Fluvial Responses to Climate and Sea-Level Change: a Review and Look Forward

Sedimentology (2000), 47 (Suppl. 1), 2±48

Fluvial responses to climate and sea-level change: a review and look forward

MICHAEL D. BLUM* and TORBJOÈ RN E. TOÈ RNQVIST² *Department of Geosciences, University of Nebraska ± Lincoln, 214 Bessey Hall, Lincoln, NE 68588-0340, USA (E-mail: [email protected]) ²Department of Earth and Environmental Sciences, University of Illinois at Chicago, 845 West Taylor Street, Chicago, IL 60607-7059, USA (E-mail: [email protected])

ABSTRACT Fluvial landforms and deposits provide one of the most readily studied Quaternary continental records, and alluvial strata represent an important component in most ancient continental interior and continental margin successions. Moreover, studies of the long-term dynamics of ¯uvial systems and their responses to external or `allogenic' controls, can play important roles in research concerning both global change and sequence-stratigraphy, as well as in studies of the dynamic interactions between tectonic activity and surface processes. These themes were energized in the ®nal decades of the twentieth century, and may become increasingly important in the ®rst decades of this millennium. This review paper provides a historical perspective on the development of ideas in the ®elds of geomorphology/Quaternary geology vs. sedimentary geology, and then summarizes key processes that operate to produce alluvial stratigraphic records over time-scales of 103±106 years. Of particular interest are changes in discharge regimes, sediment supply and sediment storage en route from source terrains to sedimentary basins, as well as changes in sea-level and the concept of accommodation. Late Quaternary stratigraphic records from the Loire (France), Mississippi (USA), Colorado (Texas, USA) and Rhine±Meuse (The Netherlands) Rivers are used toillustrate the in¯uences ofclimate change on continental interior rivers, as well as the in¯uence of interacting climate and sea-level change on continental margin systems. The paper concludes with a look forward to a bright future for studies of ¯uvial response to climate and sea-level change. At present, empirical ®eld- based research on ¯uvial response to climate and sea-level change lags behind: (a) the global change community's understanding of the magnitude and frequency of climate and sea-level change; (b) the sequence-stratigraphic community's desire to interpret climate and, especially, sea-level change as forcing mechanisms; and (c) the modelling community's ability to generate numerical and physical models of surface processes and their stratigraphic results. A major challenge for the future is to catch up, which will require the development of more detailed and sophisticated Quaternary stratigraphic, sedimentological and geochronological frameworks in a variety of continental interior and continental margin settings. There is a particular need for studies that seek to document ¯uvial responses to allogenic forcing over both shorter (102±103 years) and longer (104±106 years) time-scales than has commonly been the case to date, as well as in larger river systems, from source to sink. Studies of Quaternary systems in depositional basin settings are especially critical because they can provide realistic analogues for interpretation of the pre- Quaternary rock record. Keywords Climate change, ¯uvial sedimentology, Quaternary geology, sea-level change, sequence stratigraphy.

# 2000 International Association of Sedimentologists 2 Fluvial responses to climate and sea-level change 3

INTRODUCTION responses to climate and sea-level change, and attempts to: (a) develop a historical perspective; Studies of continental stratigraphic records, (b) outline a basic process framework, with Quaternary and ancient, were energized in the illustrations of ¯uvial responses to climate and ®nal decades of the twentieth century by two sea-level change from Late Quaternary contexts; somewhat distinct trends that emerged within the and (c) suggest key areas of emphasis for the ®rst geosciences. First, there was increasingly wide- decades of this millennium. spread recognition that human activities were altering climatic and environmental conditions over very short time-scales. Since the stratigraphic record is the only empirical record of HISTORICAL STARTING POINTS global change (Burke et al., 1990), studies of the magnitude and frequency of past global changes, Studies of ¯uvial responses to climate and sea- Quaternary and ancient, were seen as a means to level changes have evolved somewhat differently provide a context for observed historical trends within the ®elds of geomorphology and and predicted near-future conditions. Second, Quaternary geology vs. sedimentary geology. development of sequence-stratigraphic concepts and methods (e.g. Vail et al., 1977; Jervey, 1988; Geomorphology and Quaternary geology Posamentier & Vail, 1988; Posamentier et al., 1988) provided a potential unifying framework Several concepts from the late 1800s and early for much of sedimentary geology, especially 1900s have played key roles in studies of ¯uvial coastal and marine strata, but their application response to climate and sea-level change. One of to continental deposits has been controversial. these was base level, de®ned by Powell (1875) as At a basic level, global change and sequence- the lower limit to which rivers can erode their stratigraphic research have in common a focus on valleys, with his `grand' or ultimate base level changes in process through time in response to corresponding to sea-level. A second concept was external forcing, and it is hard to envisage any that of the graded stream, which Gilbert (1877) scenario for the ®rst decades of this millennium suggested was adjusted in terms of channel slope, in which these issues will be less important. In or longitudinal pro®le, so that the available addition, an important third trend has begun to discharge could transport the amount of sediment emerge, as there is now a deeper appreciation of supplied by the drainage network, and the the magnitude and frequency of tectonic activity channel bed was neither aggrading nor degrading. in plate boundary, continental interior and Mackin's (1948) discussion of the graded stream passive margin settings, and relationships be- is better known and more detailed, and also tween tectonic activity and surface processes (e.g. focused on equilibrium long pro®les that are Molnar & England, 1990; Molnar et al., 1993; adjusted to prevailing conditions of discharge Merritts & Ellis, 1994; Pinter & Brandon, 1997; and sediment load. Burbank & Pinter, 1999). This third trend will One of the ®rst empirical studies of ¯uvial further energize studies of continental records, response to climate change was by Penck & since integrative models of this kind must be BruÈ ckner (1909), whose examination of terraces tested against, and re®ned by, studies of along tributaries to the Danube in southern Quaternary and ancient strata. Germany gave birth to the concept of four Fluvial landforms and deposits provide one of Pleistocene glacial±interglacial cycles. This the most readily studied Quaternary continental model for Quaternary climate change remained records, and alluvial strata represent an important dominant until the development of oxygen- component in most ancient continental interior isotope records from ocean basins in the and continental margin successions. Moreover, 1960s and 1970s, and con®rmation of the modern ¯uvial environments are critical to hu- Milankovitch theory of glaciation (Hays et al., man activities, and ancient ¯uvial deposits are 1976). Critical tothe present discussion,Penck important repositories for hydrocarbons, ground- & BruÈ ckner (1909) linked aggradation to glacial- water and other resources. Studies of the long- period sediment supply, and incision with term dynamics of ¯uvial systems, and their ¯oodplain abandonment and terrace formation responses to external or `allogenic' controls, to interglacial periods (Fig. 1A). By contrast, therefore have a broad range of applications. Lamothe (1918) in the Somme Valley of France, This paper provides an overview of ¯uvial and Fisk (1944) in the Lower Mississippi Valley

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 4 M. D. Blum and T. E. ToÈrnqvist of the USA, focused on the role of glacio- and terrace formation during sea-level fall and eustasy. Fisk's (1944) well-known model sug- lowstand, followed by valley aggradation dur- gested valley incision, ¯oodplain abandonment ing transgression and highstand (Fig. 1B,C).

Older Deckenschotter A (Günz Glaciation)

Younger Deckenschotter (Mindel Glaciation) 500 Hochterrassen (Riss Glaciation) Niedterrassen 450 (Würm Glaciation)

River Save 400 Elevation (m) Elevation

350 0 5 10 15 Distance (km)

Mississippi River Mississippi River B uplands Trench Sea Level Sea Level low high low high

alluvial alluvial fans fans Tertiary bedrock sands and braided stream and gravels backswamp deposits

Mississippi River Mississippi River Sea Level Sea Level low high backswamp low high

alluvial alluvial alluvial alluvial fans fans fans fans sands and sands and gravels zone of Holocene gravels meanderbelt migration C Williana glacial period valley Terrace Bentley cutting episodes Terrace Montogomery Terrace Prairie Terrace Holocene floodplain

Fig. 1. (A) Penck & BruÈ ckner's (1909) model for terraces in southern Germany, linking aggradation to glacial cycles, and incision to interglacials (adapted from Bowen, 1978; original from Penck & BruÈ ckner, 1909). (B) Fisk's (1944) model for valley development during an individual eustatic cycle, illustrating cutting during sea-level fall, and the different stages of ®lling during sea-level rise and highstand. (C) Fisk's model for terraces of the Gulf Coastal Plain in Louisiana (USA), linking incision to glacial cycles, and aggradation to interglacials (B and C modi®ed from Fisk, 1944).

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 5

This `continental interior' vs. `continental mar- geographically circumscribed, nondeterministic gin' dichotomy, or a focus on upstream vs. and nonlinear. downstream controls depending on study area Many of these ideas are re¯ected in review location, was still prevalent in the last decades papers by Knox (1983, 1996) and Starkel (1987, of the twentieth century. 1991), and Bull's (1991) monograph. Knox (1983), for example, summarized the Holocene record from the USA, and argued that climate change produces time-parallel discontinuities in alluvial Upstream controls in continental interiors successions over broad areas, even though the As described by Butzer (1980), a `glacial' vs. direction of change may differ between regions. `interglacial' dichotomy, following Penck & He suggested this could be due to the way BruÈ ckner (1909), also persisted for some time, changes in global atmospheric circulation are and it was commonly assumed that present manifested regionally, and/or the intrinsic char- ¯oodplains represented the Holocene, the ®rst acteristics of each system. Starkel (1987, 1991) terrace was from the last Pleistocene glacial, the also recognized the variability of ¯uvial response second terrace was from the penultimate glacial, toclimate change overthe last 15 kyr, and and so on. Clear turning points can be traced to differentiated classes of European rivers on the the development of radiocarbon dating (Libby, basis of relationships between tectonic highlands, 1952) and tothe hydraulic geometrystudies of Pleistocene ice extent, the former periglacial Leopold & Maddock (1953). Radiocarbon dating region, isostatic rebound and sea-level change. techniques provided an opportunity to develop Starkel's papers emanated from international chronological frameworks for climate change projects that demonstrate increased interest in from palaeobiological and other indicators, ¯uvial systems and climate change during the last and, independently, for alluvial successions. two decades (Gregory, 1983; Gregory et al., 1987, Moreover, the effective limits of radiocarbon eventually served to focus attention on the last full-glacial and the Holocene (last 20 kyr). Leopold & Maddock (1953), by contrast, initiated edge of floodplain a generation of efforts that developed quantitative relations between discharge regimes, sediment modern load and measures of channel geometry (e.g. river width, depth, sinuosity, meander wavelength, slope). Several papers from the 1960s illustrate the immediate impact of hydraulic geometry. Among these are Schumm's (1965) general model for Stuart Highway ¯uvial response to climatic change, Dury's (1965) studies of `under®t' streams and Schumm's younger (1968a) discussion of the concept of channel palaeochannel metamorphosis, based on studies of the Murrumbidgee River, south-eastern Australia (Fig. 2). A subsequent series of in¯uential papers older palaeochannel by Schumm and colleagues in the 1970s and early 1980s served tocautionagainst developing general models for ¯uvial response to climate change. In concise summaries of this work, Schumm (1977, 1991) notes the importance of: (a) spatial 0 3 km and temporal scale; (b) convergence of responses given different external forcing mechanisms; (c) divergence of response following similar external Fig. 2. Concept of channel metamorphosis, introduced by Schumm (1968a) based on studies of the inputs; (d) differential sensitivity of ¯uvial sys- Murrumbidgee River, Australia. Map shows modern tems to external controls due to differing thresh- highly sinuous Murrumbidgee River, vs. older olds for change; and (e) complex response due to channel with larger meanders, and an even older low- internal system dynamics. These issues suggest sinuosity channel (adapted from Schumm & that ¯uvial responses to climate change may be Brakenridge, 1987; original from Schumm, 1968a).

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 6 M. D. Blum and T. E. ToÈrnqvist

1996; Starkel et al., 1991; Branson et al., 1996; a fault block) results in long-pro®le lowering by Benito et al., 1998). nickpoint migration at rates that diminish in the upstream direction. Slingerland & Snow (1988) developed a numerical sediment-transport model Downstream controls in continental margins to simulate long-pro®le evolution in response to base-level fall, and suggested a series of complex Fisk's (1944) investigations of the Lower autogenic oscillations of cutting and ®lling might Mississippi Valley coupled valley evolution to result from a single allogenic perturbation. At a glacio-eustasy and changes in stream gradients, broader conceptual level, Summer®eld (1985) and played a fundamental role in development of suggested the in¯uence of sea-level fall will vary models for ¯uvial response to sea-level change. with geometric relationships between the slope of According to Fisk, the valley was deeply incised the coastal plain and the slope of the shelf as it and swept clean of sediments to a point some becomes emergent (Fig. 3). For example, Brown & 1000 km upstream from the present Gulf of Wilson (1988) and Leckie (1994) discuss the Mexicoshoreline due toincreased channel Canterbury Plains of New Zealand, where aggra- gradients during the Last Glacial Maximum sea- dation occurred during glacial periods because of level lowstand. Braided-stream surfaces formed newly exposed shelf gradients that were less when gradients decreased during latest Pleistocene to middle Holocene sea-level rise, with a transition to a meandering regime during the late Holocene sea-level highstand. Fisk's A Slope of Coastal Plain < Slope of Shelf model was so in¯uential that many workers assigned successively older terraces in the lower SL 1 reaches of river valleys to successively older interglacials. Extreme examples of this type of channel extension SL 2 thinking can be found in studies along the Gulf with deep incision Coast of the USA that inferred short-lived periods of sea-level rise for terraces that did not ®t into the standard model of four glacial±interglacial cycles (e.g. Bernard & Leblanc, 1965). B Slope of Coastal Plain > Slope of Shelf Further empirical studies of continental margin systems have been less visible, perhaps because of the perceived power of Fisk's model. It is therefore appropriate that the Mississippi has SL 1 channel extension played a role in reassessment of ¯uvial responses with progradation SL 2 tosea-level change. Saucier's (1994, 1996) work, and aggradation for example, shows that the Mississippi was never completely excavated, and did not feel the C Slope of Coastal Plain = Slope of Shelf effects of sea-level change as far upstream as Fisk had envisaged. Saucier alsoassigned braided- stream surfaces tothe last glacial period,and suggested they record ¯uvial responses to gla- SL 1 cially induced changes in discharge and sediment channel extension but no significant SL 2 load, whereas the transition to a meandering incision or aggradation regime occurred in the earliest Holocene with the Fig. 3. Model for ¯uvial response to sea-level fall as a loss of a glacial source for sediment and water. function of coastal plain and shelf gradients. (A) A variety of conceptual, experimental and Incision through coastal prism (depositional shoreline theoretical models of river response to base-level break) due to steeper gradients exposed on the inner change appeared in the 1970s and 1980s. Leopold shelf as sea-level falls; (B) aggradation and & Bull (1979), for example, used observations progradation due to sea-level fall across a shelf that from modern rivers to suggest the upstream has a gradient shallower than the coastal plain; (C) effects of base-level rise were limited, whereas channel extension with little incision, other than that necessary tocontainthe channel itself, due tono Begin et al. (1981) and Begin (1988) used ¯ume gradient differences between coastal plain and studies coupled with a diffusion-based model to exposed shelf during sea-level fall. Adapted from show that vertical lowering of base-level (as along Summer®eld (1985).

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 7

98°W97°W96°W95°W94°W

30°N Trinity R. Sabine R. 30°N

Brazos R. Houston San Bernard R. Colorado R.

29°N 29°N Guadalupe R. incised valleys

Nueces R.

28°N deltas 28°N ? Fig. 4. Valleys and deltas of the Corpus ? north-west Gulf of Mexico shelf Christi fans during the last full-glacial period, as identi®ed in high- 0 50 100 km resolution seismic data (adapted ° shelf edge ° from Suter & Berryhill, 1985; 27 N 27 N Anderson et al., 1996). 98°W97°W96°W95°W94°W

steep than the permanently subaerial alluvial plate tectonics on one hand, and facies models on plains. the other (Dott, 1992), such that studies in the Most recently, workers have examined interac- 1960s and 1970s commonly focused on tectonic tions between upstream climatic and downstream activity coupled with autogenic processes to sea-level controls in the lowermost reaches of explain stratal patterns. Sedimentary geologists coastal plain rivers. As discussed later, these have also long recognized the importance of efforts have concentrated on rivers of the Texas climate in the production of sediments, as well Gulf Coastal Plain (USA) and on the Rhine± as the range of depositional systems, environ- Meuse of The Netherlands. The 1980s and 1990s ments and facies, but attempts toexplain changes also witnessed the emergence of high-resolution in alluvial successions through time with refer- seismic-stratigraphic studies on continental ence to climate change are a recent innovation. shelves. Suter & Berryhill (1985), for example, There is little question that plate tectonics demonstrated channel extension and delta pro- plays the overriding role in organization of land- gradation across the Gulf of Mexico shelf during masses into sediment source regions, transport the last sea-level fall and lowstand. Subsequent routes and sedimentary basins, and is thus work by others continued to document the extent primarily responsible for the accumulation of and variability of ¯uvial channels and deltaic thick successions of alluvial strata. The following strata on the shelves and at shelf edges (Fig. 4). discussion refers readers to Miall (1996) for a Important examples include work by Morton & comprehensive discussion of the role of tectonic Suter (1996) and Anderson et al. (1996) for the activity in the development of alluvial succes- north-western Gulf of Mexico (USA), and Tesson sions, and instead highlights alternative trends et al. (1993; also Tesson & Gensous, 1998) for the from the last three decades. Of greatest signi®- Rhone shelf of the Meditteranean. cance is the emergence of sequence-stratigraphic models and their emphasis on the interwoven concepts of sea-level, base level and accommoda- Sedimentary geology tion. Most recently, workers have begun to Eustatic sea-level change played an important examine links between alluvial successions and role in explanations of cyclicity in pre- climate change. Quaternary sedimentary rocks during the ®rst two-thirds of the twentieth century (e.g. Wanless Sea-level, base level and accommodation & Weller, 1932; Sloss, 1963; Wheeler, 1964). However, a period of waning enthusiasm for Development of sequence-stratigraphic con- sea-level change accompanied development of cepts, beginning with publications of Vail and

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 8 M. D. Blum and T. E. ToÈrnqvist colleagues from Exxon Production Research in isolated, high-sinuosity AAPG Memoir 26 (Payton, 1977) initiated a wave fluvial deposits Base Level of research into the responses of depositional low high systems to sea-level change. Publication of the ®rst Exxon cycle charts (Vail et al., 1977) gave rise to new views on the pervasiveness of sea-level changes through geological time. The origins of this wave of research, as it pertains to`¯uvial sequence stratigraphy', lagged tidally-influenced fluvial deposits behind development of the general model, and can be traced to SEPM Special Publication 42 (Wilgus et al., 1988), which contained two papers that catalysed discussion over the next decade. The amalgamated fluvial deposits ®rst was by Jervey (1988) which introduced the term àccommodation' as the space available below base level for sediments to accumulate, a function of eustatic sea-level change and subsidence (i.e. relative sea-level change). The second valley incision and formation of terrace deposits was by Posamentier & Vail (1988), in which ¯uvial response to sea-level change played a prominent role in the development of conceptual models for low-sinuosity, `systems tracts', the basic building blocks in the high-gradient rivers Exxon model. Their process framework rests on 1-10's m 1-10's km the premise that rivers adjust longitudinal pro®les in response to sea-level change, and they argued Fig. 5. Shanley & McCabe's (1993, 1994) model for that: (a) widespread ¯uvial deposition charac- development of depositional sequences in mixed terizes the early stages of relative sea-level fall; (b) ¯uvial±coastal±marine strata of the Cretaceous Western Interior basin of the USA, in response to valley incision and sediment bypass characterizes base-level change. They illustrate the concept of the late stages of sea-level fall and lowstand; and incision and sediment bypass during base-level fall (c) ®lling of valleys occurs during transgression that is common to many sequence stratigraphic and highstand. models, followed by valley ®lling, ®rst with The Exxon views on sequence stratigraphy amalgamated ¯uvial deposits, then tidally in¯uenced have been carefully scrutinized, accepted by strata and ®nally low net-to-gross ¯uvial deposits. some, and criticized by others (e.g. Galloway, 1989; Miall, 1991, 1996; Helland-Hansen & Gjelberg, 1994; Shanley & McCabe, 1994). Some criticisms are grounded on scienti®c shortcom- facies associations, and the `terms lowstand and ings, but others re¯ect an unfortunate use of highstand are not meant to imply a unique period genetic terminology for Exxon systems tracts (see of time or position on a cycle of eustatic or Van Wagoner, 1995a, for a historical perspective), relative change of sea-level'. as well as contradictions within and between Nevertheless, SEPM Special Publication 42 publications from the Exxon group as a whole. resulted in a ¯urry of interest in the concepts For example, numerous ®gures in Posamentier & of sea-level, base level and accommodation, Vail (1988) show widespread ¯uvial deposition and in how ¯uvial and other continental during sea-level fall, but the text commonly infers systems ®t within the mostly marine-derived ¯uvial incision and sediment bypass during that sequence-stratigraphic model (see the review by time. Moreover, Posamentier & Vail (1988) illus- Shanley & McCabe, 1994). Ìncised valleys' trate the development of systems tracts with emerged as a focus of attention; according to respect to eustasy, and/or various combinations Van Wagoner et al. (1990), ìncised valleys' of eustasy and subsidence. By contrast, Van contain signi®cant erosional relief, truncate Wagoner et al. (1988) provide an operational view older strata, juxtapose ¯uvial or estuarine that follows the original de®nition by Mitchum sandstone on marine deposits, and de®ne a et al. (1977), and systems tracts are `de®ned signi®cant basinward shift of facies due to objectively' on the basis of key bounding surfaces, relative sea-level fall. Dalrymple et al. (1994) position within a sequence, stratal geometry and de®ned two types of ìncised valleys', those

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 9

SB3 AST BST FST SB2 AST BST 50-100 m FST SB1 5-10 km

SB3 SB3 SB3 AST AST SB2 SB3 AST SB1 BST AST BST AST BST SB2 AST BST FST SB1 BST SB2 FST AST SB2 AST BST BST FST SB1 Relative Base-Level FST AST Change SB1

palaeosol erosional suspended-load bedload- scouring deposits dominated deposits FSTBST AST SB SB Fig. 6. Hinterland depositional sequence model of Legarreta and colleagues, based on Cretaceous nonmarine strata in Argentina. FST = forestepping systems tract, BST = backstepping systems tract, AST = aggradational systems tract, SB = sequence boundary. Downdip is to the left. Modi®ed from Legarreta & Uliana (1998).

