Geo-Mar Lett DOI 10.1007/s00367-016-0457-3
ORIGINAL
Particle shape-controlled sorting and transport behaviour of mixed siliciclastic/bioclastic sediments in a mesotidal lagoon, South Africa Burghard W. Flemming 1
Received: 19 April 2016 / Accepted: 5 July 2016 # Springer-Verlag Berlin Heidelberg 2016
Abstract This study investigates the effect of particle shape on the transport and deposition of mixed siliciclasticbioclastic sediments in the lower mesotidal Langebaan Lagoon along the South Atlantic coast of South Africa. As the two sediment components have undergone mutual sorting for the last 7 ka, they can be expected to have reached a highest possible degree of hydraulic equivalence. A comparison of sieve and settling tube data shows that, with progressive coarsening of the size fractions, the mean diameters of individual sediment components increasingly depart from the spherical quartz standard, the experimental data demonstrating the hydraulic incompatibility of the sieve data. Overall, the spatial distribution patterns of textural parameters (mean settling diameter, sorting and skewness) of the siliciclastic and bioclastic sediment components are very similar. Bivariate plots between them reveal linear trends when averaged over small intervals. A systematic deviation is observed in sorting, the trend ranging from uniformity at poorer sorting levels to a progressively increasing lag of the bioclastic component relative to the siliciclastic one as overall sorting improves. The deviation amounts to 0.8 relative sorting units at the optimal sorting level. The small textural differences between the two components are considered to reflect the influence of particle shape, which prevents the bioclastic fraction from achieving complete textural equivalence with the siliciclastic one. This is also reflected in the inferred transport behaviour of the two shape components, the bioclastic fraction moving closer to the Responsible guest editor: J.I. Cuitiño * Burghard W. Flemming
[email protected] 1
Senckenberg Institute, Suedstrand 40, 26382 Wilhelmshaven, Germany
bed than the siliciclastic one because of the higher drag experienced by low shape factor particles. As a consequence, the bed-phase development of bioclastic sediments departs significantly from that of siliciclastic sediments. Systematic flume experiments, however, are currently still lacking.
Introduction The physical properties of sediments such as mean size, sorting, particle density, particle shape and settling velocity are, together with the dynamic forces of the fluid, the most important factors controlling depositional processes (e.g. Hjulström 1939; Krumbein 1943; Menard 1950; Bagnold 1956, 1968; Sundborg 1956; Simons et al. 1961; Jopling 1963). In this connection, it has long been realized that particle density (heavy minerals) and particle shape (bioclastic debris) play particularly important roles in the segregation of sediments in nature (e.g. Rittenhouse 1943; Bridge and Bennett 1992). Whereas the hydraulic behaviour of particles of variable density and shape have been investigated over many decades in dedicated experiments using settling tubes and flumes (e.g. Rubey 1933; Krumbein 1942; Pettyjohn and Christiansen 1948; McNown and Malaika 1950; Fontein 1960; Briggs et al. 1962; Kelling and Williams 1967; Alger and Simons 1968; Maiklem 1968; Braithwaite 1973; Komar and Reimers 1978; Göğüş and Defne 2005), corresponding analyses of bulk bioclastic and mixed siliciclastic/ bioclastic (heterogeneous) sediments have received much less attention and have shifted into focus only in more recent decades (e.g. Wiberg and Dungan Smith 1987; Flemming 1992; Kench and McLean 1996; Smith and Cheung 2002, 2003, 2004, 2005; Paphitis et al. 2002; Weill 2010; Durafour et al. 2013, 2015). Nevertheless, numerous studies have investigated the dynamics and depositional character of mixed siliciclastic-bioclastic sediments in the field (e.g. Larsonneur 1975; Flemming 1977a;
Geo-Mar Lett
Mount 1984; Lee et al. 1994; Gonzales and Eberli 1997; McNeill et al. 2004; García et al. 2005; Weill 2010). In none of these publications, however, have the hydraulic relationships between the siliciclastic and bioclastic components of the mixed sediments been specifically addressed. Coastal and shallow marine sediments, in particular, are usually of a heterogeneous nature, being composed of variable mixtures of silicilastic and bioclastic particles, the main contrast between the two being their different particle shapes (e.g. Flemming 1977a, 1977b; Gussmann and Smith 2002). In the present study, the effect of particle shape on the transport and deposition of mixed siliciclastic/bioclastic sediments is investigated in a micro- to low mesotidal coastal lagoon along the South Atlantic coast of South Africa. The sediments of Langebaan Lagoon would have been recycled several times during successive Pleistocene sea-level high- and lowstands, and have most recently undergone further mutual sorting since the postglacial sea level reached its present position in the study area some 7 ka ago (Compton 2001). One can thus expect the two sediment components to have reached a highest possible degree of hydraulic equivalence in the course of such prolonged periods of mutual size and shape sorting. Within this context, the specific aim of the present study is to systematically compare the spatial distribution patterns and textural attributes of siliciclastic and bioclastic components in order to elucidate the degree of hydraulic equivalence. The comparison also aims at distinguishing potential differences between the three main sub-environments of Langebaan Lagoon (intertidal flats, subtidal flats, channels). Because the content of the bioclastic component in the lagoonal sediments ranges from 50%, an assessment of the effect of different mixing proportions is a welcomed additional objective.
