The identification of potential coastal inundation caused by future sea level rise requires not only time series records from tide gauges, but also high-quality digital elevation models (DEMs). This study assesses the importance of DEM vertical accuracy in predicting inundation by sea level rise along the Valdelagrana beach and marshes of the Bay of Cádiz (SW Spain). A presentday (2000) and a projected (2100) high tide have been spatialized over a traditional (aerial photogrammetry) regional DEM of Andalusia with a horizontal spatial resolution of 10 m and a vertical accuracy of 0.68 m RMSE (root mean square error), and a LIDAR-derived DEM of the Valdelagrana study site with the same spatial resolution but a vertical accuracy of 0.205 m RMSE. The simulations are based on a bathtub model, which accounts for the effect of vertical barriers. The results reveal that the presence of infrastructures such as roads and salterns is the key to delimit the extent of water penetration during high tides in an otherwise homogeneously flat area comprising the beach and marshes of Valdelagrana. Moreover, in comparison with the highly accurate LIDAR DEM, the inundation areas derived from the lower-resolution DEM are overestimated by 72 % and 26 % for the present-day and future scenarios respectively. These findings demonstrate that DEM vertical accuracy is a critical variable in meaningfully gauging the impacts of sea level rise.