storm erosion during the past 2000years along the north ... · storm erosion during the past 2000...
TRANSCRIPT
Geomorphology 208 (2014) 160–172
Contents lists available at ScienceDirect
Geomorphology
j ourna l homepage: www.e lsev ie r .com/ locate /geomorph
Storm erosion during the past 2000 years along the north shore ofDelaware Bay, USA
Daria L. Nikitina a,⁎, Andrew C. Kemp b, Benjamin P. Horton c,d, Christopher H. Vane e,Orson van de Plassche f,†, Simon E. Engelhart g
a Department of Geology and Astronomy, West Chester University, West Chester, PA 19383, USAb Department of Earth and Ocean Sciences, Tufts University, Medford, MA 02155, USAc Sea Level Research, Institute of Marine and Coastal Sciences, Rutgers, New Brunswick, NJ, USAd Earth Observatory of Singapore and Division of Earth Sciences, Nanyang Technological University, 639798 Singapore, Singaporee British Geological Survey, Environmental Science Centre, Keyworth, Nottingham NG12 5GG, UKf Department of Marine Biogeology, Faculty of Earth & Life Sciences, Vrije Universiteit Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlandsg Department of Geosciences, University of Rhode Island, Kingston, RI 02881, USA
⁎ Corresponding author. Tel.: +1 610 436 3103.E-mail address: [email protected] (D.L. Nikitina).
† Deceased.
0169-555X/$ – see front matter © 2013 Elsevier B.V. All rhttp://dx.doi.org/10.1016/j.geomorph.2013.11.022
a b s t r a c t
a r t i c l e i n f oArticle history:Received 10 May 2013Received in revised form 6 November 2013Accepted 23 November 2013Available online 11 December 2013
Keywords:Salt-marshTropical cyclonePaleotempestologyHoloceneDelaware Estuary
The recent impacts of tropical cyclones and severe storms on the U.S. Atlantic coast brought into focus the needfor extended records of storm activity from different geomorphologic settings. Such reconstructions are typicallydeveloped from sites that experienced repeated overwash of sand into low-energy, depositional environments.However, salt-marsh sedimentmay also preserve a record of repeated erosion from tropical cyclones and storms.We describe late Holocene sediments beneath the Sea Breeze salt marsh (Delaware Bay, New Jersey) frommore than 200 gouge cores positioned along seven transects. The stratigraphic record documents at leastseven depositional sequences consisting of salt-marsh peat and mud couplets that represent dramatic changesin sedimentation regime. There are a number of processes that could produce these stratigraphic sequencesagainst a background of rising relative sea level including: lateralmigration of tidal creeks; tidal channel networkand/or drainage ditch expansion; changes in sediment delivery rates; rapid relative sea-level change; tsunami;and formation of salt pans. The abrupt contacts between the salt-marsh peat and overlying intertidalmud suggestthat erosion of the peat was followed by rapid infilling of accommodation space. Correlation of erosional surfacesacross 2.5 km suggests a common mechanism and we propose that the erosion was caused by tropical cyclonesand/or storms. We developed a chronology of repeated salt-marsh erosion and recovery using 137Cs, metalpollution (Pb concentration and stable isotopes), and radiocarbon data. Two recent episodes of salt-marsherosion may correlate with historic tropical cyclones in AD 1903, and AD 1821/1788 that impacted the Atlanticcoast of New Jersey, but the erosive nature of the Sea Breeze site hinders definitive correlation. Prehistoricerosional sequences correlate with overwash fans preserved in the regional sedimentary record. We estimatedthat it takes from several decades to almost 200 years for salt-marshes to recover from storm erosion.
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
TheU.S. Atlantic coast is at risk from landfalling tropical cyclones andstorms that cause economic, social, and environmental devastation.Significant uncertainty surrounds projected tropical cyclone activity inresponse to climate change (e.g. Emanuel et al., 2008; Knutson et al.,2010; Lin et al., 2012; Holland and Bruyere, 2013). Instrumental andobservational records are too short to adequately describe trends intropical cyclone activity particularly for the largest and rarest events(Elsner, 2007; Landsea, 2007; Liu, 2007). Paleotempestology seeks toreconstruct and explain the spatial and temporal variability in
ights reserved.
frequency and intensity of landfalling tropical cyclones and severestorms during past centuries and millennia.
Tropical cyclone deposits are most commonly recognized as anoma-lous sand layers deposited by storm surges in low-energy environmentssuch as back-barriermarshes, lakes, and lagoonswhere “normal” condi-tions between tropical cyclones are characterized by deposition of or-ganic and fine-grained sediments (e.g. Liu and Fearn, 1993; Donnellyet al., 2001a, 2001b; Woodruff et al., 2008; Madsen et al., 2009;Williams and Flanagan, 2009). The character of these storm-surgedeposits varies by locality depending upon the source of sediment,flow characteristics and duration of the storm surge, and local topogra-phy (e.g. Morgan et al., 1958; Cahoon, 2006; Sallenger et al., 2006;Hawkes and Horton, 2012). On coastlines characterized by barrierisland geomorphologies such as the U.S. mid-Atlantic, storm surges ofsufficient height overtop barrier islands and redistribute sediment
Erosive, abrupt stratigraphic contact
Ris
ing
Sea
Lev
el
Low-marsh (Sp. alterniflora)
rapid sediment accumulation
accommodation space created by erosion
High-marsh(Sp. patens)(D. spicata)
A
B
C
D
E
Tidal flat mud
Low marsh peat
High marsh peat
Plant Communities Sediment Marsh surface 1
Paleomarsh surface 1
Marsh surface 2
Paleomarsh surface 2
Modern marsh surface
F
maximum age of erosion(Spartina alterniflora colonizationof tidal mud deposit)
minimum age ofsalt-marsh recovery
sediment core (F)
dept
h in
cor
e
year
rapid infill bytidal mud
hiatus of erodedsediment
recolonization andinfill of low-marsh peat
Fig. 1. Proposedmodel for the development of a sedimentary record of tropical cyclone- orstorm-induced erosion in a salt marsh. (A–B) Under a regime of rising relative sea level,the salt marsh accumulates sediment to preserve its elevation in the tidal frame. (C) Theaction of a storm surge erodes and removes a portion of the previously undisturbedsedimentary record. The now exposed surface is located at a lower elevation with respectto sea level. (D–E) The newly created accommodation space is rapidly infilled, first byintertidal mud and then by low-marsh and high-marsh peat as sediment accumulationraises the elevation of the surface in the tidal frame. The resulting sediment is a regressivesequence of relative sea level change. The erosive surface represents a hiatus in thesedimentary record that can be identified by a sharp contact between sedimentary unitsand a temporal gap in accumulation.
161D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
from the beach and shore front to the back barrier where it is depositedas a washover fan, often on top of a salt-marsh. Studies of back-barriersediments in southern New Jersey revealed that the timing of washoverfanspreserved in the sedimentary recordwas consistentwith depositionduring intense historical tropical cyclones and storms in AD 1938, AD1944, AD 1950, and AD 1962 (Donnelly et al., 2001b; Donnelly andWebb III, 2004). Additional overwash deposits were likely depositedby similar tropical cyclones or storms that made landfall in AD 1821(Donnelly et al., 2001b), AD 550 to 1400, and AD 500 to 600 (Donnellyet al., 2001b; Donnelly andWebb III, 2004). Although the identificationof tropical cyclone deposits in coastal lakes and salt-marshes is usuallybased on the recognition of sand units and/or microfossil evidence,organic geochemical signatures such as C/N, δ13C and δ15N provide acomplementary approach (Lambert et al., 2008).
The sedimentary record beneath salt marshes may also preservean erosive signature of tropical cyclones and storms. For example, strat-igraphic and radiocarbon data from a Connecticut salt marsh showedwidespread erosion events at 1400 to 1440 AD and prior to 1640 to1670 AD (van de Plassche et al., 2004; van de Plassche et al., 2006).Each time the erosion was followed by rapid and complete infilling ofthe accommodation space created by erosion with a regressive relativesea level (RSL) sequence of tidal mud overlain by low-marsh and thenhigh salt-marsh peat as the salt marsh recovered (Fig. 1). The erosiveevents in Connecticut occurred at the same time as deposition ofoverwash fans ~60 km away in Rhode Island (Donnelly et al., 2001a).Van de Plassche et al. (2006) concluded that the most likely cause ofthis sedimentary pattern was the erosive action of a storm surge.
We reconstruct a history of erosional events impacting the northernshore of Delaware Bay in the last ~2000 years by applying the approachof van de Plassche et al. (2006). Detailed litho-, bio- and chrono-stratigraphic investigation of a salt marsh at Sea Breeze, New Jersey,identifies at least seven episodes of substantial erosion followed byrapid sediment accumulation and salt-marsh recovery. After consider-ing a wide range of potential mechanism of salt-marsh erosion, frommicro-scale tidal creek migration to macro-scale events such as tsuna-mi, we propose that the erosion was caused by storm surges. Thisstudy aids the building of regional databases of past tropical cyclonesand/or severe storms that are necessary for comprehensive spatial andtime-series analyses and illustrates the influence of storm surges ontidal salt-marshes.
2. Study area
Sea Breeze is located in New Jersey on the northern shore of theDelaware Bay (Fig. 2A) and experienced RSL rise throughout theHolocene (Miller et al., 2009; Engelhart and Horton, 2012; Hortonet al., 2013). During the last ~2000 years (prior to AD 1900), RSL roseat a rate of ~1.3 mm/yr, primarily because of glacial isostatic adjustment(Engelhart et al., 2011). This continuous RSL rise created accommoda-tion space that was infilled by estuarine sediments, preserving a detailedrecord of RSL change and coastal processes during the Holocene. Thetidal regime in this region is semidiurnal and the great diurnal range(mean lower low water, MLLW, to mean higher high water, MHHW) atthe nearest NOAA tide gauge is 1.78 m (Reedy Point; Fig. 2A). Fine-grained, minerogenic sediments transported by the Delaware River aretrapped in the estuary and distributed to the salt marshes by tides(Sommerfield and Wong, 2011). Salt marshes on the Delaware Bay(like most on the U.S. Atlantic coast) are highly organic and include littlesand, with silt being the most dominant size fraction of clastic material(Nikitina et al., 2003).