persistent south-westerly winds in summer cause lower water levels and wave impact on the eastern tidal flats, whereas westerly to north-westerly winter storms frequently cause the southern salt marshes to be flooded by water-level setups. Surface current velocities vary from 1.3 m s–1 in the inlet channel to 2 m below mean sea level). The mean tidal ranges (long-term averages) in the southern lagoon sector vary from 0.39 m at neap tide to 1.50 m at spring tide. According to the classification scheme of Hayes (1979), the tides straddle the micro- to lower mesotidal regimes (cf. also Flemming 2005, 2012). There is a progressive time delay in the turn of the tide with distance from the inlet, amounting to 80 minutes for high tide and 50 minutes for low tide. Instantaneous water levels and wave action are also affected by the local wind regime, whereby
Sediment samples were collected in 1975 at variable intervals (150–800 m) at 220 stations distributed along 27 W–E profiles spaced at about 500 m (Fig. 2b). Sample positions were determined by means of two-angle measurements using a sextant (cf. GPS was not available at that time), a high accuracy being nevertheless documented by the consistent alignment of sample positions along the profiles and close agreement with positions identified on aerial photographs (not shown). At each station, four randomly placed short cores (6.5 cm long) were taken, which were later mixed into a single bulk sample of 450–500 g. By this procedure, the probable sampling error due to local inhomogeneities was reduced by 50% (Krumbein 1934). In the laboratory, the samples were dialysed overnight to remove the salt and then split by means of a mechanical sample splitter into subsamples of suitable mass for various analytical procedures. Grain-size distributions were determined
Geo-Mar Lett
Fig. 1 Map showing the location of Langebaan Lagoon along the South Atlantic coast about 100 km north of Cape Town
by settling tube analysis. Details on the construction and performance (reproducibility, precision) of the tube can be found in Flemming and Thum (1978). The measured settling velocity distributions were converted into equivalent spherical settling diameters by means of a computer program adapted from the original Fortran program developed at the University of California (cf. Reed et al. 1975). The conversion from settling
velocities to equivalent settling diameters made use of the equation published by Gibbs et al. (1971). In each case, one analysis was performed on a split of the total sample (ca. 1 g) and another analysis on the carbonatefree fraction after the bioclastic material had been removed by acid digestion. To remove the bioclastic fraction, the total sample was weighed and then treated with hydrochloric acid.
Fig. 2 a Physiography of Langebaan Lagoon showing salt marshes, intertidal flats, subtidal flats, subtidal bars and tidal channels (modified after Flemming 1977a, 1977b). b Sample stations in Langebaan Lagoon (modified after Flemming 1977a, 1977b)
Geo-Mar Lett
After effervescence ceased, the residue was washed in tap water, dried and weighed. The carbonate content was determined by subtracting the mass of the residue from that of the total sample, a procedure considered to be sufficiently accurate for the purpose of the present study. The grain-size distributions of the bioclastic components were calculated by subtracting the carbonate-free settling diameters from their total sediment equivalents as a function of carbonate content. In both cases, the textural parameters, i.e. mean settling diameter, standard deviation (sorting), skewness and kurtosis, were calculated according to the percentile statistics proposed by Folk and Ward (1957). In the present study, the standard deviation (sorting) values derived on the basis of the percentile statistics were transformed into dimensionless relative sorting values in an approach based on Walger (1962), as modified by Flemming (1977b). Relative sorting is defined as QH=QD/QDe where QH is the relative sorting, QD the quartile deviation and QDe the empirically determined elementary (or optimal) sorting level which a particular sediment can reach (cf. Walger 1962). This pays tribute to the fact that optimal sorting is a function of grain size, with fine sand achieving best sorting and both finer and coarser sediments achieving progressively poorer optimal sorting values (cf. Flemming 1977b). In the present case, the terrigenous component essentially consisted of subrounded to well-rounded quartz grains, whereas the bioclastic component was composed of platy to very irregularly shaped and, in part, also porous skeletal fragments. To illustrate the fundamental difference between sieving and settling, a coarse to very coarse shallow-marine bioclastic sand sample was sieved and the siliciclastic (quartz grains) and main bioclastic sediment components (mollusc, cirriped and bryozoan fragments) manually separated from each quarter-phi size fraction under a binocular microscope. In some cases, the particles from two quarter-phi fractions were pooled to obtain enough material for settling tube analysis. The grain-size distribution of each component was then determined for all size fractions by settling tube analysis (cf. Flemming 1977b, 2007). The fractional distributions were subsequently summed proportionally to also obtain the distributions of the combined fractions. The contours in all spatial distribution maps were drawn manually on the basis of the textural parameters representing the sediment at each sampling station.
