The importance of keeping river environments healthy drives the scientific community towards the improvement of sustainable and validated environmental monitoring approaches. Accurate data on the state of the ecosystems provided rapidly are key in order to correctly assess, which interventions and management decisions are suitable, and which must be avoided. This paper analyses a rapid non-intrusive approach to change detection in surface flow patterns near fish passages at hydropower dams with the goal to improve the understanding of factors influencing fish passage discoverability. This, in turn, is of great relevance to the sustainability of migrating riverine fish populations from both ecological and economical perspectives. The present study includes three unique experiments performed at a large-scale hydropower dam site with an integrated fish passage under controlled discharge conditions. The analysis is performed with the use of the freely available KLT-IV software. The use of an Unmanned Aerial System (UAS) as a camera carrier platform provides the key flexibility in terms of any study site selection. The use of KLT-IV speeds up and simplifies flow pattern analysis, especially when compared to labour-intensive modelling relying on point-based ground truth data. In this paper, we demonstrate that the selected approach can be effectively applied to identify changes in surface flow patterns both in terms of flow velocity magnitudes and in terms of flow directions. It shows that the identification of actual flow patterns near the fish passage entrance provides more information on the potential discoverability of the fish passage than traditionally measured bulk discharge values alone.