The Sea Breeze salt-marsh is a ~2 km wide platform, dissectedby tidal channels and underlain by 2–4 m of salt-marsh and tidal-flatsediments (Fig. 2B). A small group of houses existed on the seawardedge of the marsh from the late 19th century before being abandonedin AD 2010. A seawall was built in AD 2007 to protect the houses fromstormerosion, butwasdamaged the same year and is no longer repaired.
No other engineering structures were in place prior to AD 2007.The modern salt-marsh is subdivided into four, vertically-aligned sub-environments reflecting the varied tolerance of plants to the frequency
open water
salt marsh
Delaware Bay
NEW JERSEY
Philadelphia
Atlantic City
Atlantic Ocean
20km
39.5oN
39.0oN
40.0oN
74.5oW75.0oW
PENNSYLVANIA
DELAWARE
Sea Breeze (B)
N
1
2
Paleostorm Site1. Brigantine2. Whale Beach
A
Reedy Point
I’
I
II’III’
VI’
VI
24
5663 62
32
75°19'0"W
39°19'0"N
39°18'30"N
75°18'30"W
0 250 500 750 1,000125
B
Delaware Bay
transect
core
IV
V VII’
V’
IV’
59
III
II
N
30
I
II
VIIVV
5663
IV
VV
55
IIIII
IIIIIII
N
VII
town road
Fig. 2. (A) Location of the study site in New Jersey, USA. Depositional records of pastoverwash events attributed to tropical cyclones or storms are located at Brigantine andWhale Beach (Donnelly et al., 2001b). (B) Transects of cores across the Sea Breeze sitewere used to describe the stratigraphy underlying the modern salt marsh. The locationof cores used in detailed analysis and with reported radiocarbon ages is shown.
Fig. 3. Examples of sediment cores recovered using different coring devices. Each photo il-lustrates abrupt contacts between lithologic units. (A) Core recovered with 2.5-cm-wideEijkelkamp core auger. A sharp contact between salt marsh peat and tidal gray mud ofsequence 5 is at 42 cmmark. (B) Segment of core 63 recovered with 5-cm-wide Russianpeat auger. The photo shows the entire sequence 1 and an abrupt change betweenhigh-marsh peat and gray mud of sequence 2 at 355 cm depth. The core was placed inplastic pipe and sampled in the laboratory for 14C AMS dating, grain-size and LOI analysis.(C) Core 30 recovered with PVC pipe and sampled for 137Cs dating.
162 D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
anddurationof inundationbysalt orbrackishwater. Tidal-flatenvironmentsbetween MLLW and mean tide level (MTL) are not colonized by vascu-lar plants and are characterized by gray mud. Low salt-marsh envi-ronments along the banks of tidal channels correspond to elevationsbetween approximately MTL and mean high water (MHW) and arecolonized by tall form of Spartina alterniflora. Most of the modernmarsh platform is vegetated by Spartina patens, Distichlis spicata, andthe stunted formof Spartina alterniflora. This high salt-marsh communi-ty typically exists between MHW and MHHW (McKee and Patrick,1988; Oertel andWoo, 1994). Phragmites australis and Iva frutescens oc-cupy the elevated banks (levees) of tidal creeks and the borders be-tween salt marshes and freshwater uplands at elevations aboveMHHW.Meadows of Schoenoplectus spp. are also found at the boundarybetween upland salt marshes and freshwater uplands and represent
brackish, waterlogged conditions (Tiner, 1985; Stuckey and Gould,2000). Macrofossils of plant species from each of the vegetated sub-environments are commonly recognizable in buried sediments andaid reconstruction of tidal flat and salt-marsh facies.
3. Methods
3.1. Core recovery and sampling
The sediment beneath the Sea Breeze saltmarshwas described frommore than 200 hand-driven gouge cores positioned along seventransects (Fig. 2B). Cores were recovered in 1-meter segments with2.5-cm-wide Eijkelkamp gouge cores. Multiple cores with a 5 to 15 mradius were taken at many sites to confirm the stratigraphy. Sedimentsequences were described using the Tröels-Smith (1955) scheme. Thesediment type, color, texture, organic content, the presence of identifi-able plant macrofossils, and the character of transitions between litho-logic units were described in the field. Selected core sections werephotographed at the time of collection to document abrupt changesin lithology (Fig. 3A, B, C). We developed a detailed lithostratigraphyfrom all cores to document sediment accumulation patterns and re-construct paleoenvironmental changes (Fig. 4). We identified plantmacrofossils preserved in core material by comparing roots, rhizomesand plant stems with modern examples of the same species and fieldguides (Niering et al., 1977; Tiner, 1985). Core locations were recordedusing GPS and the surface elevation of each core was measured using aSokkia Total Station and referenced to North American Vertical Datum
high-marsh peat
low-marsh mud/muddy peat
tidal mud pre-Holocene sand
tidal creek
50m
10 2
78 56 6314 50
51 29 30 31 32 48 26 52 49 27 46 45 90
7
65
2 1
3 3
4
5
6
7
6
5 5
32
12 1
3
44
5
7
6
32
I I’
2
1451-1632
893-1012900-1040
214-381
BC 390-200
1319-1438
981-1147
890-1030
260-520
1665-1950
62
NW SE
1.0
0.0
-1.0
-2.0
-3.0
980-1150
776-1119
0.5
-0.5
-1.5
-2.5
4
km
201
202
Ele
vati
on (
m, N
AV
D88
)
transect I-I’
6
NWII II’
SE
53 79 8081 82
13 71 70 67 6968 66 65 58 64
7
5
43
2
4 4
5
7
56
4
32
12
1
34
3
567
6
7
1.00.50.0
-0.5-1.0-1.5-2.0-2.5-3.0
transect II-II’
NW SEIII III’
75 74 73 72 59 116 115
5
76
4 4
32
123
5
8
1428-1610
429-567
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0 transect III-III’
NW
V’SE
22 87 23 88 89 24 5 25112 111 110 109 108 107
787
876
7
56
V
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
1.5
transect V-V’
NWIV
SEIV’
76 EW
3
EW
2
EW
1
19
106
105
104
103
102
99 101
100
98 96 97 94 93 928
7
6 5
7
5
7
6
54 4
5
7
6
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0transect IV-IV’
VISW NE
VI’
82 75 83 8485 20
320
4
86 98 110
5 56
566
7
4
7
3
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
34
transect VI-VI’
53 5455 33 34 35 E
SS 4
81
106E 88 23
NESW
7
65
4
87
4
76
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
VII VII’
transect VII-VII’
radiocarbon date
37sediment core(fully recovery)
6 numbered regressive sequence
50m50m
50m50m
50m
45a46a 45b
80a80b
sediment core(partial recovery)
5
Fig. 4. Stratigraphic cross-sections from seven transects of exploratory cores described at the Sea Breeze sites. The position of radiocarbon-dated samples is shown (open circles) and theregressive sequences are numbered 1–8 in each cross section. All vertical axes are altitude relative to North American Vertical Datum 1988 (NAVD88). Dates are calibrated radiocarbonages expressed in years AD.
163D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
1988 (NAVD88). Representative cores for laboratory analysis and datingwere recovered in 0.5 m sections using a 5 cm-wide Russian-type handauger to minimize compaction and contamination from upper sectionsduring coring. The cores were sealed in plastic wrap and refrigerated.
One core (63, located on transect I–I′) was sampled at a resolution of10 cm to ensure that all stratigraphic units were adequately represent-ed. Grain-size and loss-on-ignition (LOI) analyses refined and supportedthe lithology described in the field. Grain size distribution was mea-sured using a Beckman Coulter laser particle-size analyzer following
the sediment preparation procedures of Pilarczyk et al. (2012). Grainsize classifications followed the Wentworth scale. Organic content wasdetermined by LOI following the procedures of Ball (1964).
3.2. Radiocarbon dating
We sampled identifiable plant macrofossils from six cores (30, 32,56, 59, 62, and 63) for accelerator mass spectrometer (AMS) radiocar-bon dating. Since salt-marsh plants have a known relationship to
164 D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
salt-marsh surface, corms (short, vertical, underground plant stems)preserved in growth position were picked from the sediment cores toensure the accuracy of dating of paleo-salt-marsh surfaces. Sampleswere cleaned under a microscope to remove contaminating materials,such as adhered sediment or invasive younger roots, and dried at50 °C. Radiocarbon ages were calibrated using CALIB v6.1.0. (Stuiverand Reimer, 1993) and the IntCal09 calibration data set (Reimer et al.,2011). We report original radiocarbon ages and calibrated years AD/BCwith 2σ calibrated uncertainty for 14 radiocarbon dates (Table 1). Theages constrain when changes in sedimentary environment took place,provide a means to verify correlation of regressive sequences amongcores, confirm the presence of erosive hiatuses, and estimate the timenecessary for salt-marsh recovery after erosion. The maximum age oferosion was estimated by radiocarbon dating corms of alterniflorafound beneath the upper contact of the mud unit that overliesthe high-marsh peat (Fig. 1F). The minimum age of high salt-marshrecovery was estimated by radiocarbon dating corms of patens andSchoenoplectus americanus from the contact between the low-marshand high-marsh peat (Fig. 1E).
