High Resolution Morphodynamics and Sedimentary Evolution of Estuaries

Estuaries are found along many of the world’s coastlines irrespective of geological setting, energy regime, and depositional environment (Perillo, 1995a). They also represent one of Earth’s most dynamic sedimentary environments because they lie at the interface of the terrestrial and marine spheres, and evolve in response to the interaction of fluvial, coastal (tidal) and marine (wave) processes. The genetic classification of estuaries has focused on the interaction of processes in these fluvial, coastal, and marine environments (e.g. Perillo, 1995b; Elliott and McLusky, 2002), although in practice the processes influencing estuary morphodynamics vary along the length of the estuary, with tidal state, and over different time-spans. Estuaries are therefore not homogeneous sedimentary systems: their fluvial, coastal and marine environmental regimes are all subject to change in their intrinsic characteristics and their interactions over different scales of time and space, particularly in response to changes in climate and relative sealevel (RSL) (Uncles, 2002). It can be argued, therefore, that the estuaries found along present-day coasts worldwide are both environmentallysensitive and geologically-transient phenomena.

The coast of Ireland, located on the paraglacial shelf of the north-east Atlantic (Carter, 1990), is well placed to respond dynamically to external forcing factors in the marine and onshore environments. These factors include eustatic changes in relative sea-level (RSL) driven by glacial cycles on 3^rd and 4^th order (Milankovitch) time-scales; changes in shelf, nearshore, coastal, estuarine, dune and fluvial sediment storage and supply; changes in North Atlantic wind and wave climates; and the effects of high-magnitude events such as storms, storm surges and sea floods (Delaney and Devoy, 1995). In addition, formerly paraglacial coasts and shelves in particular are subject to a range of environmental factors impacting on present-day shelf stratigraphy and sediment dynamics. These factors include sediment supply to continental shelves, and 4^th order (glacioisostatic-driven) changes in RSL (Barnhardt et al., 1997; Plag et al., 1996; Syvitski, 1991; Kelley et al., 1989). In western Britain, late Devensian (Wisconsinan; ∼ 25-13 kyr BP) ice spread outwards from dispersal centres in upland areas of Scotland, Wales, northern England, and northern and western Ireland (Bowen et al., 1986). This ice spread generally radially onto adjacent lowlands and offshore shelves in the North Atlantic, North Sea and Irish Sea (including the Northern Ireland coast) which were dry due to 4^th order eustatically-driven RSL fall (Bridgland, 2002).

Due to the increasing demand for clean water, recreation areas, and healthy ecosystems the management of watersheds has become an issue of increasing importance. This demand has increased the need for detailed inventories of the present state of watersheds and the understanding of related processes. In 1996, the New York State Department of Environmental Conservation (DEC) initiated an effort to map the benthic habitat of the Hudson River Estuary as part of a larger Hudson River Action Plan. This project includes extensive mapping using sidescan sonar, subbottom profiling, single and multi-beam bathymetry, as well as collecting ground truth data with sediment cores, grab samples, and sediment profiling imagery (SPI). The goal of the project is the creation of a comprehensive data set that includes detailed interpretive maps of sediment distribution, grain size, bed forms, and benthic habitats (Ladd et al., 2002).

Over the past 10 to 15 years the use of Geographical Information Systems (GIS) in the analysis of coastal and marine systems has expanded dramatically. One of the reasons for this is the fact that GIS is an ideal means to analyse and visualize the complex spatial and temporal evolution of morphologically complex areas (Bartlett, 1999). Most environmental data in dynamic coastal areas are complex and show variations in location, depth, and time (Li and Saxena, 1993; Kemp, 1997), raising the problem of finding suitable ways to represent environmental phenomena (Lucas, 1999).

New Inlet formed on 2 January 1987 when a northeasterly storm passed over the southern portion of Cape Cod, Massachusetts, USA (Fig. 1). Breaching of Nauset Spit during this event presented an excellent opportunity to describe and understand how tidal inlets evolve hydrodynamically and morphologically after their formation (FitzGerald and Montello, 1991, 1993; Weidman and Ebert, 1993). In this study an analysis of the evolution of a newly formed, frictionally-dominated bay is conducted and results are compared to two well-documented inlet systems.

Puerto Galván (Fig. 1) is one of the five harbors that form the BahÍa Blanca Harbor System, the largest and deepest of Argentina. The harbors are all located along the Canal Principal of the BahÍa Blanca Estuary, a mesotidal, coastal plain environment (Perillo, 1995) formed by a series of major NW-SE trending channels separating extensive tidal flats, low salt marshes and islands. The geomorphology and physical characteristics of the estuary are described in detail elsewhere (Perillo and Piccolo, 1999) including a recent review of its major environmental features (Perillo et al., 2000).