formed by relative sea-level fall, and those (Fig. 5). Posamentier & Allen (1993) proposed a formed in response to some other mechanism. variant on the standard sequence model that They argued the ®rst type should have se- applies to foreland basins, whereas Howell & quence boundaries at their base and will be Flint (1996) discussed active extensional settings. ®lled by predictable successions of ¯uvial and Van Wagoner (1995b) developed a detailed model estuarine facies due torelative sea-level rise for Cretaceous foreland basin strata of the Book (Zaitlin et al., 1994), whereas the second type Cliffs in Utah (USA) that, among other things, has will be dif®cult torecognize since they may served as a catalyst for debate in the literature (see occur within a succession of ¯uvial strata, and Yoshida et al., 1998; Van Wagoner, 1998) and, have nosigni®cance in the Exxonsequence- more informally, among the sedimentology com- stratigraphic sense. munity as a whole. A variety of explanatory models for ¯uvial Fully nonmarine models have been proposed sequence stratigraphy have been published that as well. Legarreta & Uliana (1998) propose a focus on linking ¯uvial and nearshore marine model for Mesozoic continental basins in strata. Shanley & McCabe (1991, 1993, 1994) Argentina that includes: (a) a coarsening-up- recognized depositional sequences based on wards `forestepping' systems tract that is later- interbedded ¯uvial and marine strata from the ally con®ned and dominated by amalgamated Cretaceous of the Western Interior foreland basin channel sandstones; (b) a ®ning-upwards `back- of Utah (USA). Their interpretations are grounded stepping' systems tract; and (c) an `aggrada- on correlations between alluvial strata and coeval tional' systems tract (Fig. 6). Wright & Marriott shorelines that step basinward or landward, as (1993) propose a generic model that is not well as recognition of systematic vertical changes based on actual examples, but incorporates in sandbody amalgamation and/or isolation systems tracts terminology and emphasizes

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 10 M. D. Blum and T. E. ToÈrnqvist relationships between alluvial architecture, palaeosols and changes in accommodation that 500 relate speci®cally to ¯oodplain sedimentation. Gibling & Bird (1994) and Aitken & Flint (1995), among others, use Carboniferous exam- no sandstone bodies ples from Canada and the USA to discuss overall sequence architecture, the role of pa- 450 laeosols, and the recognition, variability and importance of `inter¯uve' sequence boundaries. Most recently, Martinsen et al. (1999) use the accommodation to sediment supply ratio (A/S ratio) described in Muto & Steel (1997) to 400 interpret changes in alluvial architecture. Cycle-Bundle Climate change Studies that link ¯uvial deposition to climate 350 change in the pre-Quaternary record are relatively sparse when compared with the volume of literature on ¯uvial responses to sea-level or Mega-Cycle base-level change that has been generated follow- Cycle (min) Cycle ing introduction of sequence-stratigraphic con- 300 (max) cepts. By far the most common theme is the Height Above Base of Kap Graah Group (m) Height interpretation of ancient climate from individual features (e.g. palaeosols) within alluvial successions, as opposed to interpretations of ¯uvial response to climate change. Much of this work is 250 summarized by Parrish (1998) and not repeated here. Instead the discussion below focuses on examples where workers have speci®cally inter- 0 5 10 preted depositional responses to climate change. Thickness of Sandstone Body (m) A number of workers have inferred climate change as a control based on transitions between Fig. 7. Interpreted cyclical changes in sandbody 6 ¯uvial and other strata. Over time-scales of 10 ± thicknesses deposited by meandering rivers from the 107 years, Olsen & Larsen (1993) suggest the Late Andersson Land Formation, Devonian Kap Graah Devonian of East Greenland can be subdivided Group of East Greenland (after Olsen, 1994). Cycle into depositional complexes similar in concept to (min) indicates cycles de®ned by minimum Exxon's systems tracts. Interpreted changes from thicknesses, whereas Cycle (max) indicates the opposite de®nition. Mega-cycles are interpreted to braidplain environments to terminal ¯uvial sys- re¯ect the Milankovitch eccentricity period, tems that pass laterally to aeolian environments, calculated at 95 kyr for the Devonian, whereas then toa meandering stream-dominated setting individual cycles are interpreted torepresent were attributed to shifts towards more arid and precessional periods of 17 or 20 kyr. then more humid conditions. A number of studies alsointerpret transitionsbetween ¯uvial and other strata due to climate changes over time- strata within the Permian Cutler Group of the scales of 104±105 years (Milankovitch scales). western USA. These include Cojan's (1993) work on Upper Changes within fully alluvial successions have Cretaceous to Palaeocene ¯uvial and lacustrine been ascribed tothe in¯uence ofMilankovitch- strata in the Provence Basin of southern France, scale climate changes as well. Olsen (1990, 1994), Yang & Nio's (1993) studies of Permian for example, examined the Late Devonian Rotliegendes strata of the Dutch North Sea, Andersson Land Formation of east Greenland, Clemmensen et al.'s (1994) summary of inter- the axial meandering stream-dominated complex bedded ¯uvial and aeolian rocks of Permian to described above. He identi®ed two scales of Triassic age in the UK, and Dubiel et al.'s (1996) cyclicity in the thickness of channel sand bodies, discussion of inter®ngering ¯uvial and aeolian which were interpreted torepresent climatically

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 11 controlled changes in discharge regimes at rivers produce stratigraphic records over these Milankovitch-scale periodicities of 100 kyr and time-scales; the section concentrates on alluvial ~ 20 kyr (Fig. 7). Smith (1994) suggested that channels because of their mobile beds and banks, changes in channel sandbody geometry, sand-to- the rapidity with which such systems can mud ratios, sediment accumulation rates and respond to allogenic controls, and the fact that types of palaeosols in Plio-Pleistocene strata from such rivers provide the common link between Arizona (USA) occurred during a tectonically Quaternary and ancient records. quiescent period and might instead re¯ect climate changes. Modelling ¯uvial responses to climate change Channel geometry and depositional style in the pre-Quaternary record has not been Process-based interpretations of ¯uvial responses common practice, but a number of important toclimate and sea-level change shouldbe linked efforts can be identi®ed. Perlmutter & Matthews in some way to controls on the mutually (1989) propose a cyclostratigraphic model that dependent variables of channel geometry and emphasizes Milankovitch-scale climate change depositional style. Numerous discussions of superimposed on tectonically driven basin evolu- alluvial channel process and form, as well as tion. Key model components include weathering ¯uvial depositional processes and systems, are and sediment production, discharge regimes, and readily available elsewhere and not repeated here transport ef®ciency as they relate to variations in (e.g. Knox, 1976; Leopold & Bull, 1979; Richards, climate, as well as how these factors control the 1982; Miall, 1985, 1995, 1996; Schumm, 1985; spatial patterns of clastic environments and Ferguson, 1987; Nanson & Croke, 1992; Bridge, temporal evolution of clastic successions. Paola 1993a,b, 1995; Orton & Reading, 1993; Van den et al. (1992) use a diffusion-based model to Berg, 1995; Brierley, 1996; Nanson & Knighton, examine the roles of sediment ¯ux, rates of 1996; Knighton, 1998; Fielding, 1999). Instead, subsidence, changes in sediment grain-size four fundamental relationships can be distilled (gravel vs. sand) and water supply on patterns from this body of literature: (a) cross-sectional of deposition in foreland basins at time steps of and plan view geometry of a `graded stream' is 107 and 105 years. This study is not speci®cally adjusted toprevailing discharge regimes and concerned with causal mechanisms, but the sediment loads, and maintains a statistically authors note that sediment ¯ux, changes in constant size and shape over some interval of grain-size and variations in water supply can time (102±103 years); (b) ¯uvial facies and deposi- have several causes, including climate change. tional systems are linked to channel geometry Variations in sediment ¯ux and grain size are through the spatial distribution of depositional important in both time steps, with both resulting landforms and environments; (c) changes in the in progradation or retrogradation of gravel fronts. relationship between discharge and sediment By contrast, cyclical changes in water supply play load result in changes in one or more geometric a signi®cant role in shorter time steps only, variables, for example width, depth, width-to- resulting in progradational cycles during time depth ratio, sinuosity or slope; and (d) changes in periods when water supply is increasing. Paola channel geometry result in changes in the lateral et al. (1992) suggest that over time-scales of and vertical distribution of depositional environ- 105 years (Milankovitch time-scales), patterns of ments and facies. deposition for foreland basin settings most clo- sely re¯ect surface processes, whereas over longer time-scales basin subsidence and other tectoni- Aggradation and degradation: changes in cally controlled factors become dominant. sediment storage Aggradation or degradation often accompanies changes in channel geometry and depositional GENERAL PROCESS FRAMEWORK style, and results in changes in sediment storage per unit valley length due tochanges in channel Promising avenues of research lie at the interface bed and ¯oodplain elevation. Width-averaged between Quaternary and ancient ¯uvial systems, bedload transport rates (qs) re¯ect stream especially studies of ¯uvial responses to climate power per unit channel length, W = gQS and sea-level changes at time-scales of 106 years (where W = stream power, g = speci®c weight, or less. The following section focuses on how Q = discharge and S = channel bed slope;

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 12 M. D. Blum and T. E. ToÈrnqvist

Bagnold, 1973). Channel beds aggrade when of Schumm, 1993); (b) sediment overloading from sediment supply exceeds maximum bedload upstream and lateral sources (drainage network transport rates, degrade when the reverse is true feeding a main valley axis) will result in a and maintain stable long pro®les when there is a relatively constant thickness of channel-bed balance between stream power and sediment aggradation; and (c) base-level rise under condi- supply (Fig. 8). tions of constant sediment supply will result in Theoretical models to describe aggradation or downstream decreases in sediment transport degradation are widely published in the engineer- rates, and downstream increasing rates of chan- ing, geomorphological and sedimentological lit- nel-bed aggradation (the `back®lling' of Schumm, erature (e.g. Torres & Jain, 1984; Wyroll, 1988; 1993). Degradation can be triggered in a variety of Paola et al., 1992; Leeder & Stewart, 1996). Such ways, and from both upstream and downstream models are ultimately based on the sediment directions, but always results in channel incision continuity equation, which can be written in its and ¯oodplain abandonment, net sediment re- simplest one-dimensional form as: moval, and production of an unconformity. A simple two-dimensional view of channel-bed

dz/dt +dqs/dx = 0 (1) aggradation and degradation is shown in Fig. 9A, but may be complicated by a number of factors. where z = bed elevation, t = time, qs = width-aver- For example, under equilibrium conditions chan- aged sediment transport rate and x = distance nels could migrate laterally but undergo no along the channel. For a channel bed that is in signi®cant aggradation or degradation, which equilibrium (`graded') and neither aggrading nor translates intononet change in sediment storage. degrading, dz/dt = 0, hence dqs/dx = 0. From this This does not mean that a stable stream cannot simple relationship, three end-member situations leave behind a deposit, since lateral migration of for aggradation can be derived: (a) sediment a channel belt results in cannibalization of older overloading (holding discharge constant) from sediments and emplacement of new deposits. In an upstream point source (typical of an alluvial this case, interpretation of true aggradation and fan) will result in a downstream-tapering wedge development of multistoried channel belts vs. of channel-bed aggradation (i.e. the `down®lling' simple lateral migration is not a simple issue, and should be grounded on observations of deposit thicknesses that exceed the vertical distance between the maximum depth of autogenic scour (Salter, 1993) and the upper limit to ¯oodplain accretion as a channel belt migrates laterally. For example, recent studies demonstrate that depths coarse fine flat steep sediment size channel slope of autogenic scour at ¯ow con¯uences and at channel bends can be up to®ve times deeper than the mean channel depth (Best & Ashworth, 1997), discharge and depths of scour can be signi®cantly below contemporary sea-level positions. degradation aggradation Perhaps equally important are cases where there is little, if any, change in channel-bed Sediment Supply Stream Power elevation, but ¯oodplain height increases or decreases (e.g. Blum & Valastro, 1994; Fig. 9B). For example, ¯oodplain abandonment can take place without signi®cant channel incision due to a reduction in peak discharges and overbank Fig. 8. Balance model for aggradation and degradation ¯ooding. Lateral migration of a channel belt then of alluvial channels, emphasizing changes in the produces a basal erosional unconformity that relationship between discharge and sediment supply. traces laterally up to soils that developed on the Channels aggrade when sediment supply exceeds abandoned ¯oodplain surface. By contrast, for- transport capacity of the discharge regime, and merly stable terrace surfaces can be buried by degrade when the reverse is true. Redrawn from a widely circulated diagram that originated as an renewed overbank ¯ooding if ¯ood magnitudes unpublished drawing by W. Borland of the USA increase. Finally, changes in channel-bed and Bureau of Reclamation, and was based on an equation ¯oodplain elevation may occur at different rates, by Lane (1955). and in some instances in opposite directions (cf.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 13

Tebbens et al., 1999), due tochanges in channel aggradation and/or incision are less than straight- geometry. This variety of possible geometric forward, and should be based on evidence that responses suggests that inferences of ¯uvial bears on changes in both channel-bed and ¯oodplain elevations, wherever possible.

Controls on changes in discharge, sediment supply and sediment storage Changes in channel geometry, depositional style and net sediment storage are dependent variables over time-scales of interest here, and re¯ect prevailing conditions of discharge and sediment supply. What controls changes in discharge and sediment supply over these time-scales?

Discharge regimes Alluvial channel size re¯ects the volume of water that must be transmitted on a relatively frequent basis, with recurrence intervals for channel- forming discharges that range from 1 to 30 years (Williams, 1978; Knox, 1983; Lewin, 1989). Overviews of ¯ood characteristics in various climates can be found in Baker et al. (1988), McMahon et al. (1987, 1992) and Waylen (1996), whereas Pazzaglia et al. (1998) discuss possible relationships between rates of uplift, rates of erosion, river long pro®les and ¯ood hydro- graphs. Critical tothe present discussionis the potential impact of climate change on ¯ood magnitude and frequency, and how this might vary tothe extent that globalclimate change impacts different river systems in the same or different ways. In this context, Knox (1983) summarizes a variety of historical period data that illustrate

Fig. 9. Changes in sediment storage within valley cross-sections, and possible stratigraphic results. Each scenario begins with lateral migration of a single-story channel belt. (A) Channel-bed degradation and ¯oodplain abandonment with terrace formation, then channel-bed aggradation to produce a multistory channel belt, with ¯oodplain deposition over previously stable terrace surfaces characterized by palaeosols (hachures). (B) Reduction in channel size, ¯oodplain abandonment and terrace formation without signi®cant channel-bed degradation, then channel enlargement with ¯oodplain deposition over previously stable terrace surfaces characterized by palaeosols. (C) Signi®cant reduction in channel size accompanied by channel-bed aggradation, ¯oodplain abandonment and terrace formation, followed by channel enlargement accompanied by channel-bed degradation and ¯oodplain deposition over previously stable terrace surfaces characterized by palaeosols.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 14 M. D. Blum and T. E. ToÈrnqvist how climate change can impact ¯ood regimes of global climate change, but in different directions individual rivers. For small drainages in the mid- due to the regional manifestations of changes in latitude USA, he shows that ¯ood discharge per global atmospheric circulation. Similar in-phase unit area increases (holding recurrence intervals and out-of-phase relationships between regional constant) with decreasing precipitation due to precipitation changes and changes in atmospheric decreases in vegetation cover and corresponding circulation are well documented (e.g. Lau, 1988; increases in surface runoff. Moreover, he uses a Hurrell, 1995; Rogers, 1997; Dai et al., 1997). long time series of annual ¯oods from the upper In aggregate, these studies serve toillustrate Mississippi River toillustrate relationships be- four points that are of particular relevance. First, tween ¯ood magnitudes and independently iden- changes in atmospheric circulation through time ti®ed changes in global atmospheric circulation. have a direct impact on the magnitude and These data show that circulation changes have frequency of ¯oods. Second, these changes are signi®cant impacts on the magnitude of rare not directly proportional to increases or decreases events (recurrence intervals > 10 years), but lesser in precipitation, but are instead buffered by other impact on more frequent ¯oods (Fig. 10). More variables that in¯uence runoff rates, especially recently, Knox (1987, 1993) documents changes in vegetation. Third, ¯ood regimes are especially middle to late Holocene channel-forming and sensitive tochanges in climatic inputs, such that extreme ¯oods, and concludes that minor changes small changes in climate can signi®cantly in- in climate produced changes in bankfull dis- crease or decrease ¯ood magnitudes. Finally, charges ranging from 70 to 130% of present-day responses of ¯ood regimes to climate change values. should be geographically circumscribed since Another important approach is illustrated by global changes in circulation may result in Probst (1989), who examines the spatial structure simultaneous changes in discharge regimes every- of runoff ¯uctuations from larger European rivers where, but some regions will be out-of-phase with during the historical period, and shows periods others in terms of the direction of change. with positive and negative deviations from long- term averages. Most commonly all rivers varied in- Sediment supply phase, but at other times runoff regimes of rivers within present-day oceanic and continental re- Routing of sediments in river systems is a topic gimes (Rhone, Loire, Garonne, Seine and Rhine) that resides at the interface between studies of were out-of-phase with rivers in the core modern processes, Quaternary landscape evolu- Mediterranean region of southern Europe (Ebro, tion and development of the pre-Quaternary Guadalquivir and Po). From these relationships, it stratigraphic record. Hovius & Leeder (1998) note might be envisaged that discharge regimes in the that sedimentary basins record erosion of the two areas should respond simultaneously to continents, whereas drainage networks are the

p = 2%, RI = 50 yrs 3000 3 -1 p = 5%, RI = 20 yrs 2000 p = 10%, RI = 10 yrs

1000

p = 99%, RI ~ 1 yrs Flood Magnitude (m s ) 0 1860 1880 1900 1920 1940 1960 1980 Year Fig. 10. Changes in ¯ood magnitudes due to changes in atmospheric circulation, Mississippi River at St. Paul, Minnesota, USA (after Knox, 1983). Data represent a time series of all ¯oods that exceeded 368 m3 s±1, and was disaggregated into time periods de®ned by dominant types of atmospheric circulation (meridional vs. zonal). Flood probabilities were then calculated for each period, with p = the probability that a ¯ood with a given magnitude will be equalled or exceeded in any given year, and RI = average recurrence interval.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 15

`negative imprint' of this same record, and their where E is speci®c sediment yield (t km±2 yr±1), A existence must be inferred from stratigraphic is drainage basin area (km2), H is mean height data. Especially pertinent, then, are the allogenic (relief) of the drainage basin (m), T and TR controls on erosion from upland source terrains are mean annual temperature and annual tem- and delivery to depositional basins. perature range (°C), and R is speci®c runoff There is a long history to discussions of (mm km±2 yr±1). This equation explains only climatic controls on sediment supply, with 49% of the observed variance and, although studies by Langbein & Schumm (1958) and vegetation was not considered, Hovius suggests Fournier (1960) among the ®rst to develop that inclusion of other variables does not sub- quantitative relationships between climate, ero- stantially improve the explanatory power. Hovius sion and sediment yield. Langbein & Schumm (1998) alsouses Milliman & Syvitski's (1992) (1958), in particular, indicate that sediment larger data set toshowthat speci®c sediment yield increases with increasingly effective pre- yield varies with tectonic setting and rates of rock cipitation through the arid to semi-arid transition, uplift, with contractional orogens far outpacing then decreases towards more humid climates due other settings, and low-relief cratons yielding to the increasing importance of vegetation. very low values. Compilations by Walling & Webb (1983, 1996) Tothe extent that drainage area, tectonic suggest the Langbein & Schumm (1958) curve setting, average rates of uplift and total relief may be applicable for the midcontinent USA remain relatively constant for individual river where it was derived, but when other climate systems over time-scales of 103±105 years, Hovius' regimes, especially subtropical and tropical, are (1998) equation suggests signi®cant variations in taken intoaccountthen sediment yield increases speci®c sediment yields with changes in climatic with precipitation at values greater than parameters such as temperature and runoff. ~ 1000 mm yr±1. Because of the mitigating effect However, Leeder et al. (1998) model erosion rates of vegetation in each of these relationships, they and discharge regimes during the late Pleistocene are most applicable for that part of Earth history full-glacial and the Holocene, contrasting the when land plants have been common. In the USA Basin and Range province with the absence of land plants, sediment supply would Mediterranean region, and in doing so illustrate most likely increase as precipitation increases a key point that deserves emphasis. In much the (e.g. Schumm, 1968b). same way as changes in discharge regimes can There is an equally long history to discussions vary regionally, Leeder et al. (1998) show the of relationships between sediment yield and same global climate change can produce out-of- measures of relief or rates of uplift. Indeed, phase responses for the Basin and Range and Walling & Webb's (1983, 1996) compilations Mediterranean regions, when measured in terms show the highest rates of sediment yield in of runoff and sediment yield, sediment ¯ux to the orogenic belts that surround the Paci®c Ocean. depositional basin, relative changes in lake or sea- Moreover, speci®c sediment yields (yields per level and development of a stratigraphic record unit area) decrease as drainage area increases, (Fig. 11). In addition to the above, an important which emphasizes the importance of small and factor that cannot be gleaned readily from the steep catchments for sediment production. present interglacial world is the high rates of Milliman & Syvitski (1992), for example, show erosion and sediment production that accompa- that up to80% ofsediments delivered tothe nies widespread glaciation (e.g. Hallet et al., oceans is derived from small catchments within 1996). active orogens, and suggest that sediment yield is a log-linear function of drainage area and max- Sediment storage en-route to depositional imum elevation. basins Hovius (1998) presents the most comprehensive study of sediment supply to date, based on At any point in time, there is a disparity between examination of 97 intermediate and large drai- the liberation of sediments from uplands and nage basins around the world. He indicates that sediment yields past some point in the drainage speci®c sediment yield can be estimated using network. The difference is related to sediments the following empirical relationship: that are temporarily stored in the drainage network as colluvial, alluvial fan and ¯uvial depos- lnE = ± 0´416 lnA + 4´26 3 10±4H + 0´15T its. This amount can be considered in the context

+ 0´095TR + 0´0015R + 3´58 (2) of the sediment delivery ratio (Walling, 1983),

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 16 M. D. Blum and T. E. ToÈrnqvist

Mediterranean Great Basin Great Basin Mediterranean interglacial interglacial glacial glacial

MARINE LACUSTRINE SEA LEVEL LAKE LEVEL

Great Basin Mediterranean Mediterranean Great Basin glacial glacial interglacial interglacial Low High Low High Low High Low High SEDIMENT SEDIMENT SUPPLY SUPPLY Fig. 11. Model for changes in sediment supply in Mediterranean vs. USA Great Basin climates during glacial and interglacial periods. Diagram on left illustrates conditions that might apply to a marine basin, whereas diagram on the right illustrates conditions for a lacustrine basin. Both diagrams illustrate a succession of possible out-of- phase responses. After Leeder et al. (1998).

de®ned as Sd = Sy/ER, where Sy is sediment yield ocean basins, but up to 20% of all sediment is past a given point in the drainage network, and trapped by reservoirs constructed on a few large ER equals total erosion from the drainage basin rivers. As a result, continent-wide sediment above that point. delivery of 20±30% (Sd = 0´2±0´3) or more may Compilation of sediment delivery ratios from a be more representative. variety of settings indicates that they typically Changes in the sediment delivery ratiocan decrease as drainage area increases and stream signi®cantly imprint the stratigraphic record. gradients decrease, with very high values for very Hovius' (1998) view that drainage networks are small drainage basins, and values as low as 0´1± the negative imprint of basin ®lling would 0´2 for drainage areas > 1000 km2 (Walling, 1983). correctly imply that basin margin sediment Data from modern settings are strongly biased by delivery over geological time must approach human activities that have increased rates of 100%. It follows that over shorter periods of time, erosion, increased sediment storage within small individual segments, or perhaps entire drainage catchments and trapped sediments within arti®- networks, must have delivery ratios greater than cial reservoirs. For example, Meade et al. (1990) average values, and must in some cases be greater note that roughly 10% of all sediment eroded than 1 during periods when previously stored from the conterminous USA presently reaches the sediments are ¯ushed basinward.

