Evolution of paraglacial coasts in response to changes in fluvial sediment supply

Evolution of paraglacial coasts in response to changes

in fluvial sediment supply

CHRISTOPHER J. HEIN1,6*, D. M. FITZGERALD2, I. V. BUYNEVICH3,

S. VAN HETEREN4 & J. T. KELLEY5

1Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution,

Woods Hole, MA, 02543, USA2Department of Earth and Environment, Boston University, Boston, MA, 02215, USA

3Department of Earth and Environmental Science, Temple University,

Philadelphia, PA, 19122, USA4TNO – Geological Survey of the Netherlands, NL-3508 TA Utrecht, The Netherlands5School of Earth and Climate Sciences, University of Maine, Orono, 04469, ME, USA

6Present address: Department of Physical Sciences, Virginia Institute of Marine Science,

College of William & Mary, Gloucester Point, VA 23062, USA

*Corresponding author (e-mail: [email protected])

Abstract: Paraglacial coastal systems are formed on or proximal to formerly ice-covered terrainfrom sediments with direct or indirect glacial origin. This review addresses the roles of tectoniccontrols, glacial advances and retreats, sea-level changes, and coastal processes in sediment pro-duction, delivery and redistribution along the paraglacial Gulf of Maine coast (USA andCanada). Coastal accumulation forms are compositionally heterogeneous and found primarily atthe seaward edge of the Gulf’s largest estuaries; their existence is directly attributable to the avail-ability of glacial sediments derived from erosion of weathered plutons within coastal river basins.Multiple post-glacial sea-level fluctuations drove the redistribution of these sediments across themodern lowland and inner shelf. Central to the formation of barrier systems was the paraglacialsand maximum, a time-transgressive phase of relative sea-level fall and enhanced fluvial sandexport c. 2000–4000 years following deglaciation. Vast quantities of sand and gravel werereworked landward during the subsequent transgression and combined with additional riverinesediments to form the modern barrier systems. Today, reduced fluvial sediment loads, anthro-pogenic modifications of barrier and river systems, and sea-level rise have combined to exacerbatelong-term coastal erosion and may eventually force these barriers toward a state of rapidlandward migration.

Glaciations perturb large parts of the global land-scape to a greater degree and over a shorter timeperiodthananyothersurfaceprocess.Grosssedimen-tation rates associated with glaciations are muchhigher than during interglacial periods (Broeckeret al. 1958) and the mean rate of sediment deliveryfrom ice sheets is an order of magnitude higher thanfrom fluvial activity in some of the largest riversystems in the world (Dowdeswell et al. 2010). Gla-ciers leave behind large volumes of easily erodi-ble unconsolidated sediment that are subsequentlyredistributed by non-glacial earth-surface processes(Ryder 1971; Church & Ryder 1972).

Modern coasts located within the sphere of influ-ence of these formerly glaciated terrains are knownas ‘paraglacial coasts’ (Forbes & Syvitski 1994).The concept of a paraglacial environment was first

developed in terrestrial settings to describe the non-glacial processes that were directly conditioned byglaciation and occurred literally ‘beyond the gla-cier’, in environments located around and withinthe margins of glaciation (Ryder 1971; Church &Ryder 1972). It has since come to include not onlythe processes, but also the land systems, landscapesand sediment accumulations that are directly con-ditioned by glaciation and deglaciation. Paraglacialenvironments are unstable or metastable systemsexperiencing transient responses to a variety of non-glacial processes, acting over a number of spatial(metres to hundreds of kilometres) and temporal(years to thousands of years) scales to drive thesystems toward recovery from glaciation (Ballan-tyne 2002a; Hewitt et al. 2002; Knight & Harrison2009; Slaymaker 2009). Paraglacial environments

From: Martini, I. P. & Wanless, H. R. (eds) Sedimentary Coastal Zones from High to Low Latitudes: Similarities andDifferences. Geological Society, London, Special Publications, 388, http://dx.doi.org/10.1144/SP388.15# The Geological Society of London 2014. Publishing disclaimer: www.geolsoc.org.uk/pub_ethics

by guest on March 13, 2014http://sp.lyellcollection.org/Downloaded from

http://sp.lyellcollection.org/


are found in all regions of the globe that underwentor were directly influenced by glaciation during thePleistocene (Fig. 1) (Mercier 2009).

Paraglacial coasts include erosional and depo-sitional coastal landforms such as fjords and coarse-clastic barriers as well as formerly glaciated shelves(Forbes & Syvitski 1994). Such coasts fringe morethan 30% of Northern Hemisphere continentalshelves, and are common throughout northern NorthAmerica, northern Eurasia, and Greenland (Forbes& Syvitski 1994; Forbes et al. 1995; Forbes 2005).In the Southern Hemisphere, the southern tip ofSouth America and ice-free parts of Antarctica areprominent examples (Fig. 1). These coasts retainthe recognizable influence of glaciogenic sedimentsor morphologies (Forbes & Syvitski 1994). They aredistinguished from many of their non-paraglacialcounterparts by a unique combination of (1) gla-cially overprinted landforms (such as fjords anddrumlin fields); (2) nourishment by heterogeneoussand and gravel sources; (3) variable rates of sedi-ment supply governed by substrate erodibility andimpacted by terrestrial and marine processes; and(4) a high degree of compartmentalization (Forbes& Syvitski 1994).

Barrier islands and beaches are common alongmost paraglacial coasts. They are generally small(102–104 m long; 104–109 m3 sediment volume)and relatively isolated (FitzGerald & van Heteren1999). Many of the most prominent and continuousparaglacial barriers flank the mouths of rivers drain-ing glaciated terrains. The goal of this paper is toprovide an idealized stratigraphic and process fra-mework that describes the unique features of barrierformation over periods ranging from hundreds to

thousands of years in such river-associated paragla-cial settings. This is accomplished by comparingthree river-associated barriers along the paraglacialcoast of the Gulf of Maine (GoM), located in the NEUnited States (USA) and SE Canada (Fig. 2), and bycontrasting these systems with river-associated bar-riers formed in other paraglacial and non-glaciatedsettings. We review the dominant controls commonto the formation of these coastal systems (under-lying geology, glacial advance and retreat, relativesea-level changes, sediment redistribution by fluvial,coastal and marine processes) and their affiliateddeposits. Detailed examples of these processes anddeposits are drawn from the evolutionary historiesof the barriers associated with the Kennebec/Androscoggin rivers (Kennebec barrier chain), theSaco River (Saco Bay barrier system) and the Mer-rimack River (Merrimack Embayment barrierchain) (Table 1; Fig. 3a). These details and compari-sons are then used to refine and expand upon earlierconcepts of the paraglacial period by reviewing therole of various contributions of glacial, paraglacialand non-glacial sediment sources to the develop-ment of the GoM barriers. Finally, we assess thefuture of the GoM barrier systems given climatechange and human interference in the naturalsediment-supply pathways.

River-associated paraglacial coasts

Most barriers located along paraglacial coastsare mainland-attached spits or bayhead barriers,formed through a cyclic pattern of formation anddestruction as finite local sediment sources (such

Fig. 1. Extent of formerly glaciated coasts. Northern Hemisphere map modified from Mercier (2009). SouthernHemisphere map constructed from data of Ehlers & Gibbard (2008) and Denton (2011). LGM: Last Glacial Maximum.

C. J. HEIN ET AL.




as drumlins, till bluffs and outwash) are periodi-cally made available as relative sea-level (RSL)rises and shorelines migrate landward across for-merly glaciated terrain (Boyd et al. 1987; Forbes& Taylor 1987; Duffy et al. 1989). They are gener-ally composed of coarse-grained sand and gravel;coarse-clastic beaches up to boulder size are com-mon as fines are progressively removed from thesystem (Orford & Carter 1985; Carter et al. 1989;Forbes et al. 1995). Even tidal flats, with sedimentderived from erosion of glacial deposits, commonlycontain a substantial gravel component.

Longer and more voluminous paraglacial bar-riers directly nourished by glacial sediment aregenerally confined to zones proximal to the maxi-mum ice limit of the Last Glacial Maximum(LGM). On the coasts of outer Cape Cod, Massa-chusetts and southern Long Island, New York, forexample, RSL rise and coastal processes havereworked ample, easily eroded sediment from ter-minal moraines and expansive outwash (sandur)plains since deglaciation. Here, most barriers aremoderately long (2–12 km), 200 m to .1 kmwide, 5–25 m thick and composed primarily ofsand (Rampino & Sanders 1981; FitzGerald et al.1994; FitzGerald & van Heteren 1999).

By contrast, paraglacial coasts located at themouth of major rivers have received sediment notonly from the erosion and direct reworking oflocal glacial deposits, but also from sediments deliv-ered by rivers from the erosion of both glaciogenicand paraglacial upstream sources (Forbes &Syvitski 1994; Ballantyne 2002a; Forbes 2005). InNew England, such river-associated paraglacialcoasts have been influenced by late-Pleistocene andHolocene RSL changes. During many millennia, thesea has eroded and smoothed landforms as theypassed through the nearshore zone, reworking pre-viously deposited glacial, paraglacial and fluvialsediments into swash-aligned barrier islands thatfront extensive backbarrier marshes and tidal flats(Kelley 1987; Hein et al. 2012). These barriers arecomposed chiefly of sand with a minor gravelcomponent.

Coastal evolution in the Gulf of Maine:

dominant controls and common deposits

The GoM covers an area of c. 93 000 km2 fromCape Cod to the southern tip of Nova Scotia(Fig. 2). It is fringed by almost every type

Fig. 2. Gulf of Maine and adjacent New England. Major drainage basins of the Gulf of Maine are shown in dark grey(modified from Kelley et al. 1995a).

GULF OF MAINE PARAGLACIAL BARRIERS




of paraglacial coast and has been ice-free for12 000–17 000 years depending on position alonga north–south gradient. Thus, its shore presents anideal location to investigate glacial and paraglacialcoastal evolution, and to study coastal environ-ments and deposits formed at various stages of

post-glacial recovery. The major controls on coastalevolution in the GoM are antecedent (bedrock-governed) structure, glaciation-dominated sedimentgeneration, river-dominated sediment supply, RSLchange, and sediment redistribution by marine andcoastal processes.

Table 1. Physiography, hydraulics, and sedimentology of western Gulf of Maine estuaries

Kennebec/Androscoggin Saco Merrimack

Estuary physiographyGeological setting Peninsula/deep

embaymentBedrock valley Drowned river valley/

upper bedrock valleyParaglacial coastal

setting (FitzGerald &van Heteren 1999)

Mixed-energymainland-segmented(Type 3b)

Mixed-energy mainland-segmented (Type 3b)

Mixed-energyinlet-segmented(Type 4b)

Spring tidal range (m) 3.0 3.1 2.9Shallow-water wave

height (m)0.4 0.4 0.4

Tidal prism (m3) 101 × 106 8.1 × 106 30 × 106

Estuary type Partially to verticallymixed

Partially to verticallymixed

Partially to verticallymixed

Anthropogenic alterations Dams, dredging dams, jetties, dredging dams, jetties, dredging

Associated river hydrologyDrainage area (×103 km2) 24.9 (combined rivers) 4.6 13.5Length (km) 520 (combined rivers) 210 220Maximum elevation (m) 1200 c. 500 1600Mean discharge (Qw)

(km3/yr)12.9 (combined rivers) 2.2 6.5

Total suspended sediment(Mt/yr)

0.82 (combined rivers) N/A 0.2

Lower-river sedimentologyBedload Medium sand to granules,

megaripples, sand wavesMedium sand to pebbles Medium to coarse sand

Bedforms Transverse bars Megaripples, sand waves Megaripples, sand wavesTerrestrial sediment

sourcesEskers, outwash plains,

plutonsEskers, outwash plains,

plutonsEskers, outwash plains,

plutons

Associated barrier systemBarrier chain Kennebec Barrier Chain Saco Bay Barrier Chain Merrimack Embayment

Barrier ChainThickness (m) 5–10 3–11 5–20Length (km) 11 10 21Volume (m3) 24 × 106 22 × 106 115 × 106

Offshore depositsGlacial deposits Till, glaciomarine clay Till, glaciomarine clay Scattered drumlins and

drumlin-related lagdeposits, thin till coveron drumlins andbedrock, glaciomarineclay

Primary paraglacialsediment features

Palaeodelta lobes at20–30 m, 30–40 mand 50–60 m (totalvolume: 2.1 × 109 m3)

Scattered regressive andlowstand deposits, nolowstand delta (sedimenttrapped in uplandestuaries)

Regressive braidplain delta(volume: 0.9 × 109 m3),lowstand palaeodelta(volume: 1.3 × 109 m3)

Holocene sedimentdeposits

Thin (c. 1 m) transgressivesand and gravel deposits

Thin (c.1 m) transgressivesand and gravel deposits

1–9 m thick mobile sandsheet

Modified and updated from FitzGerald et al. (2005).

C. J. HEIN ET AL.




Structural control

The bedrock in the western GoM and on the adjacentland is dominated by granite, granitic gneiss andmetasedimentary and metavolcanic rocks rangingfrom Precambrian to Middle Palaeozoic in age(Osberg et al. 1985; Lyons et al. 1997; Robinson& Kapo 2003). The major GoM rivers flow to theSSE, generally across the structural grain of bed-rock, except locally where erosion of weaker strataby both fluvial and glacial scouring controls briefstrike-aligned courses (Kelley 1987).

Structural controls related to Palaeozoic tec-tonism have played a dominant role in shap-ing the variability between barrier systems in thewestern GoM. For example, the southward-facingKennebec/Androscoggin coastline is largely pro-tected from NE waves by bedrock headlands, result-ing in the development of swash-aligned barriers(Buynevich 2001). By contrast, the MerrimackRiver drains into structural lowlands, providingample accommodation for the development of the

associated NE-facing (drift-aligned) barriers, butless protection from dominant waves (Goldsmith1991; FitzGerald et al. 2005). Faulting and fold-ing patterns control the configuration of individ-ual coastal compartments, and the gradients of thecoastal lowland and proximal continental shelf con-trol accommodation available for backbarrier andoffshore deposits. Barriers throughout the GoMare commonly pinned to bedrock promontories,with abundant granitic plutons in SW Maine serv-ing as headlands anchoring and protecting beachesand marshes (Kelley 1987). Tidal inlets are typi-cally situated in drowned bedrock-controlled rivervalleys (FitzGerald et al. 2002).

Advancing and receding ice sheets

Quaternary glaciations resulted in significant sedi-ment and bedrock erosion across the GoM andin extensive scouring of bedrock-controlled flu-vial channels. Physical and chemical weatheringof intrusive granitic (with minor gabbro and

Fig. 3. (a) Drainage basins affiliated with the river-associated Kennebec/Androscoggin, Saco and Merrimackparaglacial barrier systems. (b) Distribution of plutons and sandy glacial deposits in the drainage basins and of thebarriers and offshore deltas associated with these rivers (modified from FitzGerald et al. 2005).





granodiorities) plutons common to inland regionsthroughout the GoM (Fig. 3b), as well as glacialexcavation of saprolite, generated much of thesand-rich sediment that was later reworked intoglacial and paraglacial deposits (Hanson & Cald-well 1989; Thompson et al. 1989; FitzGerald et al.2005). Ice sheets of the most recent glaciations,the Illinoian and Wisconsinan, left behind non-stratified glaciogenic sediments throughout theGoM and in adjacent New England. These taketwo dominant forms: (1) drumlins, which have apatchy, clustered distribution and were deposited inpart during the Illinoian glaciation, and (2) coarse-grained till of late Wisconsinan age (Stone et al.2006). This latter deposit drapes bedrock surfacesthroughout the region as a thin veneer. Contem-poraneously, meltwater streams drove the accumu-lation of extensive, quartz-feldspar-rich, stratifiedice-contact sediments throughout the region, bothunder and in front of the ice sheet. The distribu-tion of sandy eskers, outwash plains and fans, andcoarse sandy ice-marginal deltas (Fig. 3b) reflectthe ice-sheet extent and recession (FitzGeraldet al. 2005).

