Reservoirs collectively contribute 1–2 percent of global man-made greenhouse gas emissions, although their individual emissions can vary widely. Reservoir emission models have considerably advanced our understanding of the lifetime carbon impacts of reservoirs globally and inherently, offer means to inform judicious planning of new investments. Their widespread adoption is nonetheless hindered by high manual processing requirements, wide prediction uncertainties, and linkages to geospatial drivers that can be obscure for planners. Here, an automated methodology that overcomes and ameliorates these constraints is demonstrated by application to planning hydropower expansion in Myanmar. We show potential hydropower emission reductions of 0.94 MtCO2e /year, while conserving 239 km2 of forest and arable land, and reducing barriers in lower river reaches from 28 to 7. We highlight the benefits of combining detailed empirical models with clear explanations for increasing transparency towards enhancing stakeholder acceptance, informed decision-making, and effective policy design.