Logo do repositório

Percorrer por autor "Simons, Francesca Estelle"

A mostrar 1 - 1 de 1
  • Improving dredging plume prediction models using satellite imagery
    Publication . Simons, Francesca Estelle; Biezen, Tim van der; Ferreira, Óscar
    The dredging industry is essential for land reclamation, coastal protection, navigation, and environmental remediation. However, dredging activities generate suspended sediment plumes that can have economic, social, and environmental impacts. Various types of dredgers, including trailing suction hopper dredgers (TSHDs), generate these plumes, which consist of fine particles suspended in the water column. These plumes affect local ecosystems by reducing light availability and oxygen levels and damaging, for example, seagrass, and corals. As a precaution, Environmental Impact Assessments (EIAs) are carried out to analyse potential impacts and ensure compliance with regulations. Far-field sediment transport models are used to estimate total suspended solids (TSS) as a proxy for turbidity. These models, however, are subject to uncertainty due to unknown sediments conditions and the complexity of plume development processes. One factor of uncertainty is the source term, which represents the sediment flux released during dredging operations. Accurate estimation of this term is challenging due to the complex nature of sediment dynamics and the variability of dredging activities. The Becker method, which involves empirical calculations, is commonly used but introduces uncertainties because it does not take the specific local circumstances into consideration. Sediment transport in dredging plumes is classified into dynamic near-field and passive far field phases. Density and vessel interactions influence the dynamic near-field, while ambient hydrodynamics govern the passive far-field. Accurate modelling requires separate approaches for near- and far-field processes. Far-field models such as Delft3D and MIKE use large-scale simulations to predict sediment dispersion, but they face challenges in estimating source terms, dispersion factors, and fall velocities. Satellite imagery could enhance sediment plume modelling by providing real-time data on plume extents. The conversion of satellite bands to surface turbidity TSS concentration values could aid in the calibration of sediment transport models. Thus the conversion of satellite bands opens the possibilities for feedback loops. Comparing model accuracy during both planning and executable phase. By incorporating satellite data, Boskalis aims to strengthen its toolkit for managing dredge plumes. VI This research aims to improve sediment dispersion parameter selection for Delft3D far-field models by using techniques from AbuShanab (2022) and Boersma (2023). A proof of concept was developed for an existing project in the Philippines. Key objectives included: - Defining the source term, dispersion factor, and fall velocity assumptions. - Enhancing parameter definitions based on satellite imagery analysis. - Evaluating model accuracy and improvement. - Creating a feedback loop for continuous model refinement. This study demonstrated that satellite imagery can help improve the accuracy of sediment transport models. The model that would be simulated in the planning and design phase (PD model) showed high uncertainties in TSS concentrations and in the plume extent % compared to the satellite-based model (SB-model). The satellite imagery is utilised for analysis of the plume behaviour, determining the background concentration values, maximum concentration values and distinguishing the vessels displayed in the imagery with additional help of AIS vessel tracks. With this information and the use of log files, when available, the additional information from the satellite images is utilised to calibrate the SB model. Creating profile plots along the AIS vessel tracks, difference plots and analysis parameters like RMSE, R2 , bias and plume extent % calculations. The SB-model differed from the PD-model in the peak overflow source term 80% exceedance value (P80), plume positioning, IM fraction distribution and dispersion factor. The P80 value was lower than the predicted value and was set to release along the AIS vessel track every minute relative to the stationary source term of the PD-model. The IM fractions of the SB-model were IM1 33% IM2 66% whereas the PD-model had IM1 66% and IM2 33% and the dispersion factor was set to 3 m2 /s instead of 1 m2 /s. Although the SB-model is already more accurate than the original model, there is still room for improvement. This can be achieved by evaluating satellite images with a clean plume and the ship log files should be available. An attempt could be made to calculate an adjustable Peak Overflow source term using the feedback loop to improve model accuracy and environmental management in dredging projects. In conclusion, satellite imagery is a valuable tool for refinement of sediment transport models. Keywords: Satellite imagery, turbidity, sediment dispersion plume models, dredging, source term.
    2024-11-27Dissertação de mestrado Acesso aberto Ver mais