The Laurentide Ice Sheet of the Wisconsinanglaciation reached its maximum extent, beyond the

southern boundary of the GoM (Fig. 1a), between28.0 and 23.7 thousand calendar years before pres-ent (ka) (Balco et al. 2002). As it then recededrapidly northward between c.17 and 15 ka (Bornset al. 2004), rising RSL in the GoM submergedisostatically depressed areas immediately upondeglaciation (Fig. 4) (Bloom 1963). This submerg-ence culminated in a highstand of RSL severaltens of metres above modern mean sea level (MSL)(Fig. 4), prelude to a subsequent set of complexRSL changes that served to redistribute the sedi-ments produced by the glaciers.

Relative sea-level change

Late-Pleistocene and Holocene RSL changes (Fig.4) resulted from the combined forcings of globaleustatic sea-level rise and regional glacio- andhydro-isostatic adjustments. These RSL changeswere largely similar across the GoM coast, varyingonly in amplitude and timing (Fig. 4).

While near its maximum extent, the massof the Laurentide Ice Sheet depressed the litho-sphere, including that of the GoM region. Theoutward displacement of the underlying astheno-sphere formed a peripheral crustal forebulge tens

Fig. 4. Sea-level history of the western Gulf of Maine. Northern Massachusetts RSL curve is modified from Hein et al.(2012). Coastal Maine RSL curve is modified from Kelley et al. (2010).

C. J. HEIN ET AL.




to hundreds of kilometres beyond the glacial front(Daly 1934; Barnhardt et al. 1995). The ice sheetreceded northward as climate warmed, exitingpresent-day Massachusetts by c. 17–16 ka andcoastal Maine by c. 15 ka (Borns et al. 2004).Eustatic sea-level rose rapidly during this period.The continued isostatic depression of the crustbelow contemporary MSL due to delayed reboundexposed the GoM area to immediate marine flood-ing of land (Bloom 1963). Isostatic rebound haltedthis process, and the maximum marine limit wasreached at 31–33 m above modern MSL in Massa-chusetts (Fig. 4) (Stone & Peper 1982; Oldale et al.1983; Ridge 2004; Stone et al. 2004) and at 70–75 m above MSL in coastal Maine (Fig. 4) (Thomp-son et al. 1989; Kelley et al. 1992).

Continued isostatic rebound resulted in rapidRSL fall as regional uplift outpaced eustatic sea-level rise. RSL stabilized as rebound deceleratedand temporarily matched the rate of eustatic sea-level rise. This produced a relative marine lowstandacross the GoM that ranged from c. 241 m MSL at13–14 ka (Oldale et al. 1993) in northern Massa-chusetts to 260 m MSL at 12.5 ka in centralcoastal Maine (Barnhardt et al. 1997) and 265 mMSL at 11.5 ka along eastern Nova Scotia (Fig. 4)(Stea et al. 1994).

Following this regional relative lowstand, RSLrose very rapidly (c. 40 mm/yr) in coastal Mainefor c. 1000 years as eustatic sea-level rise greatlyoutpaced isostatic rebound. RSL rise then slowedabruptly to only 1–1.5 mm/yr at 11.5 ka as the fore-bulge that migrated north through coastal Mainecollapsed (Barnhardt et al. 1995). This so-calledslowstand (Fig. 4) (Kelley et al. 2010, 2013) alongthe Maine coast lasted until 7.5 ka. No such detailedinformation is available for the southern GoM,where sparse post-lowstand data indicate that sea-level rose at a time-averaged rate of c. 4 mm/yrbetween 13.5 and 6 ka.

RSL rise in the southern GoM gradually slowedto less than 2 mm/yr by 4–5 ka (Fig. 4). By con-trast, coastal Maine saw one final period of rela-tively rapid (c. 7.5 mm/yr) RSL rise between 7.5and 5.5 ka, followed by 5 ka of much slowerchange (less than 1 mm/yr) to its modern verticalposition (Fig. 4) (Barnhardt et al. 1995; Kelleyet al. 2010, 2013).

Deposits associated with changing sea-levels

In governing the changing extent of the GoM,RSL rise and fall set the stage for large-scaleerosion and redistribution of glaciogenic sedimentacross the modern coastal zone and adjacent shal-low shelf, generating a range of coastal, marineand glaciomarine landforms and deposits (Table 2;Fig. 5).

Following the LGM, receding ice sheets acrossthe GoM left behind extensive stratified and non-stratified deposits, including sandy eskers, outwashplains and fans, coarse sandy ice-marginal deltasand till (Stone et al. 2006). In the GoM region,this latter deposit is an unsorted mixture of mud,sand and gravel in various proportions and litho-logical compositions, depending on the materialeroded and on the influence of eroding and trans-porting processes.

Flooding of land immediately following degla-ciation resulted in the deposition of glaciomarinesilt and clay across the modern shelf and adjacentterrestrial environments, occasionally extendingseveral hundred kilometres inland from the mod-ern shoreline (Thompson & Borns 1985). Althoughthis deposit smoothed antecedent topography, itprovided limited sediment for the later develop-ment of barriers and beaches. Glaciomarine siltand clay was generally deposited in environmentsof low wave energy along the coast and in riverestuaries. Its strongly micaceous composition indi-cates derivation from glacially eroded metamorphicrocks (Kelley 1989).

Falling RSL following the 17–15 ka highstandforced rapid shoreline progradation. Sedimentsdelivered to the contemporary coast were reworkedcross- and along-shore by coastal processes, form-ing sandy parasequences and regressive deltas thatreflect the seaward migration of the coastal zone(Oldale et al. 1983; Barnhardt et al. 1997; Belknapet al. 2002; Kelley et al. 2003). Coastal and fluvialprocesses modified the landscape and redistrib-uted sediments across discrete segments of emer-gent glaciomarine plains. Discontinuous, disparateremnants of regressive beaches and spits and fluvialterraces have been identified between the high-stand and modern coastlines along much of theGoM (Retelle & Weddle 2001). Near several rivermouths, these regressive deposits formed exten-sive and thick strandplains and braidplain deltas(braided deltaic plains) as ample fluvially derivedsand and fine gravel was efficiently reworked along-shore by waves and tides (Fig. 3b; Table 2). TheSanford-Kennebunk braidplain delta, for example,located upstream of the Mousam River near Wells,Maine (Fig. 3b), has a surface area of 125 km2 anda thickness of 5–14 m. Its volume of 1.5 × 109 m3

(Tary et al. 2001) exceeds that of all individualmodern-day GoM barriers. The Brunswick braid-plain delta, located at the former confluence of theKennebec and Androscoggin rivers, covers an areaof 25 km2 and is 5–15 m thick. It gradually stepsdown in elevation from west to east, reflectingunderlying seaward-dipping bedrock graduallyexposed by falling RSL (Crider 1998). Similarregressive units have also been found on the shal-low shelves proximal to several river mouths. For





Table 2. Sedimentological units common to the coastal zone along river-associated paraglacial coasts of the GoM

Deposit Approximateage in GoM

Relativesea-level (RSL)

conditions duringdeposition

Sedimentology Direct sedimentsource

Environment/mechanism of deposition

Associated features Contributions tobarriers

Till (non-stratifiedice-contactdeposits)

100–16 ka Galling sea-levelduring glacialadvance, glaciallowstand, risingRSL duringdeglaciation

Non-sorted,non-stratifiedsediment with amatrix of sand andlesser amounts of siltand clay containingscattered gravel clastsand few largeboulders

Erosion of bedrockand preglacialsediments byglaciers

Direct deposition byglaciers

Drumlins; crag-and-tailice-streamlined deposits;kames; ground,washboard and deglacialmoraines

Drumlins formpinning points forbarriers, minorsedimentcontributions

Glaciofluvialdeposits

100–16 ka Glacial lowstand,rising RSL duringdeglaciation

Bedded gravel, sand andmud

Bedrock erosion byglaciers; erosionand reworking ofglacial deposits

Deposition by meltwaterin terrestrialenvironment

Eskers and outwash plains inall river basins (Fig. 3b),glaciomarine deltas,grounding-line fans

Erosion of deposits inriver basinsprovides sedimentfor barriers

Glaciomarinesilt-clay deposits

21–13 ka RSL rise andhighstand

Silty clay, fine sand andsome fine gravel,containing dropstonegravel clasts; highlycompacted anddewatered; commonlysandy in the upperfew metres, overlyingthicker silty clay

Bedrock erosion byglaciers; erosionand reworking ofglacial deposits

Transport to marineenvironment bymeltwater, depositionby settling in marineenvironment

Presumpscot Formation inMaine (Bloom 1963),Boston Blue Clay inMassachusetts (Kaye1961)

Nearly ubiquitousdeposit that formunderpinnings ofbarriers (Figs 8, 11& 13)

Regressive shorelinedeposits

17–12 ka Sea-level highstandand RSL fall

Sand and fine gravelforming coastallandforms (barrierbeaches, spits,regressive fluvialdeltas)

Erosion andreworking ofsandy glacialdeposits

Transport to highstand/regressing shorelineby meltwater andmeteoric water;reworking by waves,currents, tides andwind action alonghighstand andregressive shorelines

Progradational deltas,beaches, spits, dunes;braidplain deltas (BPD):Sandford-KennebunkBPD, Brunswick BPD,Merrimack BPD, parts ofKennebec Riverpalaeodelta

Deposits belowmodern meansea-level werepartially eroded andreworked bylate-Pleistocene/Holocenetransgression, thuscontributing coarsesediments to barriersystems

Lowstandpalaeodeltas

14–12 ka RSL fall andsea-levellowstand

Fine to coarse, stratifiedsand and silt;bottomsets dominatedby silt and clay;foresets dominated byfine, well-sorted sandand silt; topsetsdominated bymedium to coarsesand

Erosion andreworking ofsandy glacialdeposits

Transport to lowstandshoreline by meteoricwater; reworked bywaves, currents, tidesand deposited asseaward-progradingbottomset, foreset andtopset beds

Kennebec palaeodelta(Fig. 8); Merrimackpalaeodelta (Fig. 13a);Penobscot Bay deltaicdeposits

Palaeodeltas partiallyeroded andreworked bylate-Pleistocene/Holocenetransgression, thuscontributing coarsesediments to barriersystems

C.

J.H

EIN

ET

AL

.

by guest on March 13, 2014

http://sp.lyellcollection.org/D

ownloaded from



Shelf sand sheets 12 ka topresent

Rapidly, then slowlyrising RSL

Well-sorted fine tomedium sand withminor quantities ofsilt and gravel

Erosion andreworking ofregressive andlowstanddeposits; erosionof sandy glacialdeposits in riverbasins

In situ erosion andreworking of shelfdeposits; transport tolowstand shoreline bymeteoric water;reworked by waves,currents and tides

Mobile sand sheet inMerrimack Embayment;thin transgressive sandsand gravels in Kennebecand Saco Bays

Sand sheets arefraction of shelfdeposits notincorporated intobarriers during theirformation; activeexchange of shelfand barriersediment

Estuarine deposits 8 ka to present Rapidly, then slowlyrising RSL

Largely massive,moderatelywell-sorted fine sandand silt, dominated byquartz with traces oforganic material

Erosion andreworking ofregressive andlowstanddeposits; minorbedrock andupland erosion(fluvial inputs)

Onshore transport ofshelf sediments;transport tobackbarrier by tidesthrough inlets and bywaves as overwashacross barriers

Backbarrier tidal channelsand tidal flats; inlet ebb-and flood-tidal deltas;common living bivalvespecies includeMercenaria, Ostreidae,Ensis directus andPteriomorpha

Underlie barriers; fillmostaccommodationbehind barriers;active exchange ofsediment betweenestuaries andbarriers throughinlet processes

Barrier lithosome 6 ka to present Slowly rising RSL Moderately sorted, fineto very coarse sand;commonly parallellaminated; coarserlayers contain somegranules and finepebbles; finer layerscontain very fine sandand traces of silt

Erosion andreworking ofregressive andlowstanddeposits; minorbedrock andupland erosion(fluvial inputs)

Onshore migration ofshelf deposits; directfluvial contributions;reworking alongshoreby waves, tides andcurrents

Barrier beaches; dunes;sandy intertidal zones;American dunegrasscommon in supratidalareas

Some active sedimentexchange betweenbarriers

Salt-marsh 4 ka to present Slowly rising RSL Fine-grained clastic andorganic matter, fibricand hemic peatinterbedded with finesand, silt and clay;typically greater than30% organic

In situ production oforganicsediments;inorganicsediments largelyfluvially derived

In situ production;inorganic sedimentstransported tobackbarrier by tidesthrough inlets andbydirect fluvial influxand overland flowproximal to uplandareas

Marsh grasses includeSpartina alterniflora(cordgrass), S. patens(marsh hay), J. gerardii(black rush), Phragmites(common reed), andIchnocarpus frutescens(shrub) (Jacobson &Jacobson 1989)

Marsh peats underliemainland-proximalsides of barriercomplexes

GU

LF

OF

MA

INE

PA

RA

GL

AC

IAL

BA

RR

IER

S

by guest on March 13, 2014

http://sp.lyellcollection.org/D

ownloaded from



example, off the mouth of the Merrimack River,abundant coarse sediment produced a 10-km-widestrandplain that parallels the shore for 16 km andis 4–15 m thick (Barnhardt et al. 2009). Its uppersurface is marked by a lag deposit of coarse sandand fine gravel, formed during the later Holocenetransgression (Hein et al. 2013).

Deceleration and subsequent cessation of RSLfall led to the deposition of lowstand delta lobes at

the mouths of several major rivers discharging intothe contemporary GoM, most notably the Merri-mack and Kennebec/Androscoggin rivers (Oldaleet al. 1983; Barnhardt et al. 1997). No equivalentdeltaic deposits have been uncovered offshorethe Piscataqua (New Hampshire/Maine border),St. John (New Brunswick) and Saco rivers. Somedeltaic sediments were deposited in Penobscot Bay(Fig. 3a) before c. 9 ka. However, sediment supply

Fig. 5. Idealized stratigraphic sections through (a) river-associated paraglacial barrier island and (b) shallow-shelfpalaeodelta sequence offshore a river-associated paraglacial barrier. Thicknesses given for each unit are approximateand estimated from data in McIntire & Morgan (1964), Rhodes (1973), van Heteren (1996), Buynevich (2001),Buynevich & FitzGerald (2001), Stone et al. (2006), Barnhardt et al. (2009) and Hein et al. (2013).

C. J. HEIN ET AL.




was largely cut off as the Penobscot River lost itscompetence to transport sand when isostatic upliftin its headwaters led to a drainage-divide shift andassociated loss of water and sediment to the Kenne-bec River (Kelley et al. 2011). As much as 10 m ofHolocene mud now covers this pseudo-palaeodelta(Belknap et al. 2002; Kelley et al. 2011). Erodedeastward-facing fossil shoreline features abound inSaco Bay, but these are very thin (less than 1 m)(Kelley et al. 2003).

The period of rapid RSL rise following thelowstand led to the surficial erosion of regressiveand lowstand deposits and to the formation of athin (,1 m) transgressive sand-and-gravel unit onnewly formed shelves. Muddy sands started toaccumulate during the slowstand period in Maine(Fig. 4) to form wedge-shaped estuarine units overeroded glaciomarine mud, thickening toward themodern shoreline and showing large regional dif-ferences (Barnhardt et al. 1997). Off the KennebecRiver, the estuarine unit is up to 10 m thick. InSaco Bay, a similar but muddier and much thinner(up to several metres) unit marked by an intertidalto shallow-subtidal fauna extends from the pres-ent shoreline out c. 2 km offshore (Kelley et al.2005; D. Barber, pers. comm.).

