Search

[Stratigraphic forward models can reproduce distributive fluvial systems at spatial and temporal scale. The model allows full parameter control and sampling coverage with computer capacity as the only limiting factor. Key finding was that lateral variability is greatest at the transition between the proximal and distal zone. , Abstract To better understand the stratigraphic development of sedimentary systems, it is necessary to link the controls on sedimentary processes to the resulting deposits, which in turn allows predictions of stratigraphic architectures at a range of scales. We use a stratigraphic forward model to link the governing parameters to the distribution of deposits within a distributive fluvial system (DFS). The numerical model has been validated against outcrop observations to establish how the depositional processes needed to form the specific sedimentary system have been reproduced. We chose the previously studied Oligocene to Miocene Huesca DFS in northern Spain to investigate and calibrate the model. Additionally, downstream profiles from modern DFS in northern India, and hydrological measurements from the High Island Creek, Minnesota, USA, were used as input parameters for the model in addition to the outcrop data from the Huesca DFS. The resulting model adequately reproduced the real‐world system. Once validated, the analysis of the modelled DFS led to key findings, which expand our understanding of DFS stratigraphic architecture. Reservoir characteristics in radial DFS are dependent on the angle away from the meridian (straight line from the source through the apex to the distal zone of the DFS). The greater the angle is, the coarser the average grain size in the proximal zone is but the finer the average grain size in the medial and distal zones. Lateral variability of net to gross, sandbody thickness and number, and amalgamation ratio is greatest at the transition between the proximal and medial zone and is still significant in the distal part of the DFS. Stratigraphic forward modelling enhanced our understanding of DFS, which leads to reducing risk associated with exploration, production and storage of fluids in subsurface DFS.]

ACKNOWLEDGEMENTS This work was funded by the SAFARI group. We are deeply grateful to Joe Phillips, Sean Kelly, James Mullins, Ryan King and Jostein Myking Kjaerefjord for help in the field. We would also like to thank Associate Editor, Christopher Fielding, for handling the review of this paper. Additionally we thank reviewers Benjamin Cardenas and Brian Currie for their comments and suggested revisions which have greatly enhanced this paper. Open Access via the Jisc Wiley Agreement

[A new tectonic and depositional model is presented for the Lower Old Red Sandstone of the northern Midland Valley Basin—consolidating new and existing sedimentological data—reconstructing a large foreland basin DFS that records the time‐period between the late phases of the Caledonian Orogeny and the onset of post‐orogenic collapse in the mid‐Devonian , Abstract Reconstruction of the geological history of orogenic events can be challenging where basins have limited and/or fragmentary preservation. Here, we apply understanding gained from modern analogues to the sedimentological analysis of the succession of Upper Silurian to Lower Devonian Lower Old Red Sandstone (LORS), northern Midland Valley, Scotland, in order to reconstruct the foreland to the Caledonian orogeny. A new depositional model is presented which differs significantly from current understanding. Using facies analysis, grain size distribution and palaeocurrent data a large distributive fluvial system is reconstructed. Three lithofacies and nine sublithofacies are identified, forming fluvial channel and floodplain facies associations. The system was derived from an emerging mountain range in the Caledonian foreland undergoing constant tectonic rejuvenation to produce 9 km of coarse‐grained sediment, exhibiting an overall decrease in thickness towards the west and a large‐scale downstream reduction in grain size. Conglomerate sublithofacies dominate proximal areas in the east where amalgamated fluvial channel facies association is abundant, with a downstream increase in the dominance of floodplain facies. Additionally, observed grain size cyclicity is attributed to a pulsatory tectonic influence. The LORS records the time‐period between the late phases of the Caledonian Orogeny and the onset of post‐orogenic collapse in the mid‐Devonian and the presented model allows improved understanding of the north‐Atlantic Caledonian foreland.]

