**Optical Methods Applied to Hydrodynamics of Cohesive Sediments** Optical Methods Applied to Hydrodynamics of Cohesive Sediments

DOI: 10.5772/intechopen.72347

Juan Antonio Garcia Aragon, Salinas Tapia Humberto, Diaz Palomarez Victor and Klever Izquierdo Ayala Juan Antonio Garcia Aragon, Salinas Tapia Humberto, Diaz Palomarez Victor and

Additional information is available at the end of the chapter Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/intechopen.72347

#### Abstract

Klever Izquierdo Ayala

Suspended sediment transport in large rivers is constituted mainly by cohesive sediments, which form aggregates or flocs with primary particles less than 65 μm. The removal of cohesive sediments in aquaculture tanks is a difficult problem. Due to its size, density, and shape, the hydrodynamic behavior of flocs is very different from that of non-cohesive sediments as they depend on the interaction with the water column. This chapter describes the experimental results obtained in sedimentation tanks, reduced models of aquaculture recirculation tanks, and a rotating circular flume with Plexiglas walls, in which optical methods were used to determine flocs' characteristics. These methods include particle tracking velocimetry (PTV) and digital holography for particle image velocimetry (DHPIV). Fractal models for floc density were successfully developed and validated with PTV experimental results in an aquaculture recirculation tank. Also, a model for the settling velocity of the flocs was validated using a permeable drag coefficient definition. Suspended sediments from Mexico's two largest rivers, Usumacinta and Grijalva, with a mean flow rate near mouth of 1700 and 650 m3 /s, respectively, were analyzed in a rotating circular flume. The shear velocity obtained in the field was reproduced in the circular flume and size and shape of flocs were obtained. This allowed to reproduce suspended sediment concentration profiles of rivers. DHPIV techniques were developed in order to obtain the actual size of the flocs based on Fresnel approximation for the reconstruction of holographic images.

Keywords: floc, settling velocity, suspended sediments, fractal dimension, PTV, DHPIV

### 1. Introduction

Hydrodynamic behavior of cohesive sediments is important in many fields of engineering. A large part of the suspended sediment charge in large rivers is constituted by cohesive sediments

© The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and eproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

as shown in the Amazon River sampling of suspended sediments [1]. A characteristic of cohesive sediments is to form aggregates (flocs) that behave in a very different way that non-cohesive sediments. Measuring in-situ flocs settling velocities in rivers is not possible with conventional sediment sampling instruments. Recently, in-situ optical instruments are being used for floc size measurement, particularly in the field of oceanography [2, 3]. Those instruments are very expensive because they use laser illumination and are not used in common river engineering practice. In this chapter, a method based on suspended concentration sampling and laboratory particle size analysis in a rotating annular flume is used to obtain flocs size, and with an appropriate settling velocity model deduce the settling velocity to be used in the Rouse equation. The method is validated with sediments from Usumacinta and Grijalva rivers, the two major rivers in México, whose basins are located in the states of Chiapas and Tabasco near the Guatemalan border.

The settling velocity of cohesive sediments is an important design parameter in aquatic environments such as water treatment plants, storm water ponds, sediment filling in lakes, sedimentation in estuaries, dredging in rivers, and sediment removal in aquaculture devices especially when shortage of water is a concern [4]. The reuse of water is the main characteristic of the latter systems.

The efficient removal of solids is a main concern in these systems because of the accumulation of non-used food and fish excreta. These solids are generally less than 65 μm in diameter and behave as cohesive sediments [5]. These sediments form flocs or aggregates, made of water, inorganic particles, and organic particles [6–8]. To obtain adequate settling models for these particles is an active field of research [9]. Some researchers include variable fractal dimension functions that depend on a characteristic size of flocs that is difficult to obtain over a large population of flocs [10]. Other authors use geometrical parameters of flocs like floc perimeter, which is not easy to measure in engineering practice [11, 12]. The flocs' settling velocity model proposed in this study uses parameters that are possible to average, using optical methods with some floc samples.

In this study, the results were used to calibrate settling velocity models using fractal theory and by including an adequate definition of the drag coefficient for permeable flocs. The proposed model is shown to provide reproducible results if a calibration of the parameters in the density

Optical Methods Applied to Hydrodynamics of Cohesive Sediments

http://dx.doi.org/10.5772/intechopen.72347

113

Suspended sediment samples obtained near the mouth of the Grijalva and Usumacinta rivers were analyzed in a rotating annular flume using PTV. It was possible to obtain appropriate values of the Rouse parameter ZR, which was shown to be difficult to obtain with classical

2.1. Experiments using PTV and sediments coming from aquaculture recirculation tanks

Initial experiments were performed with fish food in order to have a better control of primary particles. The sediments where sieved and only those passing sieve 200 (0.075 mm), with a mean density of 1430 kg/m3 were used. A Plexiglas settling column was realized in order to allow the use of PIV and PTV optical methods. The set up consisted of a rectangular tank of cross section 15.5 15.5 cm and 100 cm height. A laser sheet was introduced from above using a double

granulometric and Coulter counter methods used in river studies [1].

vs. diameter model is properly done.

Figure 1. Reduced model of an aquaculture recirculation tank.

2. Methods

The tanks most widely used are circular [13, 14]. Water is supplied in these tanks by means of diffusers at the walls. In this project, a small scale circular water recirculation tank was used in order to study the solids behavior in the tank. There is a central settling device in order to remove the solids (Figure 1). The settling device functions according to the hydrociclons principle [15].

Two optical techniques were used in this work, Particle Image Velocimetry (PIV) and Particle Tracking Velocimetry (PTV), to measure fluid and particle velocities, respectively. Polyamide tracers 25 μm in diameter were used to obtain fluid velocities using PIV, and flocs were tracked as particles in the PTV technique. PTV also allowed us to measure particle size and shape. Digital holography for particle image velocimetry (DHPIV) has also been used to determine the size and shape of flocs considering their volumetric nature.

The attempts to model settling velocity as a function of floc size, shape, and density demonstrated that density varies with floc size. Later work demonstrated that floc density depends on the fractal nature of flocs [16]. Recently, the effect of shear rate on floc density was demonstrated [11].

Figure 1. Reduced model of an aquaculture recirculation tank.

In this study, the results were used to calibrate settling velocity models using fractal theory and by including an adequate definition of the drag coefficient for permeable flocs. The proposed model is shown to provide reproducible results if a calibration of the parameters in the density vs. diameter model is properly done.

Suspended sediment samples obtained near the mouth of the Grijalva and Usumacinta rivers were analyzed in a rotating annular flume using PTV. It was possible to obtain appropriate values of the Rouse parameter ZR, which was shown to be difficult to obtain with classical granulometric and Coulter counter methods used in river studies [1].
