Kowalewski, 1996). Time-averaging is fundamentally
determined by three largely independent parameters:
(1) the rate of sediment accumulation; (2) the characteristics of sediment mixing; and (3) the durability of the individual biological constituents being averaged (Kosnik
et al., 2009; Tomasovych and Zuschin, 2009). The sedimentation rate determines the minimum possible degree
of time-averaging and in general, higher rates of sedimentation lead to less time-averaging. Rate and depth of
mixing determines often the maximum possible timeaveraging and deeper and faster sediment mixing leads
to more time-averaging (e.g., callianassid shrimps, which
effectively sort sedimentary grains by size, shape and/or
other characteristics) (Branch and Pringle, 1987; Tudhope
and Scoffin, 1984). Durability determines the length of
time that a sedimentary grain or fragment remains intact
and recognizable. Fragile sedimentary grains/fragments
are more likely to be destroyed (i.e., eroded, or dissolved)
during mixing leading to less time-averaging, whereas
durable grains can be thoroughly mixed without breaking
leading to more time-averaging (Kosnik et al., 2009). Taxa
with different taphonomic characteristics are, therefore,
likely to withstand different intensities of mixing before
breaking, and therefore record different amounts of timeaveraging even within the same sedimentary deposit.
The potential for differential time-averaging has important
implications not only for the age model of the deposits but
also for the creation of death assemblages and the formation of the fossil record (Cummins et al., 1986; Perry,
1998; Tomasovych and Zuschin, 2009).
It is well known that pre- and post-burial concentrations
of skeletal remains pose problems for assessing population densities of individual species, in particular, the recognition of sharp changes in abundance (including
plague outbreaks). For example, studies of the abundance
of the skeletal elements of the crown-of-thorns starfish
Acanthaster planci in Great Barrier Reef sediments
(Moran, 1992; Walbran et al., 1989a) showed that bioturbation of the sediment is too great to recognize individual
historical outbreaks. A similar study following the mass
mortality of the Caribbean reef urchin Diadema
antillarum in 1983 showed that the fossil record of Antillean Diadema offers no clues as to whether die-offs had
occurred in the ancient past (Greenstein, 1993). Despite
the fact that for several weeks immediately after the
1983 mass mortality Caribbean reefs were littered with
the long black spines and disarticulated calices of this
echinoid, less than a year after the event, the sedimentary
record contained no evidence of a marked increase in the
remains of Diadema (Greenstein, 1991). This may have
been the product of the limited sampling method applied
or more likely, the extent of physical and biological
reworking and chemical breakdown of Diadema skeletal
elements. In a theoretical approach, it has been suggested
that the assessment of the abundance of a species in the
fossil assemblage is impossible to determine until the indices are scaled with the indices of other faunal constituents
(Pandolfi, 1992). Indices should also be calculated at
several different size classes to provide information on
depositional and taphonomic processes.
Species, size, and composition-dependent preservation
and mixing potential and its subsequent implication for
carbonate sediment composition in reefal environments
has received new attention with the advent of amino-acid
racemization dating techniques, that, when combined with
radiocarbon methods, enable large numbers of specimens
to be dated (e.g., Carroll et al., 2003; Kidwell et al., 2005;
Kosnik et al., 2007; Kosnik et al., 2009; Kosnik and Kaufman, 2008; Kosnik et al., 2008). The study from the mixed
carbonate-siliciclastic shelf of Brazil looking at calcitic
brachiopod shells (Bouchardia rosea) from four different
locations showed that the dated shells vary in age from
modern to 3,000 years, with a standard deviation of
690 years (Carroll et al., 2003). The data from four localities displayed significant differences in the range of
time-averaging and the structure of the age distribution
(i.e., scale and mixing of the sediment columns), implying
that environmental factors and local fluctuations in
populations of shell-producing organisms are the principal
determinants of time-averaging in marine benthic shelly
assemblages.
The study of the Rib Reef (Great Barrier Reef, Australia) lagoonal sediments documented significant half-life
differences between large and small Tellina bivalve shells
(Kosnik et al., 2007). There, the top 20 cm of sediment
contained almost exclusively living bivalves whilst the
sediments in the subsequent 100 cm depth were
homogenously mixed. The youngest shell age at 120 cm
depth was 33 years whilst at 30–35 cm depth, the oldest
shell was very old ($4,680 years). In addition, comparisons of age distributions and shell half-lives of four molluscan taxa (Ethalia, Natica, Tellina, and Turbo) from
Rib Reef supported these findings (Kosnik et al., 2009).
There, the 428 dated shells displayed the same homogenous shell stratigraphy below 20 cm depth. Shell half-lives
did not coincide with any single morphological characteristic thought to infer greater durability, but correlated to
a combined durability score based on shell density, thickness, and shape. The half-lives of the four taxa ranged
between $575 years (Tellina) and 1,230 years (Turbo
opercula). Interestingly, whilst the Rib Reef studies
showed a distinct top layer and a deeper age-homogeneous
layer, other studies using radiocarbon ages of the bulk
carbonate sediments have found stratigraphic consistency,
for example, on neighboring John Brewer Reef (Walbran
et al., 1989b), but investigations using
210
Pb (associated
with finer sediment fractions) also showed that the top
50 cm were actively mixed, whilst the next 50 cm were
less mixed (Walbran, 1996).