The Salt Wash distributive fluvial system (DFS) of the Upper Jurassic Morrison Formation consists of stacked fluvial channel bodies interbedded with overbank deposits. The Salt Wash DFS has previously been recognised as a braided fluvial fan covering an area of over 100,000 km2. However, the addition of high‐resolution satellite imagery means planform exposures are now recognisable and point bar deposits can be identified in the sand‐rich proximal to medial reaches of the system, requiring a reassessment in this area of the DFS. An individual exhumed point bar deposit has been identified within a 30,000 km2 plan view area of meander belt in central Utah, which offers a unique perspective into the preserved internal distribution of facies and architecture. Field techniques have been utilised in conjunction with LiDAR and heli‐drone three‐dimensional outcrop datasets to compare measurements of the bedform and barform architectural elements within two contrasting outcrop exposure styles. One outcrop (Caineville Wash) has extensive plan view and vertical exposures, whereas the other (Caineville Road) is semi‐restricted to 2D canyon exposures with limited planform exposure. The contrasting exposure styles have been used to develop criteria for the interpretation of sandy meandering river deposits in 2D exposures, where plan view characteristics are not available. Internally, the point bar body exposed at Caineville Wash outcrop can be subdivided into three portions from the plan view imagery: upstream, central and downstream portions of a point bar. Mapping of the internal architecture of the fluvial bar allows recognition of downstream and laterally accreting components. The upstream and downstream portions of the point bar are predominately composed of downstream migrating barform deposits; and the central portion of the point bar consists of laterally accreting elements. Sedimentary logs taken from around the outcrop display vertical profiles commonly considered characteristic of a braided fluvial system. Through understanding the internal architecture of the Caineville Wash point bar deposit, it is possible to create a planform reconstruction of a stacked multi‐storey channel body from a two‐dimensional outcrop. Results indicate that point bar deposits in sand‐rich fluvial systems may have been incorrectly interpreted as braided deposits due to: 1) the presence of a significant proportion of downstream accreting elements within the point bar deposits; and 2) 2D sedimentary logs considered characteristic of a braided fluvial system (such as vertical grain‐size trends and repeating conglomerate lags). Subsequently, sandy meander‐belt deposits may be under‐represented within the proximal – medial portion of the Salt Wash DFS.

Amalgamated sandy meander belts and their deposits are common in modern continental and marine‐connected basins yet comprise a minor constituent of the reported fluvial rock record. This suggests that either amalgamated meander‐belts are uncommon in the rock record or that the recognition criteria are lacking to identify sandy meandering river deposits. To address this apparent discrepancy, the authors document the range and occurrence of amalgamated sandy meander belts (ASMB) from modern basins and the stratigraphic record. ASMB are widely distributed throughout both present and rock record sedimentary basins occurring in foreland, extensional, cratonic, strike‐slip and passive margin basins. They occur in all climatic settings ranging from tundra to hot deserts. Three specific occurrences of ASMB are recognised in modern basins: in the proximal to medial parts of distributive fluvial systems (DFS), as laterally‐confined belts that mainly form axial fluvial systems; and as valley‐confined meander belts that may infill bedrock, alluvial or coastal plain valleys. From the limited amount of rock record examples of ASMB that are available, it is clear that they occur in similar settings to those observed in modern basins, the recognition of which provides a framework for the further prediction and identification of ASMB in the rock record. The lack of recognition of ASMB in the rock record is considered to be due to an absence of characteristics that allow clear distinction between sandy meandering and braided fluvial deposits. Characteristics considered common to both include: multi‐storey, laterally extensive (sheet‐like) amalgamated channel belts, dominance of downstream accreting bedforms, no fining upwards grain‐size profile and little or no fine‐grained sediment and/or soil preservation. In contrast, features considered characteristic of meandering rivers such as inclined heterolithic stratification, high palaeocurrent dispersion, single storey channels and fining upwards grain‐size profiles are absent. The authors suggest that no single criterion can be used to definitively identify sandy meander belt deposits in the rock record and that a combination of systematic variations in accretion direction, palaeocurrent dispersal patterns and recognition of storey scale accretion surfaces is necessary to identify clearly this fluvial style. The common occurrence and distribution of sandy meander belts in modern sedimentary basins together with their limited recognition in the rock record suggests that their true stratigraphic distribution has yet to be determined. This has important implications for palaeogeographic reconstructions, understanding the impact of plant colonisation on fluvial planform style and predicting sandstone body dimensions and internal heterogeneity distribution within hydrocarbon reservoirs and aquifers.

