21 2 General Properties of Sediments Certain properties of sediments are basic and common to many of the topics in the remainder of this book and are presented here. In Sections 2.1, 2.2, and 2.3, essential characteristics of individual sedimentary particles (particle size, settling speed, and mineralogy) are described. Although individual particles are the basic building blocks for suspended as well as bottom sediments, sedimentary particles generally do not exist in the form of isolated particles. In the overlying water, ner-grained particles are generally present as aggregates of particles, that is, as ocs; in contrast, coarser-grained particles usually do not aggregate and are present as individual particles. Properties of a oc, such as size and density, are often signicantly different from those of the individual particles making up the oc. An introduction to occulation and the properties of ocs is given in Section 2.4. As the suspended particles or ocs settle out of the overlying water, they are deposited on the sediment bed, where they may be buried by other depositing particles. As this occurs, ocs are compressed and formed into larger and denser aggregates. Due to the weight of the overlying sediments, interstitial water and gas are forced out from between the solid particles/aggregates in the sediment bed and are transported toward the sediment-water surface; differential settling and transport of different size particles may also occur. In addition, gas may be generated within the sediments due to the decay of organic matter. The bed con- solidates with time due to these processes and also due to chemical reactions in the bed. In Section 2.5, procedures for measuring the bulk densities of these bot- tom sediments are discussed, and illustrations of the variations of bulk densities with depth and time are given. 2.1 PARTICLE SIZES 2.1.1 C LASSIFICATION OF SIZES The most obvious property of a sediment is the sizes of its particles. Because most particles are irregular in shape, a unique size or diameter is difcult to dene. However, a conceptually useful and unique diameter can be dened as the diam- eter of a sphere with the same volume as the particle. Nevertheless, in practice, effective particle diameters are usually dened operationally, that is, by the tech- nique used to measure the particle sizes. Because of this, small differences in particle sizes as measured by different techniques may occur. For a typical sedi- ment of interest here, particle diameters generally range over several orders of © 2009 by Taylor & Francis Group, LLC 22 Sediment and Contaminant Transport in Surface Waters magnitude, often from less than a micrometer to as much as a centimeter. Because a particular measuring technique is only valid for a limited range of particle sizes, more than one measurement technique may be necessary to measure the complete size distribution of a sediment. Particle sizes often are classied using the Wentworth scale (Table 2.1). This scale standardizes and quanties the denitions of terms commonly used to describe sediments (e.g., clay-size, silt, sand). The basic unit of the scale is 1 mm; different size classications follow by multiplying or dividing by two. For example, very coarse sand is dened as particles with diameters between 1 and 2 mm, and very ne sand is dened as particles with diameters between 1/8 and 1/16 mm. Broader classications that will be used extensively are: clay-size particles with d<1/256mm; silts, 1/256<d<1/16mm; sands, 1/16<d<2mm; and larger par- ticles (granules, pebbles, cobbles, and boulders), d > 2 mm. In units of microm- eters (µm), these sizes are clay-size (d<3.91 µm), silts (3.91<d<62.5µm), sands (62.5<d<2000 µm), and larger particles (d>2000 µm). For convenience, a scale with other metric units is also given. Because the Wentworth scale proceeds in TABLE 2.1 Classification of Particle Sizes  $! #%   ! $  !%" !$ !% !"  !" $  !%(  !" $  !%(  !" $  !%(           mm metric phi unitsWentworth scale Particle diameter    & & & & ' ' ' ' ' ' ' '                          © 2009 by Taylor & Francis Group, LLC General Properties of Sediments 23 powers of two, the phi scale is sometimes used (usually by geologists), where G =−log 2 (particle diameter in mm), or d = 2 –G . The negative logarithm is used so that particles smaller than 1 mm (the most frequently encountered particle diam- eters) have positive values of G. The phi scale also is shown in Table 2.1. 