3.3. Identifying chronological markers
Use of radiocarbon to date macrofossils from the last ~400 years ishindered by a plateau in the calibration curve resulting in multipleages and large uncertainty. To address this limitation we identifiedchronostratigraphic pollution markers of 137Cs activity, Pb concen-trations, and stable Pb isotopes in bulk sediment to constrain recentdepositional histories. Three, 80 cm long, cores (24, 30, and 56) wererecovered and sampled at 2 cm intervals, dried and ground to homoge-nized powder. For measurement of 137Cs activity, samples were sealedin 60 ml plastic jars and stored for at least 30 days to ensure radioactiveequilibrium between Ra and Bi, and then counted for at least 24 h onCanberraModel 2020 low-energy germaniumdetectors. Measurementsof 137Cs (t1/2 = 30.1 years) activity were made by gamma spectrom-etry. Peak 137Cs activity occurred in AD 1963 as a result of above-
Table 1Radiocarbon dates from Sea Breeze, New Jersey.
Core #(Seq, #)
Labnumber
Sample depth(cm)
Sample elevation(m) NAVD88
Material dated
32 (3) Beta-240760 219–224 −1.24 S. alterniflora
32 Beta-240761 396–398 −3.98 Schoenoplectus americanus56 (3) Beta-252034 255 −1.62 S. alterniflora
62 (2) Beta-252035 280 −1.83 S. alterniflora63 (1) Beta-252036 352 −2.59 S. alterniflora
63 (2) OS-80664 282 −1.89 S. alterniflora
63 (3) OS-80663 258 −1.65 S. alterniflora
63 (4) OS-80692 190 −0.97 S. alterniflora
63 (4) OS-80662 170 −0.77 S. patens
59 (1) OS-84496 290 −2.25 Schoenoplectus americanus59 (4) OS-84495 190 −1.02 S. alterniflora
30 OS-84497 340 −2.43 Schoenoplectus americanus
30 (3) OS-84499 244 −1.47 S. patens
30 (5) OS-84498 116 −0.19 S. patens
ground testing of nuclear weapons. Activity of 137Cs was measured incores 24, 30, and 56. In each case the AD 1963 peak occurs within a sin-gle stratigraphic unit and without a step change suggesting that identi-fication of peak 137Cs activity is not distorted by erosion (Fig. 5).
Concentration and isotope ratio determinations for Pb in core 63were made using a quadrupole ICP-MS instrument (Agilent 7500c)with a conventional glass concentric nebulizer. The long-term, 2σprecision for the BCR-2 reference material (total Pb concentrationof 11 mg/kg) used for quality control was 207:206Pb = 0.0008, and208:206Pb = 0.0020, based on n = 32 replicates over 29 months and amean accuracy, relative to the published values of Baker et al. (2004),within that error. Processing consisted of: (i) removal of background;(ii) calculation of isotope ratios; (iii) determination of mass bias cor-rection factor from defined isotope ratios of SRM981; (iv) applicationof mass bias factors derived from SRM981 using external standard-sample-standard bracketing; and (v) optional further correction forlinearity of ratio with signal strength.
We assumed that the Pb inventory of the Sea Breeze salt marsh wasdeposited primarily from atmospheric deposition rather than from localupstream sources (Graney et al., 1995; Lima et al., 2005). Because resi-dence time of Pb pollutants in the atmosphere is several years and emis-sion values per unit of production and consumption likely changedthrough time, chronological markers were recognized by trends ratherthan absolute values. Downcore trends in Pb represent historical Pbemissions from industrial centers, changes in national production andconsumption, and the introduction and subsequent phasing out ofleaded gasoline. This approach is reliant upon the implicit assumptionthat measured downcore trends represent a full and complete historyof atmospheric deposition. In the case of Sea Breeze where multipleerosive events are a key part of the stratigraphic record, this assumptionis invalid and interpretations of downcore pollution profiles mustbe made with appropriate caution. Our interpretation of downcorechanges is based on that presented by Kemp et al. (2012) for a salt-marsh in central New Jersey that experienced continuous sedimentaccumulation during the last ~200 years.While Sea Breeze experienced
Radiocarbon age δ 13C(‰, PDB)
Calibrated Feature dated
810 ± 40 −11 AD 900 to 920AD 950 to 1040
Post-erosionalrecolonization of a mudflat
2240 ± 40 −24.8 BC 390 to 200 Basal peat810 ± 40 −13.7 AD 980 to 1060
AD 1080 to 1150Post-erosionalrecolonization of a mudflat
1070 ± 40 −12.8 AD 890 to 1030 High-marsh recovery1460 ± 40 −12.8 AD 260 to 290
AD 320 to 440AD 490 to 520
Post-erosionalrecolonization of a mudflat
1170 ± 30 −13.63 AD 776 to 901AD 917 to 966
Post-erosionalrecolonization of a mudflat
1010 ± 25 −12.92 AD 981 to 1045AD 1097 to 1119AD 1142 to 1147
Post-erosionalrecolonization of a mudflat
535 ± 30 −12.56 AD 1319 to 1351AD 1390 to 1438
Post-erosional recolonizationof a mudflat
365 ± 25 −11.46 AD 1451 to 1526AD 1556 to 1632
High-marsh recovery
1550 ± 25 −23.22 AD 429 to 567 Brackish marsh recovery425 ± 25 −13.18 AD 1428 to 1492
AD 1603 to 1610Post-erosionalrecolonization of a mudflat
1760 ± 25 −24.63 AD 214 to 360AD 365 to 381
Basal peat
1090 ± 25 −13.94 AD 893 to 997AD 1004 to 1012
High-marsh recovery
165 ± 25 −12.91 AD 1665 to 1696AD 1725 to 1786AD 1792 to 1814AD 1835 to 1877AD 1917 to 1952
High-marsh recovery
Loss-on-Ignition (%)10 20 30 40 50 60 70
Dep
th in
Cor
e (c
m)
0
50
100
150
200
250
300
350
400
0 20 40 60 80 100
0 10 20 30 40Median Grain Size
260-520
1319-1438
1451-1632
776-966
BC390-200 (C-32)
890-1030 (C-62)
980-1150 (C-56)900-1040 (C-32)893-1012 (C-30)
1428-1610(C 59)
1665-1950 (C-30)
1
2
3
4
5
6
7
AD 1903tropical cyclone
Overwash fansAD 1788 tropical cyclone
AD 1526-1558 or AD 1630-1643Brigantine overwash
AD 550-1400 Brigantine overwash
AD 1278-1438Whale Beach overwash
post AD 558-673 Brigantine overwash
prior AD 558-673 Brigantine overwash
Marsh Erosion and Recovery Sequence
Grain Size (volume; cumulative %)
core 63
clay silt sand
high-marsh peat
low-marsh mud/muddy peat
tidal mud
radiocarbon date (core 63)
radiocarbon date (other core)
basal peat sand
humic sand
sharp contact
981-1147
Fig. 5. Downcore lithology identified based on color, texture and macrofossils; grain size, and organic content estimated by loss on ignition (LOI) in core 63, stratigraphic sequences 1–7.Calibrated radiocarbon ages indicate timing of changes in marsh environment and sedimentation regime. Historical tropical cyclone landfalls and dated overwash fans deposited insouthern New Jersey are listed for comparison (from Donnelly et al., 2001b; Donnelly and Webb III, 2004).
165D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
accumulation punctuated by episodes of erosion, it is possible to identi-fy some chronological marker horizons based on stratigraphic positionand downcore trends that occur within a single sedimentary sequence(i.e. where trends are not inferred across erosive boundaries).
Early Pb production in the United States was centered on the UpperMississippi Valley (UMV) where Pb ores have a distinctive isotopicsignature allowing its identification in sediments bymeasuring changes
0.5 1.0 1.5 2.0 2.5 3.0 3.5
0
20
40
60
80
100
120
20 40 60 80 100 120
1
Pb Concentration (mg/kg)
Sb Concentration (mg/kg)
SbPb
Dep
th in
Cor
e (c
m)
A B
high marsh peat low marsh peat
0 10 20 30
LOI (%)
A
A
AD 1925 ± 5
AD 1974 ± 5
AD 1935 ± 5sequ
ence
7
Fig. 6.Downcore measurements of Pb concentrations and stable isotopic ratios in Core 56. (A) Oblack line) and Sb (open circles, dashed line), dashed lines marked geo-chronological horizo(D) detail of measured 206Pb:207Pb ratio between 0 and 50 cm.
in the ratio of 206Pb to 207Pb (Lima et al., 2005; Gobeil et al., 2013). UMVPb is characterized by a higher 206Pb to 207Pb ratio than other oresexploited in the United States or imported from overseas for domesticconsumption. Emissions during this early period of Pb production (AD1827–1880) were unchecked, allowing a relatively small amount of Pbproduction to leave a large isotopic signature. Prevailing winds carriedPb to New Jersey (Kemp et al., 2012), New England (Lima et al., 2005),
.19 1.20 1.21 1.22 1.23 1.24 1.25 1.1921.1941.1961.1981.2001.2021.2041.206
enla
rged
in p
anel
D
206Pb:207Pb
C
intertidal mud
10
20
30
40
50
0
D
D 1857 ± 5
D 1827 ± 5
AD 1880 ± 20
AD 1965 ± 5
AD 1980 ± 5
206Pb:207Pb
rganic content estimated by loss on ignition (LOI); (B) concentration of Pb (closed circles,ns; (C) measured ratio of 206Pb:207Pb, dashed lines marked geo-chronological horizons;
166 D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
Canada (Gobeil et al., 2013), and Bermuda (Kelly et al., 2009). Historicalrecords show that Pb production in the UMV began in AD 1827, peakedin AD 1857, and had declined to a small relative and absolute amount byAD 1880. In Sea Breeze core 56, the record of Pb deposition from theUMV is captured entirely in the intertidal mud and high salt-marshpeat of sequence 6 between erosive contacts (Fig. 6). Therefore, ouridentification of the start, peak, and decline of UMV Pb production isnot distorted by a downcore profile spanning an erosive contact and isfurther supported by the absence of step changes in 206Pb:207Pb in themeasured profile that would indicate erosion. Since ~AD 1930, changesin 206Pb:207Pb reflect the introduction, phasing out, and changing mix-ture of leaded gasoline. The post AD 1930 interval is preserved in core56 above the erosive contact at the base of sequence 7 (Fig. 6) and weidentified horizons at AD 1965 and AD 1980.