The broad time-transgressive sand sheets thatoccupy the shallow shelves between the lowstandand modern shorelines evolved from sediment notonly reworked from early transgressive barriersand intertidal/supratidal sand shoals (Oldale 1985;Oldale et al. 1993; FitzGerald et al. 1994), but alsocontributed by direct fluvial input and by wave-driven erosion of glacial and regressive depos-its exposed at the seabed. As the rate of RSLrise decreased, coastal processes associated withslowly retrograding shorelines drove sedimentsfrom coastal and river sources ever farther onshore,

eventually forming proto-barriers (McIntire &Morgan 1964; FitzGerald et al. 1994; van Heteren1996; Buynevich 2001; Buynevich & FitzGerald2003). In many areas, sand and silt derived fromfluvial and nearshore sources were deposited inbackbarrier lagoons, tidal inlets and channels, andflood-tidal deltas as these proto-barriers length-ened and widened to their modern dimensions (vanHeteren 1996; Buynevich 2001; Hein et al. 2012).

Freshwater and brackish marsh deposits ini-tially formed at the leading edge of the transgres-sion (McIntire & Morgan 1964). Most observationssuggest a rapid transition from barren tidal flats towell-developed high marsh around or after 4 ka(Oldale 1989; FitzGerald et al. 1994; Kelley et al.1995b). The cause and timing of salt-marsh expan-sion in New England estuaries has generally beenattributed to a late-Holocene decrease in the rateof RSL rise. However, local settings and processesmust not be neglected: salt-marsh developmentbehind the Saco barriers, for example, is hypoth-esized to have coincided with a reduction in thewidth of the palaeolagoon, when an inferred palaeo-barrier system was replaced by a more landwardseries of barriers. Previous to this event, a combi-nation of rapid RSL rise and strong winds acrosswide lagoons had maintained open-water conditions(Van Heteren 1996). Today, the salt-marsh systemsare dominated by Spartina alterniflora, Spartinapatens, Juncus roemerianus, Juncus gerardii andthe shrub Iva frutescens.

Riverine sediment redistribution

Rivers draining c. 160 000 km2 of land dischargeinto the GoM (Fig. 2; Table 3). Their drainagebasins contain voluminous glaciofluvial and gla-ciomarine deposits. Together, the rivers annually

Table 3. Major rivers draining into the Gulf of Maine

Location of river mouth States/provinces indrainage basin

Drainage-basinarea (km2)

Charles River Boston Harbor, MA (USA) Massachusetts (USA) 1593Merrimack River Merrimack Embayment, MA (USA) Massachusetts/New Hampshire

(USA)13 507

Saco River Saco Bay, ME (USA) New Hampshire/Maine (USA) 4610Androscoggin

RiverPopham, ME (USA) Maine (USA) 9376

Kennebec River Popham, ME (USA) Maine (USA) 15 618Penobscot River Penobscot Bay, ME (USA) Maine (USA)/New Brunswick

(Canada)23 245

St. Croix River Passamaquoddy Bay, ME (USA)/New Brunswick (Canada)

Maine (USA)/New Brunswick(Canada)

3885

St. John River St. John, New Brunswick (Canada) New Brunswick (Canada) 7601Annapolis River Annapolis Basin, Nova Scotia (Canada) Nova Scotia (Canada) 7600





deliver c. 950 × 106 m3 of freshwater and more thana million cubic metres of suspended sediment(Kelley et al. 1995a). Directly or indirectly, via tem-porary storage on the inner continental shelf, sedi-ment input from these rivers has supplied nearlyall of the sediment for the development of mostbarrier and backbarrier systems of the GoM (Fitz-Gerald et al. 2005; Kelley et al. 2005; Hein et al.2012); without them barriers in the GoM would berare or absent, even in areas marked by extensiveinland sand sources (FitzGerald et al. 2002). Sedi-ment feeding the Saco and Merrimack barriers wasdominantly provided by the rivers themselves(Kelley et al. 2003; FitzGerald et al. 2005; Heinet al. 2012). Elsewhere, marine erosion and rework-ing of shelf and post-glacial fluvial deposits, and,to a lesser extent, coastal erosion of drumlins(Chute & Nichols 1941; Dougherty et al. 2004;Hein et al. 2012), have also played a role.

Modern fluvial sediment delivery is episodic,dominated by high-discharge events associatedwith precipitation from the passage of hurricanesand extratropical storms (Hill et al. 2004), and byannual spring floods (freshets) governed by meltingsnow and enhanced rainfall (Brothers et al. 2008).Although these high-discharge events feed sedimentto barrier systems across the GoM (FitzGerald et al.2002), they are especially important at the mouth ofthe Saco River, where nearly all sediment is trappedin the estuary during normal flow conditions (Broth-ers et al. 2008).

Modern sediment redistribution by marine

and coastal processes

Estuarine sediment trapping, salt-marsh develop-ment and barrier dynamics continue to the pres-ent day. Offshore, the presence of abundant activebedforms along the shallow shelves off the Merri-mack and Kennebec rivers indicates that paragla-cial deposits undergo varying degrees of reworkingby waves, tides and currents (Dickson 1999; Heinet al. 2007). Prevailing summer wind throughoutthe GoM is from the SSE and produces low-energywave conditions and swells (Bigelow 1924; Jensen1983). Spring, autumn and winter prevailing windsare from the WNW, and storm events are associatedwith the passage of high-pressure fronts from theNW, inland low-pressure systems and NE stormsthat parallel the coast (Hill et al. 2004). Nor’easters(macrostorms driven by northeastern wind) accountfor at least 50% of all winter storms (Dolan & Davis1992) and produce the strongest wind and wavesacross the GoM. Along the mixed-energy, east- toNE-facing, drift-aligned coastline at the mouthof the Merrimack River, these nor’easters drivesoutherly longshore transport at a rate of 38 000

(Castle Neck) to 150 000 (Plum Island) m3/yr(Smith 1991). By contrast, Saco Bay is more shel-tered; here, wave refraction around headlandsleads to a net NE longshore transport estimated atonly 10 000–16 000 m3/yr (Kelley et al. 2005).The southward-facing coast near the mouth of theKennebec/Androscoggin river system is largelyprotected from NE waves by bedrock headlands.This swash-aligned system experiences a clock-wise sediment gyre driven by tidal currents andstorm waves (FitzGerald et al. 2000).

Tides in the GoM are semidiurnal and rangesgenerally increase from 2.5 m at Cape Cod in theSW to c. 17 m in the Bay of Fundy in the NE. Theshorelines at the mouths of the Kennebec/Andros-coggin, Saco and Merrimack rivers all experiencesimilar spring tidal ranges (2.9–3.1 m), althoughtidal prisms vary markedly among them as a func-tion of the area and nature of backbarrier envi-ronments (percentage covered by tidal flats, baysand/or marshes), river dimensions and anthropo-genic modifications (Table 1) (FitzGerald et al.2005). Circulation in the present GoM, which hasa mean depth of c. 140 m, is generally cyclonicand dominated by buoyancy-driven coastal currents(Beardsley et al. 1997; Lynch et al. 1997; Lentz2012) that have little impact on the coast exceptduring major storms.

River-associated paraglacial barrier

development: examples from three

barrier complexes

The combination of variable structural controls,sediment sources and supply rates, RSL changesand hydrographic regimes has produced a diverseset of barrier systems in the GoM. Three of thebest studied are the Kennebec barrier chain (Kenne-bec and Androscoggin rivers), the Saco Bay barriersystem (Saco River) and the Merrimack Embay-ment barrier chain (Merrimack River) (Fig. 1; Table1). Each has distinctive features and a distinctmiddle- to late-Holocene history of barrier forma-tion and development. Evolutionary models forthese systems are used here to contrast barrierformation along river-associated paraglacial coastsin different settings of sediment supply and accom-modation, and exhibiting diverse intrasystemvariability.

Kennebec barrier chain

Located at the mouth of the Kennebec/Androscog-gin river system, the Kennebec barrier chain (KBC)is located along a fjard-type paraglacial coast; thatis, one formed along flooded glacial valleys withmoderately shallow depths and moderate relief.

C. J. HEIN ET AL.




The KBC consists of approximately 30 coastalaccumulation forms (welded barriers and mainlandbeaches) stretching between New Meadows Bay(west of Small Point Harbor) and Sheepscot Bay(Fig. 6). These are subdivided into four physio-graphic provinces (West, Central, East–Centraland Eastern complexes) that span nearly 2508 inshoreline orientation (Fig. 6).

The KBC is fed by the Kennebec and Andros-coggin rivers, which join at Merrymeeting Bayc. 20 km north of the estuary mouth, and con-tinue toward the GoM in a narrow bedrock-carvedchannel (Figs 3a & 6). The confluent, lower Kenne-bec River is a partially mixed to stratified mesotidalparaglacial estuary with seasonal variations in riverdischarge (Fenster & FitzGerald 1996). This river

Fig. 6. Kennebec barrier chain showing sandy Holocene coastal landforms with a range of orientations. Fourphysiographic compartments are distinguished on the basis of morphology and sedimentology (Buynevich 2001): (1)western (Small Point Harbor/Cape Small barriers), (2) central (Seawall/Popham barriers), (3) east-central (SagadahocBay barriers) and (4) eastern (Reid State Park barriers).





system continues to supply coarse-grained sedimentto the coastal region, especially during spring fre-shets (Fenster et al. 2001). Most of this sedimentis derived from upland outwash deposits, withcompositions inherited from distinct bedrock lithol-ogies (Borns & Hagar 1965). Today, the KennebecRiver Estuary seaward of the Merrymeeting Bayreceives a mixture of sediments from both the Ken-nebec and Androscoggin rivers; their sources canbe differentiated on the basis of contrasting minera-logies that reflect the compositional differencesof the respective river drainage basins (FitzGeraldet al. 2002).

The KBC exemplifies an indented paraglacialcoast that has experienced an active but localizedriverine sediment contribution to Holocene accumu-lation forms. All incipient transgressive barriersproximal to the Kennebec River mouth were estab-lished c. 4.6 ka (Buynevich 2001). Away fromthe direct river influence, along sediment-starved

complexes supplied primarily from the aban-doned early Holocene deposits of the KennebecRiver palaeodelta, transgressive barriers did notform until 1.2 ka (Buynevich 2001). The proximityto fluvial and palaeodeltaic sediment reservoirs andchanges in three-dimensional accommodation spaceduring decelerating RSL rise have been the majorfactors controlling the timing of barrier emplace-ment and progradation (Buynevich & FitzGerald2001, 2005), degree of compartmentalization, andsediment volume (Barnhardt et al. 1997). Thus,despite their proximity and similarities, each com-plex of the KBC has a unique evolutionary history.

Situated on the western margin of the Kenne-bec estuary mouth, the Popham Beach System con-sists of a 4-km-long sandy barrier subdivided intothree segments (Riverside, Hunnewell and Seawallbeaches) (Fig. 7a) anchored to pegmatitic bedrockheadlands and isoclinally folded metasedimentaryformations. Intertidal shoals connect the western

Fig. 7. Stratigraphy of Hunnewell Beach (modified from Buynevich et al. 2004). (a) Satellite image of Popham Beachcomplex, with Riverside, Hunnewell and Seawall beaches, showing transect location (for overall location, see Fig. 6).(b) Stratigraphic cross-section extending from the modern beach to Silver Lake, showing the extent of bothtransgressive and regressive facies, with the latter punctuated by a series of buried erosional scarps (optical chronologyfrom Buynevich et al. 2007). (c) Analogue GSSI GPR section collected with a 200 MHz antenna and showingtransgressive barrier core overlying bedrock, in turn overlain by a prograded barrier sequence. MHW, mean high water;MSL, mean sea-level; m MSL, metres with respect to mean sea-level; VE, vertical exaggeration.

C. J. HEIN ET AL.




and eastern ends of Hunnewell Beach to Fox Islandand Wood Island, respectively (Figs 6 & 7). Exten-sive geophysical (ground-penetrating-radar, GPR)and sediment-core data show that the PophamSystem contains at least 14 × 106 m3 of Holocenesand (Buynevich & FitzGerald 2005). A texturallyand compositionally submature, fluvially derivedtransgressive unit at 21.5 to 27.0 m MSL is con-fined to the landward part of this system, andcan be traced alongshore for over 4 km, formingthe core of the Hunnewell and Seawall beaches.Changes in shoreline exposure and decreasingproximity to the estuary mouth as the transgressionproceeded led to an increase in maturity of sub-sequently deposited barrier sediments. Minimallyreworked facies of fluvial origin were coveredby younger parasequences composed of relativelymature, micaceous sand with evidence of estuarinesorting (Riverside Beach).

East of the Kennebec River Estuary, the EasternComplex of Reid State Park (Fig. 6) demonstratesthe effect of structural controls and relict offshoresediment sources on its Holocene development.This complex consists of two segments separatedby a bedrock ridge: Mile Barrier to the east andHalf Mile Barrier to the west (Fig. 6). Mile Barrieris backed by a salt-marsh along its entire length,extending up to 600 m into the bedrock re-entrants.A small, man-made bedrock-bound inlet providesexchange between the ocean and the backbarrier.Numerous ledges and islands represent offshorestructural extensions of bedrock ridges. These struc-tural extensions can be followed for more than10 km offshore in seismic-reflection profiles (Fig.8) (Belknap et al. 1989; Buynevich et al. 1999),where they likely served as anchor points and com-partmentalization elements for early barriers. Thecoarse- to very coarse-grained sand, particularlyalong the Mile Barrier, forms a steep (reflective)beach backed by a narrow dune ridge. GPR profilesand vibracores show an extensive salt-marsh unitbeneath the barrier lithosome on either side of

Todd’s Head promontory, which separates Mileand Half Mile Beaches (Buynevich & FitzGerald2002). The presence, extent and age of this unit indi-cate that the salt-marsh functioned as a supratidalbackbarrier platform for overwash and aggradationas early as 3000 years ago (Buynevich 2001).

The regressive/aggradational phases of barrierdevelopment along the KBC were marked by punc-tuated barrier progradation and dune development(Buynevich & FitzGerald 2001; Buynevich et al.2004). Barrier progradation phases, which lasted fortens to hundreds of years, are recorded in the strati-graphy of the Popham Beach System as uniform tocomplex progradational units truncated by erosionalscarps that are marked by high concentrations ofheavy minerals (Fig. 7c). By contrast, along manyof the small, sand-starved systems of the West-ern and East–Central complexes (Fig. 6), regres-sive barrier elements range from a single beachor dune ridge to widespread wind-driven aggrada-tional units. Formation of these barriers, includingsome exhibiting limited progradation, was likelystimulated by a combination of a deceleration inRSL rise and increasingly efficient alongshore sanddelivery facilitated by sediment filling of re-entrants(Buynevich 2001). Localized areas of retrograda-tion (Buynevich & FitzGerald 2005) reflect intenseseaward-side barrier erosion as indicated by exten-sive dune scarps and heavy-mineral concentrationscaused by storm surges and/or tidal-inlet migration(Fig. 7c) (Buynevich et al. 2004).

Depending on sediment supply and wind direc-tion (e.g. the large south-facing beaches of theCentral Complex v. the small north-facing pocketbeaches within the Western Complex), internalsediment reworking and landward transport haveresulted in growth and limited migration of para-bolic and transgressive (climbing) dunes (Buyne-vich & FitzGerald 2003). The dune facies thatpresently comprise up to 40–50% of the barrierlithosome represents no more than 5–15% (c.200 years) of the barrier history.