2.1.2 MEASUREMENTS OF PARTICLE SIZE Various methods are used to measure particle sizes. The most common are siev- ing, sedimentation, and the use of light diffraction instruments. Sieving is done by means of a series of stainless steel sieves with aperture sizes at 1/2 phi intervals. The sediments generally are dried, disaggregated, and passed through a series of these sieves; the residue on each sieve is then weighed. Because ne sieves tend to become clogged rather rapidly, sieving generally is useful only for coarse sedi- ments with d > 63 µm. For ne- and medium-size particles, various types of sedimentation proce- dures have been popular in the past and are still used to some extent. These pro- cedures depend on the fact that different-size particles settle at different speeds and use Stokes law (Section 2.2) to deduce particle size from a measurement of settling speed. There are numerous difculties with these procedures, including time-dependent occulation of ne-grained particles as they settle, hindered set- tling due to the upward movement of water when sediment concentrations are high, and the slow dissipation of the turbulence due to the initial mixing of the sediment-water mixture. Because of this, these sedimentation procedures are not used extensively at present. The most accurate and convenient procedure for measuring particle sizes over a wide range is by means of light diffraction instruments. In these instruments, light is passed through low concentrations of the sediment-water mixture and is diffracted by the particles. The particle size distribution is then determined from the diffraction pattern. The procedure is non-invasive and, with different lenses, can measure particle sizes from about 0.1 to 2000 µm. For example, the Malvern Particle Sizer (Mastersizer X) can measure particle sizes from 0.1 to 80 µm with one lens, from 0.5 to 180 µm with a second lens, from 1.2 to 600 µm with a third lens, and from 4.0 to 2000 µm with a fourth lens. Only sieving for large particles and the use of a light diffraction instrument for all other particles are recom- mended for accurate particle sizing. 2.1.3 SIZE DISTRIBUTIONS In a natural sediment sample, particle sizes are generally not uniform or even close to uniform; a wide distribution of sizes generally is present. As an example, the particle size distribution for a sediment from the Detroit River in Michigan is shown in Figure 2.1(a), where the fraction of particles by mass in a particular size class is plotted as a function of the logarithm of the diameter of that size class. A wide asymmetric distribution is evident, with a signicant fraction of particles less than 4 µm (clays) and also greater than 62.5 µm (sands). However, © 2009 by Taylor & Francis Group, LLC 24 Sediment and Contaminant Transport in Surface Waters even when it is known, the detailed distribution of sizes usually generates more information than is needed or can be used. Because of this, the distribution often is described in terms of various statistical parameters (e.g., the mean, standard deviation, skewness, and possibly higher moments). However, only the simplest parameters (the mean, median, and mode) are used here. The mean (also generally known as the average) is dened as the center of gravity of the area under the distribution curve on a linear scale; that is, d mean x d i ii ()£ (2.1) where d i and x i are the diameter and fraction of particles by mass in size class i, respectively. The median particle diameter is dened as that diameter for which 50% of the sediment by mass is smaller and 50% is greater, whereas the mode is dened as the diameter of the size fraction with the largest fraction of particles within it. For the size distribution shown in Figure 2.1(a), the mean, median, and mode are 18, 6.0, and 2.7 µm, respectively. In some cases, the size distribution (on a log-diameter scale) may be similar to a Gaussian distribution; this is termed “log-normal.” However, there is no known theoretical reason why this should be true. In fact, upon close examination, most sediment size distributions are not log-normal. As a specic example, a log-nor- mal distribution (with a mean of 18 µm and a standard deviation the same as that of the Detroit River sediments) is shown in Figure 2.1(a) and can be compared there with the Detroit River size distribution. The Detroit River distribution is clearly not log-normal. As additional examples, size distributions for three sediment samples from the Detroit River (different from the