Fig. 12. (A) Contrasts between `vacuum cleaner' vs. `conveyor belt' models for sediment supply to basin margins. The vacuum cleaner derives all sediment from more distal parts of the basin by means of excavation of a valley at the basin margin, whereas the conveyor belt relies on sediment delivery from a large hinterland drainage network, and does not require deep incision with complete sediment bypass during sea-level fall. (B) Calculation of relative contribution of vacuum cleaner vs. conveyor belt model, using the palaeovalley ®ll from the last glacial cycle of the Colorado River, Texas, as an example. Dip and strike dimensions and thickness of palaeovalley ®ll based on data presented in Blum (1993) and Aslan & Blum (1999), with data on sediment supply taken from Hovius (1998). For vacuum cleaner calculation, V = volume of palaeovalley ®ll, r = average density of earth materials, D = duration of glacial cycle falling stage and lowstand (kyr), and VC = mass of sediments produced by complete excavation of the valley over duration D. For conveyor belt calculation, sy = speci®c sediment yield (tons km±2 yr±1), A = drainage area (km2), SY = total sediment yield from drainage network (t yr±1), D = as above. For both calculations, CB = mass of sediments delivered from the conveyor belt during time interval D.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 17

Role of sea-level change envisage how sea-level or base-level change can have a signi®cant impact on discharge regimes The following discussion asks a simple ques- or liberation of sediments, since these are tion ± exactly how does sea-level change affect mostly upstream-controlled. Nevertheless, since a ¯uvial system? Tobegin, it is dif®cult to much of the literature implies relative sea-level

A Large Hinterland Incised Valley VACUUM Sediment Source Sediment Source CLEANER MODEL

Drainage Basin Valley Excavated Highstand Deltas CONVEYOR BELT MODEL Filled Valley

Lowstand Deltas Highstand Shoreline Lowstand Fans Lowstand Shoreline B Conveyor Belt Source

3 V = 22.5 km -3 ρ = 2700 kg m 10 km D = 60 kyrs 9 VC = 60.8 x 10 t in 60 kyrs -2 -1 Vacuum Cleaner Source sy = 120 t km yr 100 km 2 A = 100,000 km 2 SY = 12 x 10 t yr-1 D = 60 kyrs 10 CB = 72 x 10 t in 60 kyrs 20 km 30 m

Vacuum Cleaner = 8.4% Conveyor Belt VE = 250 x

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 18 M. D. Blum and T. E. ToÈrnqvist

Fig. 13. De®nition sketch for ¯uvial response to sea-level change along a continental margin with a distinct highstand depositional shoreline break. Diagram illustrates concepts of channel extension during sea-level fall and lowstand, vs. upstream limits of onlap during sea-level rise and highstand. Also, average slope of the channel above the limits of sea-level in¯uence is shown as changes, and re¯ect a long-term tectonically controlled base- level of erosion. controls over sediment supply, it is important can be said for other large ¯uvial systems of the to ®rst look at what sea-level change does not Gulf Coastal Plain of the USA (Blum et al., 1995; do. Blum & Price, 1998), for large rivers of southern Nowhere is confusion more apparent than in Europe such as the Rhone and Po (Mandier, 1988; sequence-stratigraphic models for `incised val- Amorosi et al., 1999), the Rhine±Meuse system in leys' (hereafter valleys or palaeovalleys, since all north-west Europe (e.g. Pons, 1957; Berendsen valleys are incised, by de®nition) that assume et al., 1995; Kasse et al., 1995; ToÈrnqvist, 1998) relative sea-level fall produces incision, and an and the Nile (Stanley & Warne, 1993). In fact, it is upstream-propagating wave of stream rejuvena- dif®cult to®nd an example where complete tion, which produces sediments that entirely sediment bypass in response to base-level fall bypass the coastal plain and newly emergent has actually been documented on Quaternary shelf to provide a critical volume of sediment for river systems that have signi®cant hinterland systems tracts further basinward. This is referred drainage areas and discharge to a depositional to here as a `vacuum cleaner' model for sediment basin. Hence, tothe extent that Quaternary supply, as contrasted with a `conveyor belt' systems provide ground-truth for sequence- model, where sediments are continuously deliv- stratigraphic models, the body of evidence ered tothe basin margin froma large inland actually points in a different direction. These drainage (Fig. 12A). Figure 12(B) illustrates these conclusions are supported by modelling results of models with reference to the Colorado River, Leeder & Stewart (1996), which demonstrate that Texas Gulf Coast (USA), and shows that the high rates of sediment supply can overcome base- volume of sediments produced by complete level fall, and result in progradation without `vacuum cleaner' excavation of a valley during signi®cant incision. the last glacio-eustatic falling stage and lowstand Given a continental margin setting with a would be an order of magnitude less than the total pronounced depositional shoreline break volume delivered by the `conveyor belt' during (Fig. 13), and a shelf that is as steep or steeper that same interval of time. than the coastal plain, why should rivers not The `vacuum cleaner' model has intellectual completely bypass sediments during sea-level roots in Fisk (1944), yet subsequent work in the fall? Simply put, given the in¯ux of sediments Lower Mississippi Valley clearly shows that from the `conveyor belt', complete sediment complete sediment bypass was not the case (e.g. bypass of a coastal plain and shelf during sea- Saucier, 1994, 1996; Blum et al., 2000). The same level fall would require signi®cant downstream

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 19 increases in slope and stream power (b > a in ment supply from the drainage network and local Fig. 13), and corresponding downstream in- physiographical factors. With respect to the latter, creases in rates of sediment transport. In most Leckie (1994) provides an example from the cases, some incision through a highstand deposi- Canterbury Plains of New Zealand, where inci- tional shoreline break takes place as the channel sion has occurred during Holocene sea-level rise extends across the newly exposed shelf because of steep alluvial plain gradients in (Posamentier et al., 1992; Blum, 1993; Talling, combination with high-energy wave attack. 1998; ToÈrnqvist et al., 2000), with corresponding Moreover, Schumm (1993), Wescott (1993) and net removal of previously stored sediment, but it Ethridge et al. (1998) note that ¯uvial systems seems unlikely that the depth of incision will possess several degrees of freedom by which they produce slopes that exceed those further up- can adjust channel parameters tochanges in base stream from the limits of sea-level in¯uence (i.e. level, a discussion not repeated here. b > a is not likely). Hence if the valley upstream Many recent papers (e.g. Butcher, 1990; from the limits of long pro®le adjustment ag- Nummedal et al., 1993; Posamentier & Weimer, grades due to sediment overloading, or laterally 1993; Schumm, 1993; Koss et al., 1994; Merritts migrates due to equilibrium conditions, this et al., 1994; Shanley & McCabe, 1994; Leeder & mode of behaviour is simply translated further Stewart, 1996) have emphasized the issue of how basinward. Periods of complete sediment bypass far inland sea-level actually controls ¯uvial should only occur when sediment supply from incision and aggradation. Again, Fisk's (1944) upstream sources decreases relative to maximum work in the Lower Mississippi Valley has been sediment transport rates, such that the valley extremely in¯uential, postulating eustatic control further upstream is alsoincising. more than 1000 km upstream from the modern On the other hand, sea-level rise along a shoreline. This view has been challenged by more continental margin with the con®guration de- recent studies in the same area (e.g. Saucier, 1994, scribed above (Fig. 13) results in channel short- 1996) that, in turn, are now cited as critical ening, decreases in the distance over which evidence in numerous recent reviews (Shanley & sediments can be stored and, in most cases, McCabe, 1994; Emery & Myers, 1996; Miall, 1996, ¯attening of channel slope. Discharge is con- 1997; North, 1996). Shanley & McCabe (1993) served, but reductions in slope will result in propose strong generalizations, suggesting that corresponding downstream decreases in stream sea-level control typically extends 100±150 km power and sediment transport rates. These con- updip from the coeval shoreline. However, in ditions result in net valley aggradation that in most of these studies it is not clear how the turn may play a signi®cant role in adjustments of landward limit of sea-level control is actually channel geometry and depositional style (dis- de®ned. cussed more fully below). However, even within For this paper, and for the purposes of discus- long periods of overall net valley aggradation sion and comparison, the landward limit is driven by sea-level rise (time-scales of 103± de®ned as the upstream extent of coastal onlap 104 years), signi®cant intervals of incision and/ due tosea-level rise (see Fig. 13). In late or sediment bypass can occur if sediment supply Quaternary contexts, this consists of the intersec- decreases relative to stream power, as shown by tion between the modern ¯oodplain and the Blum (1990, 1993) for the Colorado River of the ¯oodplain surface from the Last Glacial Texas Gulf Coast (USA), ToÈrnqvist (1998) for the Maximum (~ 20 ka), and is not necessarily the Rhine±Meuse and Blum et al. (2000) for the same as the landward extent of incision due to Lower Mississippi Valley. sea-level fall (cf. Ethridge et al., 1998). By con- Turning back to the original question, the trast, the basinward limit of sea-level in¯uence is fundamental responses to sea-level change appear de®ned as the maximum distance of channel to be channel extension or shortening, coupled extension and delta progradation during sea-level with changes in the elevation of channel bases fall. Table 1 summarizes a variety of data from and ¯oodplain surfaces, in order to keep pace ¯uvial systems from Europe and the Gulf Coastal with a shoreline that is advancing or retreating Plain of the USA. These data suggest the land- and changing its elevation. All other adjustments ward limit of onlap due to sea-level rise correlates take place within this context, and should be tohinterland sediment supply, and is inversely regarded as nondeterministic, since they depend related to the gradient of the onlapped ¯oodplain on alluvial valley, coastal plain, shoreface and surface. Accordingly, this distance is highly shelf gradients (see Fig. 3), discharge and sedi- variable and ranges from at least 300±400 km

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 20 M. D. Blum and T. E. ToÈrnqvist

Table 1. Relationship between drainage area, onlapped lowstand ¯oodplain gradient and aggradation distance for six late Quaternary ¯uvial systems. Channel extension and onlap distances are measured with respect to the average highstand shoreline. Data from the Rhone taken from Mandier (1988) and Tesson et al. (1993).

Drainage area LGM ¯oodplain LGM channel Holocene onlap River system (km2) gradient (cm km±1) extension (km) distance (km)

Nueces (Texas, USA) 40 000 50 > 200 40 Trinity (Texas, USA) 50 000 20 150 90 Rhone (France) 99 000 130 50±60 100 Colorado (Texas, USA) 100 000 80 100±120 90 Rhine±Meuse (The Netherlands) 254 000 25 800 150 Mississippi (Louisiana, USA) 3 344 000 15 150 ± 200* > 300 ± 400

* The present birdfoot delta is also extended close to this amount.

for the low-gradient, high-sediment-supply In the ¯uvial context, stratigraphic base-level Mississippi River, to~ 40 km forsteep-gradient, changes in response to changes in geomorphic low-sediment-supply systems like the Nueces base level (lake or sea-level), climate-related and River of the Texas Gulf Coastal Plain. Channel geomorphic parameters (¯ood magnitudes, sedi- extension during lowstand re¯ects shelf physio- ment supply, channel geometries, etc.) and rates graphy, and ranges from 800 km or so for the of uplift or subsidence. It follows from the above Rhine, 100±200 km for rivers of the Texas Gulf that accommodation is not exactly a control Coastal Plain (Colorado, Trinity, Nueces), to very either. Instead, increases or decreases in accom- short distances for the Mississippi, since it now modation are at least in part a response to changes discharges tothe shelf edge. The Rhine is a in geomorphic base level, which de®nes the lower special case, since during the last lowstand it limits to accommodation, and in ¯uvial contexts ¯owed across the very low-gradient North Sea, the upper limits are de®ned by changes in then was diverted south-westward through the discharge regimes and sediment supply as they Strait of Dover and discharged to the Atlantic control equilibrium ¯oodplain heights. (Gibbard, 1995; Bridgland & D'Olier, 1995). It also seems that accommodation, as it is commonly used, somewhat imprecisely mixes processes that operate over a range of rates and Role of accommodation temporal scales; it is dif®cult to reconcile the Few would argue that the concept of accommoda- time-scales over which sediments are deposited tion plays a critical role in sedimentology in the in the ®rst place, whether or not those deposits new millennium. This section again asks a basic will be preserved in the stratigraphic record, and question: what exactly is accommodation? the manner in which ancient alluvial successions To begin, many workers recognize that Jervey's are interpreted in terms of changes in accommo- (1988) de®nition of accommodation uses a `strati- dation or an accommodation/sediment supply graphic' (as opposed to `geomorphic') de®nition ratio. It may be more precise to consider these of base level that can be traced to Barrell (1917), issues in much the same manner as Kocurek & Sloss (1962) and Wheeler (1964) among others. Havholm (1993) and Kocurek (1998), who differ- These differing de®nitions for base level are not entiate `accumulation' vs. `preservation' space in trivial. The geomorphic view de®nes the lower the context of aeolian sequences. In a ¯uvial limits to erosion, and extends upstream from sea- context, accumulation space (`process-scale' or level intothe subaerial landscape as a `base level real-time accommodation) would be the volume of erosion' that de®nes the graded longitudinal of space that can be ®lled within present pro®le (see Bull, 1991), but has no signi®cance process regimes, and is fundamentally governed offshore beyond the ¯uvial system. By contrast, by the relationship between stream power the stratigraphic de®nition de®nes the upper and sediment load, and how this changes in limits to deposition, and was intended to be used response to geomorphic base level. However, in marine systems as well. However, as Wheeler preservation in the stratigraphic record occurs (1964) noted, stratigraphic base level does not when subsidence lowers these deposits below control anything ± it is not a cause but an effect. possible depths of incision and removal. As

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 21 accumulations subside towards this depth, the The most signi®cant development in probability of preservation increases, but it is Quaternary palaeoclimatology over the last 3±4 only assured when this depth is reached. Hence decades has been the new picture of glacial± preservation space (net accommodation?) lies interglacial cycles that emerged from studies of below the maximum depths of incision. For a oxygen-isotope records in deep-sea sediments continental margin, like that shown in Fig. 13, (e.g. Shackleton & Opdyke, 1973; Imbrie et al., preservation space would lie below the lowstand 1984; Chappell & Shackleton, 1986; Shackleton, longitudinal pro®le. 1987). Isotope records are responsible for the Accommodation as a term in the vocabulary of widespread acceptance of the Milankovitch the- sedimentary geology is certainly here to stay. ory of climate change (Hays et al., 1976; Imbrie & However, it might be most useful to use accom- Imbrie, 1979), and provide a model for major modation explicitly in the same manner as changes in global climate and ice volume, as well Kocurek & Havholm (1993) and Kocurek (1998) as the corresponding eustatic component of sea- use accumulation space, and future discussions level over the last several millions of years (see should be explicit about deposition- vs. preserva- the recent review by Pillans et al., 1998). There tion-scale processes, since they operate at differ- has been a high-amplitude 100-kyr periodicity to ent rates, as well as over different time-scales, but full glacial±interglacial cycles during the middle ultimately result in what is preserved in the to late Pleistocene, but lower amplitude 40-kyr stratigraphic record. While accommodation re- cycles dominated prior to that. One important mains a useful concept, the stratigraphic concept implication pertains to the concept of `average of base level is the same as the upper limits of conditions', since only a small percentage of the accommodation ± it is not a control on anything, last 106 years resembled full interglacial (present- has outlived its usefulness, and it should simply day) or full glacial (20 ka) conditions (with be abandoned. eustatic sea-level at ±120 to±130 m). As noted by Porter (1989), up to 80% of any middle to late Pleistocene 100-kyr glacial cycle witnessed global temperatures that were cooler than full intergla- MAGNITUDE AND FREQUENCY OF cial conditions, but not as cold as a full glacial, CLIMATE AND SEA-LEVEL CHANGE and eustatic sea-level was at ±15 to±85 m (Fig. 14A). Fluvial systems respond to climate changes with- More recently, work on ice cores and the in their catchments and/or relative sea-level marine record shows that signi®cant climate changes that occur at their mouths. Over time- changes at scales of 103 and 104 years are super- scales of interest here, climate change occurs imposed on the 100-kyr glacial±interglacial cy- because of global-scale changes in radiation cles discussed above (e.g. Johnsen et al., 1992; balance and atmospheric or ocean circulation, Bond et al., 1993, 1997; Grootes et al., 1993; Bond and sea-level changes have a globally coherent & Lotti, 1995; Behl & Kennett, 1996), and provides `eustatic' component. However, global-scale an exciting picture of the complexity of the last changes are manifested in speci®c regions in a glacial, as well as the Pleistocene±Holocene variety of ways. The following section outlines transition. `Dansgaard±Oeschger cycles' (D-O cy- key elements of climate and sea-level change, cles), for example, consist of abrupt warming over as viewed from the Quaternary vs. the a period of decades to centuries, followed by Phanerozoic. cooling trends that last 1±2 kyr. D-O cycles are themselves bundled together into cooling trends that last 10±15 kyr (`Bond cycles'), culminate The Quaternary with massive iceberg discharges and deposition The Quaternary provides a unique perspective on of large quantities of ice-rafted debris into the the dynamism of the Earth's climate and oceano- North Atlantic (`Heinrich events'), and are graphic systems, and has been the subject of followed by abrupt warming (Fig. 14B). The intense study over the past three decades. Recent causal mechanisms for these cycles are not publications that summarize records and me- completely understood, but they must in some chanisms of Quaternary climate and/or sea-level way re¯ect the internal dynamics, and feedback change include Crowley & North (1991), Broecker within, the coupled atmosphere±cryosphere± (1995), Pirazzoli (1996), Lowe & Walker (1997) ocean system (Bradley, 1999), and their effects and Bradley (1999). are propagated around the ocean basins via

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 22 M. D. Blum and T. E. ToÈrnqvist

Fig. 14. (A) Eustatic sea-level curve based on precision U±Th dating of corals from Huon Peninsula, Papua New Guinea (after Chappell et al., 1996), with oxygen isotope stages (boundaries shown by dashed lines). (B) Comparison of the proportion of the planktonic coldwater foraminifera Neogloboquadrina pachyderma (sinistral) in marine core VM 23±81 (after Bond et al., 1993).

impacts on the large-scale thermohaline circu- trends (COHMAP Project Members, 1988; Wright lation and formation of North Atlantic deep et al., 1993; Kutzbach et al., 1997). Local sea-level water (Broecker, 1997). changes also most commonly depart from eustasy Regional climate changes are more complex in both rate and magnitude, and sometimes in than the global picture due to the way global direction, due to local to regional tectonics, changes in atmospheric circulation are mani- mantle response to accumulation and melting of fested in different regions; for the last 20 kyr, ice sheets, glacio- and hydro-isostatic deforma- many regions show signi®cant departures in tion of coastlines, and migration of ocean water magnitude and/or direction from global climatic mass to geoidal lows (Tushingham & Peltier,

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 23

Global Mean Global within the Phanerozoic is between what Frakes et al. (1992) refer toas alternating `warm' and Sea Level Temperature Ma Ma low high cold warm `cool' modes with average durations of 50 Myr, and the long-wavelength cycles of relative sea- level change evident in Haq et al. (1987, 1988;

cool Fig. 15). For the purposes of this paper, such 55 climate changes are initiated by the agglomera- CENOZOIC 67 tion and break-up of supercontinents, and/or by migrations of landmasses into different latitu- 105 dinal belts, and are beyond the time-scales of

CRETAC. interest here. 144 Of great interest, however, are the externally cool forced cyclical climate and sea-level changes that 183 might go on within these longer-term modes. For

JURASSIC 208 example, there is little reason to believe that Milankovitch-scale radiation-forced changes have

TRIAS 245 not been signi®cant throughout the Phanerozoic 253 (Barron & Moore, 1994), even though their

PERM 286 periodicity, amplitudes and phase relations may have changed. Indeed, calculations by Berger cool et al. (1989) and Berger & Loutre (1994) show that 333 wavelengths of the present 100 kyr eccentricity

CARBONIF 360 cycle have varied little since 500 Ma, whereas the obliquity (currently 41 kyr) and precessional (currently 23 and 19 kyr) components were short- DEVON 408 er at 500 Ma (~ 30 kyr, 19 and 16 kyr, respectively)

