Hydropower reservoirs, like all inland waters, emit greenhouse gases (GHGs) to the atmosphere. However, GHG emission estimates are highly uncertain, in part due to spatiotemporal variation in the physical, chemical, and biological factors that affect the generation and release of GHGs from reservoirs. Here, we discuss findings from three complementary studies. First, examination of GHG emissions measured in six large hydropower reservoirs in the southeastern US revealed that CO2 diffusion was the dominant emission pathway, and that CH4 ebullition was the most variable both within and across reservoirs. Second, a study of monthly ebullitive and degassing emissions measured in one reservoir illustrated influences of reservoir operations on GHG emissions, with increased ebullitive emissions during water level drawdown in fall, and high variability in degassing emission estimates due to considerable day-to-day variation in the volume of water released from the dam. Third, an analysis of spatial variation in GHG emissions by sampling 200 sites across a 115 km2 reservoir revealed strong spatial gradients in CO2 and CH4 diffusive emissions, with higher emission rates near tributary inputs to the reservoir. Together, these studies provide insight into the spatial and temporal scales in which uncertainty in GHG emission estimates may be reduced and can inform future measurement and modeling efforts.