Page 105 - Carbonate Sedimentology and Sequence Stratigraphy

96 WOLFGANG SCHLAGER

quence of the extensive glaciations. Longer term variation sequence is distinctly asymmetric. Second and first order
in supply must have resulted from the changes in the global cycles were assumed to be progressively more symmetric.
rates of sea-floor spreading and subduction. It can be ex- First-order cycles in particular were viewed as nearly sym-
pected that this cyclic change in sediment input to the ocean metric patterns of flooding and exposure of the continental
is correlated with the first-order cycle of Haq et al. (1987) interiors and were called “continental encroachment cycles”
and Hallam (1977). by Duval et al. (1998).
In summary, the assumption of standard sequence stratig-
raphy that the sequence pattern is dominated by sea-level Critique of the concept of orders in the standard model
related changes in accommodation as opposed to supply
variations is not rooted in some basic principle of sedimen- While the principle of defining orders by duration has
tary geology but based on case studies and their interpreta- been almost universally followed, the actual figures in the
tion. Consequently, this assumption should not be accepted definitions scatter widely. Two data sets may illustrate this
a priori but tested wherever possible. Progradation and ret- point. Fig. 6.14 plots the durations of sequence cycles of 2 nd
rogradation of shelf breaks are unreliable guides to sea level. and 3 rd order determined in one publication: the sea-level
The reconstruction of sea-level changes from depositional curve of Haq et al. (1987). The two categories clearly dif-
sequences should proceed from two other sources of infor- fer in their modes but broadly overlap in range. Fig. 6.15
mation: first, the timing and extent of exposure unconfor- shows the definitions used in key publications since the in-
mities and downsteps to determine the sea-level falls; sec- troduction of the concept. In the range of 2 nd to 4 th order,
ond, the rates of vertical aggradation of shoal-water systems the discrepancies are about one order on either side. For the
whose facies indicate that water depth during sedimenta- shorter categories the discrepancies are even larger. More-
tion did not deviate significantly from sea level (such as cer- over, the values do not seem to converge with time and
tain carbonate platforms); such sections provide reasonable growing amounts of data. Finally, one notes that many au-
estimates on timing and amount of sea level rises. thors choose durations that are constant on a logarithmic
The balance of the rate of accommodation change and the scale, often coinciding with full powers of ten on a year
rate of sediment supply not only controls progradation and scale. This practice, too, suggests that the orders are sub-
retrogradation of depositional systems. It also profoundly divisions of convenience.
affects small- scale geometry and facies patterns by variation In view of these difficulties, Hardenbol et al. (1998) aban-
of water depth. We will examine this effect for carbonate doned the subdivision into 2 nd and 3 rd order cycles. They
systems in chapter 7. argue that better understanding of mechanisms is required
to justify this classification. I certainly agree that the way se-
ORDERS VERSUS FRACTALS IN THE SEQUENCE quence orders were defined and these definitions later mod-
RECORD ified leaves one with the impression that sequence orders
are subdivisions of convenience, not an expression of natu-
The concept of “orders” of sequences already appeared ral structure.
in the first monographic publication on sequence stratigra- This impression is strengthened if one looks at the char-
phy. Vail et al., (1977, p. 86) described the sequence record as acterization of sequence orders by sediment anatomy. The
a hierarchy of cycles that were ranked by their duration as papers by Vail et al. (1991) and Duval et al. (1998) presented
first, second and third order. Soon after their introduction concepts and examples but no statistical data on the pos-
as categories in time, sequence orders were also character- tulated correlation of sequence duration and depositional
ized by depositional architecture. An important step was anatomy. And to this day, dearth of statistical data char-
the recognition of parasequences - building blocks of clas- acterizes the discussion on this topic. Qualitative observa-
sical sequences that were made up of shoaling successions
tion show no clear correlation of duration and anatomy of
bounded by flooding surfaces rather than exposure surfaces 4 5
sequences. Sequences of 10 -10 ydurationare often of
(Van Wagoner et al. 1988; Van Wagoner et al. 1990). On the
the parasequence type, but there are also many sequences
other hand, it was also observed that in many instances the of the standard type, for instance in the Neogene. Among
building blocks of standard 3 rd order sequences were “sim- 6
sequences in the 10 y domain there is a significant number
ple sequences”, i.e. units bounded by exposure surfaces and
of units bounded by flooding surfaces and without recog-
built like the standard 3rd order sequence (p. 91; Vail et al. nizable lowstand tract. Finally, at all time scales we observe
1991, p. 630; Mitchum and Van Wagoner, 1991). some sequences whose anatomy is more or less symmetri-
Comprehensive classifications that assigned a specific cal, showing deepening followed by shoaling, or fining fol-
anatomy to each sequence order were advanced by Vail et lowed by coarsening trends. Thus, the notion that long cy-
al. (1991) and Duval et al. (1998), see Fig. 6.13. These au- cles are symmetric, 3 rd order and shorter cycles asymmetric
thors proposed that the cycles of fourth order and shorter needs to be examined quantitatively. The fraction of sym-
have the anatomy of parasequences, third-order cycles fol- metric cycles among long cycles may be higher, but they cer-
low the standard model of sequence stratigraphy with units tainly occur also in “third-order” and shorter domains (see
bounded by exposure surfaces and composed of lowstand,
chapter7forexamples).
transgressive and highstand systems tracts. This standard