The formation of groups for regional flood frequency analysis

Abstract

A new technique is developed for identifying groups for regional flood frequency analysis. The technique uses a clustering algorithm as a starting point for partitioning the collection of catchments. The groups formed using the clustering algorithm are subsequently revised to improve the regional characteristics based on three requirements that are defined for effective groups. The result is overlapping groups that can be used to estimate extreme flow quantiles for gauged or ungauged catchments. The technique is applied to a collection of catchments from India and the results indicate that regions with the desired characteristics can be identified using the technique. The use of the groups for estimating extreme flow quantiles is demonstrated for three example sites.     

Subaqueous liquefied and fluidized sediment flows and their deposits

A clear distinction must be made between liquefied and fluidized systems. In liquefied beds and flows, the solids settle downward through the fluid, displacing it upward, whereas, in fluidized beds, the fluid moves upward through the solids, which are temporarily suspended without net downward movement. Many recent references to fluidized sediment gravity flows refer, in fact, to flows of liquefied debris. Most uniformly liquefied beds of well-sorted sand- or gravel-sized sediment will resediment as simple two-layer systems. Liquefied flows can originate either by liquefaction followed by failure, as in many retrogressive flow slides, or by failure followed by liquefaction, as in the case of some slumps. Empirical and theoretical estimates of flow velocity, thickness, and travel distance suggest that natural laminar liquefied flows of fine-grained sand will generally resediment after moving a kilometre or less. Laminar flows of coarse-grained sand will resediment after moving only a few metres. Grain dispersive pressure is thought to be of little significance in the development or maintenance of liquefied flows. Many surficial submarine sand beds are apparently susceptible to liquefaction, including submarine canyon and continental rise deposits. Within submarine canyons and narrow fjords, steep slopes and channels promote the evolution of liquefied flows from slumps by liquefaction after failure and of high density turbidity currents from liquefied flows by the development of turbulence. Upon moving into the lower parts of submarine canyons or into proximal fan channels, liquefied flows will resediment and high density turbidity currents will tend to decline to flows transitional between liquefied flows and turbidity currents. The liquefied, coarser detritus within such transitional flows will be deposited while finer-grained debris will remain in suspension and continue downslope as dilute turbidity currents. Resedimentation of the liquefied portions of such flows may be responsible for the deposition of the A-subdivision of many turbidites and many thick, structureless ‘proximal turbidites’ or ‘fluxoturbidites’. Similar units can originate by liquefaction of the traction deposits of normal turbidity currents. Fluidized flows are probably uncommon, thin, and, where formed, originate through fluidization of the fine-grained tops of liquefied graded beds.     

CONSOLIDATION AND SEDIMENTATION-COMPRESSION STUDIES OF A CALCAREOUS CORE, EXUMA SOUND, BAHAMAS

SUMMARY
A comparison is made between the void ratio and pressure relationships resulting from a laboratory consolidation test and a sedimentation-compression computation on a short core of calcareous mud or ooze of low plasticity. Geo-technical measurements of grain size, bulk density, Atterberg limits, water content, vane shear strength, pore-water salinity, and carbonate content are graphically related to depth in the core. Results of the laboratory consolidation test on this material differ markedly from the in-place relationship between void ratio, or water content, and the effective overburden pressure, or burial depth, shown by the sedimentation-compression curve. The previous maximum consolidation pressure, based on laboratory consolidation test data, is about 60 times greater than the computed in-place effective overburden pressure. An explanation for this difference would include the different magnitudes of time available for consolidation, cementation occurring in-place, and orientation of the constituents. It is suggested that results of the consolidation test on carbonate muds or oozes should be interpreted with caution for geological and engineering purposes.     

Numerical simulation of turbidity current flow and sedimentation: I. Theory

A computer-based numerical model of turbidity current flow and sedimentation is presented that integrates geological observations with basic equations for fluid and sediment motion. The model quantifies those aspects of turbidity currents that make them different from better-understood fluvial processes, including water mixing across the upper flow boundary and the interactions between the suspended-sediment concentration and the flow dynamics and sedimentation. The model includes three numerical components: (1) a layer-averaged three-equation flow model for tracing downslope flow evolution using continuity and momentum equations, (2) a sedimentation/fluidization model for tracing sediment-size fractionation in sedimenting multicomponent suspensions and (3) a concentration-viscosity model for quantifying the changes in resistance of such suspensions toward fluid and sediment motion. The model traces the evolution of a model turbidity current in terms the layer-averaged flow velocity, flow thickness, sediment concentration distribution, and the rate of sedimentation and sediment size fractionation. It generates synthetic turbidites with downslope variations in thickness and grain-size structuring at each point along the flow path. This study represents an effort to evaluate quantitatively the effects of basin geometry, sediment supply and sediment properties on the mechanics of turbidity current flow and sedimentation and on the geometry and grain size characteristics of the resulting deposits.     