Results Sieving versus settling The results of the sieving and settling experiment on the individual sediment components are illustrated in Fig. 3. A comparison of the mean diameters of the individual size fractions
(Fig. 3a) clearly shows that the sieve counterparts of all components, also the siliciclastic ones, increasingly depart from the spherical quartz standard as the size fractions become coarser. Moreover, the bioclastic components show substantially greater deviation than the siliciclastic ones. Figure 3b illustrates the same effect by comparing the sieve and settling distributions of the three main bioclastic components. In Fig. 3c, scanning electron microscope images highlight their distinctly different particle shapes. The experimental results vividly demonstrate the hydraulic incompatibility of the sieve data, the only plausible explanation being the differences in particle shape. Indeed, differences in particle density are negligible (1.5–1.7 g cm–3; average 1.6 g cm–3) in the present case. Carbonate content The spatial distribution of carbonate contents in surficial sediments of Langebaan Lagoon is illustrated in Fig. 4. Siliciclastic-bioclastic particle mixtures occur throughout the lagoon, albeit in different proportions from north to south. At first sight, this pattern could be interpreted as reflecting the N– S energy gradient, with highest contents (slightly exceeding 50%) in the northern sector and lowest contents (3
Geo-Mar Lett
Fig. 4 Distribution of bioclastic carbonate contents in surficial sediments of Langebaan Lagoon
sorting (Fig. 6b). Considering the prolonged period of mutual size sorting in Langebaan Lagoon (cf. above), the systematically poorer sorting of the bioclastic component appears to be an inherent feature most plausibly explained by the deviation in average particle shape (see below). Skewness The similarity of the spatial distribution patterns of skewness values (Fig. 5e, f) is again, as in the case of mean diameters, very striking. It is also visually evident that a greater number of siliciclastic samples are more positively skewed than their bioclastic counterparts, as documented by the much larger area occupied by positively skewed siliciclastic sediments (Fig. 5e and f). The bioclastic component, by contrast, is on average more negatively or less positively skewed (64% of all samples) than the siliciclastic sediment. Because density differences are negligible, the observed trend is again considered to reflect the pronounced differences in particle shape characterising the two sediment components. Textural relationships In this subsection, interrelationships amongst selected textural parameters compared above in the spatial distribution maps are assessed based on bivariate diagrams, starting with the overall scatter between siliciclastic and bioclastic mean diameters (Fig. 7a). The point scatter reveals a close grouping around the diagonal line representing equality, the overall
scatter rarely exceeding 0.5 phi intervals to either side. To highlight the trends of the two components, each has been plotted against the mean diameter of the total sediment, being marked in different colours with respective regression lines added in Fig. 7a1. The regression lines now quantify the trends that had already been qualitatively identified in the spatial distribution maps. They reiterate the fact that the bioclastic fraction is slightly coarser than the siliciclastic one towards finer mean diameters, and slightly finer towards coarser mean diameters. The cross-over point of the two regression lines lies on the equality line at a mean diameter of 1.7 phi (~0.31 mm). Furthermore, the maximum deviation of each component in the one or other direction relative to the total sediment barely exceeds 0.25 phi size intervals. Averaging the mean diameters of the two components over discrete size intervals (Fig. 7b) reveals a linear trend extending from above the diagonal equality line at the finer end to below the line at the coarser end, the cross-over point again lying on the diagonal equality line at 1.7 phi. The linear relationship suggests a high degree of hydraulic equivalence between the two shape components. Furthermore, the fact that the average values for the three sub-environments also fall on the trend line shows that this applies to the lagoon as a whole. It is thus not surprising that this trend is independent of the relative proportion of each component in the total sediment (Fig. 7b1). Similarly, when averaging sorting values at discrete intervals, a near-linear relationship emerges. This suggests a systematic trend ranging from essentially uniformity at poorer sorting levels to a progressively increasing lag in the sorting
Geo-Mar Lett
Fig. 5 Spatial distribution of textural parameters in Langebaan Lagoon. a Mean diameter (phi) of the siliciclastic fraction. b Mean diameter (phi) of the bioclastic fraction. c Relative sorting of the siliciclastic fraction. d
Relative sorting of the bioclastic fraction. e Skewness of the siliciclastic fraction. f Skewness of the bioclastic fraction
of the bioclastic component relative to the siliciclastic one as overall sorting improves. The lag amounts to 0.8 relative sorting units at the theoretically best sorting level (Fig. 7c). In effect this means that, whereas the better rounded siliciclastic particle group can achieve an optimal relative sorting of 0.5, the irregularly shaped bioclastic particle group on average reaches an optimal relative sorting of only 1.3. As in the case of mean diameters, the average sorting values of the individual sub-environments are grouped very closely around the linear trend line, indicating little difference between them. Equivalent relative sorting levels of the two shape components are listed in Table 1. As both particle groups have been exposed to the same sorting process, the difference in average sorting between the two components should be regarded as a systematic feature signifying hydraulically equivalent states. This conclusion is strongly supported by the observation that, as in the