previous sample), the Fox River in Wisconsin, and the Santa Barbara Slough in California are shown in Figure 2.1(b) (Jepsen et al., 1997). The means are 12, 20, and 35 µm, respectively. The general character of the size distributions shown in Figures 2.1(a) and (b) is more or less typical of most sediments with a wide range of particle sizes and with one dominant size fraction. A more unusual distribution is shown in Fig- ure 2.1(c), where measurements of two subsamples of a single and larger sample from Lake Michigan (McNeil and Lick, 2002a) are plotted. Both measurements were made by a Malvern particle sizer and show close agreement with each other. Two peaks in each distribution are obvious, indicating that the sediment may be a mixture of sediments from two different sources. For both subsamples, the means (66.5 µm) and medians (22.0 and 20.7 µm) are in good agreement with each other. However, the modes (149.2 and 2.1 µm) are not. The reason for this is that the modes depend in a very sensitive manner on the relative heights of the two peaks, which are slightly different for the two subsamples. In this particular case, neither the mean, median, nor mode is a useful parameter for characterizing the particle size distribution. Quite clearly, the distributions shown in Figures 2.1(b) and (c) are also not log-normal. Herein, log-normality will not be assumed for any size distribution. © 2009 by Taylor & Francis Group, LLC General Properties of Sediments 25 Because transport and chemical sorption properties often and signicantly depend on particle size, some representation of the distribution of particle sizes is often necessary. For modeling purposes, this is usually done by separating the entire size distribution into three or more intervals, with the corresponding mass Percent in Size Class Particle Diameter (µm) 1 10 100 (a) 1000 0 2 4 6 8 10 Particle size distributio n Log-normal distribution 10 8 6 4 2 1 10 100 (b) Particle Size (µm) Detroit River Fox River Santa Barbara Volume Percent 1000 0 FIGURE 2.1 Particle size distributions. Percent by volume (mass) as a function of diam- eter: (a) sediments from the Detroit River. Also shown is a log-normal distribution with a mean and standard deviation the same as the Detroit River sediments; (b) sediments from the Detroit River, Fox River, and Santa Barbara Slough (From Jepsen et al., 1997. With permission.) © 2009 by Taylor & Francis Group, LLC 26 Sediment and Contaminant Transport in Surface Waters fraction of particles given in each interval — for example, 20% ne (0–10 µm), 50% medium (10–64 µm), and 30% coarse (>64 µm). The number of intervals (or size classes) and the range of particle sizes within each interval depend on the sediment and the application. Examples are given in Chapter 6. 2.1.4 VARIATIONS IN SIZE OF NATURAL SEDIMENTS THROUGHOUT A SYSTEM Sediment properties, including particle size, often vary greatly throughout a sed- iment bed, in both the horizontal and vertical directions. Sediment types can change rapidly in the horizontal direction from coarse sands (where currents and wave action are strong) to ne-grained muds (where currents and wave action are small). They also can change rapidly in the vertical direction. For example, this can happen due to strong storms that may deposit layers of coarse sands on top of ner sediments. During quiescent conditions, this coarse layer may then be covered by ne sediments. As an example of horizontal variations in particle size, consider the surcial sediments of Lake Erie. The bathymetry of the lake is shown in Figure 2.2(a). Lake Erie is a relatively shallow lake and can be conveniently separated into three basins: (1) the Western Basin, with an average depth of 6 m; (2) the Central Basin, with an average depth of 20 m; and (3) an Eastern Basin, with an average depth of 26 m. Mean particle sizes (on a phi scale) of the disaggregated surcial sedi- ments are shown in Figure 2.2(b) (Thomas et al., 1976) and range from greater than medium sand (d > 1/2 mm, G < 1) to less than coarse clay-size particles Particle Diameter (µm) 10 8 6 4 2 0 1 10 100 (c) 1000 Amount in Size Class (%) Mean = 66.5 µm, Median = 22.0 µm, Mode = 149.2 µm Mean = 66.5 µm, Median = 20.7 µm, Mode = 2.1 µm FIGURE 2.1 (CONTINUED) Particle size distributions. Percent by volume (mass) as a function of diameter: (c) sediments from Lake Michigan near the Milwaukee bluffs. (Source: From Jepsen et al., 1997. With permission.) © 2009 by Taylor & Francis Group, LLC