Downcore changes in Pb concentration identified sediment horizonsthat correspond to historical changes in U.S. production and consump-tion. The onset of national Pb production (and pollution) began in AD1875 and reached a peak around AD 1925 before declining during theGreat Depression between AD 1930, and AD 1933. A second peak inPb production and consumption occurred in AD 1974 followed by adecrease due to the phasing out of leaded gasoline (Nriagu, 1998;Kelly et al., 2009). In core 56, themaximummeasured Pb concentrationoccurs close to themiddle of the intertidal mud in sequence 7 (Fig. 6). Itwas, therefore, possible to recognize the AD 1925, AD 1935, and AD1974 pollution marker horizons because these are preserved in a singlesedimentary unit and this interpretation does not span an erosiveboundary as demonstrated by the complete increase to a clear peakand subsequent decline. We were unable to identify the AD 1875onset of Pb pollution because it was likely removed by the erosiveevent marking the base of sequence 7. This interpretation is also sup-ported by the step change in Pb and Sb concentrations at approximately0.70 mdepth. The position of the AD 1925 peak in Pb concentration anddecline out of UMV Pb suggest that the erosive contact at the base ofsequence 7 removed approximately 20–40 years of from the sedimen-tary record preserved in core 56.
4. Stratigraphy of Sea Breeze
We recognized six lithostratigraphic units in the sediment underly-ing the Sea Breeze salt marsh (Fig. 5). The depositional environmentrepresented by each unit was determined from organic content andrecognizable plantmacro-fossils. Grain size analysis of Core 63 indicateda low percentage of sand (~b15%) in the late Holocene stratigraphyabove the basal sand with negligible changes in clay and silt distribu-tions between units. These six units and associated erosive boundarieswere correlated across the site from core logs (Fig. 4):
a) The Sea Breeze salt marsh is uniformly underlain by pre-Holocenefluvial deposits of gray, fine- to medium-sized sand.
b) The sand is overlain by a black, humic, sandy mud that we interpretas a paleosol.
c) A dark brown to black basal peat with fragments of Schoenoplectusspp., is found in the majority of cores on transects I–I′ and II–II′ atelevations below −2.5 m NAVD88 (Fig. 4). This peat was likelydeposited in a brackish environment at the transition between thesaltmarsh and freshwater upland; similar to themodern transitionalenvironment at Sea Breeze. Dated fragments of Schoenoplectus spp.in core 32 indicate that brackish tidal marshes began to develop atSea Breeze around 290 BC ± 40 years (Table 1).
d) A gray-brown, fibrous peat overlies the paleosol and brackish salt-marsh deposits. The presence of S. patens and D. spicatamacrofossilswithin this peat indicates that it accumulated in a high salt-marshenvironment between MHW and MHHW.
e) A graymud described at different stratigraphic levels above the highsalt-marsh peat units represents tidal-flat and low salt-marsh depo-sitional environments. Vertical stems of S. alterniflora deposited
above tidalmud illustrate colonization by low salt-marshvegetation.f) A gray-brown muddy peat with abundant S. alterniflora fragments
was likely deposited in a low salt-marsh environment.
Following deposition of the lowermost high salt-marsh peat, the stra-tigraphy at Sea Breeze shows at least seven repeated sequences (num-bered 1 to 8, from oldest to youngest) of high salt-marsh peat overlainby gray mud with little (10–15%) organic matter, representing tidal-flatdeposits, which recovers to a low salt-marsh and then to high salt-marsh environment (Fig. 1). The contacts between the high salt-mashpeat and overlyingmud are sharp (b1 mm), suggesting an abrupt changein the sedimentation regime and environment of deposition (Fig. 3A, B).
Sequence 1. A brackish/high salt-marsh peat is abruptly overlain by a~10 cm thick gray tidal-flat mud. The upper contact ofthe mud occurs at ~−2.5 m NAVD 88 (~3.5 m belowmodern salt-marsh surface) (Figs. 3B, 4). CorrespondinglyLOI decreases in Core 63 from 68% within the peat to 15%within the mud (Fig. 5). The mud unit is overlain bygray-brown, low salt-marsh muddy peat. Rhizomes ofS. alterniflora recovered immediately below the transitionbetween mud and low salt-marsh, muddy peat in core 63indicate that the tidal flat was colonized by salt-marshvegetation between AD 260 and AD 520. The low salt-marsh peat is capped by high salt-marsh peat that accu-mulated at the elevation of -2.25 m NAVD 88 by AD 429to AD 567, based on a dated fragment of Schoenoplectusspp. (core 59; Transect III). Sequence 1 was completealong transects I, II, and III (Fig. 4).
Sequence 2. The high salt-marsh peat from sequence 1 is abruptly over-lain by a 10 to 60 cm thick tidal-flat mud. The peat has anorganic content of up to 53% where as the mud is as lowas 8%. The upper contact of the mud was documented atthe elevation of −2.0 m NAVD 88 (Figs. 3B, 4). The mudwas capped by a low salt-marsh peat. A S. alterniflorarhizome in core 63 yielded a maximum age of AD 776 toAD 966 for the recolonization of the tidal-flat surface bylow salt-marsh plants. The age of a S. alterniflora stemfound in the overlying high salt-marsh peat in core 62 indi-cates that high salt-marsh vegetation had recolonized byAD 890 to AD 1030 at an elevation of -1.8 m NAVD 88.Sequence 2 was complete along the north-west portionof Transect I (cores 62-78), and at selected sites along II(cores 81, 13, 69, and 66) and III (core 72).
Sequence 3. The high salt-marsh from sequence 2 is overlain bya 10 cm to 40 cm thick tidal-flat mud. In Core 63 theLOI rapidly decreases from 47% in the peat to 10% inthe mud. This mud unit occurs between −1.6 m and−2.0 m NAVD 88. In transects I, II, and III, the mud unitunconformably overlies mud or low salt-marsh peat ofsequence 2. Salt-marsh recovery and recolonizationof the site by low salt-marsh plants occurred from AD900 to AD 1150, as indicated by the first appearanceof S. alterniflora vegetation in the low salt-marsh peatabove the mud in cores 32, 56 and 63. A radiocarbondate from core 30 shows a recovery to a high salt-marshenvironment by AD 893 to AD 1012. High salt-marshpeat accumulation began at the elevation of ~−1.5 mNAVD 88. Sequence 3 is found in nearly all cores on tran-sects I, II and III. It was also documented in transect VI.
Sequence 4. A tidal-flat mud (10 to 60 cm thick) abruptly overliesthe high salt-marsh peat of sequence 3. Although, not asabrupt as in sequences 1, 2 and 3, the LOI does changefrom 35% in the peat to 12% in the mud in Core 63 (Fig. 5).The upper contact of mud was documented at ~−1.05 mNAVD 88. In transects I to IV the abrupt boundary ofthe mud unconformably extends into the mud unit of
167D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
sequence 3. The dated S. alterniflora stem (core 63) andrhizome (core 59) indicate that a community of lowsalt-marsh plants recolonized by AD 1319 to AD 1610(Fig. 5). A high salt-marsh plant community hadrecolonized by AD 1451 to AD 1632 (core 63) at the ele-vation of ~−0.8 m NAVD 88. Sequence 4 is found in alltransects except V.
Sequence 5. The high salt-marsh peat from sequence 4 is overlain bytidal-flat mud of varying thickness (20 cm to 1.0 m). InCore 63, the high salt-marsh organic content is relativelylow (30–27%), but there is an abrupt decrease in LOIacross the erosive contact to the mud (7%). Though themud lower contact was documented in more than 20cores and appeared to be sharp (Fig. 3A), its bottomboundary unconformably overlaps the mud of sequence4 in transects I, II, IV and VI. The upper contact of mudoccurs at ~−0.5 m NAVD 88. There is no date availableto estimate when low salt-marsh plants recolonized thetidal-flat surface. A high salt-marsh plant communityhad replaced low salt-marsh vegetation in sequence 5 byAD1665 to AD1814 (core 30). Sequence 5was document-ed in all transects. However, its occurrence is limited toone or two core sites located next to each other and there-fore, its lateral extent is limited to ~50 m except in tran-sects IV and VII where it was correlated across ~150 m.The shape and thickness of mud unit, unconformitiesand lateral discontinuity are similar to stratigraphy ofinfilled paleo-channels.
Sequence 6. The high salt-marsh peat of sequence 5 or 4 is overlain by~20 cm of tidal-flat mud. The upper boundary of the mudoccurs at 0 m NAVD 88. In Core 63, there is a notable de-crease in organic content from the peat (50%) to the over-lying mud (12–10%), although the change begins 10 cmbelow the erosive contact. The age of salt-marsh recoverywas estimated using trends inmeasured 206Pb:207Pb ratiosin core 56 (Fig. 6C). The shift toward higher 206Pb:207Pbratios marks the onset of Pb production in the UpperMississippi Valley in AD 1827. Therefore we can onlyinfer that erosion at the base of sequence 6 occurredprior to this date. Peak and declining 206Pb:207Pb ratiosshow that high salt-marsh peat accumulation began priorto AD 1857 and continued to at least AD 1880. The com-plete sequence 6 was documented in all transects exceptfor transects III and VI where a change in deposition fromhigh salt-marsh peat to low salt-marsh muddy peat atthe elevation of 0 to −0.2 m NAVD 88 may represent anincomplete sequence 6 (cores 73, 72, 59 and 116, transectIII and cores 86, 98 and 110, Transect VI). In transect I therewas only partial core recovery between cores 27 and 46a(Fig. 4).