Fig. 8. Shore-normal stratigraphic cross-section across Kennebec palaeodelta (modified from Barnhardt et al.1997).Cross-section is interpreted from high-resolution boomer seismic-reflection data, ground-truthed with fourvibracores. MSL, mean sea-level. Vertical exaggeration ¼ 25×. Transect location shown in Figure 6.





Saco Bay barrier system

Two prominent headlands composed of Palaeozoicmetamorphic rocks (Osberg et al. 1985) – ProutsNeck and Biddeford Pool – delimit the 15-km-long Saco Bay barrier system (Fig. 9). This systemis composed of relatively narrow barriers that arecompartmentalized by bedrock ridges and pinna-cles, tidal inlets and, in the south, the Saco RiverEstuary. Differential erosion of Palaeozoic base-ment rocks by Palaeogene and Neogene fluvial pro-cesses and Quaternary glacial processes has resultedin irregular topography that has given the coastalarea its islands, headlands and embayments.

The Saco River, which presently contributes10 000–16 000 m3/yr of sand to Saco Bay (Kelleyet al. 1995c), primarily during spring freshets(Brothers et al. 2008), has supplied much of thesediment to the Saco Bay barrier system overthe course of the Holocene (Kelley et al. 1995c;FitzGerald et al. 2005). During deglaciation, largeamounts of glaciofluvial sand and gravel were

deposited between the ice sheet and the ocean atmany locations in the Saco River drainage basinand on the modern shallow shelf. These easily erod-ible sediments have been a particularly importantsource for the Saco Bay dune, beach and shorefacesystems (D. Barber, pers. comm.). Additional sedi-ment has come from the erosion of glacial depositsnear Prouts Neck (Kelley et al. 2005).

GPR and core data allow the distinction of onelongshore and three cross-shore barrier-sequencetypes within the Saco Bay barrier system (Fig. 10)(Van Heteren 1996). Highly diverse morphostrati-graphies are reflective of several phases of middle-to late-Holocene barrier development. The SacoBay barrier system shows a strong imprint of over-stepping, fostered by the presence of numerousanchor points in the form of bedrock pinnacles,Pleistocene till mounds and other highs (Fig. 11)(van Heteren 1996). Backbarrier peat extendsbeneath only limited parts of the present-day bar-rier system, primarily on its landward margin, nextto the modern salt-marshes. Inorganic backbarrierfacies are much more widespread, underlying muchof the modern barrier lithosome (Fig. 10) and occur-ring in the subsurface of the present-day inner shelf.This evidence of an ancient, wide, open-waterlagoon or estuary extending well beyond the loca-tion of the modern barrier system indicates that anearly barrier system formed seaward of the mod-ern barriers during the early to middle Holocene(van Heteren 1996), possibly coincident with thesea-level slowstand of c. 11.5–7.5 ka.

The precursor barriers deteriorated between 7.0and 4.5 ka. This probably occurred in conjunctionwith the initial establishment, and subsequent long-shore accretion, of proto-barriers that formed in theapproximate location of the modern barriers (vanHeteren 1996). This barrier morphosome initiallyabutted the mainland, at the edge of the formerlagoon. In time a narrow backbarrier region devel-oped through submergence of upland induced byrising RSL, allowing salt-marsh development andexpansion (van Heteren 1996). Limited burial ofbackbarrier salt-marsh as a result of slow, RSL-induced barrier retrogradation has not compensatedsalt-marsh encroachment onto the mainland; thishas resulted in the gradual widening of the backbar-rier areas (van Heteren 1996).

The barrier system underwent limited retrogra-dation and continued longshore accretion until c.1 ka. Tidal inlets narrowed and the barrier systemgrew more continuous during this time. Finally, inthe last 1000 years, most inlets closed or narrowedsubstantially to their modern dimensions and thebarrier has primarily prograded seaward (vanHeteren 1996).

The modern Saco Bay coastline has a dis-tinct log-spiral (or zeta [z]) shape, reflecting the

Fig. 9. Distribution of barrier and backbarrier sedimentswithin the Saco Bay barrier chain. No lowstand delta ispresent in Saco Bay, owing to the trapping of glacial andparaglacial sediment in upstream river plains, lakesand wetlands.

C. J. HEIN ET AL.




refraction pattern of oblique incident waves (Farrell1970). Minor deviations from this log-spiral formare related to the presence of islands and sub-merged bathymetric highs off the present-daycoast (Bremner & LeBlond 1974). The subaerialbeach and dune system contains approximately

2.2 × 107 m3 of sand (Kelley et al. 2005). Relativelynarrow, segmented barriers are exposed to wavesfrom the easterly quadrant, driven by the region’sdominant winds. Breaker heights are ,0.5 m forthe majority of the year, but during storms theyaverage between 0.9 and 1.4 m (Farrell 1972). Net

Fig. 10. Barrier-sequence types at Saco Bay, interpreted from geophysical (GPR) and sediment-core data (modifiedfrom Van Heteren 1996). (a) Headland-beach sequences interpreted to be the result of beach stabilization at outlyingheadlands that formed pinning points for subsequent barrier-spit accretion. (b) Simple successions of barrier andbackbarrier facies in which peat and inorganic backbarrier facies have infilled irregular palaeo-topography and arepartly capped by washover and aolian sand in a retrogradational succession. (c) Successions of inlet-proximalbarrier-spit and tidal-inlet facies in which coarse sandy and gravelly lag deposits formed in inlet channels fine upwardinto platform facies, covered in turn by somewhat coarser spit-beach facies and capped by aeolian sand. This sequence,with a strong shore-parallel element of variability, is interpreted to form by longshore migration of a barrier-spit andtidal-inlet system over considerable lateral distances (cf. Heron et al. 1984). Most of the spit sequences show a northerlycomponent of net migration, which is reflected in the shape of recurved ridges along the landward barrier margin (Kelleyet al. 1989). (d) Complex juxtaposition of barrier and backbarrier facies, with marsh ridges forming near inletsduring storms.





northerly transport along the beaches, at an esti-mated rate of 17 000 m3/yr, results in long-terminfilling of Scarborough River Inlet, a sedimenttrap located at the northern end of the barriersystem (Kelley et al. 2005). In the south, construc-tion of jetties at the mouth of the Saco River in themid-1800s has greatly impacted the adjacent shore-line and may have resulted in closure of the formerLittle River Inlet and rapid progradation of PinePoint (Fig. 9). This human measure resulted in theevolution of a secondary sediment sink betweenthe Saco River jetties, which has necessitated regulardredging of the inlet at a rate of 10 000 m3/yr(Kelley et al. 2005).

Merrimack Embayment barrier system

The longest barrier chain in the GoM is located inthe Merrimack Embayment (Figs 2 & 12). Thismixed-energy inlet-segmented (FitzGerald & van

Heteren 1999) embayment contains a 34-km-longseries of barriers, tidal inlets, estuaries and backbar-rier sand flats, channels and marshes. Individual bar-riers are 2–13 km long, generally less than 1 kmwide and are backed primarily by marsh and tidalcreeks that typically expand to small bays nearinlets (Smith & FitzGerald 1994). They are com-monly pinned to bedrock or shallow glacial deposits(Fig. 13a). Each contains abundant, vegetated para-bolic dunes that reach as much as 20 m in elevation.These are best developed along central and southernPlum Island and Castle Neck, reflecting abundantquartz sand associated with high rates of longshoretransport.

Sediments for this system were dominantlyderived from the 180-km-long Merrimack River.Like the Kennebec, Androscoggin and Saco riv-ers, the Merrimack is largely bedrock-controlled.Its headwaters are in the pluton-dominated WhiteMountains of New Hampshire (Fig. 3b), ensuring

Fig. 11. Shore-perpendicular analogue GSSI GPR section across Saco barrier chain, collected with a 120 MHz antennaand showing barrier anchoring to a bedrock pinning point that is no longer recognizable at the surface. Profile location isgiven in Figure 9. Top section, GPR transect; bottom section, interpretation.

C. J. HEIN ET AL.




a steady supply of quartz-rich, sandy sediments tothe embayment. The Merrimack River has deliv-ered an average annual bedload volume of 4.16 ×104 m3/yr since at least the mid-1900s (Hein et al.2012). Assuming a stable flux over time, the vol-ume of coarse sand and gravel delivered by theMerrimack since barrier pinning at 4 ka (166 ×106 m3) can account for the entire volume of the bar-riers and tidal deltas of the Merrimack Embayment(c. 137 × 106 m3) (Hein 2012). These barrier/tidaldelta sand volumes are dwarfed by the volume offiner, sandy estuarine sediment in the backbarriersof the Merrimack barrier chain (c. 850 × 106 m3)(Hein et al. 2011a), comparable in volume tothe lowstand palaeodelta (1300 × 106 m3) (Oldaleet al. 1983). Additional Holocene sandy sedimentsdeposited across the shallow shelf as a sand sheettotal c. 650 × 106 m3 (Barnhardt et al. 2009; Hein2012). Assuming modern fluxes, sandy sedimentsupply from the Merrimack River since the 14 kalowstand would amount to no more than 500 ×106 m3, only about one-third of the combined vol-ume of the sand sheet and backbarriers, bothof which post-date the lowstand. Thus, fluvial

sediment-supply rates during the early Holocenewere likely several times higher than at present.Additional sediment for the present barrier, back-barrier and/or shelf sand sheet was also contributedfrom the erosion of regressive and lowstand depos-its; seismic-reflection profiles across the lowstanddelta demonstrate the presence of a smooth, gentlydipping erosional surface that truncates the upperparts of delta foresets (Fig. 13b), indicating com-plete removal of thin topset beds and scouring toan unknown depth during the early transgression(Oldale et al. 1983; Barnhardt et al. 2009). The rela-tive contributions from the river and from marinereworking of Upper Pleistocene coastal and deltaicdeposits are unknown.

The large supply of fluvial, glacial and para-glacial sediment available to the Merrimack Embay-ment barriers has played a dominant role in theirformation and subsequent development. Sedimentsprovided by the erosion of the lowstand palaeo-delta and the regressive braidplain delta during theperiod of relatively rapid RSL rise and shorelinetransgression (c. 12–6 ka), as well as the direct con-tribution by the Merrimack River, triggered the

Fig. 12. Merrimack Embayment barrier chain and lowstand delta (modified from Hein et al. 2012). (Lowstand shorelineand delta locations are derived from Oldale et al. (1983) and Hein et al. (2013)).





development of overstepping barriers and trans-gressive sand shoals (FitzGerald et al. 1994). Thinand mobile protobarriers were pinned to contem-poraneous emerged drumlins and bedrock outcropsproximal to modern barrier positions by 4 ka (Figs13a & 14a) (Hein et al. 2012). There is little evi-dence of overwash following this original pinn-ing phase. Abundant GPR profiles collected alongSalisbury Beach and Plum Island reveal few over-wash deposits along the proximal landward sideof the barriers (Fig. 15a, b) (Costas & FitzGerald2011; Hein et al. 2012). Although preserved wash-overs are somewhat more prevalent along theCrane Beach section of Castle Neck (Doughertyet al. 2004), southerly and seaward-dipping reflec-tions generally dominate (greater than 90%) bar-rier widths (Fig. 15c). This is indicative of theseveral thousand years of progradation, aggrada-tion and spit elongation that built the modern bar-riers (Hein et al. 2012).

Palaeo-inlet sequences are common along theMerrimack Embayment barrier chain (Fig. 14)

(Hein et al. 2012). The closure of the palaeo-ParkerInlet, located in central Plum Island, highlighted therole of the deposition of abundant fine- to medium-grained estuarine sand in barrier stabilization.Here, decelerating RSL rise at c. 6 ka led to reducedmainland shoreline transgression and diminishedthe creation of backbarrier accommodation. Back-barrier sedimentation exceeded accommodationcreation over several thousand years, leading to adecrease in tidal prism and ultimately to inlet clo-sure between 3.6 and 3.0 ka. This closure, whichcreated a single 13-km-long island, was closely fol-lowed by a rapid expansion of backbarrier marshesand by the aggradation, elongation and prograda-tion of the barrier itself (Hein et al. 2011a, 2012).

Modern sediment supplied by the MerrimackRiver is largely driven southward along PlumIsland towards Castle Neck by the dominant NEstorm waves. Southerly orientated, ebb-dominantsandwaves within the large ebb-tidal-delta com-plex corroborate sedimentological evidence of asoutherly fining trend across the ebb delta, and of

Fig. 13. Stratigraphic cross-sections of barrier chain at the mouth of the Merrimack River. Locations shown inFigure 12. (a) Shore-parallel cross-section across four barriers in Merrimack Embayment (modified and expandedfrom Hein et al. 2013). Salisbury Beach and Plum Island sections are based on more than 20 km of GPR profiles(Hein et al. 2012, 2013), ground-truthed with core data from McIntire & Morgan (1964), McCormick (1968), Rhodes(1973), Costas & FitzGerald (2011), Hein (2012) and Hein et al. (2012). Castle Neck section of cross-section is based oncores from Rhodes (1973). (b) Shore-normal cross-section (modified from Hein et al. 2013). Eastern half ofcross-section is based on high-resolution chirp seismic-reflection data (Barnhardt et al. 2009), ground-truthed withsurficial sediment samples (not shown) and one offshore vibracore. Western half is ground-truthed with core data fromMcIntire & Morgan (1964), McCormick (1968), Oldale & Edwards (1990) and Hein (2012). Offshore core data arecourtesy of G. Edwards. MSL, mean sea-level. Note that vertical exaggeration of (a) is exactly twice the verticalexaggeration of (b).

C. J. HEIN ET AL.




a general trend of increasing textural and mineralo-gical maturity in the same direction, away from theriver (FitzGerald et al. 1994, 2002). The dominantsoutherly transport regime has resulted in thegrowth of recurved spits on the downdrift ends ofCrane Beach and Plum Island (Farrell 1969), isreflected in an increase in the spacing of offshore

contours to the south along the barrier chain(Smith 1991), and has influenced the developmentof the southern part of the Merrimack Embaymentas a net sediment sink (Hubbard 1976; Barnhardtet al. 2009). Active transport across the shallow(less than c. 40 m) shelf during NE storms likelyserves to rework some of this temporarily stored

Fig. 14. Buried palaeo-inlet sequences at (a) Plum Island (shore-parallel) and (b) Coffins Beach (shore-normal), asimaged in GPR profiles. Profile locations shown in Figure 12. Plum Island GPR profile is modified from Hein et al.(2013) and was collected with a digital GSSI SIR-2000 system with a 200 MHz antenna and digitally post-processedusing the GSSI Radan software package. Coffins Beach GPR profile was collected with an analogue GSSI GPR with a200 MHz antenna. (Profile is courtesy of P. McKinlay.)





C. J. HEIN ET AL.




sediment toward the barrier system (Hein et al.2007). Proximal to the mouth of the MerrimackRiver, continued fluvial sediment inputs combinewith complex river/inlet/tidal-delta interactionsto create the most dynamic section of this coast(Fig. 15a) (FitzGerald et al. 1994; Costas &FitzGerald 2011).

Paraglacial coasts: barrier formation

in a distinctive setting

River-associated barriers of the Gulf

of Maine in a paraglacial context

Coastal barriers within a paraglacial coastalframework occur in a variety of settings, includingbraided outwash plains, estuarine re-entrant coastsand prograding deltaic systems. These settings pro-vided the context for different RSL histories, sedi-ment supplies and physical processes of sedimentreworking. Despite those differences, they have allbeen marked by the formation of barriers, tidalinlets and associated tidal sand bodies, as well asvarious types of backbarrier environments. Similar-ities among these systems and differences withinindividual systems emphasize the importance ofwave and tidal processes and RSL changes in dic-tating coastal morphology.

