A sensitivity study is undertaken to assess the utility of different onshore digital elevation models for simulating the extent of tsunami inundation using case studies from two locations in Indonesia. We compare airborne interferometric synthetic aperture radar (IFSAR), ASTER and SRTM against high resolution LiDAR and stereo-camera data in locations with different coastal morphologies. Tsunami inundation extents modelled with airborne IFSAR are comparable with those modelled with the high resolution datasets and are also consistent with historical run-up data, where available. Large vertical errors and poor resolution of the coastline in the ASTER and SRTM elevation models cause modelled inundation to be much less compared with the other datasets and observations. Therefore ASTER and SRTM should not be used to underpin tsunami inundation models. Model mesh resolution of 25 m is sufficient for estimating the inundated area when using elevation data with high vertical accuracy. Differences in modelled inundation between digital terrain models (DTM) and digital surface models (DSM) for LiDAR and IFSAR are greater than differences between the two data types. Models using DTM may overestimate inundation while those using DSM may underestimate inundation when a constant Mannings roughness value is used. We recommend using DTM for modelling tsunami inundation with further work needed to resolve the scale at which surface roughness should be parameterised.