Sequence 7. The high salt-marsh peat in sequences 6, 5, or 4 is overlainby 10 to 40 cmof tidal-flatmud (Fig. 3C). However, there isno change in LOI across the contact of Core 63 (Fig. 5). Theupper boundary of themud unit is at ~0.5 m NAVD88. Thetiming of mud deposition was cautiously estimated usingchronological horizons recognized by downcore changesin Pb concentrations, trends in the ratio of stable Pb iso-topes, and 137Cs activity in Core 56 (Figs. 6, 7). Depositionof the mud unit began prior to AD 1925 and finished afterAD 1974, when a low salt-marsh plant community colo-nized the mud flat. However, 137Cs activity in cores 24,30, and 56 indicates that low salt-marsh conditionshad reestablished by AD 1963. The use of chronologicalmarker horizons to constraint depositional histories atSea Breeze is complicated by erosion. Chronological dis-crepancies between methods and cores may reflect an
incomplete inventory of atmospheric deposition that dis-torts trends. Additionally abrupt changes between highsalt-marsh and tidal-flat environments may invalidatethe assumption that deposition is primarily from atmo-spheric sources. The low salt-marsh peat is commonlycapped by modern high salt-marsh, but at some locationson transects III, IV, V, VI and VII, modern low salt-marshpeat or tidal-flat mud was deposited on top of the se-quence at the elevation of ~0.7 m NAVD 88. Sequence 7is present in all transects.
Sequence 8. High salt-marsh peat or tidal-flat mud of sequence 7 isoverlain by 10 to 20 cmmudunit with an upper boundaryat ~0.7 m NAVD 88. The mud is overlain by low salt-marsh muddy peat. Based on the depth of maximum137Cs activity in cores 24, 30 and 56, salt-marsh plantsrevegetated themud after AD 1963. Sequence 8 is presentin nine cores in transects III, IV, V andVII. Themodern salt-marsh vegetation at these core sites is tall and short fromS. alterniflora with S. patens. Sequence 8 is not present inCore 63 (Fig. 5).
The time between first appearance of S. alterniflora on the tidalflat and recolonization of low salt-marsh surface with S. patens andD. spicata represents the period of time needed for salt-marsh recoveryfollowing an abrupt change in sedimentation. Based on available datafor sequences 7, 4, 3 and 2 we estimated that it takes between 20 and194 years for salt marshes to recover from erosion events where largeexpanses of a single salt marsh are disturbed.
5. Discussion
5.1. Proposed mechanism of salt-marsh erosion
At Sea Breeze, we identified at least seven sequences of peat andmud couplets that have abrupt contacts suggesting erosion and a dra-matic change in sedimentation regime. There are a number of processesthat could produce these stratigraphic sequences against a backgroundof rising RSL including (but not limited to): lateral migration of tidalcreeks (e.g. Stumpf, 1983); tidal channel network and/or drainageditch expansion (e.g. D'Alpaos et al., 2007); changes in sediment deliv-ery rates (e.g. Kirwan et al., 2011; Mudd, 2011); rapid relative sea-levelchange (e.g. Atwater, 1987; Long et al., 2006); tsunami (Dawson, 1994;Tuttle et al., 2004); formation of salt pans (e.g. Wilson et al., 2009); orerosion during storm events and rapid infilling of the newly created ac-commodation space (e.g. van de Plassche et al., 2004; van de Plasscheet al., 2006).
Salt-marsh tidal creeks are believed to be static features that showlittle lateral channel migration due to extensive vegetation root struc-ture that supports their banks (Redfield, 1972; Garofalo, 1980; Gabet,1998). However, mechanisms that cause the destruction of aboveand below-ground vegetation, such as deposition of dead vegetationon creek banks and subsequent development of bare banks orwave ero-sionmay lead to bank face erosion and lateralmigration, or enlargementof tidal channels (Stumpf, 1983; Philipp, 2005; Lottig and Fox, 2007;Chen et al., 2012; Deegan et al., 2012). These slow processes produce re-gressive stratigraphic sequenceswith localized channel-shapedmud fa-cies overlain by peat (Allen, 2000; Nikitina et al., 2003). Based on thefacies geometry, only sequence 5 could have been formed by infillingof incised paleo-channels.
An extensive network of mosquito control ditches was created onthe Sea Breeze salt marsh between AD 1887 and 1930 (Philipp, 2005).This anthropogenic modification of the landscape almost certainlychanged the marsh hydrology and sediment delivery rates over anextensive area of the study site. By AD 1963 most of the ditches hadnaturally filled by sedimentation, but in some cases segments of multi-ple ditches became connected and evolved into the network of more
Core 24 Core 30 Core 56
0.0 0.2 0.4 0.6 0.8
Dep
th in
Cor
e (c
m)
0
10
20
30
40
50
60
70
800.0 0.5 1.0 1.5 2.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2
AD1963 AD1963 AD1963
high marsh peatlow marsh peattidal mud
137Cs Activity (dpm/g)
Fig. 7. Downcore measurements of 137Cs activity in cores 24, 30 and 56. Maximum activity occurred in AD 1963 coincident with the peak in above ground testing of nuclear weapons.
168 D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
permanent tidal channels prior to AD 1981 (Fig. 8C, D). Ditching of saltmarshes was widespread in New Jersey during the early part of the20th century. Therefore a high proportion of potential study sites inthe region have a history of modification similar to Sea Breeze. Ifhuman modification of the landscape was the primary mechanism forexplaining the stratigraphy observed in sequences 7 and 8 it would bereasonable to expect to see similar patterns at other salt marshes inthe region. However, the Sea Breeze stratigraphy is unusual (althoughlikely not unique) and the erosive pattern we documented is notpresent at other salt marshes that we investigated or that have beendescribed in published literature (e.g., Psuty, 1986; Donnelly et al.,2001b; Donnelly et al., 2004;Miller et al., 2009) suggesting that ditching
Fig. 8. Historical map and aerial photographs of Sea Breeze. (A) First topographic map of Newimagery. (D) 1981 imagery (http://www.state.nj.us).
is unlikely to be the causalmechanism for sequences 7 and 8, but cannotbe discounted.
Processes that change RSL rapidly on the scale of a single salt marshinclude sediment compaction, tidal-range change and, large earth-quakes that cause vertical displacement of the land on tectonically-active coasts. Törnqvist et al. (2008) suggested that compaction ofHolocene strata contributes significantly to the exceptionally highrates of RSL rise and coastal wetland loss in the Mississippi Delta. Insoutheastern England, Long et al. (2006) suggested that an originallyplanar peat surface was locally lowered in the tidal frame by sedimentcompaction and subsequently overlain by tidal-flat mud. In thesestudies the stratigraphy differs from Sea Breeze, because only one late
Jersey completed 1870–1887 (http://historical.mytopo.com). (B) 1930 imagery. (C) 1963
169D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
Holocene transgressive sequence was identified, where the contact(and process) was gradual and the overlyingmud sequencewas severalmeters thick.
Hall et al. (2013) showed that tidal range in Delaware Bay increasedsteadily during the late Holocene and, thus,would not be responsible forthe sedimentary sequences described at Sea Breeze. Steady increase intidal range should not result in repeated abrupt changes in sedimenta-tion pattern. Instead, it should produce one mud unit with gradualtime-transgressive contact. However, during the 20th century tidalrange in the upper estuary rapidly increased two-fold because of dredg-ing andwetlandmodification in themid-upper estuary that began in AD1910 (DiLorenzo et al., 1993). Historical maps and aerial photos of SeaBreeze show that at least two channels were dredged after AD 1930and suggest that sediment accumulation on salt-marsh surface wasmodified by human activities (Fig. 8C). Systematic deepening of thetidal channels would cause a positive feedback of increased tidal dis-charges, further channel erosion, and increase in suspended sedimentconcentrations compensated for by sediment accumulation on tidalflats and salt marshes (Fredericks, 1995). In response, the salt-marshsurface would either accumulate sediment to maintain a stable tidalelevation and support the characteristic plant community, or experi-ence increasing frequency and duration of tidal flooding resulting inmarsh drowning (Orson et al., 1998; Morris et al., 2002). Therefore,dredging may have been responsible for the deposition of the mud inthe recent sequences 7 or 8, and possibly both if there were two distinctdredging phases and its subsequent colonization by salt-marsh vegeta-tion, but is not be a plausible causal mechanism for sequences 1 to 6.
While it is difficult to distinguish between the sedimentary signatureof tsunami and storm erosion in the salt-marsh sediments (Engel andBrüchner, 2001), it is unlikely that tsunamis producedmultiple erosion-al contacts at Sea Breeze. There have been only few tsunamigenic earth-quakes documented along the Atlantic coast of North America (Smith,1966; Bakun and Hopper, 2004). But in 1817, a Mw 7 earthquake oc-curred offshore of South Carolina, and may have been responsible fora reported ~30 cm sea-level rise on the Delaware River (Hough et al.,2013). Hough et al. (2013) tsunami simulations model predicted that~50-cm tsunami runups at the mouth of Delaware Bay would requirean earthquake of M-7.5 and greater with estimated surface slip of~3 m and ~60 km long rupture along the fault oriented such that atsunami wave moves directly into the Bay.