Closest to active glaciers, most barriers formedat the leading edge of exposed, prograding outwashplains, a setting found along the active Skeiðarar-sandur coast of SE Iceland (Hine & Boothroyd1978; Nummedal et al. 1987), the Gulf of Alaskacoast from Dry Bay west to Kayak Island (Hayes& Ruby 1994), the Hallo Bay region along theAlaskan Peninsula and the Karaginsky Gulf coastof the Kamchatka Peninsula, Siberia. Coastal accu-mulation forms of this type are marked by a widerange in grain sizes, but coarse sand and finegravel dominate most barrier and tidal-delta litho-somes. These Arctic and sub-Arctic coasts are domi-nated by spit systems; true barrier islands are rare.Tidal inlets tend to be located at the downdriftend of littoral cells. Barriers fronting active, pro-grading outwash plains are generally susceptible to

breaching during storms, a process that commonlyrepositions inlets in the middle of embayments.Flood-tidal deltas along these outwash-plain bar-riers tend to be well developed and are commonlya product of storm deposition, whereas ebb deltasare only prominent at inlets with large tidal prisms.

Farther from active ice margins, the largest para-glacial barriers are associated with fluvial systemsdraining glaciated landscapes. These have volumi-nous sand and gravel sources which are generallyreplenished less frequently as glaciers recede. Stillwithin the sphere of influence of ice caps or glaciers,some systems are nourished by sediment erodedlocally from outwash plains that are no longerfully replenished by meltwater-derived sediment,especially in the coarsest fractions, whereas othersare supplied by major transport conduits from sedi-ment supplied from farther away. These lattersystems are found at the mouths of rivers drain-ing active mountain glaciers (e.g. the CopperRiver, Alaska; Hayes & Ruby 1994). They are com-monly characterized by well-developed barrierislands and tidal inlets and by backbarriers withopen-water (lagoonal) or intertidal (marsh, tidalflats and tidal creeks) dominance, dependent inpart on antecedent setting. Sand dominates thesesystems because of the distal location of theprimary sediment source, although fine gravel mayalso be an important component. The spacing anddimensions of inlets along these coasts are a func-tion of tidal prism, whereas inlet positions andsizes on barrier coasts such as those fronting theKennebec/Androscoggin, Saco and Merrimackrivers in the GoM are usually a function of floodevents, basement controls, storm breaching and pro-minent river-discharge sites.

Beyond the area of direct glacially drivensediment replenishment, most paraglacial barriersare formed and modified using finite and shrink-ing sand and gravel sources, both on- and offshore.Where these sources are large and exposed enoughto supply transporting conduits and processes witha continuous and steady flow of sediment, theabsence of replenishment does not influence bar-rier development (such as the Kennebec, Sacoand Merrimack barrier chains). Where sources are

Fig. 15. Representative post-processed GPR profiles demonstrating the dominant formation mechanisms of theMerrimack Embayment barriers. All GPR profiles were collected using a digital GSSI SIR-2000 system with a 200 MHzantenna. Profile locations shown in Figure 12. (a) Shore-normal GPR section across Salisbury Beach, demonstrating thecontribution of foreshore drift and swash-bar welding to barrier elongation. (Profile modified from Costas & FitzGerald2011.) (b) Shore-parallel GPR section across central Plum Island showing spit progradation over intertidal backbarrierdeposits, the dominant mechanism of progradation and elongation of the barrier-spit system. (Profile modified fromHein 2012.) (c) Shore-normal GPR section across Castle Neck containing high-amplitude reflections representative ofheavy-mineral concentrations deposited during storm events (Dougherty et al. 2004). Here, barrier growth wasdominated by seaward progradation. Seaward part of unit labelled as a ‘tidal channel’ is interpreted as anonshore-migrating bar associated with the southward migration of the Parker River Inlet. (Profile modified fromDougherty et al. 2003.)





small and localized, they will eventually becomeexhausted (e.g. along the drumlin-dominated East-ern Shore of Nova Scotia; Carter et al. 1990;Forbes et al. 1991).

When viewing the GoM barriers in light ofthese proximal and distal settings, behaviouraldifferences and temporal developmental patternsrelated to sediment availability can be explained.The most voluminous sediment resources alongthe GoM are located farthest south, near the termi-nus of the Wisconsinan ice sheet at Cape Cod.Here, early barrier systems will have been domi-nated by coarse-clastic spits. Even today, sedimenteroded from exposed outwash bluffs nourishesattached barriers, some of which are periodicallybreached to become islands. On the other side ofthe spectrum, barrier systems beyond the area ofpersistent glacial sediment replenishment, and thusdependent on local sediment supplies, are moststrongly affected by the finite nature of sedimentsources. Examples include the barriers of theEastern Shore of Nova Scotia and the barriers andspits of NW Alaska. Similar to the river-associatedGoM barriers, these barriers are generally composedof medium to coarse sand and gravel (Short 1979;Boyd et al. 1987); where present, backbarrier andestuarine sediments are composed of fine sandand mud (Boyd & Honig 1992; Carter et al. 1992).However, sediment supply to river-associated bar-riers along the central Maine coast has been muchless variable. With rare exceptions, such as nearthe southern end of Plum Island where drumlinshave been eroded episodically, middle- and late-Holocene sediment supply has been marked by asteady decrease (D. Barber, pers. comm.). Here,RSL changes have maintained a dominant dif-ferential control over barrier evolution during thepast c. 8000 years. The present organization of theKennebec, Saco Bay and Merrimack Embay-ment barrier systems indicates an increasinglyhigh degree of maturity that developed during anextended period (3000–5000 years) of lateral bar-rier accretion and barrier progradation under con-ditions of near-steady fluvial sediment supply tothe coast. Earlier in their development, however,each of these systems may have shown a muchhigher morphologic diversity, going through oneor more stages of morphologic immaturity anddeterioration during periods of rapid RSL rise andbarrier retrogradation that dominated over sedimentavailability. The coexistence of immature, matureand disintegrating barriers along these and otherparaglacial coasts (Kliewe & Janke 1991; Orfordet al. 1991; Nichol & Boyd 1993) is a morpholo-gical reflection of recent diachronicity in bar-rier development. Morphologic evidence from pastdevelopmental phases, however, is easily over-printed by later events.

Features of barrier formation in

paraglacial settings

The paraglacial barriers of the GoM differ fromwell-researched lower-latitude barriers that arelocated far beyond the limits of the Pleistocene icesheets in terms of both their development andmorphostratigraphy (Table 4). Their developmenthas been affected by (1) numerous bedrock pro-montories that compartmentalized the coast andserved as pinning points for barriers; (2) spatiallyand temporally variable sediment sources (plutons,reworked glacial and paraglacial sediments) and awide range in particle sizes (clay to boulders) andsediment supply rates; and (3) multiple phasesof transgression and regression, associated withrapid, large-scale, post-glacial sea-level changes(e.g. RSL fall of 120 m over 2000 years in centralMaine; Fig. 4) that allowed the reworking ofsediment across the modern coastal plain, coastalzone and shelf. The resulting barrier systems arecharacterized by progradational dunes and beachescomposed of spatially variable sediment textures;for example, beaches are coarser proximal toglacial (till) outcrops and fluvial sediment sourcesand fine with distance. They have complex min-eralogical compositions, such as the presence ofhorizons rich in garnet derived from erosion ofplutons, and an overall lithological heterogeneity.Bioclastics within the barrier lithosomes are mid/late Holocene to modern in age. Preserved marineorganisms are largely similar within GoM coastalsediments, with differences reflecting the pres-ence of exposed bedrock. For example, barnacleplates are much more common in the bedrock-dominated KBC than in the glacially pinned Merri-mack barriers.

Paraglacial GoM barriers are more commonlycomposed of coarser-grained sediment (siliciclasticmedium-grained sand to gravel) than is found alongnon-glaciated continental trailing-edge coasts. Thisreflects their (1) proximity to major river systemswith coarse sediment loads, (2) glacial sedimentsources, and (3) continued inputs from both glacial(drumlins and other till deposits) and paraglacial(upstream terraces, outwash deposits) sedimentsources.

Variable sediment supply: the defining

feature of paraglacial barrier formation

Glacial and paraglacial settings are, by definition,transient; their existence and persistence are inex-tricably linked to the availability and supply ofglacially derived sediment. The surficial evidenceof glaciation along river-associated paraglacialcoasts may be removed within several thousand

C. J. HEIN ET AL.




years by non-glacial processes. Nonetheless, aglacial imprint remains, commonly hidden withinthe stratigraphic framework of fluvial and coastaldeposits. The sedimentological, stratigraphic andchronological frameworks of the river-associatedbarriers of the GoM contain valuable informationon the nature and timescales of coastal landscaperesponses to glaciation.

The paraglacial period, as defined by Ballantyne(2002a, b), is the timescale over which glaciogenicsediment stores are tapped and finally exhausted, orlandscapes equilibrate to reworking processes. Fol-lowing this phase, the landscape returns to a non-glacial or post-glacial state. As such, a coastline isonly paraglacial for as long as glacially excavatedlandforms and glaciogenic sediments have a recog-nizable influence on the character and evolution ofthe coast (Forbes & Syvitski 1994). The durationof the paraglacial period along any given coastlinecan be affected by regional geology (erodibility ofbedrock) and by temporally varying sedimentsupplies. Sediment-supply variations can be drivenby changes in climate (amount/seasonality of

precipitation), vegetation (stabilisation of hillslopes), proximity to meltwater (steady meltwaterfluxes, annual melting events, convulsive outburstand catastrophic flooding events (jokulhlaups)),regional coastal setting (wave, tides, river inputs)and RSL changes (e.g. erosion of untapped glacio-genic sediment sources by rising RSL) (Forbes &Taylor 1987; Forbes & Syvitski 1994; Forbes et al.1995). Continued fluvial input of sediment erodedfrom voluminous glacial sources and complex post-glacial RSL changes along the river-associatedparaglacial coast of the GoM served to lengthenthe paraglacial period through the late Holoceneand broaden its influence from inland to far off-shore. The sediment yield of all river systems feed-ing these coasts is dominated by reworked glacialand paraglacial deposits. These serve only as tem-porary storage for sand and gravel generated thou-sands of years earlier. Elevated sediment exportfrom river catchments will continue as long as gla-ciogenic or paraglacial sediments remain easilyaccessible to fluvial scouring (Church & Ryder1972; Ballantyne 2002a). The re-entrainment of

Table 4. Generalized comparison between barrier systems formed along coastal plains and those formedalong, or proximal to, formerly glaciated coasts

Glaciated coastal barriers Coastal-plain barriers

Continuity Generally short barriers; may be sand- orgravel-dominated; range from barrierislands to welded barriers; barrier typecommonly changes abruptly

Barrier type constant for 50–200 km

Relative sea-levelhistory duringbarrier formation

Complex; ranges from slowly falling toslowly rising RSL during middle to lateHolocene

Slow rate of RSL rise during middleto late Holocene

Basement controls Drumlins and other glacial deposits and/orbedrock act as pinning points for barrierdevelopment

Barriers form on interfluves, and tidalinlets stabilize in former rivervalleys

Sediment sources Multiple sources that can change spatiallyand temporally; include glacial andprimary and secondary paraglacialdeposits and fluvial sediments

Continental shelf, minor fluvial inputin locations distal to medium tolarge rivers

Sediment-supplyrates

Complex; related to fluvial and coastalerosion of glacial and paraglacialsediment sources and to RSL change

Driven by sea-level change andextreme events

Substrate lithology Barrier lithosomes overlie glacial andparaglacial deposits such as till andglaciomarine clay

Barrier lithosomes overlie Pleistocenecoastal-plain deposits or bedrock

Grain size Fine to coarse sand and gravel; can changerapidly across short distances

Fine to medium sand

Backbarrierenvironment

Lagoon to marsh or tidal flat, incised bytidal creeks; ice-rafted horizons commonin marshes

Lagoon to marsh or tidal flat, incisedby tidal creeks

Examples New England (USA), Long Island (USA),Alaska (USA), Canada, New Zealand,Ireland, United Kingdom, KamchatkaPeninsula (Russia), sections of Balticcoast

East and Gulf Coasts of USA; WestAfrica; India; northern Black Sea;Algarve of Portugal; central andsouthern Brazil





submerged paraglacial deposits within the coastalsetting is another mechanism lengthening theperiod of influence of glaciation. In the GoM, thereworking of regressive and lowstand deposits off-shore of river mouths has produced transgressivesand sheets and nourished barrier systems. Ballan-tyne (2002a, b) suggested the term ‘secondary para-glacial system’ to describe such features. In thiscase, the regressive and lowstand deposits can beconsidered primary paraglacial deposits and thelater barriers secondary paraglacial landforms.

The delivery of sediment to river-associatedcoastal paraglacial systems during and after glacia-tion occurs over multiple periods, each marked bythe deposition of distinct sedimentary units andthe formation and modification of specific land-forms (Figs 5 & 16).

Glacial period

Characteristic deposits and landforms during thisperiod are formed by glaciers during both advance

and retreat of ice sheets. These features includeglacially striated bedrock, fjords, over-steepenedembayments, drumlins, crag-and-tail ice-stream-lined deposits, kames, eskers, grounding-line fans,ground and washboard moraines and deglaciation-related moraines (lateral, terminal, recessional;submarine ice-pushed moraines) (Belknap et al.1987; Syvitski 1991; Forbes & Syvitski 1994). Tillis the dominant deposit.

Proglacial period

The proglacial period begins immediately follow-ing deglaciation (Church & Ryder 1972; Ballantyne2002a, b; Slaymaker 2009). Although the term‘proglacial’ specifically refers to an area literally‘in front of the glacier’, it is adopted here for atime period during which a landscape is located insuch a position because of retreat of an ice sheet.Sediments deposited during this period are largelyderived directly from the glacier and are thereforeglacial rather than paraglacial in origin.

Fig. 16. Schematic diagram of the pattern of sedimentation during the paraglacial period. Note that time before presentincreases to the left (modern is to the right). Conceptual model builds on ideas and models proposed by Church & Ryder(1972) and Ballantyne (2002b). The post-glacial period is only possible once all glaciogenic and primary paraglacialdeposits have been exhausted or deeply buried and can no longer contribute to barrier development. Question marksassociated with this period reflect uncertainty in the possibility that such a period is ever reached, as even indirectcontributions by the cannibalization of barrier segments formed from paraglacial sediments are excluded.

C. J. HEIN ET AL.




In the Gulf of Maine, the proglacial period wasmarked by deep isostatic depression of the crust,rapid shoreline transgression, the presence of tide-water glaciers and the deposition of outwashdeposits and coarse sandy ice-marginal deltas. Else-where, proglacial coastal-zone sediments accumu-late in sandur (outwash) plains, braided outwashfans, jokulhlaup units, glaciomarine deltas andcoastal moraines, ice-rafted debris deposits, and gla-ciomarine basins (Syvitski 1991; Slaymaker 2009).Along river-associated coasts in the GoM, fluvialsediment delivery to the coast during this periodwas characterized by finer mean grain sizes thanduring the glacial period (Fig. 5). The proglacialperiod can last for many (.10) millennia if activeglaciers continue to indirectly deliver proglacialsediments to the coast via wide braidplains (e.g.Skeiðararsandur, Iceland).

