Basin Research - Volume 4, Issue 2, 1992

We study the interplay of various factors causing vertical grain‐size changes in alluvial basins using a simple coupled model for sediment transport and downstream partitioning of grain sizes. The sediment‐transport model is based on the linear diffusion equation; by deriving this from first principles we show that the main controls on the diffusivity are water discharge and stream type (braided or single‐thread). The grain‐size partitioning model is based on the assumption that the deposit is dominated by gravel until all gravel in transport has been exhausted, at which point deposition of the finer fractions begins.

We then examine the response of an alluvial basin to sinusoidal variation in each of four basic governing variables: input sediment flux, subsidence rate, supplied gravel fraction, and diffusivity (controlled mainly by water flux). We find that, except in the case of variable gravel fraction, the form of the basin response depends strongly on the time‐scale over which the variation occurs. There is a natural time‐scale for any basin, which we call the ‘equilibrium time’, defined as the square of basin length divided by the diffusivity. We define ‘slow’ variations in imposed independent variables as those whose period is long compared with the equilibrium time. We find that slow variation in subsidence produces smoothly cyclic gravel‐front migration, with progradation during times of low sedimentation rate, while slow variation in sediment flux produces gravel progradation during times of high sedimentation rate. Slow variation in diffusivity produces no effect. Conversely, we define ‘rapid’ variations as those whose period is short compared with the equilibrium time. Our model results suggest that basins respond strongly to rapid variation in either sediment flux or diffusivity; in both cases, deep proximal unconformities are associated with abrupt gravel progradation. This progradation occurs during times of either low sediment flux or high diffusivity. On the other hand, basin response to variation in subsidence rate gradually diminishes as the time scale becomes short relative to the equilibrium time. Each of the four variables we have considered ‐ input sediment flux, subsidence, gravel fraction, and diffusivity ‐ is associated with a characteristic response pattern. In addition, the time scale of imposed variations relative to the equilibrium time acts in its own right as a fundamental control on the form of the basin response.

The concept of‘syntectonic’ conglomerate is based on the idea that gravel progradation is mainly generated by an increase in tectonic uplift and erosion of a source area with attendant increase in sediment flux supplied to a basin. However, other mechanisms, such as changes in basin subsidence rates, sorting of supplied sediment, and capability of transporting streams, can also lead to progradation and be difficult to distinguish from a syntectonic origin. Here we use our previously developed model to help understand the origin of gravel progradation in three Neogene alluvial basins ‐ the Bermejo Basin of Argentina, the Himalayan Foreland Basin, and the San Pedro Basin of southern Arizona ‐ all of which have available high‐resolution magnetostratigraphy. Interpretation of the origin of gravel progradation in these basins begins with calculation of basin equilibrium time, which is the time‐scale required for the streams to reach a steady‐state profile, assuming constant conditions.

We then compare the time‐scale of the observed changes in the basin with the equilibrium time to determine if and how the model can be applied to the stratigraphic record. Most of the changes we have studied occur on time scales longer than the equilibrium time (‘slow variations’), in which case the key to interpretation is the relationship between overall grain‐size change and sedimentation rate in vertical sections.

Of the three examples studied only one, the Bermejo Basin, is consistent with the traditional model of syntectonic progradation. Overall progradation in the two other basins is most consistent with a long‐term reduction in basin subsidence rates. In addition, short‐term variation in diffusivity or sediment flux, probably climatically driven, is the most likely control of small‐scale progradation of gravel tongues in the San Pedro Basin. These results, along with observations from other basins, suggest that subsidence is clearly an important control on clastic progradation on ‘slow’ time scales (i.e. generally a million years or more). If subsidence rates are directly linked to tectonic events, then subsidence‐driven progradation marks times of tectonic quiescence and is clearly not syntectonic in the traditional sense.

These examples show that the model can be useful in interpreting the rock record, particularly when combined with other traditional basin‐analysis techniques. In particular, our results can be used to help discriminate between clastic progradation due to tectonic origin and progradation resulting from other mechanisms in alluvial basins.