SIL 421 due to changes in Earth±Moon distance and day 438 length. Even though Milankovitch-scale changes cool 458 were present throughout the Phanerozoic, they operated within a context de®ned by plate ORDOVIC 505 tectonics and long-term evolution of the bio- sphere. Accordingly, this type of external forcing warm warm warm warm should have different manifestations in cold CAMB 570 ? 560 modes, when continental landmasses were located in more polar latitudes and glaciation was Fig. 15. Estimated eustatic sea-level and global widespread, than in warm modes when evidence temperature changes through the Phanerozoic, for signi®cant glaciation is lacking (Frakes et al., illustrating warm and cool modes. Adapted from Frakes et al. (1992). 1992). Twocaveats shouldbe kept in mind. First, Broecker (1997) stresses the possible ampli®cation of relatively limited orbital forcing by changes in the ocean thermohaline circulation 1992; Lambeck, 1993; Fleming et al., 1998; Peltier, during considerable parts of Phanerozoic his- 1998). Strong eustatic signatures may be present tory. Second, and equally important, is the at some localities, but all sea-level change is ampli®cation or suppression of ¯ood regimes essentially relative. that occurs even with minor changes in climate, and the manner by which such changes are manifested in channel geometry, Pre-Quaternary ¯uvial depositional styles, sediment transport A broad range of climate and sea-level change rates and changes through time in net sediment is evident from the Phanerozoic record (Frakes, storage. The key question is how to recognize 1979; Crowley & North, 1991; Frakes et al., pre-Quaternary ¯uvial responses to climate 1992; Parrish, 1998), and only small parts of change, sea-level change, or some combination this record may resemble the Quaternary `ice- of the two, and disentangle them from tectonic house'. Perhaps the most important distinction effects or autogenic processes.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 24 M. D. Blum and T. E. ToÈrnqvist

FLUVIAL RESPONSE TO CLIMATE Types of response AND SEA-LEVEL CHANGE: Fluvial responses to climate and sea-level change RECOGNITION can be disaggregated into stratigraphic, morphological and sedimentological components. As Potential ¯uvial responses to climate and sea- de®ned here, stratigraphic responses (strati- level change vary by position within the system, graphic architecture) are produced by aggradation and the ability to store sediment over various due toincreasing accumulationspace, degrada- lengths of time. At one extreme are the high tion as accumulation space decreases and lateral stream powers and high sediment delivery ratios migration when there is no change in accumula- common in the upper reaches of drainage net- tion space. These types of responses serve to works. To the extent that these settings often differentiate the larger-scale aspects of alluvial possess bedrock channels with discontinuous successions that mostly re¯ect allogenic controls ¯oodplains, and short residence times for sedi- (climate change, sea-level change, tectonic ment as a whole, there is little potential for activity). Lower-gradient mixed bedrock/alluvial records of ¯uvial response to climate change that valleys typically contain ¯ights of downward- span signi®cant lengths of time (> 104 years). By stepping terraces and underlying ®lls that demar- contrast, lower-gradient mixed bedrock/alluvial cate long-term valley incision punctuated by valleys with signi®cant drainage areas (> 103 km2) relatively short periods (103±104 years) of lateral commonly have continuous ¯oodplains and low migration and perhaps minor aggradation. Fluvial sediment delivery ratios. These settings typically systems in subsiding basins are characterized by show long-term rates of incision that may outpace just the opposite, long-term net aggradation, rates of rock uplift (Hovius, 1999), and serve as punctuated by relatively short periods (103± the `conveyor belts' for sediment delivery to basin 104 years) of incision and/or lateral migration. margins, with sediment residence times mea- Stratigraphic responses are perhaps best de®ned sured in multiples of 104±106 years. As a result, in terms of unconformity-bounded stratigraphic mixed bedrock/alluvial valleys commonly con- units of varying scale (i.e. the `allostratigraphic tain records of ¯uvial landscape evolution over units' of the North American Commission on late Cenozoic time-scales, but have little to no Stratigraphic Nomenclature, 1983; see also actual preservation space. Walker, 1990, 1992; Autin, 1992; Blum & Alluvial valleys within subsiding basins repre- Valastro, 1994); de®nition of allostratigraphic sent the opposite extreme, and provide the units is independent of lithological characteris- critical link between Quaternary and ancient tics, but should include: (a) basal erosional systems. Present-day continental interior basins unconformities that trace laterally and upwards, occur within tectonically active drylands, where eventually truncating older landscape or palaeo- drainage integration has not proceeded to the landscape elements that are commonly de®ned by extent that ¯uvial systems reach marine basins, or palaeosols; and (b) upper boundaries that repre- in subsiding reaches of ¯uvial systems that are sent intact or truncated former ¯oodplain surfaces en-route to a marine basin. Rift and pull-apart that are/were de®ned by palaeosols or coals basins of the arid western USA provide examples (Fig. 16). of the former, whereas the Himalayan foreland For lower-gradient mixed bedrock/alluvial val- provides an example of the latter. Continental leys, the vertical separation between allostrati- margin systems are numerous, but much of the graphic units re¯ects the frequency of responses present-day landscape is drained by a relatively to allogenic forcing, but also must correlate to few large, long-lived rivers (e.g. Potter, 1978) that long-term rates of uplift; with increasing rates of discharge to passive margins or ¯ooded forelands, uplift, the preservation of older units becomes whereas smaller ¯uvial systems typically drain to increasingly fragmentary. By contrast, for con- rifted or sheared margins. Sediments delivered to tinental margin systems, vertical separation be- basins enter preservation space when they sub- tween allostratigraphic units re¯ects the side below maximum depths of incision, but until frequency of responses, sediment supply and that time they can be remobilized by a variety of relative sea-level change (eustasy vs. subsidence), processes, and transported out of the basin and with increasing rates of subsidence, preser- (continental interior) or elsewhere within the vation in the stratigraphic record becomes in- basin (both continental interior and continental creasingly likely. Twoscenarioshighlight this margin settings). relationship. First, contrast slowly subsiding vs.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 25

Fig. 16. Concept of allostratigraphic units of varying scale (scale bar applies to both diagrams, and vertical hatchures represent soil pro®les with the relative degree of soil development represented by length of hatchure). (A) Mixed bedrock/alluvial valley setting, with ¯ights of downward-stepping terrace surfaces and underlying ®lls, and a complex valley ®ll with multiple allostratigraphic units. Older and higher terraces are commonly discontinuous and dissected, with signi®cant thicknesses of strata removed; they also may represent long periods of time, and multiple lateral migration, aggradation and incision events within a limited elevation range that are not readily differentiated due to degree of preservation. Younger mostly intact terraces and the complex valley ®ll may represent the last glacial cycle. (B) Continental margin situation, where sea-level has fallen then risen, resulting in downward-stepping then aggradational successions of allostratigraphic units bounded by palaeosols. Vertical hatchures represent palaeosols and, in each diagram, the solid grey line encloses a single 100-kyr glacial± interglacial cycle. In (B), the relative degree of palaeosol development is shown schematically by the length of the hatchure (sandy facies only, with muddy facies slightly different), and is a proxy for the duration of subaerial exposure before burial. As de®ned here, unconformities that bound allostratigraphic units are equivalent in concept to the higher-order bounding surfaces of Miall (1996), the basal valley ®ll unconformity in (B) would be a strongly time-transgressive equivalent to the Exxon sequence boundary as classically de®ned, and the well- developed alluvial plain palaeosol would be equivalent to the `inter¯uve sequence boundaries' discussed by various workers. After the work of Blum & Price (1998) on the Colorado River, Texas (USA).

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 26 M. D. Blum and T. E. ToÈrnqvist A B

p Dewey al a eochannels

Prosna River vill e Latest Pleistocene braided channels

valley wall

late Holocene meander early T loops Holocene rinity meander loops River

0 1 km 0 1 km

Fig. 17. (A) Channel changes in plan-view interpreted for the Prosna River in Poland (after Schumm & Brakenridge, 1987; original from Kozarski & Rotnicki, 1977), with latest Pleistocene braided channels, followed by large early Holocene meandering channels, and small late Holocene meandering channels. (B) Contrast between palaeochannels (Deweyville allostratigraphic units) from the last glacial cycle (OIS 2±4) and the modern channel of Trinity River, Texas Gulf Coastal Plain (see Blum et al., 1995). rapidly subsiding continental margins during within single-channel systems of the Texas Gulf eustatic sea-level fall; relative sea-level fall will Coast (Fig. 17B; Blum et al., 1995). In theory, be greater along the slowly subsiding margin, changes in cross-sectional dimensions (width, hence vertical separation of allostratigraphic depth, width-to-depth ratio) can be recognized units will be signi®cantly greater. Second, during in both Quaternary and ancient settings, although eustatic sea-level rise, relative rates of rise will be in practice a number of complicating factors arise signi®cantly greater along the rapidly subsiding (see Ethridge & Schumm, 1978; Williams, 1987; margin, as will vertical separation of allostrati- Rotnicki, 1991; Wohl & Enzel, 1996). A well- graphic units. Indeed, in rapidly subsiding studied Holocene example is provided by Knox settings, key stratigraphic markers like palaeosols (1987, 1993), who cored a large number of may not form during eustatic sea-level fall due to palaeochannel ®lls from small streams in lack of separation between units, or during Wisconsin (USA) to establish changes in channel eustatic sea-level rise due to continuous deposi- cross-sectional area through time. A promising tion and burial. technique for the future may be the estimates of Morphological responses re¯ect changes in palaeochannel depth from detailed descriptions channel geometry, and occur within the context of the heights of cross-strata and strata sets. This provided by differentiation of allostratigraphic method is based on establishment of relationships units. Clear and unambiguous recognition of between thicknesses of preserved cross-strata and plan-view changes is limited toLate Quaternary the dimensions of parent bedforms (Best & Bridge, examples, where channel patterns are preserved 1992; Bridge, 1997; Leclair et al., 1997), coupled on terrace surfaces. Examples include braided to with empirical relations that relate dune height to sinuous single-channel transitions that occurred palaeo¯ow depth. along many European rivers (Fig. 17A; Starkel, As de®ned here, sedimentological responses 1991), and changes in meander dimensions can be viewed within the context of alluvial

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 27

ABCHINJE VILLAGE

2000

1800

1600

1400

1200

1000 Metres 800

600

400

200

0 0 0.2 0.4 0.6 0.8 1.0 Proportion of Thick 0 1 km VE = 25x Sandstone Bodies

Fig. 18. (A) Proportion of thick sandstone bodies in the Siwalik succession at Chinje Village, Pakistan, relative to mudstone-dominated intervals. The diagram illustrates changes in sandstone proportions through time at several possible levels of cyclicity (after Willis, 1993b). (B) Line drawings of palaeosol-bounded stratigraphic units in the Siwalik succession. Stippled pattern illustrates sandbodies, white indicates mudstones, vertical hatchures indicate palaeosols and arrows indicate palaeocurrent directions (after Willis & Behrensmeyer, 1994). architecture, initially described as the geometry, ¯oodplain facies, and quasi-cyclical vertical proportion and spatial distribution of different changes in mean grain-size and the proportions types of ¯uvial deposits (Leeder, 1978; Allen, of thick sandstone bodies relative to muddy 1978). Sedimentological responses re¯ect ¯oodplain facies (Fig. 18). changes in channel geometry and the spatial Palaeosols deserve special mention (Wright, distribution of depositional environments, and 1986; Retallack, 1986, 1990; Mack et al., 1993; also occur within the context provided by Birkeland, 1999), and a critical issue is the degree differentiation of allostratigraphic units and to which palaeosols represent allogenic or auto- bounding unconformities, or stratigraphic archi- genic controls. There is, for example, a long tecture as de®ned here. Sedimentological re- history to the use of palaeosols to subdivide and sponses can be de®ned in a number of ways, correlate ¯ights of Quaternary terraces and under- depending on the types of data available, and lying ®lls; a key underlying premise in studies of range from simple descriptions of textural this kind is the soil chronosequence and how soil changes todetailed descriptionand quanti®ca- pro®les are increasingly well developed with tion of changes in the proportions of different increasing durations of subaerial exposure (see lithofacies, facies associations and bounding Birkeland, 1999, for a recent summary). By con- surfaces. Outstanding detailed descriptions of trast, studies of Holocene soils as analogues for alluvial strata can be found in the series of papers pre-Quaternary strata (e.g. Aslan & Autin, 1998), or that discuss the Neogene Siwalik succession of pre-Quaternary palaeosols themselves (e.g. Allen, northern Pakistan (Willis, 1993a,b; Willis & 1974; Leeder, 1975; Bown & Kraus, 1987; Kraus, Behrensmeyer, 1994; Khan et al., 1997; Zaleha, 1987; Willis & Behrensmeyer, 1994; Behrensmeyer 1997). Among the many characteristics high- et al., 1995), have more commonly focused on how lighted are successions of palaeosol-bounded palaeosol characteristics are related to aggradation

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 28 M. D. Blum and T. E. ToÈrnqvist rates, distance to active channel belts or ¯oodplain Because of the numerous uncertainties, the hydrology. However, the advent of sequence most convincing studies interpret chronologically stratigraphy has drawn attention to the use of controlled alluvial stratigraphic frameworks with- palaeosols as key stratigraphic markers that trace in the context of independent records of climate from inter¯uves into adjacent palaeovalleys, and change. A number of reviews or regional synth- therefore serve as sequence-bounding unconfor- eses have been published, with most focusing on mities (e.g. Wright & Marriott, 1993; Gibling & the last 20 kyr and often concerning small Wightman, 1994; Aitken & Flint, 1996; Blum & systems. These include Knox's (1983) summary Price, 1998; McCarthy & Plint, 1998). Moreover, it of Holocene records from the USA, Bull's (1991) is important to realize that palaeosols with lesser documentation of Quaternary histories for several degrees of development, or perhaps different ¯uvial systems in the western USA & New characteristics, may serve as boundaries to smal- Zealand, Starkel's (1991) summary of late ler-scale allostratigraphic units (e.g. Blum & Price, Pleistocene and Holocene changes in European 1998; see Fig. 16B). Indeed, most important from a rivers, Bravard's (1992) summary of late stratigraphic point of view may be differences in Pleistocene and Holocene data from France the duration of periods of nondeposition and the (especially the Rhone), Macklin & Lewin's degree of development of pedogenic characteris- (1993) discussion of Holocene records in tics, which may provide an index of the time- England, Schirmer's (1995) discussion of central scales over which different allostratigraphic units European (mostly German) rivers during the last form. Clearly, palaeosols deserve close scrutiny in 20 kyr and Knox's (1996) overview of global the future. records from the last 20 kyr. There is no reason toprovidean additional review here; instead, two case studies are used to demonstrate contrasting styles of ¯uvial responses to climate change. FLUVIAL RESPONSE TO CLIMATE CHANGE: EXAMPLES Loire River, France Recognition of ¯uvial response to climate change The Loire is one of the larger rivers in Europe, in continental interiors often becomes a question with a drainage area of ~ 120 000 km2. The Loire is of spatial and temporal scale, the degree of typical of many mid-latitude continental interior temporal resolution possible given existing dating ¯uvial systems in that it has a ¯ight of downward- methods and predilection of the investigator for stepping terraces and underlying ®lls that demar- autogenic (everything is `complex response') vs. cate long-term valley incision punctuated by allogenic (everything is climate change) controls. relatively short periods of lateral migration and There are few `smoking guns' for any particular minor aggradation. model, and within the Quaternary community, Recent work by Straf®n et al. (2000) in the standards for correlation of changes in alluvial upper Loire recognized ®ve allostratigraphic successions increase with decreasing age, such units from the last glacial period (oxygen-isotope that debate can ensue regarding discrepencies of stages 2±4; hereafter OIS 2±4) through the 101±102 years as the age of deposits approach the Holocene (OIS-1; Fig. 19). Terrace surfaces with period of historical monitoring. On the one hand braided channel traces occur at elevations of 17 it seems unlikely that autogenic mechanisms can and 12 m above the modern low-water channel result in mappable stratigraphic units that can be and de®ne the upper boundaries to two glacial- traced over signi®cant distances within, or corre- age allostratigraphic units (T6 and T5). Under- lated between, river systems, whereas on the lying ®lls are 6±10 m in thickness, and interpreted other hand there is every reason to assume that to represent high concentrations of coarse sand ¯uvial systems are sensitive toclimatically con- and gravel along valley axes under periglacial trolled changes in discharge and sediment load. climatic conditions, and moderate-magnitude However, there is also every reason to assume ¯oods produced by snowmelt runoff. Inter-vening that: (a) ¯uvial response to global climate change periods of valley incision represent diminished will vary spatially such that different regions may sediment supply due toincreased slopestability, be out of phase; and (b) the frequency of and/or perhaps increased ¯ood magnitudes adjustments toglobalclimate change will vary and sediment transport capacity associated with between river systems due todifferent thresholds moderately warm periods. The last signi®cant for change, i.e. differential sensitivity. period of valley incision corresponds to the latest

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 29

T2 T6

T3 T5 280 T2 Loire R. T8 275 fossil T4 frost wedge casts & involutions 270 T2 265 T5 T7 260 T3 T4 T5 255 ? 250 T6 ? 0 1 km 245 Metres above sea level T5 ? 240 T4 Pre-Quaternary T3 deposits T2 235 modern ? 230

225 Surface Soil B l res C

Spoil OF P OF PB

Fig. 19. (A) Aerial photo and schematic geomorphic and stratigraphic cross-section from a portion of the Loire River, France, illustrating large late Pleistocene braided stream surfaces (T5) and Holocene surfaces with clear high-sinuosity single-channel traces (modi®ed from Straf®n et al., 2000). (B) Typical glacial period facies, with sets of trough cross-strata, but no overbank facies (unit T5 in (A)). (C) Typical late Holocene facies, illustrating overbank facies with palaeosols resting on point bar gravel (unit T2 in (A)). PB = point bar gravel, OF = overbank ®ne sand, P = weakly developed palaeosol.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 30 M. D. Blum and T. E. ToÈrnqvist

Pleistocene±Holocene transition, and Holocene to the scale, complexity and dynamism of this allostratigraphic units are part of a complex continental-scale ¯uvial system, as do the chan- valley ®ll, some 5±7 m in thickness, that was ging interpretations through time. deposited by a succession of single-channel Blum et al. (2000) used stratigraphic relations systems of varying sinuosity. Early to middle between chronologically controlled loess units Holocene units display high-sinuosity meander- and subjacent ¯uvial deposits to provide a model ing planforms, whereas late Holocene units have for LMV history (Fig. 20B) that differs from both lower sinuosity channels (Fig. 19A). Fisk (1944) and Saucier (1994). In this model: (a) Stratigraphic, morphological and sedimentolo- major braided stream surfaces are either late gical responses of the Loire to climate change Illinoian (OIS 6) or late Wisconsin age (OIS 2); during the last glacial cycle can be discussed at (b) there is little evidence for preserved ¯uvial twocontrastingscales. First, as is the case for deposits from the last interglacial period (OIS 5) many larger European rivers, transition from a through much of the valley; and (c) deposits from glacial tointerglacial climate produced funda- the early Wisconsin substage (OIS 4) are preserved mental transformations in channel geometry and in protected settings only. At a ®ner level of detail, depositional style. Glacial-age units record de- a difference in elevation of up to 5 m between three position by braided channels, and systematically successive late Wisconsin terraces indicates that lack ®ne sandy and muddy overbank strata large braided stream surfaces were rapidly con- (Fig. 19B), whereas Holocene units were deposited structed, incised and abandoned in response to by sinuous single channels, and contain appreci- high-frequency ¯uctuations in the delivery of able thicknesses of ®ne-grained overbank meltwaters and outwash from the former ice facies (Fig. 19C). Second, high-amplitude, low- margin and the drainage of large proglacial lakes. frequency adjustments characterized the rela- The tremendous volumes of meltwaters during tively long glacial period, with only two extended this time alsoresulted in the simultaneous periods of sediment storage punctuated by epi- occupation of two large-scale channel courses sodes of valley incision over a period of 50±60 kyr. (Western and Eastern Lowlands), as well as By contrast, low-amplitude, high-frequency ad- breaching of a new bedrock gap (Thebes Gap) justments have been typical of the relatively short through which the modern river ¯ows. Subsurface Holocene, with little if any net valley incision or data are not suf®cient to distinguish allostrati- net changes in sediment storage, and only minor graphic units; however, from surface mapping, adjustments in channel geometry. However, rela- three major units might be inferred to represent the tively frequent changes in climatically controlled last deglaciation (a period of 10±12 kyr) within the ¯ood regimes have resulted in alternating periods main valley and in over¯ow courses (Fig. 20C). of ¯oodplain stabilization with soil development Unlike the Loire, where ¯ights of terraces vs. high-magnitude ¯ooding and burial of palaeo- record long-term valley incision punctuated by sols by overbank facies. relatively short periods (104 years) of lateral Over longer time periods, records of ¯uvial migration and minor aggradation, the strati- activity in the Loire are strongly biased in favour graphic, morphological and sedimentological of the long glacial periods, whereas short inter- record of the low-gradient LMV at any point in glacial periods like the Holocene would seem to time will always be biased in favour of the most have a low potential to be recorded in the recent period of deglaciation. Maximum ¯ood- landscape. This type of situation may be typical plain elevations of only 100 m can be found at of many continental interior river systems in the distances of 1000 km upstream from the present unglaciated mid-latitudes and subtropics. highstand shoreline, so there is little opportunity for long-term progressive downcutting and development of terrace ¯ights, and strata from the Mississippi River, USA penultimate glacial period (OIS 6) are only The Mississippi River served as the principal preserved due to the wholesale abandonment of conduit for meltwater and sediment discharging that channel course in favour of the present from the southern margins of Pleistocene ice combined Mississippi±Ohio valley. Within the sheets, and the Lower Mississippi Valley (LMV) context of these longer-term controls, large ¯uc- provides an example of the response of a large, tuations in discharge and sediment loads deliv- long-lived, low-gradient ¯uvial system to the ered from the ice margin during deglaciation growth and decay of ice sheets. Indeed, land- serve to ensure that relatively short periods of forms and deposits of the LMV (Fig. 20A) testify time (~ 10 kyr) at the end of successive 100-kyr

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 31 glacial cycles (late OIS 6 and late OIS 2) dominate that records of this kind must be more wide- the landscape, and deposits from earlier parts of a spread than presently interpreted, but reasonable glacial cycle or previous interglacial periods are con®dence in such interpretations is hampered not widely preserved. by a number of factors. Among these, the temporal resolution provided by existing dating techniques is the most obvious and most impor- Pre-Quaternary records tant limitation. High-resolution palaeomagnetic As discussed above, several examples of pre- chronologies, for example those of the Neogene Quaternary ¯uvial response to climate change Siwalik succession (Khan et al., 1997), permit have been described in the literature. It is clear calculation of average time periods represented

MO IL Palaeo-Ohio A B 0 25 km Valley Thebes Map Area N Gap Eastern Cape 0 1000 km Glacial Limit Lowlands Girardeau Western entrance Lowlands Ohio R. entrance Cairo BT

Pt-2 Ridge y Pt-1 Poplar Bluff Pt-3 River Pt-1 KY TN Pt-4 x MO AK N Pt-2 Mississippi

Crowleys C Western Eastern X Lowlands Lowlands Y Ozark Pt-4 Crowleys Plateau Ridge Uplands

Pt-2Pt-2 Pt-1 Pt-3 Pt-1

Holocene Holocene floodplain floodplain Holocene Pvl1 channel belt 10's of m Older Pleistocene or Pre-Quaternary Bedrock

Fig. 20. (A) Landsat image of northern part of the Lower Mississippi alluvial valley (USA), with braided late Pleistocene vs. sinuous meandering Holocene channel belts. (B) Sur®cial geological map of the area shown in (A). (C) Schematic cross-section of inferred stratigraphic relationships along x±y in (A). Pt-4 = loess-covered terrace and underlying glacio-¯uvial deposits from the OIS 6 glacial period, Pt-3 = terrace and underlying glacio-¯uvial deposits from the OIS 2 full-glacial period, Pt-2 and Pt-1 = braided stream surfaces and underlying deposits from the OIS 2 late-glacial period. Thicknesses of deposits are schematic only, but data in Saucier indicates 50±60 m is not uncommon. Holocene ¯oodplain in Western Lowlands represents deposits of smaller streams that head on the Ozark Plateau tothe west. After Blum et al. (2000).

