Hydropower facilities often include planned and unplanned maintenance, which require reservoir drawdowns that dewater littoral areas and lead to high mortality of native freshwater mussels (Unionidae). In effort to minimize ecological impacts, hydropower operators in Michigan (USA) are required to rescue and relocate freshwater mussels prior to any drawdown operations. However, the relocation surveys vary depending on the state jurisdiction, the hydroplant operator, and the size and accessibility of the reservoir. Additionally, surveys are effort-intensive and there is little guidance to help operators maximize their results. The objective of this study is to 1) evaluate mussel survey strategies to determine how selection of survey locations affects recovery of mussels, and 2) determine if satellite imagery can estimate bathymetry of remote reservoirs and be used to develop predictive models of species distributions to help operators maximize mussel relocations. Mussels and habitat were surveyed along 10% of the shoreline in three large (~0.5 km2) reservoirs in the Menominee River basin (Michigan, USA) using two approaches to select survey locations—easy access and stratified random. Water depth was estimated for each reservoir using WV-3 imagery and the band-ratio method, and regression models were developed between band ratio and ground-reference depth. In all reservoirs surveyed, the stratified random method was more effective and detected at least twice as many species as the easy access method, including detecting more rare species, as well as up to 8 times more individuals. Bathymetric models produced reasonably good fits for the littoral areas (R2 >0.65), but the performance decreased with depth (> 1 m).  Initial research suggests that more effective guidelines can be developed for hydropower operators to improve mussel recovery and relocation efforts, and there also is potential to use commercially available satellite data to guide surveys in remote reservoirs with little previous information on mussel populations.