This work was supported by Phase 2 of the Fluvial Systems Research Group (BP, BG, Chevron, ConocoPhillips and Total). AE thanks the University of Aberdeen for additional funding and MGMS thanks the Sao Paulo Research Foundation (FAPESP 2014/13937‐3). The authors also wish to thank numerous residents of the Bighorn Basin for their kind hospitality and access to land, and Isobel Buchanan and Alistair Swan for assistance in the field. The authors also thank reviewers Luca Colombera and Sian Davies‐Vollum and AE Christopher Fielding for helpful comments on this manuscript.

Fluvial channel geometry classification schemes are commonly restricted in relation to the scale at which the study took place, often due to outcrop limitations or the need to conduct small-scale detailed studies. A number of classification schemes are present in the literature; however, there is often limited consistency between them, making application difficult. The aim of this study is to address this key problem by describing channel body geometries across a depositional basin to ensure that a wide range of architectures are documented. This was achieved by studying 28 locations over 4000 m of vertical succession in Palaeocene-aged and Early Eocene-aged deposits within the Bighorn Basin, Wyoming, USA. Five different channel body geometries have been defined based on the external geometric form, and internal arrangement and nature of storey contacts. These include the massive channel body geometry, semi-amalgamated channel body geometry, internally amalgamated channel body geometry and offset stacked channel body geometry, which are considered to be subdivisions of the sheet geometry of many other classifications. An isolated channel body geometry has also been recognized alongside splay channel and sheet sandstone geometries in the floodplain facies associations. Field evidence, including the stacking style of storey surfaces, suggests that the different geometries form a continuum. The nature and degree of amalgamation at the storey scale are important in producing the different geometries and are related to the degree of channel migration. It is speculated that this is the result of differences in sediment supply and available accommodation. In contrast to previous schemes, the classification scheme presented here recognizes the importance of transitional geometries. This geometrical range has been recognized because of the basin-scale nature of the study.

This work has been supported by the Fluvial Systems Research Group Consortium Phase 1, supported by BG, Chevron, Conoco Philips and Total. Discussions with John Howell on prograding sequences are greatly appreciated. The authors would also like to thank Anna Kulikova, Guy Prince, Kelsey McNamara and Karen Oud for assistance in the field. We thank reviewer Brian Willis, an anonymous reviewer and Editor Sebastien Castelltort for constructive comments that improved this manuscript.

The Salar de Atacama forms one of a series of forearc basins developed along the western flank of the Central Andes. Exposed along the northwest margin of the basin, a salt-cored range, the Cordillera de la Sal, records the Mid-Miocene to recent sedimentological and structural development of this basin. Sediments of the Mid-Miocene Vilama Formation record the complex interaction between regional/local climate change, halokinesis and compressional deformation. This study reveals how these factors have controlled the facies development and distribution within the Salar de Atacama. Detailed sedimentary logging, cross-sections and present day geomorphology through the northern Cordillera de la Sal have been used to establish a lithostratigraphy, chronostratigraphy and the regional distribution of the Vilama Formation. The Vilama Formation documents an increase in aridity with a hiatus in sedimentation from Mid-Miocene to 9 Ma with initial uplift of the Cordillera de la Sal. From 9 Ma to 8.5 Ma deposition of a meandering fluvial system is recorded followed by a rapid decrease in sedimentation till 6 Ma. From 6 to 2 Ma, the deposition of extensive palustrine carbonates and distal alluvial–mudflat–lacustrine demonstrates the existence of an extensive lake within the Salar de Atacama. Post 2 Ma, the lake decreased in size and braided alluvial gravels associated with alluvial fans were widespread through the region suggesting a final shift to hyperarid conditions. By comparing the Vilama Formation with similar age facies throughout northern Chile and southern Peru, several shifts in climate are recognized. Climate signatures within northern Chile appear to be largely diachronous with the last regional event in the Mid-Miocene. Since that time, humid events have been restricted to either Precordillerian basins or the Central Atacama. Within the Central Atacama, the final switch to hyperarid conditions was not till the earliest Pleistocene, much later than previously estimated within the region.