Early paraglacial period

This period is newly defined here for river-associated paraglacial coasts, differentiating themfrom other paraglacial coastal settings. The onsetof this period coincides with the end of widespreadglaciation within the barrier-associated river drai-nage basin. Glaciogenic sediments are no longerbeing deposited within the drainage basin or alongthe coast. Sand-sized sediment export to the coastreaches a maximum: a dearth of vegetation andsoil, and large quantities of glacially liberated andparaglacial sediment stored within the river basinsat the start of this period, lead to a high rate ofsediment – notably sand – export. High fluvialsediment yields, obviously proportional to river dis-charge, continue throughout this period. Along thecoast, this early paraglacial period may be referredto as the ‘paraglacial sand maximum’, a time ofunusually high rates of sand delivery to, and depo-sition in, various contemporaneous coastal environ-ments. Sediments supplied by rivers are reworkedby coastal and fluvial processes in a regressivesetting, in response to isostatic-rebound-inducedRSL fall. In the GoM, this mechanism resulted inthe deposition of both discrete and expansive (e.g.the Sanford-Kennebunk, Brunswick and Merri-mack braidplain deltas) regressive shoreline depos-its and lowstand deltas and delta lobes. During thistime of rapid isostatic land-level adjustments inthe drainage basins of paraglacial rivers, drainagedivides of large and small streams shift and canstrongly alter both the competence of streams andthe distribution river mouths.

Middle paraglacial period

This period of coastal paraglacial evolution ismarked by a gradual exhaustion of terrestrial

glacigenic and paraglacial deposits by associatedriver systems and by a commensurate decrease influvial sediment supply to the coast. Intense rework-ing by waves, tides and currents of glaciogeniclandforms (drumlins, till bluffs, onshore outwashplains) and newly drowned primary coastal para-glacial deposits (regressive shoreline deposits, low-stand deltas) results in the formation of secondarycoastal paraglacial deposits and features such astransgressive sand sheets, estuarine fills, barriers,tidal deltas, backbarrier marshes and tidal flats. Ris-ing RSL eventually fully submerges and strands gla-ciogenic and primary paraglacial deposits beyondthe depth of reworking by waves as transgres-sion progresses. However, this same RSL rise andshoreline transgression can also tap new sedimentsources and alter the distribution and prominenceof pinning points and coastline orientation, result-ing in cyclic patterns of barrier formation anddestruction such as found along drumlin-dominatedcoasts (Rosen 1984; Forbes & Taylor 1987; Nichol& Boyd 1993). A key feature of this period alongriver-associated barrier coasts is the transition todominantly non-glaciogenic and non-paraglacialfluvial sediment export. This reflects the develop-ment of post-glacial conditions upstream, in whichsediment transport and export is in equilibriumwith erosion of primary materials (bedrock) bynon-glacial processes. Such a transition may takeas long as 5000–10 000 years as even bedrockerosion rates may remain elevated for long periodsbecause of the persistence of a weakened upperbedrock surface that had been physically fracturedand weathered by a long-departed ice sheet.

Late paraglacial period

This stage is marked by the gradual exhaustionof accessible paraglacial sediment. During thisperiod, coastal systems are fed dominantly byrivers that export sediment derived only from ero-sion of non-glacial sources, such as the bedrock inNew England. Non-fluvial sediment inputs to thecoastal system are increasingly rare. Some finalparaglacial activity is linked to occasional erosionof coastal glaciogenic or paraglacial deposits, butmost of the non-fluvial sediment will be derivedfrom shorefaces and shelves that are increasinglyin equilibrium with prevailing wave and currentconditions. Along sediment-rich coasts such as atthe mouths of the Kennebec/Androscoggin, Sacoand Merrimack rivers in the GoM, deposits asso-ciated with this period largely reflect the mat-uration of coastal features first formed during themiddle paraglacial period. By contrast, alongsediment-starved coasts or those with a high rateof creation of accommodation space for a givenRSL rise (i.e. those with a low gradient), the period





of transition to negligible (or intermittent) glacio-genic/paraglacial sediment supply can lead towidespread erosion, barrier instability and eventualdisintegration.

In the GoM, the transition to the late paraglacialperiod corresponded with a marked decrease in therate of RSL rise at c. 4 ka (Fig. 4). Relativelymature fluvial and reworked paraglacial sedimentsfilled an increasing proportion of available accom-modation along river-associated coasts in the GoM.Progradation, aggradation and the development ofan equilibrium shoreface, expansive vegetated andunvegetated dunes, marshes and equilibrium ebband flood-tidal deltas were the result.

Post-paraglacial period

This period is characterized by a coast that is con-trolled exclusively by non-glacial processes anddoes not receive any additional sediment fromany glaciogenic or paraglacial sources. The coastresponds to intrinsic and extrinsic forcings in amanner indistinguishable from a coast in a non-glaciated setting.

The timing of the transition to, and even theexistence of, the post-paraglacial period is highlycontentious for all areas falling within the sphereof influence of Quaternary glaciations. Coastal sys-tems within these areas are never fully and per-manently removed from the effects of glaciation.After many years of non-paraglacial conditions, astorm may expose a glacial source, or a chang-ing coastal or fluvial configuration may make onebarrier paraglacial again and render another bar-rier non-paraglacial for the first time. For exam-ple, the river-associated paraglacial barriers in theGoM, although formed solely by non-glacial pro-cesses, would not have had a sediment sourcelarge enough for their formation without the priordeposition of paraglacial deposits. Even after sedi-ments have been reworked multiple times, theseretain imprints, either clear or very subtle, of theirparaglacial origins. Furthermore, glaciogenic andparaglacial sediments can be released in terrestrialenvironments by processes unrelated to direct orlagged landscape response to glaciation. For exam-ple, changes in climate or human disturbance canrapidly change terrestrial landscapes, affecting ero-sion and deposition patterns and delivering freshquantities of previously unavailable glaciogenic orparaglacial sediment to a river-associated paragla-cial coast many millennia following deglaciation.Likewise, increasing rates of RSL rise, notablyalong the NE coast of the USA (Sallenger et al.2012) will force shorelines to transgress previouslyuntapped terrestrial glaciogenic sediment sources,thus contributing new glaciogenic sediments to thebarrier systems. In the Merrimack Embayment,

drumlins presently undergoing erosion at thesouthern end of Plum Island and on Castle Neckare cored by till of Illinoian age (Stone et al.2006); erosion of these would contribute glacio-genic sediments that have been stored along thecoast for more than 100 000 years. Does thisimply that the Merrimack Embayment is still in astate of paraglacial or post-paraglacial non-equili-brium dating back to c. 120 ka? A positive answerto this question implies that coasts in regions sus-ceptible to glaciations may never truly enter a post-paraglacial period and will always be paraglacial innature. Using this line of reasoning, even the majorNorth Sea barrier system extending from northernFrance to Denmark is entirely paraglacial. Alongthe Frisian Islands, a clear paraglacial overprint isvisible in eroding bluffs (Denmark and Germany)and in glacial highs serving as anchor points tobarrier islands (Denmark, Germany and the Nether-lands). Farther south, the paraglacial characteristicsare more subtle. Here, the progradational barriersystem of the western Netherlands owes much ofits size to a mid-Holocene abundance of fluvial low-stand sand formed under periglacial conditions andlater reworked by shallow-marine and coastal pro-cesses as the initially gentle shoreface steepenedtoward equilibrium.

Changing climate, human interference

and the future of river-associated

barriers in the Gulf of Maine

Coastal sediment supplies have undergone sig-nificant natural and human-induced perturbationsat local, regional and global scales and over timeperiods ranging from months to thousands of years.Natural climate-geomorphic feedbacks under chan-ging precipitation regimes have driven changes inthe rates of erosion and fluvial sediment deliveryto the coast (Leeder et al. 1998; Blum & Tornqvist2000; Goodbred 2003; Hein et al. 2011b). Overthe Anthropocene, fluvial sediment supplies havefurther varied in response to deforestation, agricul-tural expansion and contraction, urbanization, sedi-ment quarrying and mining, land reclamation, andriver engineering, impoundment and damming(Yang et al. 2010; Kirwan et al. 2011; Milliman &Farnsworth 2011). Over shorter timescales, theemplacement of artificial hard protective structures(such as jetties, groins, seawalls, bluff-stabilizationmeasures, and breakwaters) and implementation ofsoft engineering solutions (beach and shorefacenourishment, dewatering, sand-bagging, scrapingand draining) have disrupted natural pathways ofsediment within the littoral zone, resulting inmigration of accretion and erosion hotspots, modi-fication of overall beach morphodynamics, and

C. J. HEIN ET AL.




localized flooding. Given projected increases in therates of RSL rise (Church et al. 2014), net coastalerosion resulting from both human-induced andnatural changes in sediment-supply systems willlikely only accelerate.

Changing climate and enhanced anthropogenicstresses present unique challenges for the futureviability of river-associated paraglacial barriercoasts in particular. In contrast to coasts proxi-mal to actively melting glaciers that are fed bysediment-laden meltwater streams or sustained bysediment eroded from coastal bluffs composed ofthick glacial deposits, those with thinner or lessextensive glacigenic and paraglacial inland depositsare more likely to be impacted by a future reductionin terrestrial sediment supply. The GoM barriersare presently nourished by rivers with dischargessimilar to those of many moderate-sized riversalong the East and Gulf coasts of the USA (Milliman& Farnsworth 2011); however, natural depletion ofglacial and paraglacial fluvial sediment sources andthe stabilization of slope, terrace and floodplaindeposits by vegetation cover have reduced fluvialsediment supply to these paraglacial coasts moresignificantly than farther south. Unlike barriers innon-glaciated or river-distal settings, river-asso-ciated paraglacial barriers formed in regimes ofboth ample accommodation and ample fluvial sedi-ment supply, but now face a substantial naturaland anthropogenically induced reduction in thelatter. The drainage basins, tributaries and primarydownstream river segments of the Kennebec/Androscoggin, Saco and Merrimack rivers have allundergone extensive anthropogenic modificationsover the past several hundred years. Damming andre-routing of the rivers and their tributaries and thejettying and dredging of their mouths have greatlyimpacted sediment discharge and sand-dispersalpatterns (Farrell 1970; FitzGerald 1993; Kelleyet al. 2005).

Given the reductions in fluvial sediment supplyand human modifications to fluvial sediment-supplysystems, enhanced creation of backbarrier accom-modation in a regime of accelerated RSL rise islikely to cause narrowing and shortening of theriver-associated GoM barriers (FitzGerald et al.2008), with an increasing chance of barrier breach-ing. This process may reach the point where somebarriers become unstable and vulnerable to the step-wise retrogradation and overstepping that charac-terized their earlier histories. The retrogradational/aggradational pathway of the Kennebec barrierchain, for example, will likely continue in thecoming decades of accelerated RSL rise (Buynevich2001). As attested by occasional intertidal expo-sures of backbarrier sediments and tree stumpson the beach face, onshore–offshore redistribu-tion of sand and gravel during intense storms will

continue to drive longer-term barrier morphody-namics (Buynevich et al. 2004). Mobility of themature barrier-spit systems of Saco Bay and theMerrimack Embayment is likely to be limited inthe near term. However, when sediment supply nolonger compensates the effects of RSL rise, eventhese barrier spits will either migrate rapidly to anearby pinning point, as they have in the past (Fitz-Gerald et al. 1994; van Heteren 1996; Hein et al.2012), or be fragmented and destroyed, only toreform in a more landward position that is favouredby palaeotopography (Swift 1968; Boyd & Penland1984). Thus, river-associated paraglacial coastsdependent on continuous fluvial sediment inputare in a precarious situation and may be rapidlyapproaching a point of transition from regressiveto transgressive and destructive modes.

Conclusions

Paraglacial coasts are those formed on or proxi-mal to formerly ice-covered terrain and retainthe landforms and sediments derived directly orindirectly from glaciation. The river-associatedparaglacial barriers of the GoM (USA & Canada)formed along such a paraglacial coast during aperiod of decelerating RSL rise over the pastc. 5000 years. These barriers are distinguishablefrom barriers formed in coastal plain settings oreven those along other paraglacial coasts. They arecharacterized by spatially variable sediment tex-tures, complex sediment composition and lithol-ogical heterogeneity. This variability reflectsseveral unique features of barrier formation alongriver-associated paraglacial coasts: (1) the abun-dance of bedrock and glacial promontories thatcompartmentalize the coast and serve as pinningpoints for barriers; (2) the complex post-glacialRSL changes that can shift depocentres laterallytens of kilometres in hundreds of years; and (3)the variable sources, conduits and supply rates ofglacial, primary and secondary paraglacial, and non-glacial sediment sources. Sediment-supply ratesalong these river-associated paraglacial barrierswere highest within a few thousand years follow-ing deglaciation (the early and middle paraglacialperiods). Sand deposition peaks at a period hereindefined as the paraglacial sand maximum, as gla-cial and primary paraglacial deposits are erodedon land and sediments are redeposited along theregressing coast as a series of sandy shorelines,braidplain deltas and lowstand deltas.

The future stability of river-associated paragla-cial barriers in a regime of accelerated RSL rise isdependent upon the continued supply of sandy sedi-ments to the barriers and beaches, and of finer inor-ganic sediments to the backbarriers. However, a





combination of a natural depletion of glacially liber-ated sediment and anthropogenic modificationsof both the river systems delivering this sedimentand the barriers themselves threatens to enhancebarrier erosion, cause disintegration of backbarriermarshes, and, eventually, return these systems tothe retrogradational states that characterized theirearlier development.

The authors thank I.P. Martini for his careful reviews andadvice throughout the production of this manuscript, andH. Wanless, D. Forbes, plus three additional anonymousreviewers for their constructive input. C. Hein wasfunded in part by a Woods Hole Oceanographic InstitutionCoastal Ocean Institute postdoctoral fellowship andNational Science Foundation Coastal SEES Award #OCE1325430.

References

Balco, G., Stone, J. O. H., Porter, S. C. & Caffee, M.W. 2002. Cosmogenic-nuclide ages for New Englandcoastal moraines, Martha’s Vineyard and Cape Cod,Massachusetts, USA. Quaternary Science Reviews,21, 2127–2135.

Ballantyne, C. K. 2002a. Paraglacial geomorphology.Quaternary Science Reviews, 21, 1935–2017.

Ballantyne, C. K. 2002b. A general model of paraglaciallandscape response. The Holocene, 12, 371–376.

Barnhardt, W. A., Gehrels, W. R., Belknap, D. F. &Kelley, J. T. 1995. Late Quaternary relative sea-levelchange in the Western Gulf of Maine: evidence for amigrating glacial forebulge. Geology, 23, 317–320.

Barnhardt, W. A., Belknap, D. F. & Kelley, J. T.1997. Stratigraphic evolution of the inner continentalshelf in response to late Quaternary relative sea-levelchange, northwestern Gulf of Maine. GeologicalSociety of America Bulletin, 109, 612–630.

Barnhardt, W., Andrews, B., Ackerman, S.,Baldwin, W. & Hein, C. 2009. High Resolution Geo-logic Mapping of the Inner Continental Shelf: CapeAnn to Salisbury Beach Massachusetts. United StatesGeological Survey Open-File Report OFR-07-1373.

Beardsley, R. C., Butman, B., Geyer, W. R. & Smith,P. 1997. Physical oceanography of the Gulf of Maine.In: Wallace, G. T. & Braasch, E. F. (eds) Proceed-ings of the Gulf of Maine Ecosystem Dynamics Scien-tific Symposium and Workshop. Regional Associationfor Research on the Gulf of Maine (RARGOM),Report 97-1, 39–52.

Belknap, D. F., Kelley, J. T. & Shipp, R. C. 1987.Quaternary stratigraphy of representative marineestuaries: initial examination by high resolutionseismic reflection profiling. In: FitzGerald, D. M.& Rosen, P. S. (eds) Glaciated Coasts. AcademicPress, New York, NY, 143–176.

Belknap, D. F., Shipp, R. C., Kelley, J. T. & Schnitker,D. 1989. Depositional sequence modeling of the lateQuaternary geologic history, west-central Mainecoast. In: Tucker, R. D. & Marvinney, R. G. (eds)Studies in Maine Geology. Quaternary Geology.Maine Geological Survey, Augusta, ME, 5, 29–46.