case of the mean settling diameters, the average sorting trend is independent of the relative proportions of the two components (Fig. 7c1). Sediments composed of particles having high bulk shape factors (better rounded particles) thus inherently achieve better sorting levels than sediments composed of particles having low shape factors (more angular or platy particles) when exposed to the same hydrodynamic sorting process. Sorting is thus not only a function of particle size but also of particle shape. Although skewness values (not shown) scatter more widely, the general trend is similar to those of the other two parameters. Transport behaviour To test the effect of particle shape on the depositional process, a modified conceptual CM model (cf. Flemming 1977b; based on Passega 1964) defining sediment transport modes has been
Geo-Mar Lett
shape group is transported at a different average height above the bottom in the bottom boundary layer, albeit with a high degree of overlap. The pattern observed for the total sediment thus simply represents an average Bpicture^ of the combined patterns of the two constituent components (Fig. 8a, a1). An important point to note here is that the transport modes of the total sediment do not reveal the differences between the two shape components. Summarizing the textural trends of the mixed siliciclastic/ bioclastic sediments in Langebaan Lagoon, it is concluded that distinctly different shape groups respond independently to the process of hydraulic transport (Fig. 9). In each case, the individual components retain specific textural characteristics linked to the average shape of their constituent particles. The total sediment therefore reflects the proportionally averaged textural characteristics contributed by its individual shape components.
Discussion
Fig. 6 Bivariate plots of mean settling diameter versus sorting. a Siliciclastic fraction. b Bioclastic fraction
applied. The rationale is that the relation between the coarsest percentile (C) of the grain-size distribution and the mean (or median) diameter (M) is controlled by the hydrodynamic energy, in the present case reflecting the peak velocity and turbulence generated by the tidal flow. Figure 8 illustrates the transport modes of the total sediment, as well as the siliciclastic and bioclastic components (Fig. 8a–c), together with their corresponding spatial distributions in the lagoon (Fig. 8a1–c1). The different point clusters and associated spatial patterns of the siliciclastic and bioclastic components are striking, the former being predominantly transported in saltation 1 and 2 (24 and 51% respectively, or 75% combined), whereas the latter is preferentially transported in traction and saltation 1 (31 and 41% respectively, or 72% combined). This clearly highlights the fact that the bulk of the material of each
By comparing the two bulk shape groups of siliciclastic and bioclastic sediments in Langebaan Lagoon, it has been demonstrated that a functional relationship exists between particle shape, sediment texture and depositional process in naturally mixed and size-sorted sediments. This is particularly well illustrated when comparing the size distributions of bioclastic samples analysed by sieving and settling techniques (Fig. 3), the former method being unable to account for hydraulic differences caused by particle shape (and density, for that matter). Settling diameters systematically deviate from their sieve counterparts, the deviation increasing with increasing particle size (Fig. 3a). This deviation can also be expressed in terms of shape factors, whereby the empirically derived trend lines (cf. USGS 1958) increasingly deviate from the equality line as grain size increases (Fig. 10a). The increasing deviation can be explained by the increasing drag force experienced by particles having the same mean diameter but with shapes increasingly departing from that of a sphere. This relationship is nicely documented in the well-known diagram in which grain Reynolds numbers (Re) are plotted against drag coefficients (CD) as a function of particle shape factors (Fig. 10b; modified after Brezina 1979; based on Schulz et al. 1954). The grain Reynolds number is defined as Re=u*d/ν, where u* is the shear velocity, d the mean or median grain diameter and ν the kinematic viscosity. On the one hand, the diagram shows the systematic increase in drag coefficient with decreasing shape factor (i.e. high shape factor = low drag, low shape factor = high drag) and, on the other, it shows that the range in drag coefficients decreases as grain size decreases from the turbulent towards the laminar fall regime. In effect, this means that differently shaped particles of identical hydraulic mean diameter experience different drag during transport and deposition
Geo-Mar Lett
Fig. 7 Bivariate plots of various textural parameters. a Mean settling diameter of bioclastic fraction vs. mean settling diameter of siliciclastic fraction; a1 mean settling diameter of total sediment vs. mean settling diameter of siliciclastic and bioclastic fractions. b Averaged mean settling diameters of bioclastic fraction vs. averaged mean settling diameters of siliciclastic fraction; b1 averaged mean settling diameter of
bioclastic fraction vs. averaged mean settling diameter of siliciclastic fraction as a function of carbonate content. c Averaged relative sorting of bioclastic fraction vs. averaged relative sorting of siliciclastic fraction; c1 averaged relative sorting of bioclastic fraction vs. averaged relative sorting of siliciclastic fraction as a function of carbonate content
(Corey 1949; Tran-Cong et al. 2004), with higher drag coefficients resulting in greater resistance to motion (Shapiro 1961) and, hence, transport closer to the bed.