ACKNOWLEDGEMENTS M. J. Newman is gratefully acknowledged for his identification of fossil fish specimens from the Easter Ross coast which helped in confirmation of the position of the Achanarras fish bed in this region. The Inverness Field Club and the Carnegie Trust are thanked by SDA for financial support of fieldwork undertaken. Nigel Trewin is acknowledged for inspiring a long-lived interest in the Devonian of Northern Scotland which has led to some fantastic journeys. Gary Nichols and Alberto Saez are thanked for constructive reviews

The stratigraphic development of an Upper Jurassic syn-rift succession exposed at outcrop in the Inner Moray Firth Basin has been investigated using high-resolution biostratigraphy and sedimentology. A continuous 970 m thick section, exposed in the hangingwall of the Helmsdale Fault was logged in detail. The succession spans 8 Ma and contains eight lithofacies types, which indicate deposition in a deep marine setting. Boulder beds contain large, angular clasts, with bed thicknesses typically >2 m and poor sorting suggesting deposition by debris flows. An inverse clast stratigraphy is observed; the oldest boulder beds contain sandstone clasts of Upper Old Red Sandstone (ORS) with younger debris flows containing clasts of Middle ORS calcareous siltstone. A marked change from siliciclastic to carbonate dominated sedimentation occurred during the Early Tithonian, interpreted primarily as a result of change in lithologies in the footwall catchment from sandstone to calcareous siltstone, which reduced supply of siliciclastic sediment. Secondary factors are identified as increased aridity in the Early Tithonian, which reduced sand supply from the hinterland and a third-order Early Tithonian eustatic sea-level rise, which trapped coarser clastic sediment within the hinterland. Biostratigraphy allows calculation of variations in sedimentation rates with recognition of: (1) an early rift phase characterised by sandy turbidite deposition, when sedimentation rates averaged 0.08 m/ky, (2) a rift climax phase from the Early Kimmeridgian where sedimentation rates increased steadily to a maximum of 0.64 m/ky in the Early Tithonian, with strata dominated by boulder scale clast-supported debris flows and (3) a late stage of rifting from the mid Tithonian, where sedimentation rates decreased to 0.07 m/ky. Overall sedimentation rates are comparable to those of other deep marine rift basins. Unroofing a resistant lithology on the footwall of a rift has important implications for siliciclastic sediment supply in rift basins.

Long-term river flow data and one year of isotopic tracer data in a nested 749 km2 catchment were analysed conjunctively to evaluate the relationships between hydrometric statistics, transit times, and catchment characteristics. The catchment comprised two distinct geomorphic provinces; upland headwaters draining glaciated landscapes underlain by crystalline geology and lowland headwaters draining a major regional sandstone aquifer. In the uplands, flow regimes were ‘flashy’ with high runoff coefficients for storm hydrographs, steep recession curves and strong nonlinearity in event responses. In the lowlands, runoff coefficients were low, recessions less steep, and event responses more linear. Flow data from the catchment outfall showing damping of these extremes, but was most strongly influenced by the upland headwaters where precipitation was highest. The damping of variability in stable water isotopes between precipitation inputs and streamflow outputs reflected this; with upland tributaries least damped and lowland tributaries most damped. Attempts to quantify the mean transit times of the sampling points met with mixed success; partly reflecting the short run (1 year) of data, but mainly as a result of the marked damping in lowland sites. As a consequence, MTT estimates can only be said to be in the order of a few years in upland sites, but are probably decadal or greater in lowland tributaries. Again, the catchment outfall averages these extremes, but is more similar to the upland headwaters. Despite the difficulties in quantifying MTTs, it is clear that they, like the hydrological response, primarily reflect the dominant control of catchment soil cover, which in turn is determined by geology and glacial history. Copyright © 2011 John Wiley & Sons, Ltd.