The study from Caribbean reefal environments in Panama (Kidwell et al., 2005) compared time-averaging and
bivalve shell loss in both carbonate and siliciclastic environments and showed that siliciclastic sands and muds
contain significantly older shells (median. 375 year, up
to $5,400 years) than nearby carbonate seafloors (median
72 year, up to $2,900 year). This led to the conclusion that
160
BIOTURBATION
determined by three largely independent parameters:
(1) the rate of sediment accumulation; (2) the characteristics of sediment mixing; and (3) the durability of the individual biological constituents being averaged (Kosnik
et al., 2009; Tomasovych and Zuschin, 2009). The sedimentation rate determines the minimum possible degree
of time-averaging and in general, higher rates of sedimentation lead to less time-averaging. Rate and depth of
mixing determines often the maximum possible timeaveraging and deeper and faster sediment mixing leads
to more time-averaging (e.g., callianassid shrimps, which
effectively sort sedimentary grains by size, shape and/or
other characteristics) (Branch and Pringle, 1987; Tudhope
and Scoffin, 1984). Durability determines the length of
time that a sedimentary grain or fragment remains intact
and recognizable. Fragile sedimentary grains/fragments
are more likely to be destroyed (i.e., eroded, or dissolved)
during mixing leading to less time-averaging, whereas
durable grains can be thoroughly mixed without breaking
leading to more time-averaging (Kosnik et al., 2009). Taxa
with different taphonomic characteristics are, therefore,
likely to withstand different intensities of mixing before
breaking, and therefore record different amounts of timeaveraging even within the same sedimentary deposit.
The potential for differential time-averaging has important
implications not only for the age model of the deposits but
also for the creation of death assemblages and the formation of the fossil record (Cummins et al., 1986; Perry,
1998; Tomasovych and Zuschin, 2009).
It is well known that pre- and post-burial concentrations
of skeletal remains pose problems for assessing population densities of individual species, in particular, the recognition of sharp changes in abundance (including
plague outbreaks). For example, studies of the abundance
of the skeletal elements of the crown-of-thorns starfish
Acanthaster planci in Great Barrier Reef sediments
(Moran, 1992; Walbran et al., 1989a) showed that bioturbation of the sediment is too great to recognize individual
historical outbreaks. A similar study following the mass
mortality of the Caribbean reef urchin Diadema
antillarum in 1983 showed that the fossil record of Antillean Diadema offers no clues as to whether die-offs had
occurred in the ancient past (Greenstein, 1993). Despite
the fact that for several weeks immediately after the
1983 mass mortality Caribbean reefs were littered with
the long black spines and disarticulated calices of this
echinoid, less than a year after the event, the sedimentary
record contained no evidence of a marked increase in the
remains of Diadema (Greenstein, 1991). This may have
been the product of the limited sampling method applied
or more likely, the extent of physical and biological
reworking and chemical breakdown of Diadema skeletal
elements. In a theoretical approach, it has been suggested
that the assessment of the abundance of a species in the
fossil assemblage is impossible to determine until the indices are scaled with the indices of other faunal constituents
(Pandolfi, 1992). Indices should also be calculated at
several different size classes to provide information on
depositional and taphonomic processes.
Species, size, and composition-dependent preservation
and mixing potential and its subsequent implication for
carbonate sediment composition in reefal environments
has received new attention with the advent of amino-acid
racemization dating techniques, that, when combined with
radiocarbon methods, enable large numbers of specimens
to be dated (e.g., Carroll et al., 2003; Kidwell et al., 2005;
Kosnik et al., 2007; Kosnik et al., 2009; Kosnik and Kaufman, 2008; Kosnik et al., 2008). The study from the mixed
carbonate-siliciclastic shelf of Brazil looking at calcitic
brachiopod shells (Bouchardia rosea) from four different
locations showed that the dated shells vary in age from
modern to 3,000 years, with a standard deviation of
690 years (Carroll et al., 2003). The data from four localities displayed significant differences in the range of
time-averaging and the structure of the age distribution
(i.e., scale and mixing of the sediment columns), implying
that environmental factors and local fluctuations in
populations of shell-producing organisms are the principal
determinants of time-averaging in marine benthic shelly
assemblages.
The study of the Rib Reef (Great Barrier Reef, Australia) lagoonal sediments documented significant half-life
differences between large and small Tellina bivalve shells
(Kosnik et al., 2007). There, the top 20 cm of sediment
contained almost exclusively living bivalves whilst the
sediments in the subsequent 100 cm depth were
homogenously mixed. The youngest shell age at 120 cm
depth was 33 years whilst at 30–35 cm depth, the oldest
shell was very old ($4,680 years). In addition, comparisons of age distributions and shell half-lives of four molluscan taxa (Ethalia, Natica, Tellina, and Turbo) from
Rib Reef supported these findings (Kosnik et al., 2009).
There, the 428 dated shells displayed the same homogenous shell stratigraphy below 20 cm depth. Shell half-lives
did not coincide with any single morphological characteristic thought to infer greater durability, but correlated to
a combined durability score based on shell density, thickness, and shape. The half-lives of the four taxa ranged
between $575 years (Tellina) and 1,230 years (Turbo
opercula). Interestingly, whilst the Rib Reef studies
showed a distinct top layer and a deeper age-homogeneous
layer, other studies using radiocarbon ages of the bulk
carbonate sediments have found stratigraphic consistency,
for example, on neighboring John Brewer Reef (Walbran
et al., 1989b), but investigations using
210
Pb (associated
with finer sediment fractions) also showed that the top
50 cm were actively mixed, whilst the next 50 cm were
less mixed (Walbran, 1996).
The study from Caribbean reefal environments in Panama (Kidwell et al., 2005) compared time-averaging and
bivalve shell loss in both carbonate and siliciclastic environments and showed that siliciclastic sands and muds
contain significantly older shells (median. 375 year, up
to $5,400 years) than nearby carbonate seafloors (median
72 year, up to $2,900 year). This led to the conclusion that
160
BIOTURBATION