Belknap, D. F., Kelley, J. T. & Gontz, A. M. 2002.Evolution of the glaciated shelf and coastline of theNorthern Gulf of Maine, USA. Journal of CoastalResearch, SI 36, 37–55.

Bigelow, H. B. 1924. Plankton of the offshore watersof the Gulf of Maine. Bulletin of the US Bureau ofFisheries, 40, 1–509.

Bloom, A. L. 1963. Late-Pleistocene fluctuations of sealevel and postglacial crustal rebound in coastalMaine. American Journal of Science, 261, 826–879.

Blum, M. D. & Tornqvist, T. E. 2000. Fluvial responsesto climate and sea-level change: a review and lookforward. Sedimentology, 47, 2–48.

Borns, H. W. & Hagar, D. J. 1965. Late-glacial stratigra-phy of a northern part of the Kennebec River valley,western Maine. Geological Society of America Bulle-tin, 76, 1233–1250.

Borns, H. W. Jr., Doner, L. A. et al. 2004. The deglacia-tion of Maine, USA. Developments in QuaternarySciences, 2, 89–109.

Boyd, R. & Honig, C. 1992. Estuarine sedimentation onthe Eastern Shore of Nova Scotia. Journal of Sedimen-tary Research, 62, 569–583.

Boyd, R. & Penland, S. 1984. Shoreface translation andthe Holocene stratigraphic record: examples fromNovaScotia, theMississippiDeltaandeasternAustralia.In: Greenwood, B. & Davis, R. A. Jr. (eds) Hydro-dynamics and Sedimentation in Wave-DominatedCoastal Environments. Marine Geology, 60, 391–412.

Boyd, R., Bowen, A. J. & Hall, R. K. 1987. An evol-utionary model for transgressive sedimentation on theEastern shore of Nova Scotia. In: FitzGerald,D. M. & Rosen, P. S. (eds) Glaciated Coasts. Aca-demic Press, New York, NY, 87–114.

Bremner, J. M. & LeBlond, P. H. 1974. On the plani-metric shape of Wreck Bay, Vancouver Island.Journal of Sedimentary Petrology, 44, 1155–1165.

Broecker, W. S., Turekian, K. K. & Heezen, B. C.1958. The relation of deep sea [Atlantic Ocean] sedi-mentation rates to variations in climate. AmericanJournal of Science, 256, 503–517.

Brothers, L. L., Belknap, D. F., Kelley, J. T. &Janzen, C. D. 2008. Sediment transport and dispersionin a cool-temperate estuary and embayment, SacoRiver estuary, Maine, USA. Marine Geology, 251,183–194.

Buynevich, I. V. 2001. Fluvial–marine interaction andHolocene evolution of sandy barriers along anindented paraglacial coastline. PhD thesis, BostonUniversity.

Buynevich, I. V. & FitzGerald, D. M. 2001. Styles ofcoastal progradation revealed in subsurface recordsof paraglacial barriers, New England, USA. Journalof Coastal Research, SI 34, 194–208.

Buynevich, I. V. & FitzGerald, D. M. 2002.Organic-rich facies in paraglacial barrier lithosomesof northern New England: preservation and paleoenvir-onmental significance. Journal of Coastal Research, SI36, 109–117.

Buynevich, I. V. & FitzGerald, D. M. 2003. High-resolution subsurface (GPR) profiling and sedimen-tology of coastal ponds, Maine, USA: implicationsfor Holocene back-barrier evolution. Journal of Sedi-mentary Research, 73, 559–571.

C. J. HEIN ET AL.




Buynevich, I. V. & FitzGerald, D. M. 2005. Sedimen-tological signatures of riverine-dominated phases inestuarine and barrier evolution along an embayedcoastline. In: FitzGerald, D. M. & Knight, J. (eds)High Resolution Morphodynamics and SedimentaryEvolution of Estuaries. Springer, Dordrecht, 315–334.

Buynevich, I. V., FitzGerald, D. M. & Parolski, K. F.1999. Geophysical Investigation of the Nearshore Geo-logic Framework, Eastern Casco Bay – Reid StatePark, Maine. Data Interpretation and Implications forLate Quaternary Coastal Evolution. US GeologicalSurvey Open-File Report 99-0380.

Buynevich, I. V., FitzGerald, D. M. & van Heteren,S. 2004. Sedimentary records of intense storms inHolocene barrier sequences, Maine, USA. MarineGeology, 210, 135–148.

Buynevich, I. V., FitzGerald, D. M. & Goble, R. J.2007. A 1500 yr record of North Atlantic storm activitybased on optically dated relict beach scarps. Geology,35, 543–546.

Carter, R. W. G., Forbes, D. L., Jennings, S. C., Orford,J. D., Shaw, J. & Taylor, R. B. 1989. Barrier andlagoon coast evolution under differing relative sea-levelregimes: examples from Ireland and Nova Scotia. In:Ward, L. G. & Ashley, G. M. (eds) Physical Processesand Sedimentology of Siliciclastic-Dominated Lagoo-nal Systems. Marine Geology, 88, 221–242.

Carter, R. W. G., Orford, J. D., Forbes, D. L. &Taylor, R. B. 1990. Morphosedimentary developmentof drumlin-flank barriers with rapidly rising sea level,Story Head, Nova Scotia. Sedimentary Geology, 69,117–138.

Carter, R. W. G., Orford, J. D., Jennings, S. C., Shaw,J. & Smith, J. P. 1992. Recent evolution of a paragla-cial estuary under conditions of rapid sea-level rise:Chezzetcook Inlet, Nova Scotia. Proceedings of theGeological Association of London, 103, 167–185.

Church, M. & Ryder, J. M. 1972. Paraglacial sedimen-tation: a consideration of fluvial processes conditionedby glaciation. Geological Society of America Bulletin,83, 3059–3072.

Church, J. A., Clark, P. U. et al. 2014. Sea level change.In: Stocker, T. F., Qin, D. etal. (eds) Climate Change2013: The Physical Science Basis. Contributionof Working Group I to the Fifth Assessment Reportof the Intergovernmental Panel on Climate Change.Cambridge University Press, Cambridge, 1137–1216.http://www.climatechange2013.org/images/report/WG1AR5_Chapter13_FINAL.pdf.

Chute, N. E. & Nichols, R. L. 1941. The Geology of theCoast of Northeastern Massachusetts. MassachusettsDepartment of Public Works/US Geological SurveyCooperative Geology Project, Bulletin 7.

Costas, S. & FitzGerald, D. 2011. Sedimentary archi-tecture of a spit-end (Salisbury Beach, Massachusetts):the imprints of sea-level rise and inlet dynamics.Marine Geology, 284, 203–216.

Crider, H. B. 1998. Late Pleistocene development of theBrunswick, Maine sand plain and adjacent paleochan-nel. PhD thesis, Boston University.

Daly, R. A. 1934. The Changing World of the Ice Age.Yale University Press, New Haven.

Denton, G. H. 2011. East Antarctic retreat. NatureGeoscience, 4, 135–136.

Dickson, S. M. 1999. The Role of Storm-generated Com-bined Flows in Shoreface and Inner Continental ShelfSediment Erosion, Transport, and Deposition. PhDthesis, School of Marine Sciences, University of Maine.

Dolan, R. & Davis, R. E. 1992. An intensity scale forAtlantic coast northeast storms. Journal of CoastalResearch, 8, 840–853.

Dougherty, A. J., FitzGerald, D. M., Johnston, S. J.& Ferullo, B. 2003. Ground-penetrating radar pro-vides a detailed view of the complexities of barrierprogradation, Castle Neck, Massachusetts, USA. In:Proceedings Coastal Sediments ’03. CD-ROM, WorldScientific Publishing Corporation/East Meets WestProductions, Corpus Christi, Texas, USA.

Dougherty, A. J., FitzGerald, D. M., Pendleton, E. &Buynevich, I. V. 2004. Evidence for storm-dominatedearly progradation of Castle Neck Barrier, Massachu-setts, USA. Marine Geology, 210, 123–134.

Dowdeswell, J. A., Ottesen, D. & Rise, L. 2010. Ratesof sediment delivery from the Fennoscandian ice sheetthrough an ice age. Geology, 38, 3–6.

Duffy, W., Belknap, D. F. & Kelley, J. T. 1989. Mor-phology and stratigraphy of small barrier–lagoonsystems in Maine. Marine Geology, 88, 243–262.

Ehlers, J. & Gibbard, P. 2008. Extent and chronology ofquaternary glaciation. Episodes, 31, 211–218.

Farrell, S. C. 1969. Growth cycle of a small recurvedspit, Plum Island, Massachusetts. In: UNIVERSITYOF MASSACHUSETTS COASTAL RESEARCHGROUP (ed.) Coastal Environments: NortheasternMassachusetts and New Hampshire. University ofMassachusetts, Geology Department, Amherst,Field Trip Guidebook 1969 (Contribution CG-1),316–336.

Farrell, S. C. 1970. Sediment Distribution and Hydro-dynamics, Saco River and Scarborough Estuaries,Maine. Coastal Research Center, University of Massa-chusetts, Amherst, Contribution 6.

Farrell, S. C. 1972. Present coastal processes, recordedchanges, and the post-Pleistocene geologic record ofSaco Bay, Maine. PhD thesis, University ofMassachusetts.

Fenster, M. S. & FitzGerald, D. M. 1996. Morphody-namics, stratigraphy, and sediment transport patternsof the Kennebec River estuary, Maine, USA. Sedimen-tary Geology, 107, 99–120.

Fenster, M. S., FitzGerald, D. M., Kelley, J. T.,Belknap, D. F., Buynevich, I. V. & Dickson, S.M. 2001. Net ebb sediment transport in a rock-bound,mesotidal estuary during spring freshet conditions:Kennebec River estuary, Maine. Geological Societyof America Bulletin, 113, 1522–1531.

FitzGerald, D. M. 1993. Origin and stability if tidal inletsin Massachusetts. In: Aubrey, D. G. & Giese, G. S.(eds) Formation and Evolution of Multiple Tidal InletSystems. American Geophysical Union, Coastal andEstuarine Science Series, 44, 1–61.

FitzGerald, D. M. & van Heteren, S. 1999. Classifi-cation of paraglacial barrier systems: coastal NewEngland, USA. Sedimentology, 46, 1083–1108.

FitzGerald, D. M., Rosen, P. S. & van Heteren, S.1994. New England barriers. In: Davis, R. A. Jr.

(ed.) Geology of Holocene Barrier Island Systems.Springer-Verlag, Berlin, 305–394.



http://www.climatechange2013.org/images/report/WG1AR5_Chapter13_FINAL.pdf





FitzGerald, D. M., Buynevich, I. V., Fenster, M. S. &McKinlay, P. A. 2000. Sand dynamics at the mouth ofa rock-bound, tide-dominated estuary. SedimentaryGeology, 131, 25–29.

FitzGerald, D. M., Buynevich, I. V., Davis, R. A. Jr. &Fenster, M. S. 2002. New England tidal inlets withspecial reference to riverine-associated inlet systems.Geomorphology, 48, 179–208.

FitzGerald, D. M., Buynevich, I. V., Fenster, M. S.,Kelley, J. T. & Belknap, D. F. 2005. Coarse-grained sediment transport in northern New EnglandEstuaries: a synthesis. In: FitzGerald, D. M. &Knight, J. (eds) High Resolution Morphodynamicsand Sedimentary Evolution of Estuaries. Springer,New York, NY, 195–214.

FitzGerald, D. M., Fenster, M. S., Argow, B. A. &Buynevich, I. V. 2008. Coastal impacts due to sea-level rise. Annual Reviews of Earth and PlanetarySciences, 36, 601–647.

Forbes, D. L. 2005. Paraglacial coasts. In: Schwartz,M. L. (ed.) Encyclopedia of Coastal Science. Springer,Dordrecht, 760–762.

Forbes, D. L. & Syvitski, J. P. M. 1994. Paraglacialcoasts. In: Carter, R. W. G. & Woodroffe, C. D.(eds) Coastal Evolution: Late Quaternary ShorelineMorphodynamics. Cambridge University Press, Cam-bridge, 373–424.

Forbes, D. L. & Taylor, R. B. 1987. Coarse-grainedbeach sedimentation under paraglacial conditions,Canadian Atlantic coast. In: FitzGerald, D. M. &Rosen, P. S. (eds) Glaciated Coasts. AcademicPress, New York, NY, 51–86.

Forbes, D. L., Taylor, R. B., Orford, J. D., Carter,R. W. G. & Shaw, J. 1991. Gravel barrier migrationand overstepping. Marine Geology, 97, 305–313.

Forbes, D. L., Orford, J. D., Carter, R. W. G., Shaw, J.& Jennings, S. C. 1995. Morphodynamic evolution,self-organization, and instability of coarse-clastic bar-riers on paraglacial coasts. Marine Geology, 126,63–85.

Goldsmith, R. 1991. Stratigraphy of the Milford-DedhamZone, eastern Massachusetts: an Avalonian terrane.In: Hatch, N. L. Jr. (ed.) The Bedrock Geology ofMassachusetts. US Geological Survey ProfessionalPaper 1366–E–J.

Goodbred, S. L. 2003. Response of the Ganges dispersalsystem to climate change: a source-to-sink view sincethe last interstade. Sedimentary Geology, 121, 239–258.

Hanson, L. S. & Caldwell, D. W. 1989. The lithologicand structural controls on the geomorphology of themountainous areas in north-central Maine. In:Tucker, R. D. & Marvinney, R. G. (eds) Studiesin Maine Geology – Vol. 5: Quaternary Geol-ogy. Maine Geological Survey, Augusta, ME,147–167.

Hayes, M. O. & Ruby, C. H. 1994. Barriers of PacificAlaska. In: Davis, R. A. Jr. (ed.) Geology of Holo-cene Barrier Island Systems. Springer-Verlag, Berlin,395–433.

Hein, C. J. 2012. Styles of coastal evolution in response toHolocene changes in sea level and sediment supply.PhD thesis, Boston University.

Hein, C. J., FitzGerald, D. & Barnhardt, W. 2007.Holocene reworking of a sand sheet in the Merrimack

Embayment, Western Gulf of Maine. Journal ofCoastal Research, SI 50, 863–867.

Hein, C. J., FitzGerald, D. M., Stone, B. D., Car-

ruthers, E. A. & Gontz, A. M. 2011a. The role ofbackbarrier infilling in the formation of barrier islandsystems. In: Kraus, N. C. & Rosati, J. D. (eds)Coastal Sediments ’11, Proceedings of the 8th Inter-national Symposium on Coastal Engineering andScience of Coastal Sediment Processes. World Scien-tific, Hackensack, NJ, USA, 1203–1216.

Hein, C. J., FitzGerald, D. M., Milne, G. A., Bard, K.A. & Fattovich, R. 2011b. Evolution of a Pharaonicharbor on the Red Sea: implications for coastalresponse to changes in sea level and climate.Geology, 39, 687–690.

Hein, C. J., FitzGerald, D. M., Carruthers, E. A.,Stone, B. D., Barnhardt, W. A. & Gontz, A. M.2012. Refining the model of barrier island formationalong a paraglacial coast in the Gulf of Maine.Marine Geology, 307–310, 40–57.

Hein, C. J., FitzGerald, D. M., Barnhardt, W. A. &Stone, B. D. 2013. Onshore–Offshore SurficialGeologic Map of the Newburyport East and NorthernHalf of the Ipswich Quadrangles, Massachusetts.Scale 1:24 000. Massachusetts Geological Survey.Geologic Map 13–01.

Heron, S. D. Jr., Moslow, T. F., Berelson, W. M.,Herbert, J. R., Steele, G. A., III & Susman, K. R.1984. Holocene sedimentation of a wave-dominatedbarrier-island shoreline: Cape Lookout, NorthCarolina. In: Greenwood, B. & Davis, R. A. Jr.