Determining temporal and spatial variations of suspended sediments and other water column physical properties (e.g. temperature, salinity, turbidity) in estuarine systems require high-resolution observations over several scales of space and time (Uncles et al., 1988; Dyer, 2000; Grabemann and Krause, 2001; Schmidt and Luther, 2002). Although obtaining these types of measurements can be difficult due to time, equipment and monetary constraints, they are important for developing a fundamental scientific understanding of many estuarine processes, such as primary and secondary productivity, the transport and fate of contaminants, nutrient cycling, or sedimentation (Pritchard and Schubel, 1981; Ward et al., 1984; Fisher et al., 1988; Bilgili et al., 1996; Allen et al., 1998; Lee and Cundy, 2001; Sanford et al., 2001; Johnston et al., 2002; Verity, 2002). Accordingly, numerous studies have been conducted over the last several decades that seek to describe and quantify basic estuarine physics and sedimentological processes (see Kennedy, 1984; Nichols and Biggs, 1985; Eisma, 1993, and Dyer, 2000 for reviews). For instance, it has been long understood that the combination and balance of freshwater input from rivers and tidal energy controls or strongly influences net non-tidal circulation (density driven), water column stratification, and sedimentation (Pritchard, 1952; Schubel and Biggs, 1969; Biggs, 1970; Schubel, 1972; Allen et al., 1980; Biggs and Cronin, 1981; Ward and Twilley, 1986; Dyer, 2000; Sanford et al., 2001; Schmidt and Luther, 2002).

Research into the dynamics of estuary morphology has been stimulated by increasing commercial, environmental and legislative pressures and by the accumulated impacts of human intervention (Roman and Nordstrom, 1996; Soulsby, 1997). Of particular concern in the UK is the impact on estuaries of sea-level rise and large-scale interventions associated with dredging and port development, flood defence and habitat restoration. Prediction of estuary response to such changes requires an understanding of present-day processes and their interaction with morphologies that are often shaped by past human activities. As Pye and Allen (2000) note, estuarine research has hitherto been characterised by narrow disciplinary foci, such that research fronts in engineering, geomorphology and Quaternary science have rarely converged. The UK Estuaries Research Programme (EMPHASYS, 2000; French et al., 2002) has advocated a more holistic and interdisciplinary perspective, combining ‘bottom up’ studies of short-term hydrodynamics and sediment movement with ‘top down’ models of larger-scale morphodynamic behaviour, such that the predictive power of physically-based simulation may be realised within a conceptual framework provided by geomorphological analysis of longer-term sedimentary function. Few estuaries have been monitored in the spatial and temporal detail needed for integrated modelling of this kind and there is a need for intensive studies encompassing a greater variety of natural and anthropogenic contexts.

Over the past several decades, estuaries have earned a reputation as sediment sinks through the theoretical and empirical works of many scientists (e.g. Postma, 1967; Pritchard, 1967; Meade, 1969, 1972, 1982; Biggs, 1970; Biggs and Howell, 1984; Schubel, 1984; Knebel, 1989; Dalrymple et al., 1990). These studies have documented the combined roles of sediment influx rates, sea-level rise, climate, and estuarine circulation as the dominant controls on estuarine infilling (Schubel, 1984). However, aspects of most of these models (e.g. distance-velocity asymmetry and settling lag and scour lag) only consider the movement of fine-grained sediments (<100 µm) capable of suspension or transport-limited systems in which estuarine sediment supply is greater than the transport capacity (Milliman and Meade, 1983). Much less is known about the dynamics (i.e. estuarine processes and time scales responsible for sediment fluxes) within fluvial-estuarine transition zones with respect to bedload sediment transport (Milliman and Meade, 1983)

Although it is widely stated in the literature that estuarine river mouths are sediment sinks, northern New England estuaries are an exception to this model because they export coarse-grained sediment to the nearshore. The traditional view is that estuaries fill with sediment ranging from mud to gravel derived from fluvial or upland sources as well as from the inner continental shelf and adjacent shorelines. Fluvial sediments are deposited primarily in the inner and central portions of an estuary, although fluvial mud can be deposited in the outer estuary in some tide-dominated systems (e.g. sections of the Gironde River (France), Allen, 1991; Fly River (Papua New Guinea), Harris et al., 1993). The deposition of sediment in the inner and central portions of an estuary is due to the combined influences of a downstream decrease in the riverine current strength and clay flocculation produced by fresh and saltwater mixing. In estuaries having high sediment loads, fluidized mud can be an important component of estuarine sedimentation (Wells, 1983, 1995). Marine sediments enter the outer estuary due to residual, flood-oriented bottom currents and stronger flood than ebb tidal currents. This former flow pattern is caused by the seaward-flowing freshwater advecting the underlying saltwater producing a mass balance deficit of saltwater (Dyer, 1973).

Morphological monitoring is fundamental to the understanding of coastal morphodynamics, and should provide a comprehensive awareness of coastal behaviour in response to storms, climate, sea-level change and human activities on different scales, when supported by historical (meso-) scale examinations of coastal change.

