Back to search

Institutes outside Greece

Developing a multi-scale parallelised coupled system for wave-current interactions at regional scales

Abstract

dc:description

At coastal areas, the interplay between waves and currents is crucial. This interaction impacts many phenomena and applications, highlighting the necessity for accuracy and speed in the numerical representation of Wave-Current Interactions (WCI). These applications encompass a wide spectrum, including coastal morphology, sediment transport, offshore structure scouring, pollutant mixing, infrastructure design, marine energy projects, and storm surges. The complexity in representing WCI stems from incorporating multi-scale processes with diverse temporal and spatial scales. For example, wind wave periods range from seconds to hours, while the wavelengths span from centimetres to kilometres. In contrast, tides showcase much larger scales with periods in the order of hours and wavelengths in the order of thousands of kilometres. Practically, reconciling all these processes and scales within a single model is improbable, leading to the need for coupled systems to address this challenge. This study presents the development of a Python-interfaced multi-scale parallelised coupled modelling system for WCI. It is formed by coupling the spectral wave model Simulating WAves Nearshore (SWAN) with the 2-D shallow-water equation hydrodynamics model Thetis. The coupling is facilitated by the Basic Model Interface (BMI), a lightweight generic coupling interface. The impact of waves on current is introduced via the radiation stress formulation, accompanied by the integration of wave-roller effects. Two coupling options are offered: online and offline. The online choice supports both one-way and two-way coupling, while the offline alternative is focused on one-way coupling. Considering that only few existing WCI models report on validation in controlled environments, a suite of benchmarking scenarios is established consisting of analytical and experimental scenarios in quasi 1-D and 2-D configurations. In these cases, sensitivity analyses are performed spanning various parameters in both models. The results underscore the importance of customising each coupled configuration when WCI are prominent, rather than solely relying on recommended or “default” values. Calibrated results align well with the data and often showcase the same level of accuracy as other 3-D WCI. This efficiency means less computational cost, as the developed model converges faster and requires less CPU time compared to alternative options. A month-long numerical representation of the field configuration located in Duck, North Carolina, investigates the coupled system's performance under moderate wind conditions. This scenario serves to assess the influence of various coupling approaches on its predictions. Since this area is primarily influenced by waves and features low current speeds, the coupling modes have minor impact on wave predictions. However, with coupling modes transitioning from no to two-way coupling, the hydrodynamics predictions exhibit substantial improvement in regions where WCI are evident. The improved accuracy does not encompass areas characterised by rip currents or other processes that require a vertical discretisation for their hydrodynamics. Discrepancies between online and offline one-way coupling configurations are evident, with the most pronounced differences observed in the SWAN-to-Thetis coupling. They can be attributed to different interpolation methodologies. Ultimately, the WCI system is applied in a regional configuration within the Orkney archipelagos, UK. Specifically, the model simulates the waters of Westray Firth, a region known for its energetic tidal conditions, to assess its capacity for effectively depicting WCI phenomena in regional scales. Our predictions correlate well with the observations, accurately mirroring the sinusoidal pattern of the measured wave parameters, usually attributed to tidal effects. Furthermore, our model showcases similar precision to a 3-D WCI coupled system implemented in the same region at lower computational cost. The coupled system developed during this thesis presents an efficient tool for incorporating WCI phenomena across various scales, exhibiting performance comparable to its 3-D counterparts. Its efficiency is highlighted by: (a) minimising computational resource usage, as evidenced by a 38% reduction in the number of cores employed during the Westray Firth application; (b) reducing elapsed real times; and (c) accelerating convergence, such as achieving convergence 1.4 to 18 times faster in benchmarking scenarios. It provides a crucial foundation for researchers and stakeholders that seek to adopt a precise and efficient solution, independent of the 3-D nature of WCI. This unlocks new opportunities for its versatile employment in a range of applications spanning from initial research and decision-making stages to optimisation studies and to the development of forecasting systems.

Degree

thesis:*
Grantor dc:publisher
Institutes outside Greece
Year dc:date
2023

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Fragkou, Anastasia
  • Φράγκου, Αναστασία

Subjects

dc:subject × 23

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/57493

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Fragkou, Anastasia; Φράγκου, Αναστασία. Developing a multi-scale parallelised coupled system for wave-current interactions at regional scales. Institutes outside Greece, 2023. http://hdl.handle.net/10442/hedi/57493