The west slope of the central Andes exhibits a pronounced rainshadow effect. Precipitation between 15° and 27 °S is dominated by summer convective activity from Amazonia, and data analysis shows that the increase in precipitation with elevation due to the rainshadow effect best fits an exponential correlation. Coupling with limited data from high elevations suggests that the correlation is accurate to 4500 m above sea level (m a.s.l.) and perhaps to 5500 m a.s.l., suggesting that increased precipitation goes unrecorded over the peaks of the western Cordillera. South of 27 °S the precipitation is dominated by winter frontal sources and shows no well-defined relationship with elevation. The core zone of hyper-aridity in the Atacama Desert extends from 15 to 30 °S at elevations from sea level to 3500 m a.s.l. Although the Atacama Desert has existed since at least 90 Ma, it is considered that the initial onset of hyper-aridity was most likely to have developed progressively with the uplift of the Andes as they reached elevations between 1000 to 2000 m a.s.l. coupled with the intensification of a cold, upwelling Peruvian Current between 15 and 10 Ma. Also apparent in the palaeogeographic record are subsequent fluctuations between (semi-) arid to hyper-arid conditions that were probably largely controlled by changes in orbital and oceanic forcing. Copyright © 2003 Royal Meteorological Society

Gypcretes of Miocene age are preserved beneath a 9·53 ± 0·36 Ma ignimbrite along the eastern margin of the Oligo-Pleistocene Calama Basin, northern Chile. They are restricted to a single stratigraphic horizon developed within laterally extensive (>35 km) coalesced alluvial fan deposits, developed along the margin of an endorheic basin. Two types of gypcrete are recognized. Type 1 comprises almost completely gypsum-cemented sandstones containing alabastrine nodules and columns, sub-vertical and horizontal veins of fibrous gypsum and ‘v-shaped’ cracks infilled by clastic material, and are interpreted as surface weathered gypsic crusts. Type 2 gypcretes are composed of massive, reddened poikilitic and mesocrystalline gypsum (up to 80% of the rock) with isolated bedding-parallel, clast-rich lenses (200 × 30 cm) and sub-vertical veins of fibrous gypsum. The massive texture resembles that of well developed B horizons in Quaternary alluvial desert soils. The crystal forms suggest an origin as a subsurface gypsic crust formed by a combination of hydromorphic (poikilitic) and illuvial (mesocrystalline) processes with the fibrous gypsum veins suggestive of periodic surface exposure. Gypcrete horizons are up to 25 m thick and composed of both gypcrete types. They represent superimposed phases of surface and subsurface gypcrete development. Quaternary gypcretes are developed in arid climatic regimes, but are not considered to develop under hyper-arid climates. An arid climate is considered to have prevailed in the study area up to 9·5 Ma after which a change to hyper-aridity favoured gypcrete preservation.

The turbidite dominated, Oligo-Miocene Pindos foreland basin of western mainland Greece contains two thick (60–72 m), matrix supported conglomerates. The conglomerates are ungraded and contain three clast types: (1) polymict, rounded, extrabasinal clasts (long axes 3–50 cm); (2) tightly folded, intrabasinal clasts (long axes 1–10 m); and (3) tabular, largely undeformed, intrabasinal blocks (long axes 18–300 m). Clasts are isolated within a slit dominated matrix. These chaotic, matrix supported conglomerates are interpreted as mega-debris flow deposits. During transport, extrabasinal clasts were supported by a combination of matrix cohesion and clast dispersive pressure, folded intrabasinal clasts were supported by a combination of buoyancy (Archimedes principle) and clast dispersive pressure. The large tabular clasts were transported by gravity sliding/gliding within the flow on films at high pore fluid pressure. These different clast support mechanisms were active simultaneously within the Pindos mega-debris flow deposits. As a result, the deposits have no systemic vertical stratigraphy, in contrast to many described large scale mass flow deposits. The mega-debris flow deposits are significantly thicker than most described ancient siliciclastic debris flow deposits and provide an ancient analogue for the thick Recent siliciclastic debris flow deposits on continental margins.

