Forward energy grade line analysis for tsunami inundation mapping

A simplified model using 1D topobathymetric profiles for generating tsunami inundation maps is implemented and evaluated. The approach is a modification of the ASCE Energy Grade Line Analysis, that allows estimation of the maximum inundation distances using an iterative method. The modified methodology is implemented in three coastal cities in central Chile and compared with a database of 5400 full tsunami simulations obtained from a Nonlinear Shallow Water Equations solver. The key parameter of the model is based on the Froude number, for which three parameterizations and a range of values are tested. Results show that errors in the estimation of the areal extent of the inundation can be as low as 4%, after calibration. However, calibration is site specific and the optimal solution depends on the geographical characteristics of the area of interest. A sensitivity analysis based on the aleatoric sampling of the full tsunami simulation database show that as little as 100 inundation maps are required to perform the calibration of the model. This is a manageable number that offers reduced computational costs when compared with full tsunami simulations, and even those required to train other surrogate models using machine learning.