(eds) Hydrodynamics and Sedimentation in Wave-Dominated Coastal Environments. Marine Geology,60, 413–434.

Hewitt, K., Byrne, M.-L., English, M. & Young, G.(eds) 2002. Landscapes of Transition: Landform andSediment Associations in Cold Regions. Kluwer,Amsterdam.

Hill, H. W., Kelley, J. T., Belknap, D. F. & Dickson, S.M. 2004. The effects of storms and storm-generatedcurrents on sand beaches in Southern Maine, USA.Marine Geology, 210, 149–168.

Hine, A. C. & Boothroyd, J. C. 1978. Morphology, pro-cesses, and recent sedimentary history of a glacial-outwash plain shoreline, southern Iceland. Journal ofSedimentary Research, 48, 901–920.

Hubbard, D. K. 1976. Changes in inlet offset due tostabilization. In: ASCE (ed.) Proceedings of the 15thInternational Conference on Coastal Engineering.ASCE, New York, NY, 1805–1823.

Jacobson, G. L. & Jacobson, H. A. 1989. An inventory ofdistribution and variation in salt marshes from differentsettings along the Maine coast. In: Anderson, W. A. &Borns, H. W. (eds) Neotectonics of Maine. MaineGeological Survey Bulletin, 40, 69–83.

Jensen, R. E. 1983. Atlantic Coast Hindcast, Shallow-Water, Significant Wave Information. USACE, Vicks-burg, WIS Report 9.

Kaye, C. A. 1961. Pleistocene stratigraphy of Boston,Massachusetts. In: Short Papers in the Geologic andHydrologic Sciences. US Geological Survey Pro-fessional Paper, 424-B, B73–B76.

Kelley, J. T. 1987. Sedimentary environments alongMaine’s estuarine coastline. In: FitzGerald, D. M.

C. J. HEIN ET AL.




& Rosen, P. S. (eds) Glaciated Coasts. AcademicPress, New York, NY, 151–176.

Kelley, J. T. 1989. A preliminary mineralogical analy-sis of glaciomarine mud from the western margin ofthe Gulf of Maine. Northeastern Geology, 11,141–150.

Kelley, J. T., Shipp, R. C. & Belknap, D. F. 1989. Geo-morphology and Late Quaternary evolution of the SacoBay region. In: Tucker, R. D. & Marvinney, R. G.(eds) Studies in Maine Geology. Maine GeologicalSurvey, Augusta, ME, Quaternary Geology, 5, 47–65.

Kelley, J. T., Dickson, S. M., Belknap, D. F. &Stuckenrath, R. Jr. 1992. Sea-level change andlate Quaternary sediment accumulation on the south-ern Maine inner continental shelf. In: Fletcher, C.& Wehmiller, J. (eds) Quaternary Coasts of theUnited States: Marine and Lacustrine Systems. Soci-ety for Sedimentary Geology, Special Publications,48, 23–34.

Kelley, J. T., Kelley, A. R. & Apollonio, S. 1995a.Landforms of the Gulf of Maine. In: Conkling, P.(ed.) From Cape Cod to the Bay of Fundy: An Environ-mental Atlas of the Gulf of Maine. MIT Press, Cam-bridge, MA, 19–39.

Kelley, J. T., Gehrels, W. R. & Belknap, D. F. 1995b.Late Holocene relative sea-level rise and the geologicaldevelopment of tidal marshes at Wells, Maine, U.S.A.Journal of Coastal Research, 11, 136–153.

Kelley, J. T., Belknap, D. F. et al. 1995c. A SandBudget for Saco Bay, Maine: Report to Maine andNew Hampshire Sea Grant Programs. Maine Geologi-cal Survey Open File Report 95-1.

Kelley, J. T., Dickson, S. M., Belknap, D. F., Barn-

hardt, W. A. & Barber, D. C. 2003. Distributionand volume of sand bodies on the rocky, glaciatedinner continental shelf of the northwestern Gulf ofMaine. Journal of Coastal Research, 19, 41–56.

Kelley, J. T., Barber, D. C., Belknap, D. F.,FitzGerald, D. M., van Heteren, S. & Dickson,S. M. 2005. Sand budgets at geological, historicaland contemporary time scales for a developed beachsystem, Saco Bay, Maine, USA. Marine Geology,214, 117–142.

Kelley, J. T., Belknap, D. F. & Claesson, S. H. 2010.Drowned coastal deposits with associated archaeologi-cal remains from a sea-level ‘slowstand’: NorthwesternGulf of Maine, USA. Geology, 38, 695–698.

Kelley, A. R., Kelley, J. T., Belknap, D. F. & Gontz,A. M. 2011. Coastal and terrestrial impact of the isosta-tically forced late Quaternary drainage divide shift,Penobscot and Kennebec Rivers, Maine, USA.Journal of Coastal Research, 27, 1085–1093.

Kelley, J. T., Belknap, D. F., Kelley, A. R. & Claes-

son, S. H. 2013. A model for drowned terrestrial habi-tats with associated archeological remains in thenorthwestern Gulf of Maine, USA. Marine Geology,338, 1–16.

Kirwan, M. L., Murray, A. B., Donnelly, J. P. &Corbett, D. R. 2011. Rapid wetland expansionduring European settlement and its implication formarsh survival under modern sediment delivery rates.Geology, 39, 507–510.

Kliewe, H. & Janke, W. 1991. Holozaner Kustenaus-gleich im sudlichen Ostseegebiet bei besonderer

Berucksichtigung der Boddenausgleichskuste Vor-pommerns. Petermanns Geographische Mitteilun-gen, 135, 1–15.

Knight, J. & Harrison, S. 2009. Periglacial and paragla-cial environments: a view from the past into the future.In: Knight, J. & Harrison, S. (eds) Periglacial andParaglacial Processes and Environments. GeologicalSociety, London, Special Publications, 320, 1–4.

Leeder, M. R., Harris, T. & Kirkby, M. J. 1998. Sedi-ment supply and climate change: implications forbasin stratigraphy. Basin Research, 10, 7–18.

Lentz, S. 2012. Buoyant coastal currents. In: Chassignet,E., Cenedese, C. & Verron, J. (eds) Buoyancy-DrivenFlows. Cambridge University Press, Cambridge,164–202.

Lynch, D. R., Holboke, M. J. & Nairrie, C. E. 1997. TheMaine coastal current: spring climatological circula-tion. Continental Shelf Research, 17, 605–634.

Lyons, J. B., Bothner, W. A., Moench, R. H. & Thomp-

son, J. B. 1997. Bedrock Geology Map of New Hamp-shire. Scale 1:500 000. US Geological Survey. http://ngmdb.usgs.gov/Prodesc/proddesc_37338.htm.

McCormick, C. L. 1968. Holocene stratigraphy of themarshes at Plum Island, Massachusetts. PhD thesis,University of Massachusetts.

McIntire, W. G. & Morgan, J. P. 1964. Recent Geo-morphic History of Plum Island, Massachusetts, andAdjacent Coasts. Louisiana State University Press,Baton Rouge, LA, Louisiana State University CoastalStudies Series, 8.

Mercier, D. 2009. Paraglacial and paraperiglacial land-systems: concepts, temporal scales and spatial distri-bution. Geomorphologie, 4/2008, 223–233.

Milliman, J. D. & Farnsworth, K. L. 2011. River Dis-charge to the Coastal Ocean. Cambridge UniversityPress, Cambridge.

Nichol, S. L. & Boyd, R. 1993. Morphostratigraphy andfacies architecture of sandy barriers along the Easternshore of Nova Scotia. Marine Geology, 114, 59–80.

Nummedal, D., Hine, A. C. & Boothroyd, J. C. 1987.Holocene evolution of the south-central coast ofIceland. In: FitzGerald, D. M. & Rosen, P. R.(eds) Glaciated Coasts. Academic Press, New York,NY, 115–150.

Oldale, R. N. 1985. A drowned Holocene barrier spit offCape Ann, Massachusetts. Geology, 13, 375–377.

Oldale, R. N. 1989. Timing and mechanisms for thedeposition of the glacimarine mud in and around theGulf of Maine: a discussion of alternative models. In:Tucker, R. D. & Marvinney, R. G. (eds) Studies inMaine Geology. Quaternary Geology. Maine Geologi-cal Survey, Augusta, ME, 5, 1–10.

Oldale, R. N. & Edwards, G. B. 1990. Cores fromMarine Geologic Features in the Western Gulf ofMaine. US Geological Survey Field Study MapMF-2147.

Oldale, R. N., Wommack, L. E. & Whitney, A. B. 1983.Evidence for a postglacial low relative sea-levelstand in the drowned delta of the Merrimack River,western Gulf of Maine. Quaternary Research, 19,325–336.

Oldale, R. N., Colman, S. M. & Jones, G. A. 1993.Radiocarbon ages from two submerged strandline fea-tures in the western Gulf of Maine and a sea-level curve



http://ngmdb.usgs.gov/Prodesc/proddesc_37338.htm





for the northeastern Massachusetts coastal region.Quaternary Research, 40, 38–45.

Orford, J. D. & Carter, R. W. G. 1985. Storm-generated dune armouring on a sand-gravel barriersystem, southeastern Ireland. Sedimentary Geology,42, 65–82.

Orford, J. D., Carter, R. W. G. & Jennings, S. C. 1991.Coarse clastic barrier environments: evolution andimplications for Quaternary sea-level interpretation.Quaternary International, 9, 87–104.

Osberg, P. H., Hussey, A. M. II & Boone, G. M. 1985.Bedrock Geologic Map of Maine. Scale 1:500 000.Maine Geological Survey. http://ngmdb.usgs.gov/Prodesc/proddesc_16547.htm.

Rampino, M. R. & Sanders, J. E. 1981. Evolution of thebarrier islands of southern Long Island, New York.Sedimentology, 28, 37–47.

Retelle, M. J. & Weddle, T. K. 2001. Deglaciationand relative sea-level chronology, Casco BayLowland and lower Androscoggin River valley,Maine. In: Weddle, T. K. & Retelle, M. J. (eds)Deglacial History and Relative Sea-Level Changes,Northern New England and Adjacent Canada, Geo-logical Society of America, Special Papers, 351,191–214.

Rhodes, E. G. 1973. Pleistocene–Holocene SedimentsInterpreted by Seismic Refraction and Wash-BoreSampling, Plum Island-Castle Neck, Massachusetts,Coastal Engineering Research Center. US ArmyCorps of Engineers, Vicksburg, USACE TechnicalMemorandum 40.

Ridge, J. C. 2004. The Quaternary glaciations of westernNew England with correlations to surrounding areas.In: Ehlers, J. & Gibbard, P. L. (eds) Quaternary Gla-ciations, Extent and Chronology Part II: NorthAmerica. Elsevier, Amersterdam, 169–199.

Robinson, G. R. & Kapo, K. E. 2003. Generalized Lithol-ogy and Lithogeochemical Character of Near-SurfaceBedrock in the New England Region. US GeologicalSurvey Open File Report 03–225.

Rosen, P. S. 1984. Gravel spit processes, ThompsonIsland, Boston Harbor, Massachusetts. In: Hanson,L. (ed.) Geology of the Coastal Lowlands: Boston,MA to Kennebunk, ME. Salem State College, Salem,New England Intercollegiate Geological ConferenceGuidebook 76, 25–38.

Ryder, J. M. 1971. Some aspects of the morphometry ofparaglacial alluvial fans in south-central BritishColumbia. Canadian Journal of Earth Sciences, 8,1252–1264.

Sallenger, A. H. Jr., Doran, K. S. & Howd, P. A. 2012.Hotspot of accelerated sea-level rise on the Atlanticcoast of North America. Nature Climate Change, 2,884–888.

Short, A. D. 1979. Barrier island development along theAlaskan-Yukon coastal plains. Geological Society ofAmerica Bulletin, Part II, 90, 77–103.

Slaymaker, O. 2009. Proglacial, periglacial or paragla-cial?. In: Knight, J. & Harrison, S. (eds) Periglacialand Paraglacial Processes and Environments. Geo-logical Society, London, Special Publications, 320,71–84.

Smith, J. B. 1991. Morphodynamics and Stratigraphy ofEssex River Ebb-Tidal Delta: Massachusetts. USArmy Engineer Waterways Experiment Station, USArmy Corps of Engineers, Vicksburg, USACE, Tech-nical Report CERC-91-11.

Smith, J. B. & FitzGerald, D. M. 1994. Sediment trans-port patterns at the Essex River Inlet ebb tidal delta,Massachusetts, U.S.A. Journal of Coastal Research,10, 752–774.

Stea, R. R., Boyd, R., Fader, G. B. J., Courtney, R. C.,Scott, D. B. & Pecore, S. S. 1994. Morphology andseismic stratigraphy of the inner continental shelf offNova Scotia, Canada: evidence for a –65 m low-standbetween 11 650 and 11 250 14C yr B.P. MarineGeology, 117, 135–154.

Stone, B. D. & Peper, J. D. 1982. Topographic control ofthe deglaciation of eastern Massachusetts. In: Larson,G. J. & Stone, B. D. (eds) Late Wisconsin Glaciationof New England. Kendall/Hunt, Dubuque, 145–166.

Stone, B. D., Stone, J. R. & McWeeney, L. J. 2004.Where the glacier met the sea: Late Quaternarygeology of the northeast coast of Massachusetts fromCape Ann to Salisbury. In: Hanson, L. (ed.) Geologyof the Coastal Lowlands: Boston, MA to Kennebunk,ME. Salem State College, Salem, New England Inter-collegiate Geological Conference Guidebook 76, TripB-3.

Stone, B. D., Stone, J. R. & DiGiacomo-Cohen, M. L.2006. Surficial Geologic Map of the Salem-Newburyport East-Wilmington-Rockport QuadrangleArea in Northeast Massachusetts. U.S. GeologicalSurvey Open-File Report, 2006-1260-B. U.S. Geologi-cal Survey, Reston, Virginia.

Swift, D. J. P. 1968. Coastal erosion and transgressivestratigraphy. Journal of Geology, 76, 444–456.

Syvitski, J. P. 1991. Towards an understanding of sedi-ment deposition on glaciated continental shelves. Con-tinental Shelf Research, 11, 897–937.

Tary, A. K., FitzGerald, D. M. & Buynevich, I. V.2001. Late Quaternary morphogenesis of a marine-limit delta plain in southwest Maine. In: Weddle,T. K. & Retelle, M. J. (eds) Deglacial History andRelative Sea-Level Changes, Northern New Englandand Adjacent Canada. Geological Society ofAmerica, Special Papers, 351, 125–149.

Thompson, W. B. & Borns, H. W. Jr 1985. Surficial Geo-logic Map of Maine. Scale: 1:500 000. Maine Geologi-cal Survey, Augusta, ME.

Thompson, W. B., Crossen, K. J., Borns, H. W., Jr &Andersen, B. G. 1989. Glaciomarine deltas ofMaine and their relation to late Pleistocene–Holocenecrustal movements. In: Andersen, W. A. & Borns,H. W. Jr. (eds) Neotectonics of Maine: Studies in Seis-micity, Crustal Warping, and Sea-Level Change.Maine Geological Survey, Augusta, ME, Bulletin 40,43–67.

Van Heteren, S. 1996. Preserved records of coastal-morphologic and sea-level changes in the stratigraphyof paraglacial barriers. PhD thesis, Boston University.

Yang, S. L., Milliman, J. D., Li, P. & Xu, K. 2010.50 000 dams later: Erosion of the Yangtze River andits delta. Global and Planetary Change, 75, 14–20.

C. J. HEIN ET AL.







Date post:	26-Nov-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

Evolution of paraglacial coasts in response to changes in fluvial sediment supply

Documents