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 32 M. D. Blum and T. E. ToÈrnqvist by speci®c unconformity-bounded intervals of abandonment with formation of palaeosols, as deposition that are well within the Milankovitch well as changes in channel morphology and scale, and would seem to offer the best chance to depositional style that occur across unconformi- evaluate ¯uvial responses to climate change. ties (especially when de®ned in part by palaeo- Within relatively well-constrained successions sols), are excellent candidates for interpreting such as these, periods of incision and ¯oodplain ¯uvial responses to climate change.

Bedrock Valley Walls or Older Pleistocene Bedrock Valley Walls or A Colorado River Older Pleistocene

10 m ELA CBA-1 CBA-2 0 0 2 km Scale CBA-3

Beaumont Colorado Beaumont B Surface Caney Creek Channel Surface

Floodplain Floodplain CBA-3 CBA-1 CBA-2 ELA

104° 96°

36° Austin C 160 Col 32° Dra or A in ad a o B 140 ge Bastrop

28° Fig. 22

Smithville Gulf of 120 Mexico 0 150 km

100 La Grange

80 Columbus CBA-3 Eagle Lake 60 CBA-2 Elevation above MSL (m) above MSL Elevation ELA Wharton 40 upstream limits of onlap and sea-level control on stratigraphic architecture 20 250 225 200175 150 125 100 75 50 25 Distance upstream from highstand shoreline (km)

Fig. 21. Schematic cross-sections of the late Pleistocene and Holocene valley ®ll of the Colorado River, Texas Coastal Plain, illustrating downstream persistence of allostratigraphic units with downstream changes in stacking patterns. (A) The bedrock-con®ned degradational valley near La Grange, Texas, and (B) the lower valley near Wharton. Beaumont surface is the last interglacial highstand surface (see Fig. 21A), ELA = Eagle Lake Alloformation of late Pleistocene full-glacial age, whereas CBA-1±3 = Columbus Bend Alloformation Members 1±3, of latest Pleistocene and Holocene age. Caney Creek was the active channel until 200±300 years ago, when the lower 90 km avulsed to its present position. (C) Long pro®les for ELA, CBA-2 and CBA-3, illustrating crossover point, and upstream limits of onlap. Inset map shows locations for (A) and (B), with long pro®le in (C) extending between these two locations. Modi®ed from Blum & Price (1998).

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 33

FLUVIAL RESPONSES TO CLIMATE overbank facies spread laterally and bury soils AND SEA-LEVEL CHANGE: that developed on the downdip margins of the EXAMPLES previous highstand alluvial plain. Complete valley ®lling promotes avulsion, with relocation At some point upstream, ¯uvial systems of valley axes before the next sea-level fall, such become completely independent of high-fre- that successive 100-kyr palaeovalley ®lls have a quency (< 106 years) sea-level change. How- distributary pattern (Fig. 22A). Individual palaeo- ever, the reverse is not true, and there is no valleys range from 15 to 20 m in thickness at the point moving in the downstream direction updip limits of sea-level in¯uence, to > 35 m at where the in¯uence of climate and climatically the present highstand shoreline, whereas they controlled changes in discharge and sediment extend 15±40 km along strike. supply can go to zero ± if it does not rain, Within individual unconformity-bounded nothing happens. Moreover, since glacio- palaeovalley sequences, alluvial architecture re- eustasy is a function of climate change, it ¯ects interactions between changes in climate seems likely that some sort of autocorrelation and sea-level (Fig. 22B). During the glacial period, will arise between the effects of climate vs. sea- regional climate was more temperate than today level change. It follows from the above that and the shoreline was in a midshelf position or continental margin ¯uvial systems must in further basinward. Allostratigraphic units from some way re¯ect the in¯uence of both control- this time period consist of amalgamated channel- ling mechanisms, although in some cases the belt sands with few overbank facies, and are in¯uences of sea-level change and subsidence bounded by well-drained palaeosols; these units may overwhelm the recognizable signature of can be 10±15 m in thickness, and extend 2±5 km other factors. Nevertheless, Quaternary systems along strike. Transition to the interglacial period in continental margin settings provide unique resulted in increases in sea surface size and opportunities to isolate the nature of these temperature, `subtropical' land surface tempera- interactions, as discussed in the two case tures, and deeper inland penetration of tropical studies below. moisture, that in turn produced increases in the depth and ¯ashiness of ¯oods. Superimposed on this overall trend were higher-frequency climate Colorado River, Texas changes as well (Toomey et al., 1993). As a result, From studies of late Pleistocene and Holocene multiple episodes of aggradation, lateral migra- strata of the Colorado River, Gulf Coastal Plain of tion and some incision with sediment bypass Texas (USA), Blum (1993) de®ned interactions occurred contemporaneously with the late stages between climate and sea-level change in terms of: of transgression and sea-level highstand, but in (a) the downstream continuity of allostratigraphic response to climatic controls on discharge re- units and component facies as a re¯ection of gimes and sediment supply. Even in upstream climatic controls; and (b) the corresponding reaches, beyond any in¯uence of sea-level, downstream changes in stratigraphic architecture Holocene units contain an abundance of ®ne- as a re¯ection of the increasing in¯uence of sea- grained overbank facies (as was the case for the level change (Fig. 21). Subsequent work con®rms Loire as well), and sand bodies are less amalga- these relationships for palaeovalleys that span the mated (Blum & Valastro, 1994; Blum et al., 1994). last 300±400 kyr (Blum & Price, 1998). Individual These characteristics are enhanced within the palaeovalleys correspond to 100-kyr glacial± lower valley by a forced shortening of channels interglacial cycles, and are partitioned by low- and ¯attening of gradients. ering of sea-level below highstand depositional Aslan & Blum (1999) alsoshowthat in far shoreline breaks, when channels incise and downstream reaches, rates of valley ®lling and valley axes become ®xed in place as they extend style of avulsion during sea-level rise and high- across the subaerially exposed shelf. During stand play a strong role in alluvial architecture. falling stage and lowstand, multiple episodes of During early transgression, when the shoreline is lateral migration, minor aggradation and degrada- still some distance basinward, avulsion occurs by tion occur within the extended valleys in reoccupation of abandoned falling stage and response to climatic controls on discharge and lowstand channels, with erosion and reworking sediment supply. Transgression and highstand of older channel-belt sands (Fig. 23A). By con- then results in channel shortening, and valleys trast, accumulation space is rapidly generated as begin to®ll; as valley ®lling nears completion, the shoreline moves closer, rates of valley ®lling

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 34 M. D. Blum and T. E. ToÈrnqvist increase and avulsion occurs by repeated diver- individual avulsions are separated by massive to sion into ¯oodplain depressions. This produces slickensided muds or buried soil horizons that successions of massive to laminated ¯oodbasin represent periods of slow sediment accumulation muds that encase thin (< 5 m) ribbon-like crevasse or ¯oodplain stability between episodes of avul- channels and thin (< 2±3 m) sheet-like splay sive deposition. As ®lling of accumulation space sands (Fig. 23B), and comprise up to 50% of the nears completion, avulsion by channel reoccupa- total valley ®ll (comparable to the `avulsion tion again becomes the dominant process, and deposits' of Smith et al., 1989). Deposits of again results in amalgamated channel belts

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 35

Active Avulsion channel node (Fig. 23C). During the most recent avulsion, the Colorado River abandoned its valley from the last A glacial cycle, and reoccupied a Pleistocene channel belt from the previous interglacial highstand (see Fig. 22A).

Amalgamated channelbelt Rhine±Meuse Rivers, The Netherlands sand bodies falling stage and lowstand Studies of the Rhine±Meuse Rivers during the palaeochannels last glacial cycle alsoclearly document interactions between upstream and downstream B Floodbasin controls. During much of the last glacial (OIS muds 2±4) the Rhine mouth was located beyond the Strait of Dover, up to 800 km downstream from

Alluvialridge the present shoreline (Gibbard, 1995; Bridgland Crevasse & D'olier, 1995), and direct sea-level in¯uence sheet sands was limited toreaches that are nowsubmerged "Ribbon" in the English Channel. Throughout the glacial sands cycle as a whole, periglacial conditions with Active sparse vegetation and continuous permafrost alluvial alternated with a temperate climate comparable ridge Avulsion C path topresent-day, and it is generally assumed that sediment supply tothe Rhine and Meuse was considerably higher when periglacial conditions Abandoned prevailed (e.g. Vandenberghe, 1995). Recent alluvial ridge work by ToÈrnqvist et al. (2000) suggests that Floodbasin muds Isolated channel during the sea-level fall associated with OIS 4, belt sands substantial deposition occurred near the present Dutch shoreline despite net incision, Fig. 23. Model for changes in avulsion styles during similar to the Colorado River, as described valley ®lling in response to sea-level rise. (A) above. The Rhine and Meuse were also pre- Avulsion by reoccupation of falling stage and dominantly aggrading braided systems during lowstand channels during early stages of valley the Last Glacial Maximum sea-level lowstand ®lling, when rates of ¯oodplain aggradation are very (OIS 2, 25±13 ka; Kasse et al., 1995), and low. (B) Avulsion by frequent crevassing into climate change continued to play a paramount ¯oodplain depression in response to rapid increases role during deglaciation as sea-level rose rain accumulation space that correspond to high rates of sea-level rise. (C) Avulsion by reoccupation pidly. As examples, during the relatively warm of older channel belts, when sea-level highstand Bùlling±Allerùd interstadial (the last D-O has been reached, and accumulation space is nearly Cycle), which corresponds to extreme rates of ®lled (after Aslan & Blum, 1999). eustatic sea-level rise (meltwater pulse 1 of

Fig. 22. (A) Perspective map view and schematic cross-section of Pleistocene (Beaumont Formation) and Holocene alluvial plains of the Colorado River, Texas, illustrating large-scale stratigraphic architecture dominated by multiple cross-cutting incised valley ®lls. Individual valley ®lls are interpreted to represent 100 kyr glacial± interglacial cycles based on physical stratigraphic relations, mostly downdip onlap and burial of successively older alluvial plain palaeosols, especially the onlap of ¯oodplain facies and palaeosols from one highstand by ¯oodplain facies and palaeosols from successively younger highstands. A series of thermoluminescence ages support these age relations. The Holocene alluvial plain surface was recently abandoned, when the channel avulsed to reoccupy an older Pleistocene meanderbelt. Location shown in inset map of Fig. 21. After Blum & Price (1998). (B) Interpreted post-Beaumont valley ®ll cross-section along line A±A¢, based on a series of continuous cores with 2 km spacing. Lowstand channelbelt sandbody equivalent to ELA in Fig. 20, whereas strata that overlie well-drained palaeosols are equivalent to CBA (individual members not differentiated). After Aslan & Blum (1999).

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 36 M. D. Blum and T. E. ToÈrnqvist W E 0 Lobith 00 +14 1 1ka 22 ka +12 33 ka +10 Present-day 44 ka +8 floodplain surface 5 +6 5 ka +4 Rotterdam Age (ka) +2 6 6 5 4 3 2 1 6 ka MSL -2 The -4 Netherlands 7 ka -6 7

Altitude (m) -8 Ribbon-like Rotterdam N -10 channel belts Lobith -12 Last Glacial Maximum Rhine -14 floodplain surface -16 8ka 8 ka 0 30 km Meuse 0 50 km -18

Fig. 24. Holocene rise of the (ground)water table in the Rhine±Meuse Delta as a function of relative sea-level rise (relative sea-level curve indicated tothe left; mainly after Van de Plassche, 1982), primarily based on~ 100 14C ages of basal peats (Van Dijk et al., 1991). Spacing of the (ground)water gradient lines (isochrons with ages) de®ne rates of creation of accumulation space. Ribbon-like channel belts encased in thick, organic-rich overbank successions dominate downdip areas, whereas sheet-like channel belts are typical of conditions with lower rates of increase in accumulation space. Note the intersection point between the Last Glacial Maximum and modern longitudinal pro®les to the far east, representing the landward limit of onlap.

Fairbanks, 1989), as well as during the earliest encased in lacustrine muds and organics (Van part of the Holocene (meltwater pulse 2), the der Woude, 1984), and avulsion frequency was Rhine±Meuse incised in response to reductions very high (ToÈrnqvist, 1994). By contrast, mean- of sediment supply due to re-establishment of dering single-channel distributaries with sheet- dense vegetation (Pons, 1957; Kasse et al., 1995; like channel belts (width-to-thickness ratios Berendsen et al., 1995; ToÈrnqvist, 1998). > 50) and more homogeneous overbank deposits Well-documented rates of rise in (ground)- have dominated the late Holocene sea-level water tables in the present Rhine±Meuse delta highstand, and avulsion frequency was signi®- during the middle to late Holocene are inter- cantly less. There are striking similarities preted as a proxy for sea-level rise (Van Dijk between the middle to late Holocene Rhine± et al., 1991; Fig. 24), which is in turn inter- Meuse, the Texas Gulf Coast examples de- preted to have played a prominent role in the scribed above and recent re-evaluations of the development of alluvial architecture. The data- middle to late Holocene of the Lower base for interpretation of the Holocene record Mississippi Valley (Aslan & Autin, 1999). includes > 200 000 boreholes (an average of 50± 100 boreholes km±2) plus some 1000 radio- Pre-Quaternary records carbon ages (e.g. Berendsen, 1982; ToÈrnqvist, 1993; Berendsen et al., 1994), and permits How can interactions between climate and sea- unparalleled documentation of changes in level change be recognized in the pre- alluvial architecture through time. For the Quaternary record? The bad news is that middle Holocene, prior to 4 ka, rates of sea- interactions de®ned by the downstream con- level controlled (ground)water-table rise were tinuity of allostratigraphic units and component very high (typically > 1´5 mm yr±1). The record facies, with corresponding downstream changes from this time consists of a succession of in stratigraphic architecture, may be dif®cult to anastomosed (ToÈrnqvist, 1993) or straight identify. For example, the intersection between (Makaske, 1998) distributaries with ribbon-like lowstand and highstand ¯oodplain surfaces channel belts (width-to-thickness ratios < 15) de®ned along both the Texas Coast and in

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 37

The Netherlands will not be preserved, which SYNTHESIS AND A LOOK FORWARD may be typical of most passive margin settings where subsidence rates decrease landward. The discussion above has addressed the history of Indeed, as a general rule, stratigraphic units research on ¯uvial responses to climate and sea- deposited during falling stage to lowstand level change, outlined a basic process framework should occur deeper within palaeovalley ®lls, for investigations of this kind, and provided and stand the best chance of preservation, examples of ¯uvial response to climate and sea- whereas strata deposited during highstand will level change from work on Quaternary systems. more commonly be removed during subsequent The Mississippi River, which has long played an sea-level falls (cf. ToÈrnqvist, 1995; Blum & important role in the ¯uvial community, can be Price, 1998; ToÈrnqvist et al., 2000). However, used toillustrate a key concept that may play a the good news is there is no reason to assume role in the future. In a previous summary of that interactions between climate and sea-level Mississippi River response to glaciation and change will not occur, and downstream con- deglaciation, Wright (1987) noted that an idea- tinuity of strata should be the rule, not the lized model would predict some crossover point exception. Moreover, future interpretations of where aggradation due to glacial sediment load- these interactions will at least in part rely on ing gives way downstream to incision due to the well-tested method of developing modern glacio-eustatic lowstand. Wright recognized that analogues, only at `sequence' time-scales (see this crossover is more conceptual than real; the discussion below). synthesis here takes this a step further and In this context, it is worthy of note that models suggests this crossover is best viewed as a developed from the pre-Quaternary record (e.g. metaphor for how workers in different settings, Shanley & McCabe, 1991, 1993, 1994; Wright & Quaternary and ancient, have focused on causal Marriott, 1993; Olsen et al., 1995; Currie, 1997; mechanisms that reside in their own backyard. A Zhang et al., 1997) infer relationships between bright future for studies of ¯uvial response to alluvial architecture and sea-level change that are climate and sea-level change can be envisaged for very similar towhat is observed in Late the new millennium, one that will become Quaternary successions. Although details differ increasingly integrative with respect toboth (compare Shanley & McCabe, 1993; Wright & upstream and downstream controls, as well as Marriott, 1993), a common pattern would be how they operate in tandem to produce the amalgamated high net-to-gross channel belts that landscape and stratigraphic record. This review rest on what is interpreted to be a sequence concludes by suggesting some key issues for the boundary, and overlain by isolated ribbon-like future. sand bodies that are encased in laterally extensive mudstone-dominated successions. Many workers Quaternary systems have interpreted the sequence boundary to re¯ect incision and sediment bypass during sea-level The ®rst 2±3 decades of this millennium should fall, and the amalgamated character of overlying be an exciting time tostudy Quaternary ¯uvial sand bodies re¯ects low accommodation, whereas systems, with challenges that await at both ends the more isolated sand bodies upsection re¯ect of the temporal scale. On the one hand, it will be increasing accommodation due to relative sea- important to keep pace with research in high- level rise and/or highstand. For the Late frequency climate and sea-level change. For Quaternary examples discussed above, the example, how do ¯uvial systems respond to the basal palaeovalley unconformity would be inter- Dansgaard±Oeschger events that are soprominent preted as a composite erosion surface, cut in records from Greenland and the North during multiple periods of incision alternating Atlantic? Holocene records clearly show that with lateral migration and minor aggradation river systems respond in dramatic fashion to during falling stage and lowstand. Moreover, relatively subtle climate changes, but are there the contrast between laterally amalgamated and upper limits tothe frequency ofresponsesthat more isolated sand bodies that inter®nger with can be interpreted from the late Pleistocene and mud-rich successions re¯ect fundamentally dif- Holocene stratigraphic record, and were river ferent glacial vs. interglacial climate regimes, systems during the glacial period as sensitive to respectively, upon which are superimposed sea- rapid climate changes as their interglacial coun- level controls on overall palaeovalley-®ll archi- terparts? To date, much of the high-pro®le, well- tecture. funded work has focused on documentation of

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 38 M. D. Blum and T. E. ToÈrnqvist climate changes per se. However, humans live on, GPR has, for example, been applied in a number and interact with, the Earth's surface, which has of ¯uvial contexts, but mostly at the alluvial received considerably less attention and funding. architecture scale, and more speci®cally with Understanding the spatial dimensions and reference tointernal structure at the channel bar signi®cance of high-frequency climate and sea- scale (e.g. Huggenberger, 1993; Bridge et al., 1995, level changes, and their effects on surface pro- 1998). Future applications of high-resolution cesses, will require more precise, systematically geophysical techniques in Late Quaternary con- de®ned, whole-system (from source to sink) and/ tinental strata are not without problems, espe- or regional stratigraphic and geochronological cially at the stratigraphic architecture scale, but frameworks than are now widely available. may prove to be essential for developing linkages On the other hand, it is equally important to between continental strata and the high-resolu- study longer-wavelength cycles of ¯uvial land- tion records that have been, and will continue to scape evolution. It is fair to say that most detailed be, documented for alluvial±deltaic strata on the research on ¯uvial response to climate change now submerged shelves. This, in turn, must play during the last 3±4 decades has focused on the a critical role in developing an understanding of last 20 kyr. While important, this relatively short the relative importance of upstream vs. down- period encompasses only the tail end of a single stream controls at both the stratigraphic and 100-kyr glacial tointerglacial cycle. Very little is alluvial architecture scales, and the evolution of actually known about ¯uvial responses to com- ¯uvial systems from source to sink. plete glacial±interglacial cycles, or how the ®rst Second, and equally important, is a corre- 70±80% of the last cycle have preconditioned sponding need to develop, re®ne and apply new ¯uvial processes during the last 20±30% of the dating techniques. Radiocarbon has served as the cycle. A number of studies have begun to address clear reference standard for almost ®ve decades, ¯uvial responses to climate change at and has had a profound impact on understanding Milankovitch scales. Antoine (1994), for example, the dynamics of terrestrial systems over the last uses stratigraphic relationships between loess 20±30 kyr. However, many records remain poorly sheets and ¯uvial deposits in the Somme Valley constrained because they lack organic matter in of France to develop a relative-age geochronolo- key stratigraphic contexts, and the time period gical framework, and then infers cycles of covered by radiocarbon represents only 30±40% aggradation, degradation and terrace formation of a single glacial±interglacial cycle. On the near at the 100-kyr scale. Bridgland (1994) alsoinfers horizon, application and re®nement of the var- 100-kyr cyclicity for terraces of the Thames in ious luminescence techniques seem to offer the England. Moreover, terraces of the Meuse in the most promise. In theory, luminescence techni- southern Netherlands span the entire Quaternary, ques measure the time elapsed since sand- and and have been used to develop numerical models silt-sized grains were exposed to solar radiation of long-term ¯uvial response to climate change during transport, and can provide numerical ages and tectonic uplift, suggesting 100-kyr cycles of for deposits within the middle to late Pleistocene aggradation and incision (Veldkamp & Van den and Holocene. For ¯uvial deposits, optically Berg, 1993). Investigations such as these also stimulated luminescence (OSL) has considerable provide an opportunity to link studies of ¯uvial advantages over the older technique of thermo- response to climate and sea-level change over luminescence (TL), since only minutes of expo- longer time-scales with new ideas on relation- sure are required during transport (Duller, 1996; ships between tectonics and surface processes Aitken, 1998). Both TL and OSL have been (e.g. Burbank & Pinter, 1999). used in ¯uvial contexts; examples include Twoissues emerge as fundamental tofuture TL dating of ¯uvial deposits in the Lake Eyre advancements in understanding the responses of basin of Australia (Nanson et al., 1988, 1992), Quaternary ¯uvial systems toclimate and sea- Murrumbidgee River palaeochannels in south- level change over both shorter and longer time- eastern Australia (Page et al., 1996), Saharan wadi scales. First, there is a clear need todevelopand systems (Blum et al., 1998) and Texas Gulf Coast apply more sophisticated geophysical techniques ¯uvial deposits (Blum & Price, 1998), as well as to explore three-dimensional relationships at OSL dating of ¯uvial deposits of the Ebro both the stratigraphic and alluvial architecture drainage in Spain (Fuller et al., 1998), the Loire scales. Applications of ground-penetrating radar in France (Straf®n et al., 2000) and the Rhine± (GPR) and shallow high-resolution seismic will Meuse in The Netherlands (ToÈrnqvist et al., become increasingly important in this regard. 2000).