The stratigraphic record from Plum Island estuary in Massachusettsreported by Kirwan et al. (2011) is similar to the one at Sea Breeze,with the exception that only one sequence consisting of a mud unitoverlain by low and high salt-marsh peat was recorded. Kirwan et al.(2011) suggested that the Plum Island stratigraphy records infilling ofthe estuary and lateral progradation of the salt-marsh due to increasedsediment delivery associated with European settlement and land clear-ance that mobilized sediment, which ultimately was transported to thecoast and deposited in estuaries. Priestas et al. (2012), however, arguedthat the salt-marsh stratigraphic sequence at Plum Island predatesEuropean settlement and is a consequence of salt-marsh expansionrather than loss. Humans have modified the Delaware Bay watershedfor over 400 years (Philipp, 2005) and regional land clearance couldbe responsible for one of the sequences from the historic period (i.e.,6, 7, or 8).
While themore recent changes in salt-marsh sedimentologymay berelated to land-use change and humanmodification of the salt-marshes,this is not a plausible mechanism for the sequences that were createdprior to human settlement. The uniformdepth of the upper contacts be-tweenmud and low-marsh peat correlated (inmost cases, over 2.5 km)suggests a common cause for the erosion. Similarly, the gradual changebetween the low andhigh salt-marsh sedimentation documented at thetopof each sequence indicates synchronousmarsh recovery. There are twopossible explanations: salt pan development and/or storm surge erosion.
Salt pans are shallow, water-filled depressions that are common onsalt marshes along the Atlantic coast of North America (e.g. Redfield,
1972; Turner et al., 1997; Adamowicz and Roman, 2005). Some saltpans form on tidal flats and later become surrounded by salt-marshvegetation, while others form as result of vegetation disturbance onthe salt-marsh surface (Wilson et al., 2009). The stratigraphic signatureof salt pans often consists of dark-gray mud overlain by low salt-marshpeat and is therefore similar to the regressive sequences described atSea Breeze. If the stratigraphic record is conformable then the unitunderlying the salt pan sequence represents the high salt-marsh envi-ronment where the pool originally formed. The bottom contact ofsuch sequences may be erosional as a result of geochemical (sulfatereduction) rather than physical processes (van Huissteden and van dePlassche, 1998). Salt pans are usually of limited extent and effect onlyrelatively small parts of the modern salt-marsh at any given time caus-ing salt-marsh fragmentation rather than synchronous, site-widechanges (Boston, 1983; Day et al., 2000). The stratigraphic signature ofsalt pan expansion and infill would therefore be discrete rather than lat-erally continuous mud units (Wilson et al., 2009). However, remote-sensing in coastal Louisiana documented formation of geomorphologi-cal features resembling salt pans after the landfall of tropical storms(Morton and Barras, 2011). These storm-impacted salt-marsh featuresvary in size from closely spaced small scours (plucked salt-marshes)to 103 m long orthogonal-elongated ponds, or amorphous pans up to1500 m across (Barras, 2007). Once formed, erosional salt pans canpersist in the landscape for decades, enlarge in width, and increase indepth after repeated storms or through processes of positive feedback(Morton and Barras, 2011). The stratigraphic signature of storm-inducedsalt-pan expansion and infill would be synchronous, continuous mudunits overlain by organic-rich tidal-flat mud and salt-marsh peat, simi-lar to the Sea Breeze stratigraphy. Historical observations, however,show that salt pan development did not create sequences 7 or 8 at SeaBreeze. These sequenceswere created in themiddle 19th to 20th centu-ries. The earliest map of the study area (AD 1887) and historical aerialphotos from the 1930s show a series of small lakes separated by a poor-ly developed network of short and narrow tidal channels rather than apresence of a large openwater body (Fig. 8A, B). The spatially restrictednature of the salt pans indicates that they cannot be responsible forthe site-wide stratigraphic nature of sequences 7 and 8 at Sea Breeze,although as secondary processes they likely overprinted and modifiedthe principal sequence. For example, mosquito control ditches thatcould be responsible for draining small salt pans documented on theAD 1870–1887 map, but are absent on the AD 1930 image (Fig. 8A, B).
Erosion of salt-marsh peat by the physical action of strong wavesand currents during salt-marsh inundation under storm-surge con-ditions can explain the erosive contacts and regressive sequencesdescribed at Sea Breeze (van de Plassche et al., 2004; van de Plasscheet al., 2006; Fig. 1). Salt-marsh erosion may occur during the storm orcould be a post-storm process triggered by disturbance of vegetation.Numerical modeling and field observations document that temporaldisturbance of salt-marsh vegetation led to platform deepening, expan-sion of tidal creek network, and temporal decrease in marsh accretionrates (Kirwan and Murray, 2008). All of the above processes may beresponsible for sequences 1, 2, 3, 4 and 6 at Sea Breeze. During stormsurges enhanced by runoff and stream discharge, bed shear stressexceeds the shear strength of sediments. This causes erosion, undercut-ting and slumping of banks, and deepening of tidal channels (Stefanonet al., 2010). This process may be responsible for the stratigraphy wedescribe for sequence 5.
Significant loss of salt marshes due to wave erosion, inundation andvegetation loss occurred along the coast of Louisiana after the 1995hurricane season (Day et al., 2007; Feagin et al., 2009; Ravens et al.,2009). Vulnerability of salt marshes to storm surge and wave erosiondepends on the type of vegetation colonizing the marsh surface.Howes et al. (2010) reported that low-salinity salt marshes predom-inantly vegetated by S. patens were susceptible to erosion, while high-salinity salt marshes where S. alterniflora with extensive horizontalrhizomes is the most abundant species remained unchanged during
170 D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
tropical cyclone landfalls in Louisiana. The Sea Breeze stratigraphicsequences document repeated erosion or decrease in the elevation ofhigh or brackish marshes during the last 2000 years. Models predictthat once a portion of a salt marsh is disturbed and submerged, ittakes hundreds of years to fill the accommodation space with sedimentto depths capable of supporting vegetation (Kirwan andMurray, 2008).We estimated that salt-marsh recovery following storm erosion isbetween 20 and 200 years.
5.1.1. Sedimentary record of storm erosionThe sedimentary record of salt-marsh erosion from storm surges at
Sea Breeze can be correlated with geological records of tropical cycloneoverwash in New Jersey despite the large (radiocarbon) age uncer-tainties for some events (Fig. 5). Back-barrier salt marshes at BrigantineandWhale Beach in southernNew Jersey preserved six anomalous sandunits deposited since AD 550 (Donnelly et al., 2001b; Donnelly andWebb, 2004; Fig. 2A). The most recent overwash fan was deposited bythe Ash Wednesday storm in AD 1962 and historical accounts suggestthat sand deposition was widespread on back-barrier salt marshes inNew Jersey (Stewart, 1962). Although sequence 8 has variable pres-ervation along transects at Sea Breeze, the mud was revegetated byS. alterniflora after AD 1963 in core 24 (Fig. 7). However, inconclusivestratigraphy and lack of geochronologic markers from other sites donot allow for correlation of most recent mud deposition with anyknown storms of the 20th century.
Sequence 7 at Sea Breeze represents erosion that took place betweenAD 1875 and AD 1925 and we propose that it was caused by a tropicalcyclone that made landfall in the Delaware Bay in AD 1903 and wasthe only tropical cyclone of the 20th century to make landfall in NewJersey (Neumann et al., 1993). This could be the same tropical cyclonethat deposited overwash sand at Brigantine in either AD 1893 or AD1903 (Donnelly and Webb, 2004). Erosion and subsequent depositionof sequence 6 at Sea Breeze occurred prior to AD 1827 and could corre-late with overwash fans at Brigantine and Whale Beach that were de-posited by either the AD 1821 or AD 1788 tropical cyclone (Donnellyet al., 2001b; Donnelly and Webb III, 2004).
If the high salt-marsh peat at the top of sequence 5was eroded in AD1788, the deposition of sequence 5 began prior to AD 1665–1696, whichcould be associated with the same storm that overwashed Brigantinearound AD 1526–1558 or AD 1630–1643 (Donnelly et al., 2001b;Donnelly andWebb III, 2004). The thickness of themud unit, unconfor-mities, and its patchy presence in a limited number of core sites suggestthat sequence 5 is likely associatedwith erosion and subsequent infill ofpaleo-channels. In comparison with the modern drainage network, thestratigraphy in sequence 5 shows oversized (width and depth) paleo-tidal creeks, which could be related to channel erosion due to increaseddischarge under storm surge hydrology. Furthermore, within the un-certainties of radiocarbon dating sequence 5 at Sea Breeze and theBrigantine overwash fan of AD 1630–1643 could be the result of thesame tropical cyclone that eroded salt-marsh sediment in Connecticutand deposited an overwash fan in Rhode Island in AD 1635 or AD1638 (van de Plassche et al., 2006).
Radiocarbon dates indicate that overwash fans were deposited atBrigantine and Whale Beach between AD 550–1400 and AD 1278–1438respectively (Donnelly et al., 2001b; Donnelly andWebb III, 2004). Dur-ing this timeperiod sequences 2, 3, and 4were formed at Sea Breeze. Se-quence 4 includes a 30–60 cm mud unit that was colonized by salt-marsh vegetation by AD 1319–1351 in transect I and AD 1428–1492in transect III. Deposition of this sequence could be result of one ormore storms. It is possible that salt-marsh erosion documented by thesharp contact in sequence 4 is the result of the same event that depos-ited overwash fans along the coasts of New Jersey and Rhode Island, anderoded a salt marsh in Connecticut (AD 1400–1440). If the latter, thisstorm had a regional impact, not only along the open coast, but alsoon sheltered salt marshes (e.g. Sea Breeze, New Jersey and PattagansettRiver Marsh, Connecticut). However, the deposition of ~1 m thick
overwash fan in Brigantine could be the result of an earlier storm asthe sand overlays salt-marsh peat that dates back to AD 558–673. Se-quence 3 documented in transects I–III was deposited prior to AD950–1040. The storm that eroded up to 0.6 m of peat and offset deposi-tion of sequence 2 occurred sometime between AD 429 and AD 966.