Concepts pertaining to our understanding of estuarine dynamics have been heavily influenced by work carried out on the east and west coasts of the United States and western Europe (Pritchard, 1967). Antecedent geological controls have played an important role in predetermining the dominant type of estuaries along these coasts, namely drowned river valleys on coastal plains and fjord type systems tuned to moderate/high tidal regimes. Along the northern Gulf of Mexico (Fig. 1), however, estuaries are predominantly bar-built where the latest Holocene “stillstand” in sea level has permitted waves to build barrier islands/spits/beaches supplied by sediment from updrift and offshore sand sources (Stone et al., 1992; Stapor and Stone, 2004). Tides in the Gulf of Mexico are microtidal (0–0.3 m), predominantly diurnal and mixed (Marmer, 1954). Characteristically broad regions of low bathymetric relief result in minimal bathymetric steering of the otherwise low-frequency flow (Schroeder and Wiseman, 1999). Due to a high incidence of tropical cyclones in the northern Gulf (Stone et al., 1997; Muller and Stone, 2001), low profile barriers are susceptible to multiple breaches and inlet development. Such occurrences play an important role in estuarine circulation patterns due to phase lags in tidally driven waves. These interlinkages have, however, yet to be fully explored (Schroeder and Wiseman, 1999).

Delta-building in the Holocene Mississippi River system is characterized by the successive construction and abandonment of delta lobes (Fisk, 1944; Kolb and Van Lopik, 1958; Frazier, 1967). Each major delta-building episode is accompanied by a rather orderly and predictable set of events starting with stream capture followed by filling of an interdistributary basin with lacustrine deltas and swamp deposits, building of a bayhead delta at the coast, and finally construction of a major shelf delta. The process of “delta switching” involves the initiation of a new major delta while the previously active delta is systematically abandoned. These changes associated with shifting fluvial input are commonly referred to as the “delta cycle” (Roberts, 1997). Each major delta lobe in the Mississippi River system is active for about 1000–1500 years.

Renowned as the site of the world’s most infamous railway accident, the so-called Tay Bridge Disaster of 1879 (Duck and Dow, 1994), the Tay Estuary of eastern Scotland is one of the most widely studied in the country, especially in terms of geological and geomorphological processes. Over the past four decades in particular, numerous studies of the sedimentology and hydrodynamics of this complex, macrotidal system have been undertaken, largely by researchers from the University of Dundee using the boats and equipment of the dedicated Tay Estuary Research Centre. The database (http://www.dundee.ac.uk/crsem/TEF/review.htm - currently containing over 300 entries) recently compiled by the Tay Estuary Forum (a voluntary partnership established in 2000 “to promote the wise and sustainable use of the Tay Estuary and adjacent coastline”), provides an impressive but still incomplete inventory of the published and unpublished research that has been undertaken on this water body.

Embayed coastlines with fluvial bedload contribution are found in many parts of the world. Due to limited longshore sediment transport, fluvial sediment supply, sea-level history and changes in accommodation space are the primary factors controlling the formation and evolution of a variety of coastal accumulation forms (Barnhardt et al., 1997; FitzGerald and van Heteren, 1999; FitzGerald et al., 2000; 2002; Storlazzi and Field, 2000; Ballantyne, 2002). However, few studies have addressed the sedimentological relationships between fluvial systems and associated barrier sequences at millennial time scales, particularly along formerly glaciated coasts (FitzGerald et al., 1994; Forbes and Syvitski, 1994; van Heteren, 1996; Barnhardt et al., 1995; 1997; Buynevich, 2001; Belknap et al., 2002). In areas where the geological record of an initial fluvial vs. inner shelf sediment contribution is ultimately related to a common fluvial source, it may be difficult to establish the link between sediment transport pathways and coastal depocenters. In addition, such records are often confined to the deeper parts of the barrier sequences and require extensive coring efforts. The mouth of the Kennebec River, Maine, USA, and the associated Holocene barrier systems of Popham and Seawall Beaches (Fig. 1) provide an ideal setting for evaluating the use of bulk sedimentological properties of recent fluvial-estuarine and nearshore sediments in barrier-stratigraphic research.

The bedrock framework of the northern Gulf of Maine coast, USA (Fig. 1), controls the geometry of headlands and embayments (Shipp et al., 1985, 1987; Kelley, 1987). Quaternary continental glaciers sculpted this paraglacial coast, culminating in the latest Wisconsinan Laurentide Ice Sheet, which reached its maximum extent in the region 20–22 ka (Hughes et al., 1985). This ice sheet was marine-based in much of the Gulf of Maine 20–15 ka (Schnitker et al., 2001) and during later stages of retreat through the Maine coastal lowlands (Stuiver and Borns, 1975; Dorion et al., 2001). Sediments of a wide variety of (Thompson and Borns, 1985) were deposited duringeglacial retreat, interpreted in a sequence-stratigraphic model by Belknap and Shipp (1991) and Barnhardt et al. (1997). Sediment sources to the evolving Holocene coast included reworking from glacial and glaciomarine outcrops, as well as limited fluvial inputs.