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 39

Pre-Quaternary systems present views on the magnitude, frequency and signi®cance of climate change. Twofundamental issues can be identi®ed that should play a role in interpretations of ¯uvial Importance of modelling responses to high-frequency (time-scales of 104± 106 years) climate and sea-level change in the pre- Although modelling efforts are discussed else- Quaternary record. First, future interpretations where in this issue (Paola, 2000), it seems clear must at least in part rely on the development of that integration of different `¯uvial' models will `sequence-scale' Quaternary analogues. There is, be a major challenge for the near future (cf. at present, a paucity of good analogues that: (a) Nystuen, 1998). As appropriately stated by Paola extend beyond the relatively short latest in this issue, presently available numerical Pleistocene and Holocene period into geologi- models should be considered temporary building cally resolvable time-scales (multiples of 104± blocks that will ultimately merge together. Four 106 years); (b) include depositional basin settings broad classes of models can be distinguished: (a) where preservational processes operate; and (c) the predominantly autogenic and exclusively describe ¯uvial responses to allogenic controls in aggradational numerical models of alluvial archi- sedimentological terms that facilitate recognition tecture (e.g. Bridge & Leeder, 1979; Mackey & in the ancient record. Further detailed study of Bridge, 1995; Heller & Paola, 1996); (b) sequence- Quaternary analogues may also provide a me- stratigraphic models that focus on incision and/or chanism by which cross-fertilization of methods aggradation in downdip parts of the ¯uvial and perspectives from geomorphology and sedi- system (e.g. Nummedal et al., 1993; Leeder & mentology can occur, and traditional disciplinary Stewart, 1996); (c) numerical models that focus and other barriers can be dismantled. primarily on long-term ¯uvial incision and/or Second, and perhaps more dif®cult, will be sediment production in the hinterland changes in the philosophy of interpretation. It (Veldkamp & Van den Berg, 1993; Leeder et al., seems clear that the development of sequence 1998); and (d) the unscaled and scaled physical stratigraphy resulted in a paradigm shift with (analogue) models that have provided important respect to how sedimentologists view the re- qualitative insight into¯uvial responsestovary- sponses of ¯uvial systems to sea-level change. ing rates and directions of sea-level change (Wood However, the same intellectual hurdle has yet to et al., 1993, 1994; Koss et al., 1994; Paola, 2000), be crossed with respect to the signi®cance of or how alluvial architecture varies with aggrada- climate change, which both causes and covaries tion rate (e.g. Ashworth et al., 1999). with sea-level, and is one of the dominant In the future, linkage of models that focus on controls on changes in sediment delivery to both continental interiors and continental mar- depositional basins through time. Clear recogni- gins (e.g. Veldkamp & Van Dijke, 1998), as well as tion and interpretation of ¯uvial responses to integrating model results with real-world data climate change certainly remains dif®cult, and for (e.g. Leeder et al., 1996), must become increas- the time being may require well-constrained ingly important so as to more clearly understand geochronological frameworks that demonstrate relations between sediment production, transport rates of change comparable to known climate and deposition at time-scales that reside at the forcing functions, coupled with an approach interface between the traditional interests of where possible alternatives are eliminated and geomorphologists/Quaternary geologists and se- the only one that remains must be the most dimentary geologists (103±106 years). As noted by viable. However, part of the hurdle will involve a Paola in this issue, model development can willingness toexplicitly consider climate change proceed at higher rates than ®eld studies; hence as a forcing mechanism; for example, many the ®rst decades of this millennium offer sig- features described in ancient successions are ni®cant opportunities for new and creative ®eld- remarkably similar in scale and kind tothat based investigations that are designed to con- observed in Late Quaternary records, but they are strain and validate quantitative models. commonly interpreted in terms of autogenic variation superimposed on tectonic controls. No analogues clearly demonstrate, as opposed to ACKNOWLEDGEMENTS assume, this particular model, and the assump- tion of autogenic variation may not be as justi®ed We appreciate the invitation from the today as it might have been in the past given Sedimentology editorial board to write this paper.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 40 M. D. Blum and T. E. ToÈrnqvist

Comments by Peter Friend, Mike Leeder and Autin, W.J. (1992) Use of alloformations for de®nition of editor Jim Best were very helpful. We also Holocene meanderbelts in the middle Amite River, south- appreciate numerous discussions with our var- eastern Louisiana. Geol. Soc. Am. Bull., 104, 233±241. Bagnold, R.A. (1973) The nature of saltation and bedload ious collaborators, students and colleagues, but transport in water. Proc. Royal Society, London, 322A, 473± stress that this is a selective review that re¯ects 504. our own perspectives on the subject matter. Baker, V.R., Kochel, R.C. and Patton, P.C. (Eds). (1988) Flood Geomorphology. John Wiley, New York. Barrell, J. (1917) Rhythms and the measurement of geological time. Geol. Soc. Am. Bull., 28, 745±904. REFERENCES Barron, E.J. and Moore, G.T. (1994) Climate Model Application in Palaeoenvironmental Analysis. SEPM Short Course no. 33. Aitken, J.F. and Flint, S.S. (1995) The application of high- Begin, Z.B. (1988) Application of diffusion-erosion model to resolution sequence stratigraphy to ¯uvial systems: a case alluvial channels which degrade due tobase-level lowering. study from the Upper Carboniferous Breathitt Group, east- Earth Surf. Proc. Landf., 13, 487±500. ern Kentucky, USA. Sedimentology, 42, 3±30. Begin, Z.B., Meyer, D.F. and Schumm, S.A. (1981) Aitken, J.F. and Flint, S.S. (1996) Variable expressions of Development of longitudinal pro®les of alluvial channels inter¯uvial sequence boundaries in the Breathitt Group in response to base-level lowering. Earth Surf. Proc. Landf., (Pennsylvanian), eastern Kentucky, USA. In: High Resolution Sequence Stratigraphy: Innovations and 6, 49±68. Applications (Ed. by J. A. Howell and J. F. Aitken), Spec. Behl, R.J. and Kennett, J.P. (1996) Brief interstadial events in Publ. Geol. Soc., 104, 193±206. the Santa Barbara basin, NE Paci®c, during the past 60 kyr. Aitken, M.J. (1998) An Introduction to Optical Dating. Oxford Nature, 379, 243±246. University Press, Oxford. Behrensmeyer, A.K., Willis, B.J. and Quade, J. (1995) Allen, J.R.L. (1974) Studies in ¯uviatile sedimentation: Floodplains and palaeosols of Pakistan Neogene and implications of pedogenic carbonate units, Lower Old Red Wyoming Palaeogene deposits: a comparative study. Sandstone, Anglo-Welsh outcrop. Geol. J., 9, 181±208. Palaeogeogr. Palaeoclimatol. Palaeoecol., 115, 37±60. Allen, J.R.L. (1978) Studies in ¯uviatile sedimentation: an Benito, G., Baker, V.R. and Gregory, K.J. (Eds). (1998) exploratory quantitative model for the architecture of Palaeohydrology and Environmental Change. John Wiley, avulsion-controlled suites. Sedim. Geol., 21, 129±147. Chichester. Amorosi, A., Colalongo, M.L., Fusco, F., Pasini, G. and Berendsen, H., Hoek, W. and Schorn, E. (1995) Late Fiorini, F. (1999) Glacio-eustatic control of continental- Weichselian and Holocene river channel changes of the shallow marine cyclicity from Late Quaternary deposits rivers Rhine and Meuse in the Netherlands (Land van Maas of the southeastern Po Plain, northern Italy. Quat. Res., 52, en Waal). PalaÃoklimaforschung, 14, 151±171. 1±13. Berendsen, H.J.A. (1982) De genese van het landschap in het Anderson, J.B., Abdulah, K., Sarzalejo, S., Siringin, F. and zuiden van de provincie Utrecht, een fysisch-geogra®sche Thomas, M.A. (1996) Late Quaternary sedimentation and studie. Utr. Geogr. Studies, 25, 1±255. high-resolution sequence stratigraphy of the East Texas Berendsen, H.J.A., Faessen, E.L.J.H. and Kempen, H.F.J. Shelf. In: Geology of Siliciclastic Shelf Seas (Ed. by M. De (1994) Zand in Banen. Zanddiepte-Attentiekaarten van het Batist and P. Jacobs), Spec. Publ. Geol. Soc., 117, 95±124. Gelders Rivierengebied. Provincie Gelderland, Arnhem. Antoine, P. (1994) The Somme valley terrace system (northern Berger, A., Loutre, M.F. and Dehant, V. (1989) Astronomical France); a model of river response to Quaternary climatic frequencies for pre-Quaternary palaeoclimate studies. Terra variations since 800,000 BP. Terra Nova, 6, 453±464. Nova, 1, 474±479. Ashworth, P.J., Best, J.L., Peakall, J. and Lorsong, J.A. (1999) Berger, A. and Loutre, M.F. (1994) Astronomical forcing The in¯uence of aggradation rate on braided alluvial through geologic time. In: Orbital Forcing and Cyclic architecture: ®eld study and physical scale-modelling of Sequences (Ed. by P. L. de Boer and D. G. Smith), Spec. the Ashburton River gravels, Canterbury Plains, New Publ. Int. Assoc. Sediment., 19, 15±24. Zealand. In: Current Research in Fluvial Sedimentology: Bernard, H.A. and Leblanc, R.J. (1965) Resume of the Proceedings of 6th International Conference on Fluvial Quaternary Geology of the Northwestern Gulf of Mexico Sedimentology (Ed. by N. D. Smith and J. J. Rogers), Spec. Province. In: The Quaternary of the United States (Ed. by H. Publ. Int. Assoc. Sediment., 28, 333±346. E. Wright and D. G. Frey), pp. 137±185. Princeton Aslan, A. and Autin, W.J. (1998) Holocene ¯ood-plain soil University Press, Princeton. formation in the southern lower Mississippi Valley: Best, J.L. and Ashworth, P.J. (1997) Scour in large braided Implications for interpreting alluvial paleosols. Geol. Soc. rivers and the recognition of sequence stratigraphic bound- Am. Bull., 110, 433±449. aries. Nature, 387, 275±277. Aslan, A. and Autin, W.J. (1999) Evolution of the Holocene Best, J.L. and Bridge, J.S. (1992) The morphology and Mississippi River ¯oodplain, Ferriday, Louisiana: insights dynamics of low amplitude bedwaves upon upper stage on the origin of ®ne-grained ¯oodplains. J. Sedim. Res., 69, plane beds and the preservation of planar laminae. 800±815. Sedimentology, 39, 737±752. Aslan, A. and Blum, M.D. (1999) Contrasting styles of Birkeland, P.W. (1999) Soils and Geomorphology. Oxford Holocene avulsion, Texas Gulf Coastal Plain. In: Current University Press, New York. Research in Fluvial Sedimentology: Proceedings of 6th Blum, M.D. (1990) Climatic and eustatic controls on Gulf International Conference on Fluvial Sedimentology (Ed. by Coastal Plain ¯uvial sedimentation: an example from the N. D. Smith and J. J. Rogers), Spec. Publ. Int. Assoc. Late Quaternary of the Colorado River, Texas. In: Sequence Sediment., 28, 193±209. Stratigraphy as an Exploration Tool: Concepts and Practices

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 41

in the Gulf Coast (Ed. by J. M. Armentrout and B. F. Perkins) Bridge, J.S. (1993a) Description and interpretation of ¯uvial Gulf Coast Section of the Society of Economic deposits: a critical perspective. Sedimentology, 40, 801±810. Palaeontologists and Mineralogists Foundation, Bridge, J.S. (1993b) The interaction between channel geome- Proceedings of the 11th Annual Research Conference, pp. try, water ¯ow, sediment transport and deposition in 71±83. braided rivers. In: Braided Rivers (Ed. by J. L. Best and C. Blum, M.D. (1993) Genesis and architecture of incised valley S. Bristow), Spec. Publ. Geol. Soc. Lond., 75, 13±71. ®ll sequences: a Late Quaternary example from the Colorado Bridge, J.S. (1995) Description and interpretation of ¯uvial River, Gulf Coastal Plain of Texas. In: Siliciclastic Sequence deposits: a critical perspective. Sedimentology, 42, 382±389. Stratigraphy: Recent Developments and Applications (Ed. Bridge, J.S. (1997) Thickness of sets of cross-strata and planar by P. Weimer and H. W. Posamentier), Mem. Am. Assoc. strata as a function of formative bedwave geometry and Petrol. Geol., 58, 259±283. migration, and aggradation rate. Geology, 25, 971±974. Blum, M.D. and Price, D.M. (1998) Quaternary alluvial plain Bridge, J.S. and Leeder, M.R. (1979) A simulation model of construction in response to interacting glacio-eustatic and alluvial stratigraphy. Sedimentology, 26, 617±644. climatic controls, Texas Gulf Coastal Plain. In: Relative Role Bridge, J.S., Alexander, J., Collier, R.E.L., Gawthorpe, R.L. of Eustasy, Climate, and Tectonism in Continental Rocks and Jarvis, J. (1995) Ground-penetrating radar and coring (Ed. by K. W. Shanley and P. J. McCabe), Spec. Publ. Soc. used to document the large-scale structure of point-bar Econ. Paleont. Miner., 59, 31±48. deposits. Sedimentology, 42, 839±852. Blum, M.D. and Valastro, S. Jr (1994) Late Quaternary Bridge, J.S., Collier, R.E.L. and Alexander, J. (1998) Large- sedimentation, Lower Colorado River, Gulf Coastal Plain scale structure of Calamus River deposits (Nebraska, USA) of Texas. Geol. Soc. Am. Bull., 106, 1002±1016. revealed using ground-penetrating radar. Sedimentology, Blum, M.D., Toomey III, R.S. and Valastro, S. Jr (1994) Fluvial 45, 977±986. response to Late Quaternary climatic and environmental Bridgland, D.R. (1994) Quaternary of the Thames. Geological change, Edwards Plateau of Texas. Palaeogeogr. Conservation Review Series 7. Chapman & Hall, London. Palaeoclimatol. Palaeoecol., 108, 1±21. Bridgland, D.R. and D'olier, B. (1995) The Pleistocene Blum, M.D., Morton, R.A. and Durbin, J.M. (1995) evolution of the Thames and Rhine drainage systems in `Deweyville' terraces and deposits of the Texas Gulf the southern North Sea Basin. In: Island Britain: a Coastal Plain. Trans. Gulf Coast Assoc. Geol. Soc., 45, 53± Quaternary Perspective (Ed. by R. C. Preece), Spec. Publ. 60. Geol. Soc., 96, 27±45. Blum, M.D., Kocurek, G., Deynoux, M., Swezey, C., Lancaster, Brierley, G.J. (1996) Channel morphology and element N., Price, D. and Pion, J.-C. (1998) Quaternary wadi- assemblages: A constructivist approach to facies modelling. lacustrine-aeolian depositional cycles and sequences, In: Advances in Fluvial Dynamics and Stratigraphy (Ed. by Chott Rharsa Basin, Tunisia. In: Quaternary Deserts and P. A. Carling and M. R. Dawson), pp. 263±298. John Wiley, Climatic Change (Ed. by A. S. Alsharhan, K. Glennie, G. L. Chichester. Whittle and C. G. St. C. Kendall), pp. 539±552. Balkema Broecker, W.S. (1995) The Glacial World According to Wally. EldigioPress, Palisades. Press, Rotterdam. Broecker, W.S. (1997) Thermohaline circulation, the achilles Blum, M.D., Guccione, M.J., Wysocki, D.A. and Robnett, P.C. heel of our climate system: Will man-made CO2 upset the (2000) Late Pleistocene evolution of the Central Mississippi current balance? Science, 278, 1582±1588. Valley, Southen Missouri to Arkansas. Geol. Soc. Am. Bull., Brown, L.J. and Wilson, D.D. (1988) Stratigraphy of Late 112, 221±235. Quaternary deposits of the northern Canterbury Plains. NZ Bond, G., Broecker, W., Johnsen, S., McManus, J., Labeyrie, J. Geol. Geoph., 31, 305±335. L., Jouzel, J. and Bonani, G. (1993) Correlations between Bull, W.B. (1991) Geomorphic Responses to Climate Change. climate records from North Atlantic sediments and Oxford University Press, Oxford. Greenland ice. Nature, 365, 143±147. Burbank, D.W. and Pinter, N. (1999) Landscape evolution: the Bond, G., Showers, W., Cheseby, M., Lotti, R., Almasi, P., interactions of tectonics and surface processes. Basin Res., Demenocal, P., Priore, P., Cullen, H., Hajdas, I. and Bonani, 11, 1±6. G. (1997) A pervasive millennial-scale cycle in North Burke, K., Francis, P. and Wells, G. (1990) Importance of the Atlantic Holocene and glacial climates. Science, 278, geological record in understanding global change. 1257±1266. Palaeogeogr. Palaeoclimatol. Palaeoecol. (Glob. Plan. Bond, G.C. and Lotti, R. (1995) Iceberg discharges intothe Change Sect.), 89, 193±204. North Atlantic on millennial time scales during the last Butcher, S.W. (1990) The nickpoint concept and its implica- glaciation. Science, 267, 1005±1010. tions regarding onlap to the stratigraphic record. In: Bowen, D.Q. (1978) Quaternary Geology. Pergamon Press, Quantitative Dynamic Stratigraphy (Ed. by T. A. Cross), Oxford. pp. 375±385. Prentice Hall, Englewood Cliffs. Bown, T.M. and Kraus, M.J. (1987) Integration of channel and Butzer, K.W. (1980) Holocene alluvial sequences: problems of ¯oodplain suites, I. Developmental sequence and lateral dating and correlation. In: Timescales in Geomorphology relations of alluvial paleosols. J. Sedim. Petrol., 57, 587±601. (Ed. by R. A. Cullingford, D. A. Davidson and J. Lewin), pp. Bradley, R.S. (1999) Paleoclimatology. Academic Press, San 131±142. John Wiley, New York. Diego. Chappell, J. and Shackleton, N.J. (1986) Oxygen isotopes and Branson, J., Brown, A.G. and Gregory, K.J. (Eds). (1996) sea-level. Nature, 324, 137±140. Global Continental Changes: The Context of Chappell, J., Omura, A., Ezat, T., McCulloch, M., Pandol®, J., Palaeohydrology. Spec. Publ. Geol. Soc. London, 115. Ota, Y. and Pillans, B. (1996) Reconciliation of late Bravard, J.-P. (1992) Rapport no. 2: Les rythmes d'evolution Quaternary sea-levels derived from coral terraces at Huon morphologique des valleÁes Francaises au Tardiglaciaire et Peninsula with deep sea oxygen isotope records. Earth l'Holocene. Bull. Assoc. Geographes Francais. Planet. Sci. Lett., 141, 227±236.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 42 M. D. Blum and T. E. ToÈrnqvist

