Automatic, unstructured mesh optimization for simulation and assessment of tide- and surge-driven hydrodynamics in a longitudinal estuary: St. Johns River
A localized truncation error analysis with complex derivatives (LTEA+CD) is applied recursively with advanced circulation (ADCIRC) simulations of tides and storm surge for finite element mesh optimization. Mesh optimization is demonstrated with two iterations of LTEA+CD for tidal simulation in the lower 200 km of the St. Johns River, located in northeast Florida, and achieves more than an over 50% decrease in the number of mesh nodes, relating to a twofold increase in efficiency, at a zero cost to model accuracy. The recursively generated meshes using LTEA+CD lead to successive reductions in the global cumulative truncation error associated with the model mesh. Tides are simulated with root mean square error (RMSE) of 0.09–0.21 m and index of agreement (IA) values generally in the 80s and 90s percentage ranges. Tidal currents are simulated with RMSE of 0.09–0.23 m s−1 and IA values of 97% and greater. Storm tide due to Hurricane Matthew 2016 is simulated with RMSE of 0.09–0.33 m and IA values of 75–96%. Analysis of the LTEA+CD results shows the M2 constituent to dominate the node spacing requirement in the St. Johns River, with the M4 and M6 overtides and the STEADY constituent contributing some. Friction is the predominant physical factor influencing the target element size distribution, especially along the main river stem, while frequency (inertia) and Coriolis (rotation) are supplementary contributing factors. The combination of interior- and boundary-type computational molecules, providing near-full coverage of the model domain, renders LTEA+CD an attractive mesh generation/optimization tool for complex coastal and estuarine domains. The mesh optimization procedure using LTEA+CD is automatic and extensible to other finite element-based numerical models. Discussion is provided on the scope of LTEA+CD, the starting point (mesh) of the procedure, the user-specified scaling of the LTEA+CD results, and the iteration (termination) of LTEA+CD for mesh optimization.
KeywordsTides Currents Circulation Finite element Mesh generation Estuaries
D. Michael Parrish and Matthew V. Bilskie provided assistance with the implementation of LTEA+CD. The author thanks Scott C. Hagen for inspiration of the concept and providing detailed advice. The author also thank the three anonymous reviewers for providing insightful comments that served to vastly improve the paper. The data used in this study are available from the cited references.
The author did not receive any funding support for this work.
Compliance with ethical standards
Conflict of interest
The author declares that he has no competing interests.
- Advanced Circulation (ADCIRC) (2017) The official ADCIRC website. http://adcirc.org/. Accessed online 17 Mar 2017
- Bourgerie R (1999) Currents in the St. Johns River, Florida: spring and summer of 1998. NOAA technical report NOS CO-OPS 025, Washington, DC, 67 pGoogle Scholar
- Chang S, Haworth DC (1995) Kinetic-energy-balance based solution-adaptive mesh refinement. In: Johnson RW, Hughes ED (eds) Quantification of uncertainty in computational fluid dynamics—Volume 213. American Society of Mechanical Engineers, New York, pp 7–12Google Scholar
- Danish Hydraulic Institute (DHI) (2016) Mike 21: 2D modeling of coast and sea. https://www.mikepoweredbydhi.com/products/mike-21. Accessed online October 10, 2016
- Deltares (2016) Delft3D 4 suite (structured). https://www.deltares.nl/en/software/delft3d-4-suite/. Accessed online October 10, 2016
- Franca AS, Haghighi K, Oliveira LS (1995) Error estimation and adaptivity in finite element analysis of transport problems. In: Johnson RW, Hughes ED (eds) Quantification of uncertainty in computational fluid dynamics—Volume 213. American Society of Mechanical Engineers, New York, pp 19–24Google Scholar
- Luettich RA, Westerink JJ, Scheffner NW (1992) ADCIRC: an advanced three-dimensional circulation model for shelves, coasts and estuaries, I: theory and methodology of ADCIRC-2DDI and ADCIRC-3DL. Technical report DRP-92-6, Washington, DC, 144 pGoogle Scholar
- National Oceanographic and Atmospheric Administration-National Ocean Service (NOAA-NOS) (2017) Center for Operational Oceanographic Products and Services. http://tidesandcurrents.noaa.gov/. Accessed online March 6, 2017
- Open Telemac-Mascaret (2016) The mathematically superior suite of solvers. http://www.opentelemac.org/. Accessed online October 10, 2016
- Parrish DM, Hagen SC (2007) 2D unstructured mesh generation for oceanic and coastal tidal models from a localized truncation error analysis with complex derivatives. Int J Comput Fluid Dyn 21:277–296Google Scholar
- Toth DJ (1993) Volume 1 of the Lower St. Johns River basin reconnaissance: Hydrogeology. Technical Report SJ93-7, St. Johns River Water Management District, 58 pGoogle Scholar
- XMS Wiki (2017) SMS: Paving. http://www.xmswiki.com/wiki/SMS:Paving. Accessed online March 14, 2017