Palaeomagnetic and geochronological data from an Early Palaeozoic high grade metamorphic complex (Jorgina Formation) and Jurassic layered basic intrusion (Moreno Complex) are reported from the Mejillones Peninsula of northern Chile (23–23°30'S). 40Ar–39Ar dates from the Lower Palaeozoic Jorgina Formation and the Moreno Complex are between 170 and 158 Ma, coincident with a phase of emplacement of the north Chilean coastal batholith. This suggests that intrusion and magnetization of the Moreno Complex and the metamorphism and remagnetization of the Jorgina Formation were related to batholith emplacement. Extracted stable components of magnetization from all units (17 sites) define site-mean directions with a scattered distribution. The scatter in site-mean directions is interpreted as being due to minor, localized, non-uniform, block-fault related (normal or strike-slip, or both) rotation after 158 Ma. The palaeomagnetic and geochronological data indicate that no significant large-scale latitudinal translation of crustal blocks has taken place in this part of northern Chile since the Late Jurassic. In addition, they indicate that the uniform clockwise rotation after the mid-Cretaceous which affected the adjacent Cordillera de la Costa either did not extend into the Mejillones Peninsula or took the form of localized block-fault rotations. The restriction of palaeomagnetically defined styles of rotation to discrete areas within the north Chilean forearc indicates that forearc wide block-fault rotation models are not applicable to the Pacific margin of northern Chile.

Analcime (NaAlSi2O6.H2O) forms a characteristic authigenic phase in Andean alluvium of northern Chile. Three stratigraphic units ranging from the lower Cretaceous to the Miocene were studied in order to investigate the presence of analcime in diagenetic assemblages of different ages. The units were deposited in forearc, foreland/back-arc, and intra-arc basinal settings. Characteristics of other sedimentary analcime occurrences were examined in order to define conditions favourable for analcime precipitation. Zeolite zonation and evidence for the original presence of volcanic glass, common to many sedimentary analcime occurrences, are absent in the studied units. Analcime precipitation requires a high Na+/H+ ratio, relatively low Si4+/Al3+ activity and low H2O activity. Diagenetic examination of the three stratigraphic units revealed early quartz and feldspar overgrowths, mechanical clay infiltration, and dissolution of labile rock and mineral fragments, particularly intermediate plagioclase (An30–80). Conditions for these processes are favoured by oxidizing groundwater at near neutral pH. Later diagenesis involved the precipitation of analcime, non-ferroan calcite, quartz, and authigenic clays as the main cementing phases. These formed due to the concentration of oxidizing groundwater and resulted from its continued reaction with detrital feldspars and rock fragments of andesitic and dacitic composition. Analcime precipitation in Andean alluvium resulted from the evolution of saline, alkaline brines under an arid, evaporative climate in closed hydrographic basins with circulating groundwater systems.

A series of small-scale gravity induced synsedimentary faults are described from the Middle Coal Measures (Westphalian B) of the South Wales Coalfield. The synsedimentary nature of the faults is indicated by abrupt changes in sediment thickness across faults, eroded fault scarps and the restriction of faulting to 8·75 m of sediment bounded by laterally persistent black shales above and a seatearth below. Evidence for the non-tectonic origin of the faults is given by the presence of passive footwalls. Channel sandstone bodies occurring within the hanging walls of individual faults were deposited by discrete overbank flood events. Drainage patterns during flood events were controlled by topographic lows adjacent to the faults such that palaeocurrent data within the channel sandstone bodies indicate a S/SW transport direction parallel to fault strike. Synsedimentary fault movements resulted in the diagonal stacking of successively younger sandstone bodies adjacent to the faults, and an increase in tilt of the bodies with depth from 0 26°. The architecture of channel sandstone bodies was controlled by (1) the amount of fault movement and (2) fault block dimensions (or fault spacing). The interaction of these controlling factors resulted in deposition of symmetrical channel sandstone bodies over narrow fault blocks (12 m wide) with large fault displacements (7 m), asymmetrical sandstone bodies over wide fault blocks (30 m) with large fault displacement (7.5 m), and limited channel sandstone body development over fault blocks with small fault displacements (1.5 m). Synsedimentary faulting is thought to have been initiated by either seismic shock and/or overpressuring of pore fluids within buried mudstones.