Clemmensen, L.B., éxnevad, I.E.I. and De Boer, P.L. (1994) geologists of an expanded reality. In: Varieties of Fluvial Climatic controls on ancient sedimentation: some late Form (Ed. by A. J. Miller and A. Gupta), pp. 497±505. John Palaeozoic and Mezosoic examples from NW Europe and Wiley, Chichester. the Western Interior of the USA. In: Orbital Forcing and Fisk, H.N. (1944) Geological Investigation of the Alluvial Cyclic Sequences (Ed. by P. L. de Boer and D. G. Smith), Valley of the Lower Mississippi River. Mississippi River Spec. Publ. Int. Assoc. Sediment., 19, 439±457. Commission, Vicksburg. COHMAP Project Members. (1988) Climatic changes of the Fleming, K., Johnston, P., Zwartz, D., Yokoyama, Y., last 18,000 years: Observations and model simulations. Lambeck, K. and Chappell, J. (1998) Re®ning the eustatic Science, 241, 1043±1052. sea-level curve since the Last Glacial Maximum using far- Cojan, I. (1993) Alternating ¯uvial and lacustrine sedimenta- and intermediate-®eld sites. Earth Planet. Sci. Lett., 163, tion: tectonic and climatic controls (Provence Basin, S. 327±342. France, Upper Cretaceous/Palaeocene). In: Alluvial Fournier, F. (1960) Climat et Erosion: la relation entre Sedimentation (Ed. by M. Marzoand C. Puigdefabregas), lõÂerosion du sol par lõÂeau et les precipitations atmospher- Spec. Publ. Int. Assoc. Sediment., 17, 425±438. iques. Presse Universitaire de France. Crowley, T. and North, G.R. (1991) Palaeoclimatology. Oxford Frakes, L.A. (1979) Climates Throughout Geologic Time. University Press, New York. Elsevier, Amsterdam. Currie, B.S. (1997) Sequence stratigraphy of nonmarine Frakes, L.A., Francis, J.E. and Syktus, J.I. (1992) Climate Jurassic±Cretaceous rocks, central Cordilleran foreland- Modes of the Phanerozoic. Cambridge University Press, basin system. Geol. Soc. Am. Bull., 109, 1206±1222. Cambridge. Dai, A., Fung, I.Y. and Del Genio, A.D. (1997) Surface Fuller, I.C., Macklin, M.G., Lewin, J., Passmore, D.G. and observed global land precipitation variations during 1900± Wintle, A.G. (1998) River response to high-frequency 88. J. Clim., 10, 2943±2962. climate oscillations in southern Europe over the past 200 Dalrymple, R.W., Boyd, R. and Zaitlin, B.A. (Eds) (1994) k.y. Geology, 26, 275±278. History of research, types and internal organisation of Galloway, W.E. (1989) Genetic stratigraphic sequences in incised-valley systems: Introduction to the Volume. In: basin analysis I: architecture and genesis of ¯ooding-surface Incised-Valley Systems: Origin and Sedimentary Sequences bounded depositional units. Am. Assoc. Petrol. Geol. Bull., (Ed. by R. W. Dalrymple, R. Boyd and B. A. Zaitlin), Spec. 73, 125±142. Publ. Soc. Econ. Paleont. Miner., 51, 3±10. Gibbard, P.L. (1995) The formation of the Strait of Dover. In: Dott, R.H. Jr (1992) An introduction to the ups and downs of Island Britain: a Quaternary Perspective (Ed. by R. C. eustasy. In: Eustasy: the Historical Ups and Downs of a Preece), Spec. Publ. Geol. Soc., 96, 15±26. Major Geological Concept (Ed. by R. H. Dott, Jr), Geol. Soc. Gibling, M.R. and Bird, D.J. (1994) Late Carboniferous Am. Mem., 180, 1±16. cyclothems and alluvial paleovalleys in the Sydney Basin, Dubiel, R.F., Huntoon, J.E., Condon, S.M. and Stanesco, J.D. Nova Scotia. Geol. Soc. Am. Bull., 106, 105±117. (1996) Permian deposystems, palaeogeography, and palaeo- Gibling, M.R. and Wightman, W.G. (1994) Palaeovalleys and climate of the Paradox Basin and vicinity. In: Palaeozoic protozoan assemblages in a Late Carboniferous cyclothem, Systems of the Rocky Mountain Region (Ed. by M. W. Sydney Basin, Nova Scotia. Sedimentology, 41, 699±719. Longman and M. D. Sonnenfeld), pp. 427±444. Rocky Gilbert, G.K. (1877) Report on the Geology of the Henry Mountain Section, SEPM. Mountains. Government Printing Of®ce, Washington, DC. Duller, G.A.T. (1996) Recent developments in luminescence Gregory, K.J. (Ed). (1983) Background to Palaeohydrology. dating of Quaternary sediments. Progr. Phys. Geogr., 20, John Wiley, Chichester. 127±145. Gregory, K.J., Lewin, J. and Thornes, J.B. (Eds). (1987) Dury, G.H. (1965) Principles of Under®t Streams. United Palaeohydrology in Practice. John Wiley. States Geol. Surv. Prof. Pap., 452-A. Gregory, K.J., Starkel, L. and Baker, V.R. (Eds). (1996) Global Emery, D. and Myers, K.J. (Eds). (1996) Sequence Continental Palaeohydrology. John Wiley. Stratigraphy. Blackwell Science, Oxford. Grootes, P.M., Stuiver, M., White, J.W.C., Johnsen, S. and Ethridge, F.G. and Schumm, S.A. (1978) Reconstructing Jouzel, J. (1993) Comparison of oxygen isotope records from palaeochannel morphologic and ¯ow characteristics: meth- the GISP2 and GRIP Greenland ice cores. Nature, 366, 552± odology, limitations and assessment. In: Fluvial 554. Sedimentology (Ed. by A. D. Miall), Can. Soc. Petrol. Geol. Hallet, B., Hunter, L. and Bogen, J. (1996) Rates of erosion and Mem., 5, 703±721. sediment evacuation by glaciers: a review of ®eld data and Ethridge, F.G., Wood, L.J. and Schumm, S.A. (1998) Cyclic their implications. Global Planet. Change, 12, 213±235. variables controlling ¯uvial sequence development: Haq, B.U., Hardenbol, J. and Vail, P.R. (1987) Chronology of Problems and perspectives. In: Relative Role of Eustasy, ¯uctuating sea-levels since the Triassic. Science, 235, 1156± Climate, and Tectonism in Continental Rocks (Ed. by K. W. 1167. Shanley and P. J. McCabe), Spec. Publ. Soc. Econ. Paleont. Haq, B.U., Hardenbol, J. and Vail, P.R. (1988) Mesozoic and Miner., 59, 17±29. Cenozoic chronostratigraphy and cycles of sea-level change. Fairbanks, R.G. (1989) A 17,000-year glacio-eustatic sea-level record: in¯uence of glacial melting rates on the Younger In: Sea-Level Changes: an Integrated Approach (Ed. by C. K. Dryas event and deep-ocean circulation. Nature, 342, 637± Wilgus, B. S. Hastings, C. G. S. C. Kendall, H. W. 642. Posamentier, C. A. Ross and J. C. Van Wagoner), Spec. Ferguson, R. (1987) Hydraulic and sedimentary controls of Publ. Soc. Econ. Paleont. Miner., 42, 71±108. channel pattern. In: River Channels: Environment and Hays, J.D., Imbrie, J. and Shackleton, N.J. (1976) Variations in Process (Ed. by K. S. Richards), pp. 17±29. Blackwell the earth's orbit: Pacemaker of the ice ages. Science, 194, Science, Oxford. 1121±1132. Fielding, C.R. (1999) Varieties of ¯uvial form: the relevance to Helland-Hansen, W. and Gjelberg, J.G. (1994) Conceptual

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 43

basis and variability in sequence stratigraphy: a different R. C. Kochel and P. C.Patton), pp. 279±300. John Wiley and perspective. Sedim. Geol., 92, 31±52. Sons, New York. Heller, P.L. and Paola, C. (1996) Downstream changes in Knox, J.C. (1993) Large increases in ¯ood magnitude in alluvial architecture: an exploration of controls on channel- response to modest changes in climate. Nature, 361, 430± stacking patterns. J. Sedim. Res., 66, 297±306. 432. Hovius, N. (1998) Controls on sediment supply by large rivers. Knox, J.C. (1996) Fluvial systems since 20,000 yrs BP. In: In: Relative Role of Eustasy, Climate, and Tectonism in Global Continental Palaeohydrology (Ed. by K. J. Gregory, L. Continental Rocks (Ed. by K. W. Shanley and P. J. McCabe), Starkel and V. R. Baker), pp. 87±108. John Wiley and Sons, Spec. Publ. Soc. Econ. Paleont. Miner., 59, 3±16. New York. Hovius, N. (1999) Macro scale process-systems of mountain Kocurek, G. (1998) Aeolian system response to external belt erosion. In: Geomorphology and Global Tectonics (Ed. forcing factors ± a sequence stratigraphic view of the by M. A. Summer®eld), pp. 77±105. John Wiley, Chichester. Saharan region. In: Quaternary Deserts and Climatic Hovius, N. and Leeder, M. (1998) Clastic sediment supply to Change (Ed. by A. S. Alsharhan, K. Glennie, G. L. Whittle basins. Basin Res., 10, 1±5. and C. G. S. C. Kendall), pp. 327±337. Balkema Press. Howell, J.A. and Flint, S.S. (1996) A model for high resolution Kocurek, G. and Havholm, K.G. (1993) Eolian sequence sequence stratigraphy within extensional basins. In: High stratigraphy: A conceptual framework. In: Siliciclastic Resolution Sequence Stratigraphy: Innovations and Sequence Stratigraphy: Recent Developments and Applications (Ed. by J. A. Howell and J. F. Aitken). Spec. Applications (Ed. by P. Weimer and H. W. Posamentier), Publ. Geol. Soc., 104, 129±138. Am. Assoc. Petrol. Geol. Mem., 58, 393±410. Huggenberger, P. (1993) Radar facies: recognition of facies Koss, J.E., Ethridge, F.G. and Schumm, S.A. (1994) An patterns and heterogeneities within Pleistocene Rhine experimental study of the effects of base-level change on gravels, NE Switzerland. In: Braided Rivers (Ed. by J. L. ¯uvial, coastal plain and shelf systems. J. Sedim. Res., B64, Best and C. S. Bristow), Spec. Publ. Geol. Soc., 75, 163±176. 90±98. Hurrell, J.W. (1995) Decadal trends in the North Atlantic Kozarski, S. and Rotnicki, K. (1977) Valley ¯oors and changes Oscillation regional temperatures and precipitation. of river channel patterns in the north Polish Plain during Science, 269, 676±679. the late WuÈ rm and Holocene. Quaestiones Geographicae, 4, Imbrie, J. and Imbrie, K. (1979) Ice Ages. Harvard University 51±93. Press, Cambridge, MA. Kraus, M.J. (1987) Integration of channel and ¯oodplain Imbrie, J., Hays, J.D., Martinson, D.G., McIntyre, A., Mix, suites, II. Vertical relations of alluvial paleosols. J. Sedim. A.C., Moreley, J.J., Pisias, N.G., Prell, W.L. and Shackleton, Petrol., 57, 602±612. N.J. (1984) The orbital theory of Pleistocene climate: Kutzbach, J.E., Gallimore, R., Harrison, S.P., Behling, P., 18 support from a revised chronology of the marine d O Selin, R. and Laarif, F. (1997) Climate and biome simula- record. In: Milankovitch and Climate (Ed. by A. L. Berger, J. tions for the last 21,000 years. Quat. Sci. Rev., 17, 473±506. Imbrie, J. Hays, G. Kukla and B. Saltzman), pp. 269±306. Lambeck, K. (1993) Glacial rebound and sea-level change: an Reidel. example of a relationship between mantle and surface Jervey, M.T. (1988) Quantitative geological modeling of processes. In: Relationships Between Mantle Processes and siliciclastic rock sequences and their seismic expression. Geological Processes at or Near the Earth's Surface (Ed. by In: Sea-Level Changes: an Integrated Approach (Ed. by C. K. R. Wortel, U. Hansen and R. Sabadini), Tectonophysics, Wilgus, B. S. Hastings, C. G. S. C. Kendall, H. W. 223, 15±37. Posamentier, C. A. Ross and J. C. Van Wagoner), Spec. Lamothe, L. (1918) Les anciennes nappes alluviales et lignes Publ. Soc. Econ. Paleont. Miner., 42, 47±69. de Rivage du bassin de la Somme et leurs rapports avec Johnsen, S.J., Clausen, H.B., Dansgaard, W., Fuhrer, K., celles de la Meditarranee occidentale. Bull. Soc. Geol., Fr., Gundestrup, N., Hammer, C.U., Iversen, P., Jouzel, J., 18, 3±58. Stauffer, B. and Steffensen, J.P. (1992) Irregular glacial Langbein, W.B. and Schumm, S.A. (1958) Yield of sediment in interstadials recorded in a new Greenland ice core. Nature, relation to mean annual sediment. Trans. Am. Geophys 359, 311±313. Union., 39, 1076±1084. Kasse, K., Vandenberghe, J. and Bohncke, S. (1995) Climatic Lane, E.W. (1955) The importance of ¯uvial morphology in change and ¯uvial dynamics of the Maas during the late hydraulic engineering. Am. Soc. Civil Eng. Proc., 81 (745), Weichselian and early Holocene. PalaÃoklimaforschung, 14, 1±17. 123±150. Lau, N.-C. (1988) Variability of the observed midlatitude storm Khan, I.A., Bridge, J.S., Kappelman, J. and Wilson, R. (1997) tracks in relation to low-frequency changes in the circula- Evolution of Miocene ¯uvial environments, eastern Potwar tion pattern. J. Atmos. Sci., 45, 2718±2743. plateau, northern Pakistan. Sedimentology, 44, 221±251. Leckie, D.A. (1994) Canterbury Plains, New Zealand ± Knighton, D. (1998) Fluvial Forms and Processes. Edward Arnold, London. Implications for sequence stratigraphic models. Am. Knox, J.C. (1976) Concept of the Graded Stream. In: Theories Assoc. Petrol. Geol. Bull., 78, 1240±1252. of Landform Development (Ed. by W. N. Melhorn and R. C. Leclair, S.F., Bridge, J.S. and Wang, F. (1997) Preservation of Flemel), pp. 169±198. Publications in Geomorphology, State cross-strata due to migration of subaqueous dunes over University of New York at Binghamton, Binghamton, NY. aggrading and non-aggrading beds: comparison of experi- Knox, J.C. (1983) Responses of river systems to Holocene mental data with theory. Geosci. Canada, 24, 55±66. climates. In: Late Quaternary Environments of the United Leeder, M.R. (1975) Pedogenic carbonates and ¯ood sediment States, Vol. 2, the Holocene (Ed. by H. E. Wright and S. C. accretion rates: a quantitative model for alluvial arid-zone Porter), pp. 26±41. University of Minnesota Press. lithofacies. Geol. Mag., 112, 257±270. Knox, J.C. (1987) Climatic in¯uence on Upper Mississippi Leeder, M.R. (1978) A quantitative stratigraphic model for Valley ¯oods. In: Flood Geomorphology (Ed. by V. R. Baker, alluvium, with special reference tochannel depositdensity

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 44 M. D. Blum and T. E. ToÈrnqvist

and interconnectedness. In: Fluvial Sedimentology (Ed. by Srikanthan, R. (1992) Global Runoff. Continental A. D. Miall), Can. Soc. Petrol. Geol. Mem., 5, 587±596. Comparisons of Annual Flows and Peak Discharges. Leeder, M.R. and Stewart, M.D. (1996) Fluvial incision and Catena Paperback, Catena-Verlag. sequence stratigraphy: alluvial responses to relative sea- Meade, R.H., Yuzyk, T.R. and Day, T.J. (1990) Movement and level fall and their detection in the geologic record. In: storage of sediment in rivers of the United States and Sequence Stratigraphy in British Geology (Ed. by S. P. Canada. In: Surface Water Hydrology (Ed. by M. G. Wolman Hesselboand D. N. Parkinson), Spec. Publ. Geol. Soc., 103, and H. C. Riggs) Geological Society of America, Geology of 25±39. North America Series, Vol. 0-1, pp. 255±280. Leeder, M.R., Mack, G.H., Peakall, J. and Salyards, S.L. Merritts, D. and Ellis, M. (1994) Introduction to special section (1996) First quantitative test of alluvial stratigraphic on tectonics and topography. J. Geophys. Res., 99, 12135± models: Southern Rio Grande rift, New Mexico. Geology, 12141. 24, 87±90. Merritts, D.J., Vincent, K.R. and Wohl, E.E. (1994) Long river Leeder, M.R., Harris, T. and Kirkby, M.J. (1998) Sediment pro®les, tectonism, and eustasy: a guide to interpreting supply and climate change: implications for basin strati- ¯uvial terraces. J. Geophys. Res., 99, 14031±14050. graphy. Basin Res., 10, 7±18. Miall, A.D. (1985) Architectural element analysis: a new Legarreta, L. and Uliana, M.A. (1998) Anatomy of hinterland method of facies analysis applied to ¯uvial deposits. Earth- depositional sequences: Upper Cretaceous ¯uvial strata, Sci. Rev., 22, 261±308. Neuquen Basin, West-Central Argentina. In: Relative Role of Miall, A.D. (1991) Stratigraphic sequences and their chronos- Eustasy, Climate, and Tectonism in Continental Rocks (Ed. tratigraphic correlation. J. Sedim. Petrol., 61, 497±505. by K. W. Shanley and P. J. McCabe), Spec. Publ. Soc. Econ. Miall, A.D. (1995) Description and interpretation of ¯uvial Paleont. Miner., 59, 83±92. deposits: a critical perspective. Sedimentology, 42, 370±384. Leopold, L.B. and Bull., W.B. (1979) Base level, aggradation, Miall, A.D. (1996) The Geology of Fluvial Deposits. Springer, and grade. Proc. Am. Phil. Soc., 123, 168±202. Berlin. Leopold, L.B. and Maddock, T. (1953) The Hydraulic Miall, A.D. (1997) The Geology of Stratigraphic Sequences. Geometry of Stream Channels and Some Physiographic Springer, Berlin. Implications. Prof. Pap. United States Geol. Surv., 252. Milliman, J.D. and Syvitski, J.P.M. (1992) Geomorphic / Lewin, J. (1989) Floods in ¯uvial geomorphology. In: Floods: tectonic control on sediment discharge to the ocean: the Hydrological, Sedimentological, and Geomorphological importance of small mountainous streams. J. Geol., 100, Implications (Ed. by K. Beven and P. Carling), pp. 265± 525±544. 284. John Wiley, Chichester. Mitchum, R.M. Jr, Vail, P.R. and Thompson III, S. (1977) Libby, W.F. (1952) Radiocarbon Dating. University of Chicago Seismic stratigraphy and global changes of sea-level, part 2: Press, Chicago. the depositional sequence as a basic unit for stratigraphic Lowe, J.J. and Walker, M.J.C. (1997) Reconstructing analysis. In: Seismic Stratigraphy ± Applications to Quaternary Environments. Addison-Wesley Longman, Hydrocarbon Exploration (Ed. by C. E. Payton), Am. Harlow. Assoc. Petrol. Geol. Mem., 27, 53±62. Mack, G.H., James, W.C. and Monger, H.C. (1993) Molnar, P. and England, P. (1990) Late Cenozoic uplift of Classi®cation of paleosols. Geol. Soc. Am. Bull., 105, 129± mountain ranges and global climate change: chicken or egg? 136. Nature, 346, 29±34. Mackey, S.D. and Bridge, J.S. (1995) Three-dimensional Molnar, P., England, P. and Martinod, J. (1993) Mantle model of alluvial stratigraphy: theory and application. J. dynamics, uplift of the Tibetan Plateau, and the Indian Sedim. Res., B65, 7±31. monsoon. Rev. Geophys., 31, 357±396. Mackin, J.H. (1948) Concept of the graded river. Geol. Soc. Morton, R.A. and Suter, J.R. (1996) Sequence stratigraphy and Am. Bull., 59, 463±512. composition of Late Quaternary shelf margin deltas, north- Macklin, M.G. and Lewin, J. (1993) Holocene river alluviation ern Gulf of Mexico. Am. Assoc. Petrol. Geol. Bull., 80, 505± in Britain. Z. Geomorph. N. F. (Suppl.)-Bd., 88, 109±122. 530. Makaske, B. (1998) Anastomosing rivers. Nederl. Geogr. Muto, T. and Steel, R.J. (1997) Principles of regression and Studies, 249, 1±287. transgression: the nature of the interplay between accom- Mandier, P. (1988) Le Relief de la Moyenne VallEÁe du Rhone modation and sediment supply. J. Sedim. Res., 67, 994± au Tertiaire et au Quaternaire. Doc. du BGRM, No 151, 1000. Paris. Nanson, G.C. and Croke, J.C. (1992) A genetic classi®cation of Martinsen, O.J., Ryseth, A., Helland-Hansen, W., Flesche, H., ¯oodplains. Geomorphology, 4, 459±486. Torkildsen, G. and Idil, S. (1999) Stratigraphic base level Nanson, G.C. and Knighton, A.D. (1996) Anabranching rivers; and ¯uvial architecture. Ericson Sandstone (Campanian), their cause, character and classi®cation. Earth Surf. Proc. Rock Springs Uplift, W. Wyoming, USA. Sedimentology, 46, Landf., 21, 217±239. 235±260. Nanson, G.C., Young, R.Y., Prices, D.M. and Rust, B.R. (1988) McCarthy, P.J. and Plint, A.G. (1998) Recognition of inter¯uve Stratigraphy, sedimentology, and Late Quaternary chronol- seqeunce boundaries: Integrating paleopedology and sequence stratigraphy. Geology, 26, 387±390. ogy of the channel country of western Queensland. In: McMahon, T.A., Finlayson, B.L., Haines, A. and Srikanthan, Fluvial Geomorphology of Australia (Ed. by R. F. Warner), R. (1987) Runoff variability: a global perspective. In: The pp. 151±175. Academic Press, Sydney. In¯uence of Climate Change and Climate Variability on the Nanson, G.C., Price, D.M. and Short, S.A. (1992) Wetting and Hydrologic Regime and Water Resources (Proceedings of the drying of Australia over the past 300 ka. Geology, 20, 791± Vancouver Symposium, August 1987). IAHS Publ., 168, 794. 3±11. North, C.P. (1996) The prediction and modelling of subsurface McMahon, T.A., Finlayson, B.L., Haines, A.T. and ¯uvial stratigraphy. In: Advances in Fluvial Dynamics and

