Overview
What is ADCIRC Subgrid?
The ADCIRC Subgrid Preprocessor is a Python package designed to enhance the accuracy of ADCIRC hydrodynamic models by generating subgrid correction terms. These corrections account for sub-mesh scale features that cannot be explicitly resolved in the computational mesh but significantly affect the flow dynamics.
Background and Motivation
Traditional hydrodynamic modeling faces a fundamental trade-off between computational efficiency and spatial resolution. Higher resolution meshes can capture more detailed topographic features but come at significantly increased computational cost. Subgrid methods provide an elegant solution by:
Maintaining computational efficiency with coarser mesh resolutions
Incorporating fine-scale topographic effects through correction terms
Improving model accuracy without proportional increases in computational cost
The subgrid approach is particularly valuable for:
Storm surge modeling where small-scale topographic features affect flooding patterns
Coastal applications with complex bathymetry and varying land cover
Large-domain simulations where uniform high resolution is computationally prohibitive
Key Concepts
Subgrid Correction Terms
The subgrid method modifies the standard shallow water equations by introducing correction terms that account for:
Sub-grid scale bathymetry variations: Capturing the effect of topographic features smaller than the mesh resolution
Manning’s roughness heterogeneity: Accounting for sub-mesh scale variations in bottom friction
Wetting and drying dynamics: Improving the representation of complex shoreline interactions
Mathematical Framework
The subgrid corrections are implemented through modification of the continuity and momentum equations:
Averaged Continuity Equation:
Averaged Momentum Equations (simplified form):
Where: - \(\langle \cdot \rangle_G\) indicates grid-averaged quantities - \(\langle \cdot \rangle_W\) indicates wet-averaged quantities - \(C_{UU}, C_{VU}\) are advection correction coefficients - \(C_{M,f}\) is the bottom friction correction coefficient - \(\phi\) is the wet area fraction (0 ≤ φ ≤ 1)
The φ Factor
The central concept in subgrid modeling is the φ (phi) factor, which represents the fraction of a mesh element that is wet at a given water level. This factor:
Varies from 0 (completely dry) to 1 (completely wet)
Is computed based on high-resolution topographic data within each mesh element
Accounts for complex sub-mesh topography and wetting patterns
Data Requirements
The subgrid preprocessor requires several input datasets:
Essential Inputs:
ADCIRC Mesh File (
fort.14
): Defines the computational grid and bathymetryDigital Elevation Model (DEM): High-resolution topographic/bathymetric data
Land Cover Data: Classification data for Manning’s roughness assignment
Manning’s Roughness Lookup: Mapping from land cover classes to roughness values
Data Quality Considerations:
Coordinate System Consistency: All datasets should use consistent coordinate reference systems (WGS84 recommended)
Resolution Requirements: DEM resolution should be significantly higher than mesh resolution for effective subgrid corrections
Coverage: Input datasets should completely cover the computational domain
Workflow Overview
The typical subgrid generation workflow consists of:
Data Preparation: Ensuring all input datasets are properly formatted and georeferenced
Configuration: Creating YAML configuration files specifying inputs and processing options
Processing: Running the preprocessor to compute subgrid corrections
Quality Control: Using visualization tools to verify results
Model Integration: Incorporating the subgrid file into ADCIRC simulations
Advantages and Limitations
Advantages:
Improved accuracy for flow over complex topography
Computational efficiency compared to high-resolution meshes
Flexible configuration options for various applications
Integration with existing ADCIRC workflows
Limitations:
Requires high-quality input data (DEM and land cover)
Additional preprocessing step in the modeling workflow
May not capture all physical processes affected by small-scale features
Effectiveness depends on the quality of underlying assumptions about sub-grid processes
Applications
The ADCIRC Subgrid Preprocessor is particularly well-suited for:
Hurricane storm surge modeling in areas with complex coastal topography
Flood inundation studies requiring accurate representation of urban infrastructure
Coastal restoration project assessment where small-scale features affect flow patterns
Climate change impact studies requiring efficient large-domain simulations
This overview provides the foundation for understanding how the subgrid preprocessor works and when to apply it effectively. The following sections provide detailed guidance on installation, configuration, and usage.