4:30 PM - 4:45 PM
[AOS19-05] Numerical study on coagulation of particles and their size distribution in tide-induced bottom nepheloid layer
Keywords:bottom nepheloid layer, turbulence, suspended particle, mean radius
The resuspension and advection of suspended sediment particles (SSPs) in estuary are key processes transporting material from land to open ocean, and they are closely related to coastal environment. Thus, to discuss the amount of material transport by SSPs, it is necessary to understand the mechanisms of these processes.
The radius of the SSP is crucial because it affects particle characteristics such as sinking velocity and surface area, and there are observational studies focusing on the size distribution of SSP (e.g., Furuichi et al., 2017). There are also numerical studies investigating the mechanisms of the transport (e.g., Simionato and Moreira, 2018), however, they have not fully considered the coagulation process, which is thought to be important for the radius of SSP in the bottom nepheloid layer.
Therefore, we aim to make more precise estimation on the amount of material transport by considering coagulation of SSP in the turbulent nepheloid layer. As the first step to this goal, here we numerically studied on how the distribution of SSP responses to the turbulent flow with coagulation. To simulate the turbulence, we employed the LESs for calculating flow fields. We employed the Lagrange-type particle model, which was modified from that of Riechelmann et al. (2012). In this model, groups of particles were tracked. Particles in two groups coagulate when distance between the groups become less than a critical value, and this led to changes in their radii and particle numbers. The motion of the particle group was approximated as linear combination of advection (by background flow) and sinking. Three elementary coagulation processes – advection shear, differential sinking, and Brownian motion – were considered, and they all were parameterized by coagulation kernels (e.g., Burd and Jackson, 2009).
In the present study, the bottom turbulence was excited by the tidal volume force (period = 1/2 day) oscillating in x-direction imposed on the whole domain of the model ocean on a f-plane with vertical temperature gradient. The model domain is cubic with periodic in horizontal boundaries, slippery at the top and no-slip at the bottom. The groups of SSPs (density = 2 g/cm3) with a constant radius were injected at a constant number flux at the bottom. These SSPs were advected, coagulated, sunk, and finally reached to the bottom again. Then, these SSPs were immediately removed. Calculations were lasted until the turbulent flow fields and the particle distributions become statistically steady. Two cases in the radius of input SSP (10 µm, 30 µm) and two cases in the amplitude of tidal flow (50 cm/s, 25 cm/s) were considered.
As a result, consideration of coagulation led to increase in the mean radius and decrease in the number of SSPs. These changes consequently enlarged the sum of individual surface areas and volumes of SSPs, which have large impact on the amount of material transport. The number and the mean radius of SSPs were oscillating at the same frequency as the magnitude of tidal flow, and their time averages varied with the amplitude of tide and the input radius. This indicated that the input radius affected the response time of particle distribution on background flow.
Our results suggested that incorporating coagulation process into particle simulation is important to estimate the amount of material transport.
Other detailed results and discussions will be displayed at the presentation.
The radius of the SSP is crucial because it affects particle characteristics such as sinking velocity and surface area, and there are observational studies focusing on the size distribution of SSP (e.g., Furuichi et al., 2017). There are also numerical studies investigating the mechanisms of the transport (e.g., Simionato and Moreira, 2018), however, they have not fully considered the coagulation process, which is thought to be important for the radius of SSP in the bottom nepheloid layer.
Therefore, we aim to make more precise estimation on the amount of material transport by considering coagulation of SSP in the turbulent nepheloid layer. As the first step to this goal, here we numerically studied on how the distribution of SSP responses to the turbulent flow with coagulation. To simulate the turbulence, we employed the LESs for calculating flow fields. We employed the Lagrange-type particle model, which was modified from that of Riechelmann et al. (2012). In this model, groups of particles were tracked. Particles in two groups coagulate when distance between the groups become less than a critical value, and this led to changes in their radii and particle numbers. The motion of the particle group was approximated as linear combination of advection (by background flow) and sinking. Three elementary coagulation processes – advection shear, differential sinking, and Brownian motion – were considered, and they all were parameterized by coagulation kernels (e.g., Burd and Jackson, 2009).
In the present study, the bottom turbulence was excited by the tidal volume force (period = 1/2 day) oscillating in x-direction imposed on the whole domain of the model ocean on a f-plane with vertical temperature gradient. The model domain is cubic with periodic in horizontal boundaries, slippery at the top and no-slip at the bottom. The groups of SSPs (density = 2 g/cm3) with a constant radius were injected at a constant number flux at the bottom. These SSPs were advected, coagulated, sunk, and finally reached to the bottom again. Then, these SSPs were immediately removed. Calculations were lasted until the turbulent flow fields and the particle distributions become statistically steady. Two cases in the radius of input SSP (10 µm, 30 µm) and two cases in the amplitude of tidal flow (50 cm/s, 25 cm/s) were considered.
As a result, consideration of coagulation led to increase in the mean radius and decrease in the number of SSPs. These changes consequently enlarged the sum of individual surface areas and volumes of SSPs, which have large impact on the amount of material transport. The number and the mean radius of SSPs were oscillating at the same frequency as the magnitude of tidal flow, and their time averages varied with the amplitude of tide and the input radius. This indicated that the input radius affected the response time of particle distribution on background flow.
Our results suggested that incorporating coagulation process into particle simulation is important to estimate the amount of material transport.
Other detailed results and discussions will be displayed at the presentation.