As mentioned above, the spatial distribution of carbonate contents in the lagoonal sediments could be controlled by either the energy gradient or grain size, or indeed both. In fact, Force (1969) demonstrated that skeletal break-down is reflected in the micro-architecture of the bioclastic material, whereby larger bioclasts are more frequent than smaller ones. This is consistent with the inverse relationship between carbonate content and grain size in Langebaan Lagoon. Nevertheless, the size sorting of the sediment in the lagoon is an energy-related phenomenon, implying that both factors are involved in the observed carbonate distribution pattern. All other factors being equal, the independent and differential response of the two particle groups to the mechanism of hydraulic transport is evidently controlled by the different bulk shape factors of the two sediment fractions. Individual shape groups of mixed sediments should, therefore, be treated as independent hydraulic subpopulations rather than mere components of a single hydraulic population. Importantly, the size-frequency distributions of the total sediment do not reveal the different transport behaviours of individual shape
Table 1 Equivalent relative sorting values of hydraulically equilibrated siliciclastic and bioclastic sands Relative sorting (QH) of subrounded Relative sorting (QH) of to well-rounded quartz sand irregularly shaped bioclastic sand 0.5 1.0
1.30 1.70
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
2.10 2.50 2.90 3.30 3.75 4.15 4.55 5.00
Geo-Mar Lett
Fig. 8 Transport modes of the total sediment (a), the siliciclastic fraction (b) and the bioclastic fraction (c) based on the approach of Passega (1964), together with their respective spatial distribution patterns in Langebaan Lagoon (a1, b1, c1)
components. The subtle differences observed in the textural characteristics of the two shape groups reflect the optimal degree of hydraulic equivalence which can be achieved. This is considered a fundamental feature of mixed sediments which have undergone prolonged mutual sorting, although the textural differences would in addition be a function of the average shape factor of each component. Only in the case
where the shape factors (and densities) of two sediment components are similar (e.g. well-rounded spherical quartz particles and oolitic carbonate sands) would one expect to find near-identical textural characteristics. A final point of discussion concerns experimental studies on sediments composed of low shape factor particles—e.g. bioclastic sediments. To date, no systematic flume experiments
Geo-Mar Lett Fig. 9 Schematic conceptual model of near-bed transport modes differentiated between siliciclastic fraction (a), bioclastic fraction (b) and total sediment (a+ b). Note the different proportions of the siliciclastic and bioclastic fractions transported at different levels in the bottom boundary layer
on bed-phase development of bioclastic sands have been undertaken with the aim of producing a bed-phase diagram comparable to that of, for example, Southard (1971) for wellrounded fluvial sands. A preliminary experiment simulating tidal flows has been conducted by Flemming (1992), who used bioclastic sands of two different mean diameters (sieve diameters 0.6 and 1.23 mm, settling diameters 0.49 and 0.67 mm respectively) and bulk shape factors (0.55 and 0.4 respectively). That study revealed two important differences compared to well-rounded quartz sands, one concerning threshold velocities, the other bed-phase development (Fig. 11; cf. Flemming 1992; phase diagram modified after Southard 1971). As evident from
Fig. 10 a Diagram of mean settling diameter vs. mean sieve diameter showing the increasing deviation with increasing grain size of selected particle groups relative to the trend lines of selected shape factors extracted from USGS (1958). b Diagram of grain Reynolds number (Re) vs. drag coefficient (CD) as a function of hydraulic shape factors (modified after Brezina 1979)
Fig. 11 Diagram illustrating bed-phase development in well-rounded quartz sands as a function of current velocity in ca. 40-cm-deep flows (modified after Flemming 1992). Superimposed are the data of both sieve and settling analyses of two bioclastic sands. Note the underestimation of transport thresholds for the sieve data (red dots) and the vastly different bed-phase evolution of the bioclastic sands (vertical blue bars) relative to that of quartz sand
Geo-Mar Lett
Fig. 11, the sediments in both cases began to move below threshold with respect to their mean sieve diameters (red dots in Fig. 11), but transport began above threshold with respect to their mean settling diameters (yellow dots in Fig. 11). Whereas the latter situation (yellow dots) is in line with the observations and deductions made in the present study, namely that the high drag on low shape factor particles inhibits transport, the former situation (blue dots) could be conceived to mean that bioclastic sediments have lower transport thresholds. This, however, would be a misconception because the low shape factors of the bioclastic sediments render these to be hydraulically incompatible with the well-rounded ones. The second difference concerns the bed-phase development of the bioclastic material (vertical blue bars in Fig. 11). Both sediments went through lower plane bed phases which extended well into the ripple and dune bed phases (respectively) of wellrounded quartz sands, thereby documenting an inherently different initial transport behaviour in comparison to well-rounded quartz sands. Again this is entirely in line with the observations reported in this study and, at the same time, emphasizes the fact that systematic flume experiments with bioclastic and variable mixtures of siliciclastic-bioclastic sands are overdue.
