DeltaRCM+ADCIRC

The hydrodynamic simulations were conducted with version 53.04 of the advanced circulation model, ADCIRC (https://adcirc.org/). A Matlab script (AutoSwapEdges.m) is utilized to convert the structured set of gridded points generated by DeltaRCM to a finite element mesh in ADCIRC format (*.14) that optimizes bathymetric connectivity (see Appendix A). The ADCIRC input files include the finite element mesh (*.14) representations of the different delta morphologies, the input parameters (*.15) and the nodal attributes (*.13). The primary ADCIRC output files include the global velocities in time-series (*.63) and harmonic (*.54) formats. Two Fortran codes are utilized sequentially to generate export volumes from the ADCIRC-simulated velocities (*.63): 1) mass_residuals.f; and 2) tidal_prism.f. The Surface-water Modeling System (SMS) by Aquaveo (https://old.aquaveo.com/software/sms-surface-water-modeling-system-introduction) was utilized to read the ADCIRC harmonically decomposed velocities (*.54) and process the data for categorization of tidal bands. The ADCIRC input files (*.13, *.14 and *.15) are consolidated into a directory named InputsADCIRC. The file naming scheme is SLRX_f#, where X is 000, 005, 010 or 020 for sea-level rise rates of 0, 5, 10 and 20 mm year–1, respectively, and # is 25, 50, 75 or 100 for sand contents of 25, 50, 75 and 100%, respectively. The Matlab script used for converting DeltaRCM-generated gridded points to an ADCIRC finite element mesh optimized for bathymetric connectivity are consolidated into a directory named AutoSwapEdges. The two Fortran codes used for data processing of velocities-to-export volumes are consolidated into a directory named ScriptsProcessing. The codes are compatible with GNU Fortran (gfortran). The entirety of the above directories can be extracted from a TAR package housed on Zenodo, doi: 10.5281/zenodo.14911850. Appendix AThe problem begins with a structured set of gridded points where each point (x,y) contains a scalar value, z, representative of the local elevation (positive values = topography above datum; negative values = bathymetry below datum). The objective is to triangulate the gridded points – (x,y,z)j,k, where j and k are counters for the number of rows Nj and columns Nk in the structured points set – in a way that maximizes bathymetric connectivity in the subsequent finite element mesh. Performing on a local basis over the full set of gridded points (Nj, Nk) generates a triangulation that globally maximizes bathymetric connectivity. To illustrate an example of local evaluation, consider four gridded points structured in Cartesian space, (0,0,–1), (0,1,1), (1,1,–1) and (1,0,1), where the notation is (x,y,z). Now, this local set of four gridded points can be defined by two triangular elements in one of two ways, with the diagonal set as y = x (upward sloping) or y = –x (downward sloping). The logic takes the average of the z-values of the two diagonal endpoints for both potential triangulation arrangements. Then the lower of the two average-z values determines the triangulation that is prescribed for the local set of four gridded points. This local evaluation embedded within a loop that spans the full set of gridded points results in a complete triangulation that maximizes bathymetric connectivity. In the example, the averages would be applied as: 1) average-z[(0,0,–1), (1,1,–1)] = –1; and 2) average-z[(0,1,1), (1,0,1)] = 1. Therefore, the two triangular elements would be defined with the diagonal set as y = x (upward sloping) and the endpoints at (0,0,–1) and (1,1,–1). The essence of the algorithm is to ensure maximal conveyance of water (alternatively, minimal artificial blockage of water conveyance) in the solution of the shallow-water equations over the respective finite element mesh.