General Properties of Sediments 27 ERIEAU N TOLEDO CLEVELAND ERIE N.Y. PENN. OHIO OHIO MICHIGAN PENN. Point Pelee 20 20 40 40 20 20 20 20 20 20 40 40 20 80 40 80 80 100 100 180 160 120 140 140 120 40 20 60 20 40 60 40 40 20 40 20 60 60 60 60 60 40 20 60 WESTERN BASIN MEAN DEPTH 24' Depth (feet) MEAN DEPTH 60' MEAN DEPTH 80' LAKE ERIE LONGITUDINAL CROSS-SECTION CENTRAL BASIN EASTERN BASIN MILES 0 0 20 40 60 80 100 120 140 160 180 200 220 10 20 30 40 Long Point Detroit River Maumee River FIGURE 2.2(a) Lake Erie: bathymetry, depth in feet (Source: From Thomas et al., 1976. With permission.) © 2009 by Taylor & Francis Group, LLC 28 Sediment and Contaminant Transport in Surface Waters (d<2µm, G > 9). Particles tend to be coarser in the shallower parts and ner in the deeper parts of each basin. This variation in particle size depends on the primary source of the sediments (major primary sources for Lake Erie are the Maumee River, the Detroit River, and shore erosion, especially along the north- ern shore of the Central Basin); differential erosion of the bottom sediments due primarily to wave action (greater in shallow waters than in deep); transport of the suspended sediments by currents and wave action; and differential settling of the particles. This is discussed more quantitatively in Chapter 6. As an example of vertical variations in particle size, Figure 2.3 shows the mean diameter, d, as a function of depth for a sediment core from the Kalamazoo River in Michigan (McNeil and Lick, 2004). An irregular variation of d from 85 µm at the surface to 200 µm at a depth of 10 cm, to 70 µm at 15 cm, and to 180 µm at 19 cm is shown. Although vertical variations in d can be relatively small, the rapid and irregular variation shown here is not atypical. Also shown in Figure 2.3 is the variation of sediment bulk density with depth. A close correlation between bulk density and particle size is evident. However, other quantities in addition to particle size also inuence the bulk density so that the correlation is not exact. The increase in particle size and density between 8 and 14 cm is probably due to a large ood transporting coarse sediments from upstream; these then deposit near the end of the ood as the ow velocity decreases. After this ood, the ow velocity is relatively low, and only ner sedi- ments can be transported from upstream and deposited in this area. In general, the mean size and size distribution of suspended sediments will be different from those of the bottom sediments at the same location. This is due not only to transport of the suspended sediments from other locations where sediments have different sizes but also to the dependence of suspended particle size on the magnitude of the shear stress causing the resuspension. This is illus- trated in Figure 2.4, which gives the disaggregated particle size distribution for D e t r o i t           Maumee R. FIGURE 2.2(b) Lake Erie: mean grain size of surcial sediments. (Source: From Thomas et al., 1976. With permission.) © 2009 by Taylor & Francis Group, LLC General Properties of Sediments 29 Core Depth (cm) Bulk Density (g/cm 3 ) 1 1.3 1.6 1.9 0 20 15 10 5 2001601208040 2.2 Particle Size (µm) = Bulk density = Particle size FIGURE 2.3 Mean grain size and bulk density as a function of depth for a sediment from the Kalamazoo River. (Source: From McNeil and Lick, 2004. With permission.)                     !  "     !  "    FIGURE 2.4 Size distributions of a bottom sediment and the same sediment resuspended at shear stresses of 1.0 and 0.4 N/m 2 . (Source: From Lee et al., 1981. With permission.) © 2009 by Taylor & Francis Group, LLC 30 Sediment and Contaminant Transport in Surface Waters well-mixed bottom sediments and for these same sediments when they have been resuspended at shear stresses of 1.0 and 0.4 N/m 2 (Lee et al., 1981). It can be seen that the suspended sediments have a smaller average particle size than the bottom sediments and that this average size depends on the shear stress — that is, the greater the shear stress, the greater the average suspended particle size. 2.2 SETTLING SPEEDS When a particle is released into quiescent water, there will be an initial transient in the vertical speed of the particle until a steady state is reached. This steady state is due to a balance between the gravitational force, F g , and the drag force, F d , on the particle. The distance until the steady state is reached depends on the properties of the particle and of the uid but, for