Conclusions Based on the main findings of this study, the following conclusions are drawn: –
–
–
–
Sieve grain-size data are, as a matter of principle, hydraulically incompatible with corresponding settling tube data, especially in the case of heavy mineral sands and bioclastic sediments composed of low shape factor particles. In the case of bioclastic sediments, the deviation increases with increasing particle size. In hydraulically equilibrated sediments, mean settling diameters of differently shaped particle groups are very similar, although slightly out of phase. Towards coarser size classes, the lower shape factor particle group (bioclastic fraction) is on average somewhat finer, whereas towards finer size classes it becomes on average coarser, the cross-over point lying around 1.7 phi (0.31 mm). This trend is independent of the proportion that each shape group contributes to the total sediment. When exposed to the same hydrodynamic flow conditions, the better rounded siliciclastic particle group achieves better sorting than the irregularly shaped bioclastic one, the lag of the latter increasing as overall sorting improves. Different particle shape groups are transported at different levels in the bottom boundary layer and thus respond as independent, though closely related, hydraulic subpopulations. This feature is ascribed to the higher drag effect,
–
–
–
i.e. higher resistance to motion, of the irregularly shaped particles. Grain-size distributions of mixed sediments consisting of more than one particle shape group simply reflect the average textural characteristics of the combined fractions; analyses of the total sediment are unable to reveal the textural characteristics and transport behaviour of the individual shape components. No sediment transport model is currently able to handle the variable transport behaviour of sediments comprising different particle shapes. The recent work of Durafour et al. (2015), however, is a step in this direction. Systematic flume experiments with bioclastic and mixed bioclastic-siliciclastic sediments with the aim of establishing the bed-phase evolution in sediments composed of irregularly shaped particles, and mixtures of these with well-rounded quartz particles, are overdue and hence strongly recommended.
Acknowledgments The author wishes to thank the former Department of Geology, University of Cape Town, and the former National Research Institute for Oceanology (CSIR), Stellenbosch, South Africa, for funding, work space and laboratory facilities. Several colleagues and technicians provided generous assistance on numerous occasions. Dr. Gordon Moir (at the time stationed at the University of California) is thanked for kindly supplying the computer program for the conversion of settling velocities into equivalent settling diameters, Drs. Andy Duncan and David Reid of the former Department of Geochemistry, University of Cape Town, for their help in debugging and adapting the program to the local hardware, and Nicol Mahnken, Technical Assistant at the Marine Research Department of Senckenberg in Wilhelmshaven, Germany, for generating the SEM images. The article benefitted from constructive assessments by M. Durafour and an anonymous reviewer. Compliance with ethical standards Conflict of interest The author declares that there is no conflict of interest with third parties.
