博士 / 國立中山大學 / 海洋環境及工程學系研究所 / 106 / Flocculation of Suspended Particulate Matter (SPM) in estuarine and coastal environments is a complex process that is influenced by physical, biological, and chemical mechanisms. The bio-mineral flocculation model of Maggi (2009) used in this study was adapted to simulate flocculation under various weather conditions and during different seasons. The adaptation incorporated the effect of tide–wave-combined turbulence on floc dynamics and the choice of model parameter in order to simulate biological effects that are responsible for the observed seasonal variation in floc size. The model was calibrated and validated using in situ data of floc size, current velocity, turbulence and SPM concentration from the high turbid Belgian nearshore area (southern North Sea). The results show that tide-wave-combined turbulence needs to be incorporated when simulating flocculation in a tide-wave-dominated environment. Additionally, the results confirmed that floc strength has a seasonal influence on floc development.
The flocculation model has nine parameters that have to be determined. In order to investigate the sensitivity of these parameters on the model output and to find an optimized parameter set for the model, the Design of Experiment method was used. The results have shown that the most sensitive parameters are the primary particle size, fractal dimension, aggregation, breakage and floc strength. The aggregation parameter and the floc strength, which both enhance aggregation, are more dominant in summer than in winter. On the contrary, the breakage parameter is more important in winter. A stronger floc-binding strength was observed in the summer season (April–September), during which flocculation was influenced by abundant sticky organic substances, compared with the weak-biomass winter season (October–March). This seasonal variation in floc size (Fettweis et al., 2014) was reproduced in the model using varying values for various floc strengths in different seasons. The results revealed that the biological effect should be incorporated in the model by using floc strength values optimized for seasons.
The flocculation model uses a single characteristic diameter (i.e. D50) as time-dependent variable. Because of this the model performs better when the Particle Size Distribution (PSD) of the flocs is unimodal. In case of multimodal PSDs the model output is less accurate. Using a multimodal on flocculation model could possibly improve results during certain conditions but may cost computation time. On the other hand, the LISST instrument, which was used to measure the in situ PSD, may, besides of the measuring uncertainty, also capture particles that are not flocs (sand grains) and that can show up as additional modes in PSD during e.g. storm conditions.
The model was further validated using long-term in situ data. The data set shows variations in floc size that depend on weathers and seasons. In order to simulate the floc behavior, season and weather optimized parameter sets need to be included. Currently the biological effects are parameterized as a function of time (summer-winter). This could be refined in future studies by using e.g. light intensity, temperature or the coupling with a biological (nutrient dynamics) model.
| Identifer | oai:union.ndltd.org:TW/106NSYS5282007 |
| Date | January 2018 |
| Creators | Pei-hung Chen, 陳沛宏 |
| Contributors | Jason C.S. Yu, 于嘉順 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 130 |
Page generated in 0.0188 seconds