a typical sedimentary particle in water, will generally be on the order of a few particle diameters. The steady-state speed is known as the settling speed, w s . For a solid particle, w s will depend on the size, shape, and density of the particle as well as the density and viscosity of the uid. For aggregates of particles, w s will depend not only on these same parameters for the aggregate and the uid but also on the permeability of the aggregate. Because of the complex dependence of w s on the above parameters, w s gener- ally cannot be determined theoretically but must be determined experimentally. However, in certain limiting cases, theoretical determinations of w s can be made. A most important case is the simplest case, that is, a solid spherical particle with diameter d, volume V = Qd 3 6, and density S s settling at constant speed in a uid of density S w and kinematic viscosity µ. In this case, the gravitational force is just due to the immersed weight of the particle, or FV g d g gsw sw   () () RR P RR 3 6 (2.2) where g is the acceleration due to gravity and is equal to 980 cm/s 2 . For slow, viscous ow past a sphere, the drag force has been determined by Stokes and is (Schlichting, 1955): Fdw ds  3PM (2.3) By equating the gravitational and drag forces as given by the above two equations, one obtains w gd ssw  2 18M RR() (2.4) which is known as Stokes law. The formula is valid for settling speeds such that the Reynolds number (Re = S s w s d/µ) is less than about 0.5. For solid sedimentary © 2009 by Taylor & Francis Group, LLC [...]... Clay Kaolinite Orthoclase Microcline Quartz Albite Flint Calcite 1.3–1.5 2. 2 2. 6 2. 2 2. 6 2. 5 2. 6 2. 5 2. 6 2. 5 2. 8 2. 6 2. 7 2. 6 2. 7 2. 6 2. 8 Anorthite Dolomite Muscovite Biotite Apatite Limonite Magnetite Pyrite Hematite 2. 7 2. 8 2. 8 2. 9 2. 7–3.0 2. 8–3.1 3 .2 3.3 3.5–4.0 4.9–5 .2 4.9–5 .2 4.9–5.3 Source: From Kohnke, 1968 From the definition of xw , it follows that xw w =W , or xw = W / w, and therefore 2. 6 1... d2, or ws © 20 09 by Taylor & Francis Group, LLC d2 (2. 6) 32 Sediment and Contaminant Transport in Surface Waters This is a convenient and easy-to-remember equation for ws(d), especially for finegrained sediments whose diameters are often expressed in micrometers In the derivation of Stokes drag (Equation 2. 3), it is assumed that the flow is steady and laminar and that there is no wake behind the body... the less-than 2- m size fraction The bulk sediments consist primarily of clay minerals (30%) and quartz (45%), with the remainder (25 %) being other minerals The size fraction less than 2 µm consists primarily of clay minerals (93%) and contains relatively little quartz (3%) Kaolinite is the dominant clay mineral for all size classes The subject of clay mineralogy is quite diverse and is a fascinating area... subdivided into silica minerals (quartz and opaline silica); the clay minerals (usually kaolinite, illite, montmorillonite, and chlorite, but others as well); and other silicates (feldspars, zeolites) The © 20 09 by Taylor & Francis Group, LLC 34 Sediment and Contaminant Transport in Surface Waters mineralogy and size of an inorganic particle tend to be related Large particles with a diameter greater than 62. 5... procedure for measuring the bulk density of bottom sediments (hereafter called the wet-dry procedure), sediment cores are frozen, sliced © 20 09 by Taylor & Francis Group, LLC 40 Sediment and Contaminant Transport in Surface Waters into 1- to 4-cm sections, and weighed (mwet) They then are dried in an oven at approximately 75°C for 2 days and weighed (mdry) Because mw = mwet – mdry and mw + ms = mwet,... coatings © 20 09 by Taylor & Francis Group, LLC General Properties of Sediments 35 TABLE 2. 2 Mineralogy of a Composite Sediment Sample from the Kalamazoo River Sample Mineral Bulk (%) 2 15 µm (%) . 27 ERIEAU N TOLEDO CLEVELAND ERIE N.Y. PENN. OHIO OHIO MICHIGAN PENN. Point Pelee 20 20 40 40 20 20 20 20 20 20 40 40 20 80 40 80 80 100 100 180 160 120 140 140 120 40 20 60 20 40 60 40 40 20 40 20 60 60 60 60 60 40 20 60 WESTERN. TABLE 2. 3 Densities (g/cm 3 ) of Various Sediment Components Humus 1.3–1.5 Anorthite 2. 7 2. 8 Clay 2. 2 2. 6 Dolomite 2. 8 2. 9 Kaolinite 2. 2 2. 6 Muscovite 2. 7–3.0 Orthoclase 2. 5 2. 6 Biotite 2. 8–3.1 Microcline. diameter. © 20 09 by Taylor & Francis Group, LLC 32 Sediment and Contaminant Transport in Surface Waters This is a convenient and easy-to-remember equation for w s (d), especially for ne- grained sediments