References Alger GR, Simons DB (1968) Fall velocity of irregular shape particles. ASCE J Hydraul Eng 94:721–737 Bagnold RA (1956) The flow of cohesionless grains in fluids. Philos Trans R Soc Lond Ser A 249:234–297 Bagnold RA (1968) Deposition in the process of hydraulic transport. Sedimentology 10:45–56 Braithwaite JCR (1973) Settling behaviour related to sieve analysis of skeletal sands. Sedimentology 20:251–262 Brezina J (1979) Particle size and settling rate distributions of sand-sized materials. In: PARTEC (ed) Proc 2nd European Symp Particle Characterization, 24–26 September 1979, Nürnberg, Germany Bridge JS, Bennett SJ (1992) A model for the entrainment and transport of sediment grains of mixed sizes, shapes and density. Water Resour Res 28:337–363 Briggs LI, McCulloch DS, Moser F (1962) The hydraulic shape of sand particles. J Sediment Petrol 32:645–656
Geo-Mar Lett Compton JS (2001) Holocene sea-level fluctuations inferred from the evolution of depositional environments of the southern Langebaan Lagoon salt marsh, South Africa. The Holocene 11:395–405 Corey AT (1949) Influence of shape on the fall velocity of sand grains. MS thesis, Colorado A & M College, Fort Collins, CO Day JH (1959) The biology of Langebaan Lagoon: a study of the effect of shelter from wave action. Trans R Soc S Afr 35:475–547 Durafour M, Jarno A, Le Bot S, Blanpain O, Lafite R, Marin F (2013) Insitu study of the influence of size and shape of sediments on bedload transport. In: Bonneton P, Garlan T, Sottolichio A, Castelle B (eds) Proc Coastal Dynamics 2013. 7th Int Conf Coastal Dynamics, 24– 28 June 2013, Arcachon Convention Centre, France, pp 541–552 Durafour M, Jarno A, Le Bot S, Lafite R, Marin F (2015) Bedload transport for heterogeneous sediments. Environ Fluid Mech 15:731–751 Flemming BW (1977a) Langebaan Lagoon – a mixed carbonate/ siliciclastic tidal environment in a semi-arid climate. Sediment Geol 18:61–95 Flemming BW (1977b) Depositional processes in Saldanha Bay and Langebaan Lagoon. National Research Institute for Oceanology (NRIO), Stellenbosch, South Africa, CSIR Research Report 362 Flemming BW (1977b) Distribution of Recent sediments in Saldanha Bay and Langebaan Lagoon. Trans R Soc S Afr 42:317–340 Flemming BW (1988) Process and pattern of sediment mixing in a microtidal coastal lagoon along the west coast of South Africa. In: de Boer PL, van Gelder A, Nio SD (eds) Tide-influenced sedimentary environments and facies. D. Reidel, Dordrecht, pp 275–288 Flemming BW (1992) Bed phases in bioclastic sands exposed to unsteady, non-equilibrated flows: an experimental flume study. Senckenberg Marit 22:95–108 Flemming BW (2005) Tidal environments. In: Schwartz M (ed) Encyclopedia of Coastal Science. Springer, Berlin, pp 1180–1185 Flemming BW (2007) The influence of grain-size analysis methods and sediment mixing on curve shapes and textural parameters: implications for sediment trend analysis. Sediment Geol 202:425–435 Flemming BW (2012) Siliciclastic back-barrier tidal flats. In: Davis RA Jr, Dalrymple RW (eds) Principles of tidal sedimentology. Springer, Dordrecht, pp 231–267 Flemming BW, Thum AB (1978) The settling tube - a hydraulic method for the grain size analysis of sands. Kieler Meeresforsch Spec Issue 4:82–95 Folk RL, Ward WC (1957) Brazos River bar: a study in the significance of grain size parameters. J Sediment Petrol 27:3–26 Fontein FJ (1960) Separation of solid particles according to size, shape or specific gravity in a wet medium. Geol Mijnb 39:227–243 Force LM (1969) Calcium carbonate size distribution on the West Florida shelf and experimental studies on the microarchitectural control of skeletal breakdown. J Sediment Petrol 39:902–934 García T, Velo A, Fernandez-Bastero S, Gago-Duport L, Santos A, Alejo I, Vilas F (2005) Coupled transport-reaction pathways and distribution patterns between siliciclastic-carbonate sediments at the Ria de Vigo. J Mar Syst 54:227–244 Gibbs RJ, Matthews MD, Link DA (1971) The relationship between sphere size and settling velocity. J Sediment Petrol 41:7–18 Göğüş M, Defne Z (2005) Effect of shape on incipient motion of large solitary particles. ASCE J Hydraul Eng 131:38–45 Gonzales R, Eberli GP (1997) Sediment transport and bedforms in a carbonate tidal inlet; Lee Stocking Island, Exhumas, Bahamas. Sedimentology 44:1015–1030 Gussmann OA, Smith AM (2002) Mixed siliciclastic-skeletal carbonate lagoon sediments from a high volcanic island, Viti Levu, Fiji, southwest Pacific. Pac Sci 56:169–189 Hayes MO (1979) Barrier island morphology as a function of tidal and wave regime. In: Leatherman SP (ed) Barrier islands. Academic Press, New York, pp 1–27 Hjulström F (1939) Transportation of debris by moving water. In: Trask PD (ed) Recent marine sediments. Dover, New York, pp 5–31
