Calibration and assessment of two process-based numerical models, XBeach and CShore, for predicting beach profile evolution in southern California

Prediction of cross-shore beach profile response to storms is complex due to incomplete understanding of physics, imperfect knowledge of in-situ conditions and numerical error accumulation over morphological time scales. Process based morphological models are typically semi-empirical and require site specific calibration. However, cost and effort have historically limited calibration efforts over a range of conditions and locations. Here, six events with varying duration, wave intensity and beach slopes are used to calibrate and assess two process-based cross-shore models, CShore and XBeach, at two southern California beaches (Cardiff and Torrey Pines). Model performance is quantitatively evaluated and presented using high resolution spatiotemporal survey observations along with Brier Skill Scores (BSS) and a novel Beach Elevation Change Error (BECE) metric that considers alongshore-averaged upper beach volume. XBeach is tested using both default parameters and site-specific calibrated parameters. Calibration improved results, but skill often remained low and the offshore beach bar formation was not predicted accurately. Calibration revealed a sensitivity to the depth-extent of the input beach profiles; upper beach profiles produced different optimal parameters compared to beach profiles extending to the depth of closure. CShore was tested using both the 'Atlantic' and 'Pacific' parameter sets, and the best-performing set ('Atlantic') showed better skill in predicting the offshore beach bar formation, but overall model skill was also generally low.