Deposition of the thin, lower sand layer at Brigantine may coincidewith the earliest erosion event documented in Sea Breeze (sequence1). Though only ~10 cm of mudwas deposited, the evidence of a hiatusin core 63 indicates that the elevation of paleo-marsh surface waslowered by erosion. The marsh recovery between AD 260–520 resultedin deposition of low salt-marsh peat while undisturbed brackish condi-tions persisted along transect I–I′ around AD 214–381 (core 30). Themarsh recovery was completed by AD 429–567, which correlates wellwith AD 558–673 age of salt-marsh surface recovered at Brigantineafter deposition of the oldest fan.
We interpreted the salt-marsh stratigraphy at Sea Breeze as an ero-sive record assuming that only fragments of local depositional historyremained preserved in it. Proposed unconformities between mudunits (Fig. 4) suggest that sequences were destroyed by later storms.We assume that thick units of mud (N1 m) were formed as multiplestorms eroded sequences over and over, followed by infilling of accom-modation spacewithmudunits that stacked on top of each other (Fig. 4:Transect I cores 14, 50, 26, 46; Transect II cores 82, 13, 70, 67; Transect IVcores 96–104). Nevertheless this record is relevant to paleotempestologyas it identifies storm-related deposition and may contribute to buildinga regional database necessary for comprehensive spatial and time-series analyses of former storm activity.
6. Conclusions
The historical record of landfalling tropical cyclones and storms islimited, because of the relatively small number of large events thathave been documented. Coastal sediments preserve a longer, but frag-mentary record of prehistoric events that are typically recognized bydeposition of anomalous overwash sands in low-energy environmentssuch as salt marshes. Alternatively, the sedimentary record may pre-serve the impact of tropical cyclones and storms as repeated sequencesof salt-marsh erosion and recovery. The stratigraphic record at SeaBreeze, NJ includes seven complete (and one incomplete) sequencesof erosion and salt-marsh recovery that we attribute to storm surges.This record of abrupt, erosive boundaries and infilling of newly createdaccommodation space with mud correlates well with historical andgeological records of tropical cyclone activity in New Jersey and NewEngland and indicates that at least seven tropical cyclones and/or stormsimpacted the north shore of Delaware Bay in the past ~2000 years. Thisrecord compliments and extends the limited geologic data of tropicalcyclone erosion for southern New Jersey. It also demonstrates thatsalt-marsh sedimentary sequences can be used as a tool for storm-impact risk assessment. Lateral continuity of erosive boundariesmapped in the study area indicates that estuarine salt marshes are asvulnerable to storm erosion as the open coast. It takes from severaldecades to almost 200 years for complete salt-marsh recovery afterstorm erosion.
Acknowledgments
We pay tribute to Orson van de Plassche, whose pioneeringwork ontropical cyclone erosion of salt marshes in New England paved the wayfor this study. Funding for this study was provided by NICRR grant DE-FC02-06ER64298, NOAA grantNA11OAR4310101, 2007 PASSHEFacultyDevelopment Award andWest Chester University of Pennsylvania seedgrants. 137Cs dating was conducted by Dr. Christopher Sommerfield,College of Earth, Ocean and Environment, University of Delaware. Wethank the enthusiastic help with fieldwork from participants in theEarthwatch Student Challenge Awards Program. This paper is a contri-bution to IGCP project 588’Preparing for coastal change’ and PALSEA.
171D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
We thank two anonymous reviewers and Editor Andrew Plater for com-ments that greatly improved the quality of the manuscript.
References
Adamowicz, S., Roman, C., 2005. New England salt marsh pools: a quantitative analysis ofgeomorphic and geographic features. Wetlands 25 (2), 279–288.
Allen, J.R.L., 2000. Morphodynamics of Holocene salt marshes: a review sketch from theAtlantic and Southern North Sea coasts of Europe. Quat. Sci. Rev. 19 (12), 1155–1231.
Atwater, B.F., 1987. Evidence for great Holocene earthquakes along the outer coast ofWashington state. Science 236 (4804), 942–944.
Baker, J., Peate, D., Waight, T., Meyzen, C., 2004. Pb isotopic analysis of standards andsamples using a 207Pb–204Pb double spike and thallium to correct for mass biaswith a double-focusing MC–ICP-MS. Chem. Geol. 211 (3–4), 275–303.
Bakun,W.H., Hopper,M., 2004. Catalog of Significant Historical Earthquakes in the CentralUnited States. U.S. Geological Survey Open-File Report 2004-1086, 142p.
Ball, D.F., 1964. Loss-on-ignition as an estimate of organic matter and organic carbon innon-calcareous soils. Eur. J. Soil Sci. 15 (1), 84–92.
Barras, J.A., 2007. Land area changes in coastal Louisiana after Hurricanes Katrina and Rita.In: Farris, G.S., Smith, G.J., Crane, M.P., Demas, C.R., Robbins, L.L., Lavoie, D.L. (Eds.),Science and the Storms: The USGS Response to the Hurricanes of 2005. U.S. Geolog-ical Survey Circular 1306, pp. 98–113.
Boston, K.G., 1983. The development of salt pans on tidal marshes, with particularreference to south-eastern Australia. J. Biogeogr. 10 (1), 1–10.
Cahoon, D.R., 2006. A review of major storm impacts on coastal wetland elevations.Estuar. Coasts 29 (6), 889–898.
Chen, Y., Thompson, C.E.L., Collins, M.B., 2012. Saltmarsh creek bank stability:biostabilisation and consolidation with depth. Cont. Shelf Res. 35, 64–74.
D'Alpaos, A., Lanzoni, S., Marani, M., Bonometto, A., Cecconi, G., Rinald, A., 2007. Sponta-neous tidal network formation within a constructed salt marsh: observations andmorphodynamic modelling. Geomorphology 91 (3), 186–197.
Dawson, A.G., 1994. Geomorphological effects of tsunami run-up and backwash. Geomor-phology 10, 83–94.
Day Jr., J.W., Britsch, L.D., Hawes, S.R., Shaffer, G.P., Reed, D.J., Cahoon, D., 2000. Pattern andprocess of land loss in the Mississippi Delta: a spatial and temporal analysis of wet-land habitat change. Estuaries 23 (4), 425–438.
Day, J.W., Boesch, D.F., Clairain, E.J., Kemp, G.P., Laska, S.B., Mitsch,W.J., Orth, K., Mashriqui,H., Reed, D.J., Shabman, L., Charles, A., Simenstad, C.A., Streever, B.J., Twilley, R.R.,Watson, C.C.,Wells, J.T.,Whigham, D.F., 2007. Restoration of theMississippi Delta: les-sons from Hurricanes Katrina and Rita. Science 315 (5819), 1679–1684.
Deegan, L.A., Johnson, D.S., Warren, R.S., Peterson, B.J., Fleeger, W., Fagherazzi, S.,Wollheim, W.M., 2012. Coastal eutrophication as a driver of salt marsh loss. Nature490, 388–392.
DiLorenzo, J., Huang, P., Thatcher, M.L., Najarian, T.O., 1993. Dredging Impacts onõDelaware Estuary Tides, Estuarine and Coastal Modeling III, Proceedings of the 3rdInternational. Coastal and Ocean Div./ASCE, Oak Brook, IL, pp. 86–104.
Donnelly, J.P., Webb III, T., 2004. Backbarrier sedimentary records of intense hurricanelandfalls in the northeastern United States. In: Murnane, R.J., Liu, K.B. (Eds.),õHurricanes and Typhoons: Past, Present, and Future. Columbia University Press,New York City, pp. 58–95.
Donnelly, J.P., Bryant, S.M., Butler, J., Dowling, J., Fan, L., Hausmann, N., Newby, P., Shuman,B., Stern, J., Westover, K., Webb III, T., 2001a. 700 yr sedimentary record of intensehurricane landfalls in southern New England. Geol. Soc. Am. Bull. 113 (6), 714–727.
Donnelly, J.P., Roll, S., Wengren, M.J., Butler, R. Lederer, Webb III, T., 2001b. Sedimentaryevidence of intense hurricane strikes from New Jersey. Geology 29 (7), 615–618.
Elsner, J.B., 2007. Granger causality and Atlantic hurricanes. Tellus A 59 (4), 476–485.Emanuel, K., Sundararajan, R., Williams, J., 2008. Hurricanes and global warming:
results from downscaling IPCC AR4 simulations. Bull. Am. Meteorol. Soc. 89 (3),347–367.
Engel, M., Brüchner, H., 2001. The identification of paleo-tsunami deposits—a major chal-lenge in coastal sedimentary research. In: Karius, Hadler, Deicke, von Eynatten,Brückner and Vött (Eds). Coastal Reports 17, 65–80.
Engelhart, S.E., Horton, B.P., 2012. Holocene sea level database for the Atlantic coast of theUnited States. Quat. Sci. Rev. 54, 12–25.
Engelhart, S.E., Peltier, W.R., Horton, B.P., 2011. Holocene relative sea-level changes andglacial isostatic adjustment of the U.S. Atlantic coast. Geology 39 (8), 751–754.
Feagin, R.A., Lozada-Bernard, S.M., Ravens, T.M., Möller, I., Yeager, K.M., Baird, A.H., 2009.Does vegetation prevent wave erosion of salt marsh edges? Proc. Natl. Acad. Sci. 106(25), 10109–10113.
