Manning's n values for open channel (river) flow have been studied by hydraulic engineers since the late 19th century, and a rich literature exists on the topic including large-scale laboratory experiments and actual field measurements. Both river flood models and shallow water equation tsunami inundation models incorporate the importance of varying equivalent roughness values with the large-scale roughness elements present for different land use types. However, many tsunami models (especially in Japan) use n values based on a very limited set of small-scale model laboratory experiments with inappropriate Reynolds and Weber numbers, instead of using Manning's n values from the open channel flow literature. Due to this, equivalent Manning's n values for vegetated and urban areas in these tsunami inundation models are too small, causing the mitigating effect of forests and urban regions to be underestimated. This paper presents a review of Manning's n research applied to both river flood and tsunami inundation models, and suggests values to improve the reliability of the latter.