Jopling AV (1963) Hydraulic studies on the origin of bedding. Sedimentology 2:115–121 Kelling G, Williams PF (1967) Flume studies of the reorientation of pebbles and shells. J Geol 75:243–267 Kench PS, McLean RF (1996) Hydraulic characteristics of bioclastic deposits: new possibilities for environmental interpretation using settling velocity fractions. Sedimentology 43:561–570 Komar PD, Reimers CE (1978) Grain shape effects on settling rates. J Geol 86:193–209 Krumbein WC (1934) The probable error of sampling for mechanical analysis. Am J Sci 27:204–214 Krumbein WC (1942) Settling velocity and flume behavior of nonspherical particles. Trans Am Geophys Union 23:621–633 Krumbein WC (1943) Fundamental attributes of sedimentary particles. Univ Iowa Stud Eng Bull 27:318–331 Larsonneur C (1975) Tidal deposits, Mont Saint-Michel Bay, France. In: Ginsburg RN (ed) Tidal deposits. Springer, New York, pp 21–30 Lee HJ, Chun SS, Chang JH, Han S-J (1994) Landward migration of isolated shelly sand ridge (chenier) on the macrotidal flat of Gomso Bay, west coast of Korea: controls of storms and typhoon. J Sediment Res A64:886–893 Maiklem WR (1968) Some hydraulic properties of bioclastic carbonate grains. Sedimentology 10:101–109 McNeill DF, Cunningham KJ, Guertin IA, Anselmetti FS (2004) Depositional themes of mixed carbonate-siliciclastics in the south Florida Neogene: application to ancient deposits. AAPG Mem 80: 23–43 McNown JS, Malaika J (1950) Effects of particle shape on settling velocity at low Reynolds Numbers. Trans Am Geophys Union 31:74– 82 Menard HW (1950) Sediment movement in relation to current velocity. J Sediment Petrol 20:148–160 Mount JF (1984) Mixing of siliciclastic and carbonate sediments in shallow shelf environments. Geology 12:432–435 Paphitis D, Collins MB, Nash LA, Wallbridge S (2002) Settling velocities and entrainment thresholds of biogenic sands (shell fragments) under unidirectional flow. Sedimentology 49:211–225 Passega R (1964) Grain size representation by CM patterns as a geological tool. J Sediment Petrol 34:830–847 Pettyjohn ES, Christiansen EB (1948) Effect of particle shape on freesettling rates of isometric particles. Chem Eng Prog 44:157–172 Reed WE, Le Fever R, Moir GJ (1975) Depositional environment interpretation from settling velocity (Psi) distributions. Geol Soc Am Bull 86:1321–1328 Rittenhouse G (1943) The transportation and deposition of heavy minerals. Geol Soc Am Bull 54:1725–1780 Rubey WW (1933) Settling velocities of gravel, sand and silt particles. Am J Sci 25:325–338 Schulz EF, Wilde RH, Albertson ML (1954) Influence of shape on the fall velocity of sedimentary particles. US Army Corps of Engineers, Missouri River Division, Sedimentation Series Report no 5 Shapiro AH (1961) Shape and flow: The fluid dynamics of drag. Anchor Books (Doubleday & Co), New York Simons DB, Richardson EV, Albertson M (1961) Flume studies using medium sand (0.45 mm). US Geol Survey Water Supply Papers 1498-A Smith DA, Cheung KF (2002) Empirical relationships for grain size parameters of calcareous sand on Oahu, Hawaii. J Coast Res 18:82–93 Smith DA, Cheung KF (2003) Settling characteristics of calcareous sand. J Hydraul Eng 129:479–483 Smith DA, Cheung KF (2004) Initiation of motion of calcareous sand. J Hydraul Eng 130:467–472 Smith DA, Cheung KF (2005) Transport rate of calcareous sand in unidirectional flow. Sedimentology 52:1009–1020 Southard JB (1971) Representation of bed configurations in depthvelocity-size diagrams. J Sediment Petrol 41:903–915
Geo-Mar Lett Sundborg A (1956) The River Klarälven: a study of fluvial processes. Geogr Ann 38:127–316 Tran-Cong S, Gay M, Michaelides EE (2004) Drag coefficients of irregularly shaped particles. Powder Technol 139:21–32 USGS (1958) Some fundamentals of particle size analysis. Subcommittee on Sedimentation, Inter-Agency Committee on Water Resources, US Geological Survey, Report no 12
Walger E (1962) Die Korngrößenverteilung von Einzellagen sandiger Sedimente und ihre genetische Bedeutung. Geol Rundsch 51:494– 507 Weill P (2010) Formation et évolution de cheniers en contexte macrotidal. Approches expérimentales et in-situ. PhD thesis, University of Caen, France Wiberg PL, Dungan Smith J (1987) Calculations of the critical shear stress for motion of uniform and heterogeneous sediments. Water Resour Res 23:1471–1480