The deepest part of the Canary Basin, the Madeira Abyssal Plain, receives allochthonous sediments derived from a large drainage basin which, if its subaerial continuation is included, covers an area of 3.36 times 10⁶ km². An international research effort over the last 10 years has recovered over 160 sediment cores from the plain, and the development of a high‐resolution stratigraphy has enabled individual turbidites to be correlated layer by layer. Sedimentation on the Madeira Abyssal Plain during the late Quaternary is dominated by thick turbidite muds separated by thin pelagic intervals. The core density has allowed the mapping of each sedimentary unit throughout the abyssal plain, thus building up a layer by layer picture of sediment accumulation. Over the last 300 kyr, 600 km³ of turbidites compared to 60 km³ of pelagic sediments have been deposited on the plain. Sedimentary structures developed in the coarse basal facies of the larger turbidites are more complex than simple models predict due to surging flows, fluctuating flow velocities and reflection from adjacent high ground. Over the last 300 kyr, there has been a switching of entry points for turbidity currents entering the abyssal plain. From 300 ka to 200 ka, organic‐rich turbidites were emplaced predominantly from the south but around 200 ka this source switched off and subsequent organic‐ and volcanic‐rich turbidites, which included units deposited by giant, possibly hyperconcentrated flows, were emplaced from northern or eastern sources. Although restricted to the late Quaternary, the data presented provides a detailed case study of the evolution of an oceanic basin fill.

A process‐based dynamic‐slope approach is incorporated in a two‐dimensional (2D) stratigraphic basin simulation software. It is based on the classical topographical diffusion concept, but in contrast to previous models it predicts down‐slope sorting of two grain‐size classes (here sand and mud). The effect of compaction is included in the mass balance, as well as arbitrary depth‐dependent transport coefficients (difrusivities). Each lithology will have its own transport‐coefficient function, which can be tuned to predict realistic lithology dispersion (e.g. trapping of sand in shallow‐marine environments). The set of partial‐differential equations is solved numerically by using a fully implicit finite difference method, applying a Newton iteration scheme on the non‐linear equations.

Simulations show that the mass balance due to compactional effects is taken care of. A case from a technically quiescent margin demonstrates, as commonly observed in nature, that the sand fraction increases on the alluvial plain when accommodation space is limited. In the marine part, sand‐rich coarsening‐upward sequences form during progradation, while a condensed and partly erosional maximum flooding surface (MFS) forms during transgression. The MFS is associated with very shaly lithology, a feature that is observed both in field work and on geophysical well logs. If compaction and isostasy modelling are included in this case, the overall architecture is altered. The deposits become proximal, and the timing of maximum regression, etc. is modified. Alternatively, if the time span of simulation is doubled, the result is lower slope angles and increased penetration distance of the sediment wedge into the ljasin. Thus, it is demonstrated that a process‐based dynamic‐slope approach with lithology sorting is a rich alternative to more geometrically based models when the formation of large‐scale stratigraphic architecture is investigated.

The gravitational compaction of sediments is an important process in forward basin modelling. This paper presents a mathematical model for the one‐dimensional compaction of an accreting layer of argillaceous sediments. Realistic constitutive laws for the clay compressibility and the clay permeability, based on soil mechanics tests, were incorporated into the model. The governing equations were put in dimensionless form and the extent of abnormal pore fluid pressure development was found to depend on the sedimentation parameter, a dimensionless group representing the ratio of the sediment hydraulic conductivity to the sediment accumulation rate. The effects of clay compressibility were studied and highly colloidal clays such as montmorillonite developed higher overpressures than less compressible materials. The results also showed that overpressuring developed in shales for cases in which the clay permeability did not go to zero in the limit of zero porosity. Linear models based on simplifying assumptions inappropriate for sedimentary basins were found to give significantly different estimates for the conditions leading to overpressuring. Using reasonable parameters, the model adequately reproduced porosity and pore pressure profiles measured in the sand‐shale sequences of the South Caspian Sea.

Book reviewed in this article:

Sequence Stratigraphy Applications to Shelf Sandstone Reservoirs: Outcrop to Subsurface Examples: J. C. Van Wagoner, D. Nummedal, C. R. Jones, D. R. Taylor, D. C. Jennette & G. W. Riley: American Association of Petroleum Geologists, Tulsa. AAPG Field Conference, September 21–28,1991. $28 ($20 for AAPG Members) ISBN 0–089181‐815‐4.

AAPG Atlas of Oil and Gas Fields–Stratigraphic Traps II & Structural Traps V: N. H. Foster&E. A. Beaumont (Eds): American Association of Petroleum Geologists, Tulsa, 360 pp & 305 pp., 1991. US$60 ($39 for AAPG members) each (hdk) ISBN 0–089181‐585‐6 & 0–089181‐586‐4.

Principles, Methods, and Application of Particle Size Analysis: James P. M. Syvitski(Ed.): Cambridge University Press, Cambridge, 368 pp., 1991. $70 (hbk). ISBN 0–521‐36472‐8.