hyperspectral sensors it is even possible to discriminate cyanobacteria species based on their pigments and thus infer potential toxicity. A generalisation of these methods on a regional or continental scale not only yields an early warning account for a larger region, e.g. state wide, but can yield a risk estimation based on weather forecast. In combination with hydrologic modelling tools EO is applied in ecological impact studies of flood inundation, e.g., the generation of hypoxic conditions in lowland rivers, or the spread of a carp virus for pest eradication in a large basin. Although there is a large spectrum of water quality issues where EO can lead to better insight, spatial and temporal resolution of satellite sensors limits their application. Other techniques of remote sensing are necessary to fill these gaps.