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 45

Stratigraphy (Ed. by P. A. Carling and M. R. Dawson), pp. sea-level changes and sequence stratigraphy. Sedim. Geol., 395±508. John Wiley, Chichester. 122, 5±21. North American Commission on Stratigraphic Nomenclature. Pinter, N. and Brandon, M.T. (1997) How erosion builds (1983) North American Stratigraphic Code. Am. Assoc. mountains. Sci. Am., 276, 74±79. Petrol. Geol. Bull., 67, 841±875. Pirazzoli, P.A. (1996) Sea-Level Changes. John Wiley, Nummedal, D., Riley, G.W. and Templet, P.L. (1993) High- Chichester. resolution sequence architecture: a chronostratigraphic Pons, L.J. (1957) De geologie, de bodemvorming en de model based on equilibrium pro®le studies. In: Sequence waterstaatkundige ontwikkeling van het Land van Maas en Stratigraphy and Facies Associations (Ed. by H. W. Waal en een gedeelte van het Rijk van Nijmegen. Versl. Posamentier, C. P. Summerhayes, B. U. Haq and G. P. Landbouwk. Onderz., 63-11, 1±156. Allen), Spec. Publ. Int. Assoc. Sediment., 18, 55±68. Porter, S.C. (1989) Some geological implications of average Nystuen, J.P. (1998) History and development of sequence Quaternary glacial conditions. Quat. Res., 32, 245±262. stratigraphy. In: Sequence Stratigraphy ± Concepts and Posamentier, H.W. and Allen, G.P. (1993) Siliclastic sequence Applications (Ed. by F. M. Gradstein, K. O. Sandvik and N. stratigraphic patterns in foreland ramp-type basins. J. Milton), pp. 31±116. Elsevier, Amsterdam. Geology, 21, 455±458. Olsen, H. (1990) Astronomical forcing of meandering river Posamentier, H.W. and Vail, P.R. (1988) Eustatic controls on behavior: Milankovitch cycles in Devonian of East clastic deposition II ± Sequence and systems tract models. Greenland. Palaeogeogr. Palaeoclimatol. Palaeoecol., 79, In: Sea-Level Changes: an Integrated Approach (Ed. by 99±115. C. K. Wilgus, B. S. Hastings, C. G. S. C. Kendall, H. W. Olsen, H. (1994) Orbital forcing on continental depositional Posamentier, C. A. Ross and J. C. Van Wagoner), Spec. Publ. systems ± lacustrine and ¯uvial cyclicity in the Devonian of Soc. Econ. Paleont. Miner., 42, 125±154. East Greenland. In: Orbital Forcing and Cyclic Sequences Posamentier, H.W. and Weimer, P. (1993) Siliciclastic (Ed. by P. L. de Boer and D. G. Smith), Spec. Publ. Int. sequence stratigraphy and petroleum geology nÄ Where to Assoc. Sediment., 19, 429±438. from here? Am. Assoc. Petrol. Geol. Bull., 77, 731±742. Olsen, H. and Larsen, P.H. (1993) Structural and climatic Posamentier, H.W., Jervey, M.T. and Vail, P.R. (1988) Eustatic controls on ¯uvial depositional systems: Devonian of East controls on clastic deposition I ± Conceptual frameworks. Greenland. In: Alluvial Sedimentation (Ed. by M. Marzoand In: Sea-Level Changes: an Integrated Approach (Ed. by C. Puigdefabregas), Spec. Publ. Int. Assoc. Sediment., 17, C. K. Wilgus, B. S. Hastings, C. G. S. C. Kendall, H. W. 401±423. Posamentier, C. A. Ross and J. C. Van Wagoner), Spec. Publ. Olsen, T., Steel, R.H., Hùgseth, K., Skar, T. and Rùe, S.-L. Soc. Econ. Paleont. Miner., 42, 109±124. (1995) Sequential architecture in a ¯uvial succession: Posamentier, H.W., Allen, G.P., James, D.P. and Tesson, M. sequence stratigraphy in the Upper Cretaceous Mesaverde (1992) Forced regressions in a sequence stratigraphic Group, Price Canyon. Utah. J. Sedim. Res., B65, 265±280. framework: Concepts, examples, and exploration signi®- Orton, G.J. and Reading, H.G. (1993) Variability of delatic cance. Am. Assoc. Petrol. Geol. Bull., 76, 1687±1709. processes in terms of sediment supply, with particular Potter, P.E. (1978) Signi®cance and origin of big rivers. J. Geol., emphasis on grain size. Sedimentology, 40, 475±512. 86, 13±33. Page, K., Nanson, G. and Price, D. (1996) Chronology of Powell, J.W. (1875) Exploration of the Colorado River of the Murrumbidgee River palaeochannels on the Riverine Plain, West and its Tributaries. Government Printing Of®ce. southeastern Australia. J. Quat. Sci., 11, 311±326. Washington, DC. Paola, C. (2000) Quantitative models of sedimentary basin Probst, J.-L. (1989) Hydroclimatic ¯uctuations of some ®lling. Sedimentology, 47 (Suppl. 1), 121±178. European Rivers since 1800. In: Historical Changes of Paola, C., Heller, P.L. and Angevine, C.L. (1992) The large- Large Alluvial Rivers, Western Europe (Ed. by G. E. Petts), scale dynamics of grain size variation in alluvial basins, 1. pp. 41±55. John Wiley and Sons, New York. Theory. Basin Res., 4, 73±90. Retallack, G.J. (1986) Fossil soils as grounds for interpreting Parrish, J.T. (1998) Interpreting Pre-Quaternary Climate from long-term controls on ancient rivers. J. Sedim. Petrol., 56, 1± the Geologic Record. Columbia University Press, New York. 18. Payton, C.E. (Ed). (1977) Seismic Stratigraphy ± Applications Retallack, G.J. (1990) Soils of the Past ± An Introduction to to Hydrocarbon Exploration. AAPG Mem., 27. Palaeopedology. Unwin Hyman, Boston. Pazzaglia, F.J., Gardner, T.W. and Merritts, D.J. (1998) Richards, K. (1982) Rivers. Methuen, London. Bedrock ¯uvial incision and longitudinal pro®le develop- Rogers, J.C. (1997) North Atlantic storm track variability and ment over geologic time scales determined by ¯uvial its association to the North Atlantic Oscillation and climate terraces. In: Rivers Over Rock: Fluvial Processes in variability of Northern Europe. J. Clim., 10, 1635±1647. Bedrock Channels (Ed. by K. J. Tinkler and E. E. Wohl), Rotnicki, K. (1991) Retrodiction of palaeodischarges of mean- AGU Monogr., 107, 207±235. dering and sinuous alluvial rivers and its palaeohydrocli- Peltier, W.R. (1998) Post-glacial variations in the level of the matic implications. In: Temperate Palaeohydrology (Ed. by sea: implications for climate dynamics and solid-earth L. Starkel, K. J. Gregory and J. B. Thornes), pp. 431±471. geophysics. Rev. Geophys., 36, 603±689. John Wiley, Chichester. Penck, A. and BruÈ ckner, E. (1909) Die Alpen im Eiszeitalter. Salter, T. (1993) Fluvial scour and incision: models for their Tauchnitz, Leipzig. in¯uence on the development of realistic reservoir geome- Perlmutter, M.A. and Matthews, M.D. (1989) Global cyclos- tries. In: Characterization of Fluvial and Aeolian Reservoirs tratigraphy: a model. In: Quantitative Dynamic Straitgraphy (Ed. by C. P. North and D. J. Prosser), Spec. Publ. Geol. Soc., (Ed. by T. A. Cross), pp. 233±260. Prentice Hall, Englewood 73, 33±51. Cliffs. Saucier, R.T. (1994) Geomorphology and Quaternary Geologic Pillans, B., Chappell, J. and Naish, T.R. (1998) A review of the History of the Lower Mississippi Valley. Mississippi River Milankovitch climatic beat: template for Plio-Pleistocene Commission, Vicksburg.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 46 M. D. Blum and T. E. ToÈrnqvist

Saucier, R.T. (1996) A contemporary appraisal of some key complex response. In: Palaeohydrology in Practice (Ed. by Fiskian concepts with emphasis on Holocene meander belt K. J. Gregory, J. Lewin and J. B. Thornes), pp. 333±339. John formation and morphology. Engineering Geol., 45, 67±86. Wiley, Chichester. Schirmer, W. (1995) Valley bottoms in the late Quaternary. Z. Starkel, L. (1991) Long-distance correlation of ¯uvial events in Geomorph. N. F. (Suppl.)-Bd., 100, 27±51. the temperate zone. In: Temperate Palaeohydrology (Ed. by Schumm, S.A. (1965) Quaternary palaeohydrology. In: The L. Starkel, K. J. Gregory and J. B. Thornes), pp. 473±495. Quaternary of the United States (Ed. by H. E. Wright and D. John Wiley, Chichester. G. Frey), pp. 783±794. Princeton University Press. Starkel, L., Gregory, K.J. and Thornes, J.B. (Eds). Schumm, S.A. (1968a) River Adjustment toAltered (1991) Temperate Palaeohydrology. John Wiley, Chichester. Hydrologic Regimen: the Murrumbidgee River and Straf®n, E.C., Blum, M.D., Colls, A. and Stokes, S. (2000) Palaeochannels, Australia. Prof. Pap. United States Geol. Alluvial stratigraphy of the Loire and Arroux Rivers, Surv., 598. Burgundy, France. Quaternaire, 10, 271±282. Schumm, S.A. (1968b) Speculations concerning palaeohydro- Summer®eld, M.A. (1985) Plate tectonics and landscape logic controls of terrestrial sedimentation. Geol. Soc. Am. development on the African continent. In: Tectonic Bull., 79, 1573±1588. Geomorphology (Ed. by M. Morisawa and J. Hack), pp. 27± Schumm, S.A. (1985) Explanation and extrapolation in 51. Allen and Unwin, Boston. geomorphology: seven reasons for geologic uncertainty. Suter, J.R. and Berryhill, H.L. Jr (1985) Late Quaternary shelf- Trans. Jap Geomorphol. Union., 6, 1±18. margin deltas, northwest Gulf of Mexico. Am. Assoc. Petrol. Schumm, S.A. (1977) The Fluvial System. John Wiley, New Geol. Bull., 69, 77±91. York. Talling, P.J. (1998) How and where do incised valleys form if Schumm, S.A. (1991) To Interpret the Earth. Ten Ways to Be sea-level remains above the shelf edge? Geology, 26, 87±90. Wrong. Cambridge University Press. Cambridge. Tebbens, L.A., Veldkamp, A., Westerhoff, W. and Schumm, S.A. (1993) River response to baselevel change: Kroonenberg, S.B. (1999) Fluvial incision and channel Implications for sequence stratigraphy. J. Geol., 101, 279± downcutting as a response to Lateglacial and Early 294. Holocene climate change: the River Meuse lower reach. J. Schumm, S.A. and Brakenridge, G.R. (1987) River responses. Quat. Sci., 14, 59±75. In: North America and Adjacent Oceans During the Last Tesson, M., Allen, G.P. and Ravenne, C. (1993) Late Deglaciation (Ed. by W. F. Ruddiman and H. E. Wright), The Pleistocene shelf-perched lowstand wedges on the Rhone Geology of North America. K-3, pp. 221±240. Geological continental shelf. In: Sequence Stratigraphy and Facies Society of America. Associations (Ed. by H. W.Posamentier, C. P. Summerhayes, Shackleton, N.J. (1987) Oxygen isotopes, ice volume and sea- B. U. Haq and G. P. Allen), Spec. Publ. Int. Assoc. level. Quat. Sci. Rev., 6, 183±190. Sediment., 18, 183±196. Shackleton, N.J. and Opdyke, N.D. (1973) Oxygen isotope and Tesson, M. and Gensous, B. (1998) LõÂenregistrement des cycles palaeomagnetic stratigraphy of equatorial Paci®c core V28± climatiques et eustatiques quaternaires de marges recentes 238: oxygen isotope temperatures and ice volumes on a 105 du bassin Nord-Mediterraneen. C. R. Academie des and 106 year scale. Quat. Res., 3, 39±55. Sciences, Paris, Sciences de la Terre Des Planetes, 326, Shanley, K.W. and McCabe, P.J. (1991) Predicting facies 121±127. architecture through sequence stratigraphy: an example Toomey, R.S., Blum, M.D. and Valastro, S. Jr. (1993) Late from the Kaiparowits Plateau. Utah. Geology, 19, 742±745. Quaternary climates and environments of the Edwards Shanley, K.W. and McCabe, P.J. (1993) Alluvial architecture Plateau, Texas. Glob. Plan. Change, 7, 299±320. in a sequence stratigraphic framework: a case study from the ToÈrnqvist, T.E. (1993) Holocene alternation of meandering and Upper Cretaceous of southern Utah, U.S.A. In: Quantitative anastomosing ¯uvial systems in the Rhine-Meuse Delta Modelling of Clastic Hydrocarbon Reservoirs and Outcrop (central Netherlands) controlled by sea-level rise and Analogs (Ed. by S. S. Flint and I. D. Bryant), Spec. Publ. Int. subsoil erodibility. J. Sedim. Petrol., 63, 683±693. Assoc. Sediment., 15, 21±55. ToÈrnqvist, T.E. (1994) Middle and late Holocene avulsion Shanley, K.W. and McCabe, P.J. (1994) Perspectives on the history of the River Rhine (Rhine-Meuse delta, sequence stratigraphy of continental strata: report of a Netherlands). Geology, 22, 711±714. working group at the 1991 NUNA Conference on High ToÈrnqvist, T.E. (1995) Discussion: alluvial architecture of the Resolution Sequence Stratigraphy. Am. Assoc. Petrol. Geol. Quaternary Rhine±Meuse river system in the Netherlands. Bull., 74, 544±568. Geol. Mijnb., 74, 183±186. Slingerland, R.L. and Snow, R.S. (1988) Stability analysis of a ToÈrnqvist, T.E. (1998) Longitudinal pro®le evolution of the rejuvenated ¯uvial system. Z. Geomorph. N. F. (Suppl.)-Bd., Rhine±Meuse system during the last deglaciation: interplay 67, 93±102. of climate change and glacio-eustasy? Terra Nova, 10, 11± Sloss, L.L. (1962) Stratigraphic models in exploration. Am. 15. Assoc. Petrol. Geol. Bull., 46, 1050±1057. ToÈrnqvist, T.E., Wallinga, J., Murray, A.S., De Wolf, H., Sloss, L.L. (1963) Sequences in the cratonic interior of North Cleveringa, P. and De Gans, W. (2000) Response of the America. Geol. Soc. Am. Bull., 74, 93±113. Rhine-Meuse system (west-central Netherlands) tothe last Smith, G.A. (1994) Climatic in¯uences on continental deposi- Quaternary glacio-eustatic cycles: a ®rst assessment. Global tion during late-stage ®lling of an extensional basin, south- Planet. Change, in press. eastern Arizona. Geol. Soc. Am. Bull., 106, 1212±1228. Torres, W.F.J. and Jain, S.C. (1984) Aggradation and Smith, N.D., Cross, T.A., Duf®cy, J.P. and Clough, S.R. (1989) Degradation of Alluvial Channel Beds. Iowa Institute for Anatomy of an avulsion. Sedimentology, 36, 1±23. Hydraulic Research Report 274, University of Iowa, Iowa Stanley, D.J. Warne, A.G. (1993) Nile Delta: recent geological City. evolution and human impact. Science, 260, 628±634. Tushingham, A.M. and Peltier, W.R. (1992) Validation of the Starkel, L. (1987) The evolution of European rivers ± a ICE-3G model of WuÈrm-Wisconsin deglaciation using a

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 Fluvial responses to climate and sea-level change 47

global database of relative sea-level histories. J. Geophys. Walker and N. P. James), pp. 1±14. Geological Association of Res., 97, 3285±3304. Canada, St. John's. Vail, P.R., Mitchum, R.M. and Thompson III, S. (1977) Walling, D.E. (1983) The sediment delivery problem. J. Seismic stratigraphy and global changes of sea-level, part Hydrol., 65, 209±237. 4: global cycles of relative changes of sea-level. In: Seismic Walling, D.E. and Webb, B.W. (1983) Patterns of sediment Stratigraphy ± Applications to Hydrocarbon Exploration yield. In: Background to Palaeohydrology (Ed. by K. J. (Ed. by C. E. Payton), Am. Assoc. Petrol. Geol. Mem., 26, 83± Gregory), pp. 69±100. John Wiley, Chichester. 97. Walling, D.E. and Webb, B.W. (1996) Erosion and sediment Van den Berg, J.H. (1995) Prediction of alluvial channel yield: a global overview. In: Erosion and Sediment Yield: pattern of perennial rivers. Geomorphology, 12, 259±279. Global and Regional Perspectives (Ed. by D. E. Walling and Vandenberghe, J. (1995) Postglacial river activity and climate: B. W. Webb), Int. Assoc. Hydrol. Sci. Publ., 236, 3±20. state of the art and future prospects. PalaÃoklimaforschung, Wanless, H.R. and Weller, J.M. (1932) Correlation and extent 14, 1±9. of Pennsylvanian cyclothems. Geol. Soc. Am. Bull., 43, Van de Plassche, O. (1982) Sea-level change and water-level 1003±1016. movements in the Netherlands during the Holocene. Meded. Waylen, P.R. (1996) Global hydrology in relation to palaeohy- Rijks Geol. Dienst, 36, 1±93. drological change. In: Global Continental Palaeohydrology Van der Woude, J.D. (1984) The ¯uviolagoonal palaeoenvir- (Ed. by K. J. Gregory, L. Starkel and V. R. Baker), pp. 61±86. onment in the Rhine/Meuse deltaic plain. Sedimentology, John Wiley, Chichester. 31, 395±400. Wescott, W.A. (1993) Geomorphic thresholds and complex Van Dijk, G.J., Berendsen, H.J.A. and Roeleveld, W. (1991) response of ¯uvial systems ± some implications for Holocene water level development in The Netherlands' sequence stratigraphy. Am. Assoc. Petrol. Geol. Bull., 77, river area; implications for sea-level reconstruction. Geol. 1208±1218. Mijnb., 70, 311±326. Wheeler, H.E. (1964) Baselevel, lithostratigraphic surface, and Van Wagoner, J.C. (1995a) Overview of sequence stratigraphy time stratigraphy. Geol. Soc. Am. Bull., 75, 599±610. of foreland basin deposits: Terminology, summary of Wilgus, C.K., Hastings, B.S., Kendall, C.G.S.C., Posamentier, papers, and glossary of sequence stratigraphy. In: H.W., Ross, C.A. and Van Wagoner, J.C. (Eds) (1988) Sea- Sequence Stratigraphy of Foreland Basin Deposits (Ed. by Level Changes. An Integrated Approach. Spec. Publ. Soc. J. C. Van Wagoner and G. T. Bertram), Am. Assoc. Petrol. Econ. Paleont. Miner., 42, 1±407. Geol. Mem., 64, ix±xxi. Williams, G.P. (1978) Bankfull discharge of rivers. Water Res. Van Wagoner, J.C. (1995b) Sequence stratigraphy and marine Res., 14, 1141±1154. to nonmarine facies architecture of foreland basin strata, Williams, G.P. (1987) Palaeo¯uvial estimates from dimensions Book Cliffs, Utah, U.S.A. In: Sequence Stratigraphy of of former channels and meanders. In. Flood Geomorphology Foreland Basin Deposits (Ed. by J. C. Van Wagoner and G. T. (Ed. by V. R. Baker, R. C. Kochel, and P. C. Patton), pp. 321± Bertram), Am. Assoc. Petrol. Geol. Mem., 64, 137±223. 334. John Wiley and Sons. Van Wagoner, J.C. (1998) Sequence stratigraphy and marine to Willis, B. (1993a) Ancient river systems in the Himalayan nonmarine facies architecture of foreland basin strata, Book foredeep, Chinji village area, northern Pakistan. Sedim. Cliffs, Utah, U.S.A.: Reply. Am. Assoc. Petrol. Geol. Bull., Geol., 88, 1±76. 82, 1607±1618. Willis, B. (1993b) Evolution of Miocene ¯uvial systems in the Van Wagoner, J.C., Posamentier, H.W., Mitchum, R.M., Vail, Himalayan foredeep through a two kilometer thick succes- P.R., Sarg, J.F., Loutit, T.S. and Hardenbol, J. (1988) An sion in northern Pakistan. Sedim. Geol., 88, 77±122. overview of the fundamentals of sequence stratigraphy and Willis, B.J. and Behrensmeyer, A.K. (1994) Architecture of some key de®nitions. In: Sea-Level Changes: an Integrated Miocene overbank deposits in northern Pakistan. J. Sedim. Approach (Ed. by C. K. Wilgus, B. S. Hastings, C. G. S. C. Res., B64, 60±67. Kendall, H. W. Posamentier, C. A. Ross and J. C. Van Wohl, E.E. and Enzel, Y. (1996) Data for palaeohydrology. Wagoner), Spec. Publ. Soc. Econ. Paleont. Miner., 42, 39±45. In: Global Continental Palaeohydrology (Ed. by K. J. Gregory, L. Starkel and V. R. Baker), pp. 23±60. John Van Wagoner, J.C., Mitchum, R.M., Campion, K.M. and Wiley, Chichester. Rahmanian, V.D. (1990) Siliciclastic Sequence Wood, L.J., Ethridge, F.G. and Schumm, S.A. (1993) The Stratigraphy in Well Logs, Cores, and Outcrops: Concepts effects of rate of base-level ¯uctuation on coastal-plain, for High-Resolution Correlation of Time and Facies: AAPG shelf and slope depositional systems: an experimental Methods in Exploration 7. AAPG, Tulsa. approach. In: Sequence Stratigraphy and Facies Veldkamp, A. and Van den Berg, M.W. (1993) Three- Associations (Ed. by H. W. Posamentier, C. P. dimensional modelling of Quaternary ¯uvial dynamics in Summerhayes, B. U. Haq and G. P. Allen), Spec. Publ. a climo-tectonic dependent system. A case study of the Int. Assoc. Sediment., 18, 43±53. Maas record (Maastricht, The Netherlands). Global Planet. Wood, L.J., Ethridge, F.G. and Schumm, S.A. (1994) An Change, 8, 203±218. experimental study of the in¯uence of subaqueous shelf Veldkamp, A. and Van Dijke, J.J. (1998) Modelling long-term angles on coastal plain and shelf deposits. In: Siliciclastic erosion and sedimentation processes in ¯uvial systems: A Sequence Stratigraphy. Recent Developments and case study for the Allier/Loire system. In: Palaeohydrology Applications (Ed. by P. Weimer and H. W. Posamentier), and Environmental Change (Ed. by G. Benito, V. R. Baker Am. Ass. Petrol. Geol. Mem., 58, 381±391. and K. J. Gregory), pp. 53±66. John Wiley, Chichester. Wright, H.E. Jr (1987) Synthesis: Land south of the ice sheets. Walker, R.G. (1990) Facies modeling and sequence stratigra- In: North America and Adjacent Oceans During the Last phy. J. Sedim. Petrol., 60, 777±786. Deglaciation (Ed. by W. F.Ruddiman and H. E. Wright), GSA Walker, R.G. (1992) Facies, facies models and modern Geology of North America, Vol. K-3, pp. 479±488. stratigraphic concepts. In: Facies Models (Ed. by R. G. Wright, H.E., Jr, Kutzbach, J.E., Webb, T., Ruddiman, W.F.,

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48 48 M. D. Blum and T. E. ToÈrnqvist

Street-Perrott, F.A. and Bartlein, P.J. (Eds). (1993) Global stratigraphy and marine to nonmarine facies architecture of Climates Since the Last Glacial Maximum. University of foreland basin strata, Book Cliffs, Utah, U.S.A. Am. Assoc. Minnesota Press, Minneapolis. Petrol. Geol. Bull., 82, 1596±1606. Wright, V.P. (Ed). (1986) Paleosols. Princeton University Zaitlin, B.A., Dalrymple, R.W. and Boyd, R. (1994) The Press, Princeton. stratigraphic organization of incised-valley systems asso- Wright, V.P. and Marriott, S.B. (1993) The sequence strati- ciated with relative sea-level change. In: Incised-Valley graphy of ¯uvial depositional systems: the role of ¯oodplain Systems: Origin and Sedimentary Sequences (Ed. by R. W. sediment storage. Sedim. Geol., 86, 203±210. Dalrymple, R. Boyd and B. A. Zaitlin), Spec. Publ. Soc. Wyroll, K.-H. (1988) Determining the causes of Pleistocene Econ. Paleont. Miner., 51, 45±60. stream aggradation in the central coastal areas of western Zaleha, M.J. (1997) Fluvial and lacustrine palaeoenvironments Australia. Catena, 15, 39±51. of the Miocene Siwalik Group, Khaur area, northern Yang, C.-S. and Nio, S.-D. (1993) Application of high Pakistan. Sedimentology, 44, 349±368. resolution sequence stratigraphy to the Upper Rotliegend Zhang, Z., Sun, K. and Yin, J. (1997) Sedimentology and in the offshore. In: Siliciclastic Sequence Stratigraphy (Ed. sequence stratigraphy of the Shanxi Formation (Lower by P. Weimer and H. W. Posamentier), Am. Assoc. Petrol. Permian) in the northwestern Ordos Basin, China: an Geol. Mem., 58, 285±316. alternative sequence model for ¯uvial strata. Sedim. Geol., Yoshida, S., Miall, A.D. and Willis, A. (1998) Sequence 112, 123±136.

# 2000 International Association of Sedimentologists, Sedimentology, 47 (Suppl. 1), 2±48