Numerical simulation of turbidity current flow and sedimentation: II. Results and geological applications

A process-based, forward computer model of turbidity current flow and sedimentation, termed the TCFS model, has been developed to trace the downslope evolution of individual turbidity flows. Details of the model itself have been presented in a preceding paper. We here outline a series of tests of the TGFS model. The sensitivity tests of the TCFS model to general geological controls reveal the quantitative relationship between these controls and the behaviour of turbidity flows and the geometry and textural features of the resulting turbidites. Experimental turbidity currents on relatively steep slopes accelerate more rapidly and reach higher velocities than those on gentle slopes. Flows with larger initial volumes have higher initial velocities, travel further downslope, and form beds of greater thickness and downslope extent than smaller flows. Experimental high-concentration flows with suspended-sediment concentrations of 25% accelerate more rapidly and reach higher downslope velocities than dilute flows with 5% suspended sediment. The higher velocities and enhanced hindered-settling effects of the high-concentration flows lead to much greater transport distances and reduced vertical and lateral sediment size grading in the resulting turbidites. Beds formed by experimental high-concentration flows are massive or show coarse-tail grading whereas beds formed by low-concentration flows show distribution-grading. Experimental flows fed by coarse sediment sources tend to deposit the bulk of their suspended sediment loads on the proximal slope, resulting in more rapid flow deceleration and sedimentation than flows fed by silt-rich, fine-grained sediment sources. Turbidites formed by coarse-sediment flows tend to have a wedge-shaped geometry, with low downslope extent and high surface relief, whereas turbidites formed by fine-sediment flows tend to have a tabular geometry, with greater downslope extent and lower surface relief. A specific geological test of the TCFS model is based on studies of modern turbidity currents in Bute Inlet, British Columbia, Canada. With the input initial and boundary conditions estimated from Bute Inlet, the model predicts the downslope velocity evolution of turbidity currents comparable to those of modern and ancient turbidity flows measured in Bute Inlet. Model-calculated vertical and downslope grain-size properties of turbidites are similar to those exhibited by surface and cored Bute Inlet turbidites. Model flows tend to decelerate more rapidly than some stronger turbidity currents in the Bute Inlet system, and model beds tend to decrease in grain-size downslope more rapidly than observed bottom sediments. This is probably because the TCFS model flows lacked clay, which is abundant in Bute Inlet; they do not fully simulate turbulent mixing of suspended sediments; and they better represent the unsteady, depositional stage of turbidity-currents than the preceding stage of more-or-less steady-flow conditions. These tests demonstrate that the TCFS model provides a semi-quantitative method to study the growth patterns of submarine turbidite systems. It can serve as a predictive tool for analysing the facies architecture of ancient turbidite systems through simulating multi-depositional events by improving its erosion function, and the compatibility between its numerical components.     

Anorthosites and Related Megacrystic Units in the Evolution of Archean Crust

Archean anorthositic complexes occur in essentially all Archeancratons and contain large equidimensional plagioclase crystals(up to 30 cm. diam.) with highly calcic compositions (An₈₀ toAn₉₀) but are not readily amenable to determination of theirparent melt compositions. However, insight into petrogenesisof the complexes is provided by megacrysts of plagioclase thatare identical to those in the complexes and occur in many Archeanflows, sills, and dikes whose matrices display REE and fractionationpatterns that indicate tholeiitic trends and are compatiblewith prior subtraction of plagioclase during earlier evolutionof the melts. Included blocks of anorthosite and megacrystswith very thin rims that approach the more sodic compositionof lathy plagioclase in the matrices indicate an earlier stageof cryst formation under different conditions of crystallizationthan the matrices. The megacrystic units occur both in greenstonebelts that have oceanic affinities and stable cratonic dikeswarms that have continental affinities. Both major and traceelement contents of the matrices of the megacrystic units differbetween greenstone and cratonic dike environments; the dikesbeing higher in Si0₂, TiO₂ FeO, Na₂, K₂O, and light REEs butlower in Al₂O₃ and CaO. The matrices of both environments followseparate but parallel tholeiitic fractionation with high Fe-enrichmenttrends similar to Skaergaard liquids suggesting relatively lowvolatiles and fo₂. Experimental data and projections in CMAFspace suggest a multistage petrogenesis involving a relativelyhigh-pressure fractionation of olivine and/or orthopyroxenefrom a primitive mafic melt followed by ascension of the fractionated,less-dense melt, probably in several pulses, to a low-pressurechamber, probably at 1 to 2 kb. The depressurization accompaniedby cooling could easily place the melt composition in the plagioclasefield and significantly below the liquidus resulting in severalcrystallization cycles of plagioclase in the low pressure chamber.The melts would crystallize as anorthositic complexes and periodicallyexpel pulses that would form the observed megacrystic flows,sills, and dikes.     

Environmental control on diverse stromatolite morphologies in the 3000 Myr Pongola Supergroup, South Africa

A unique outcrop of partly silicified dolomite in the White Umfolozi section of the Pongola Supergroup, South Africa indicates that stromatolites were diverse and adapted to a range of shallow, tidal depositional settings 3000 Myr ago. Composite columnar stromatolitic bioherms 0.7-1.6m high and 0.4-1.0m in diameter formed along the margins of a tidal channel. They were flanked, away from the channel, by flat stratiform and small domical stromatolites growing in low energy tidal flat environments. Conical stromatolites, 0.05-0.30m high and 0.03-0.10m in diameter, accreted in high-energy coarse-grained carbonate sand along the bottom of the tidal channel. The stromatolites probably formed through the activities of filamentous, oxygen-producing, photoautotrophic cyanobacteria.