Fredericks, C.T., 1995. Stability, shear stress and equilibrium cross-sectional geometry ofsheltered tidal channels. J. Coast. Res. 11 (4), 1062–1074.
Gabet, E.J., 1998. Lateral migration and bank erosion in a saltmarsh tidal channel in SanFrancisco Bay, California. Estuaries 21, 745–753.
Garofalo, D., 1980. The influence of wetland vegetation on tidal stream channel migrationand morphology. Estuaries 3 (4), 258–270.
Gobeil, C., Tessier, A., Couture, R.-M., 2013. Upper Mississippi Pb as a mid-1800s chrono-stratigraphic marker in sediments from seasonally anoxic lakes in Eastern Canada.Geochim. Cosmochim. Acta 113, 125–135.
Graney, J.R., Halliday, A.N., Keeler, G.J., Nriagu, J.O., Robbins, J.A., Norton, S.A., 1995. Isoto-pic record of lead pollution in lake sediments from northeastern United States.Geochim. Cosmochim. Acta 59, 1715–1728.
Hall, G.F., Hill, D.F., Horton, B.P., Engelhart, S.E., Peltier,W.R., 2013. A high-resolution studyof tides in the Delaware Bay: past conditions and future scenarios. Geophys. Res. Lett.40 (2), 338–342.
Hawkes, A.D., Horton, B.P., 2012. Sedimentary record of storm deposits from HurricaneIke, Galveston and San Luis Islands, Texas. Geomorphology 171–172, 180–189.
Holland, G., Bruyere, C.L., 2013. Recent intense hurricane response to global climatechange. Clim. Dyn. http://dx.doi.org/10.1007/s00382-013-1713-0.
Horton, B.P., Engelhart, S.E., Hill, D.F., Kemp, A.C., Nikitina, D., Miller, K.G., Peltier, W.R.,2013. Influence of tidal-range change and sediment compaction on Holocene relativesea-level change in New Jersey, USA. J. Quat. Sci. 28 (4), 403–411.
Hough, S.E., Munsey, J., Ward, S.N., 2013. Reverberations on the watery element: a signif-icant tsunamigenic historical earthquake offshore the Carolina coast. Seismol. Res.Lett. 84 (5), 891–898.
Howes, N.C., FitzGerald, D.M., Hughes, Z.J., Georgiou, I.Y., Kulp, M.A., Miner, M.D., Smith,J.M., Barras, J.A., 2010. Hurricane-induced failure of low salinity wetlands. Proc.Natl. Acad. Sci. 107 (32), 14014–14019.
Kelly, A.E., Reuer, M.K., Goodkin, N.F., Boyle, E.A., 2009. Lead concentrations and isotopesin corrals and water near Bermuda, 1780–2000. Earth Planet. Sci. Lett. 283, 93–100.
Kemp, A.C., Sommerfield, C.K., Vane, C.H., Horton, B.P., Chenery, S., Anisfekd, S., Nikitina,D., 2012. Use of lead isotopes for developing chronologies in recent salt-marsh sedi-ments. Quat. Geochronol. 12, 40–49.
Kirwan, M.L., Murray, A.B., 2008. Tidal marshes as the disequilibrium landscapes: lagsbetween morphology and Holocene sea level change. Geophys. Res. Lett. 35, L24401.
Kirwan, M.L., Murray, A.B., Donnelly, J.P., Corbett, D.R., 2011. Rapid wetland expansionduring European settlement and its implication for marsh survival under modernsediment delivery rates. Geology 39 (5), 507–510.
Knutson, T.R., Sirutis, J., Garner, S., Vecchi, G., 2010. Tropical cyclones and climate change.Nat. Geosci. 3 (3), 157–163.
Lambert, W.J., Aharon, P., Rodriguez, A., 2008. Catastrophic hurricane history revealed byorganic geochemical proxies in coastal lake sediments: a case study of Lake Shelby,Alabama (USA). J. Paleolimnol. 39 (1), 117–131.
Landsea, C.W., 2007. Counting Atlantic tropical cyclones back to 1900. EOS 88 (18),197–202.
Lima, A.L., Bergquist, B.A., Boyle, E.A., Reuer, M.K., Dudas, F.O., Reddy, C.M., Eglinton, T.I.,2005. High-resolution historical records from Pettaquamscutt River basin sediments:2. Pb isotopes reveal a potential new stratigraphic marker. Geochim. Cosmochim.Acta 69 (7), 1813–1824.
Lin, N., Emanuel, K., Oppenheimer, M., Vanmarcke, E., 2012. Physically based assessmentof hurricane surge threat under climate change. Nat. Clim. Chang. 2 (6), 462–467.
Liu, K., 2007. Paleotempestology, Encyclopedia of Quaternary Science. Elsevier, Oxford1978–1986.
Liu, K., Fearn, M.L., 1993. Lake-sediment record of late Holocene hurricane activities fromcoastal Alabama. Geology 21 (9), 793–796.
Long, A.J., Waller, M.P., Stupples, P., 2006. Driving mechanisms of coastal change: peatcompaction and the destruction of late Holocene coastal wetlands. Mar. Geol. 225(1–4), 63–84.
Lottig, N.R., Fox, J.M., 2007. A potential mechanism for disturbance-mediated channel mi-gration in a southeastern United States salt marsh. Geomorphology 86 (3–4), 525–528.
Madsen, A.T., Duller, G.A.T., Donnelly, J.P., Roberts, H.M., Wintle, A.G., 2009. A chronologyof hurricane landfalls at Little Sippewissett Marsh, Massachusetts, USA, using opticaldating. Geomorphology 109 (1–2), 36–45.
Mckee, K.L., Patrick,W.H., 1988. The relationship of smooth cordgrass (Spartina alterniflora)to tidal datums—a review. Estuaries 11 (3), 143–151.
Miller, K.G., Sugarman, P.J., Browning, J.V., Horton, B.P., Stanley, A., Kahn, A., Uptegrove, J.,Aucott, M., 2009. Sea-level rise in New Jersey over the past 5000 years: implicationsto anthropogenic changes. Glob. Planet. Chang. 66 (1–2), 10–18.
Morgan, J.P., Nichols, L.G.,Wright, M., 1958. Morphological Effects of Hurricane Audrey onthe Louisiana Coast, Technical Report 10(A). Louisiana State University, CoastalStudies Institute.
Morris, J.T., Sundareshwar, P.V., Nietch, C.T., Kjerfve, B., Cahoon, D.R., 2002. Responses ofcoastal wetlands to rising sea level. Ecology 83, 2869–2877.
Morton, R.A., Barras, J.A., 2011. Hurricane impacts on coastal wetlands: a half-centuryrecord of storm-generated features from Southern Louisiana. J. Coast. Res. 27 (6A),23–43.
Mudd, S.M., 2011. The life and death of salt marshes in response to anthropogenic distur-bance of sediment supply. Geology 39 (5), 511–512.
Neumann, C.J., Jarvinen, B.R., McAdie, C.J., Elms, J.D., 1993. Tropical cyclones of the NorthAtlantic Ocean, 1871–1992. NCDC/NHC Historical Climatology Series 6–2, 193.
Niering,W.A.,Warren, R.S.,Weymouth, C.G., 1977. Our dynamic tidalmarshes: vegetationchanges as revealed by peat analysis. Bulletin-Connecticut Arboretum 22, 12.
Nikitina, D., Pizzuto, J.E., Martin, R.E., Hippensteel, S.P., 2003. Transgressive valley-fillstratigraphy and sea-level history of the Leipsic River, Bombay Hook NationalWildlife refuge, Delaware, USA. In: Olson, H.C., Leckie, R.M. (Eds.), MicropaleontologicProxies for Sea-level Change and Stratigraphic Discontinuities, pp. 51–62.
Nriagu, J.O., 1998. Tales told in lead. Science 281 (5383), 1622–1623.Oertel, G.F., Woo, H.J., 1994. Landscape classification and terminology for marsh in deficit
coastal lagoons. JCR 10, 919–932.Orson, R.A., Warren, R.S., Niering, W.A., 1998. Interpreting sea-level rise and rates of
vertical marsh accretion in a southern New England tidal salt marsh. Estuar. Coast.Shelf Sci. 47 (4), 419–429.
Philipp, K.R., 2005. History of Delaware and New Jersey salt marsh restoration sites. Ecol.Eng. 25 (3), 214–230.
Pilarczyk, J.E., Horton, B.P., Witter, R.C., Vane, C.H., Chagué-Goff, C., Goff, J., 2012. Sedimen-tary and foraminiferal evidence of the 2011 Tohoku-oki tsunami on the Sendaicoastal plain, Japan. Sediment. Geol. 282, 78–89.
Priestas, A.M., Fagherazzi, S., Wilson, C.A., FitzGerald, D.M., 2012. Rapid wetland expan-sion during European settlement and its implication for marsh survival under mod-ern sediment delivery rates: COMMENT. Geology 40, 284–285.
Psuty, N.P., 1986. Holocene Sea Level in New Jersey. 7, 7. 2 156–167.
172 D.L. Nikitina et al. / Geomorphology 208 (2014) 160–172
Ravens, T.M., Thomas, R.C., Roberts, K.A., Santschi, P.H., 2009. Causes of salt marsh erosionin Galveston Bay. J. Coast. Res. 25, 265–272.
Redfield, A.C., 1972. Development of a NewEngland saltmarsh. Ecol.Monogr. 42 (2), 201–237.Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Ramsey, C.B.,
Buck, C.E., Burr, G.S., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P.,Hajdas, I., Heaton, T.J., Hogg, A.G., Hughen, K.A., Kaiser, K.F., Kromer, B., McCormac,F.G., Manning, S.W., Reimer, R.W., Richards, D.A., Southon, J.R., Talamo, S., Turney,C.S.M., van der Plicht, J., Weyhenmeyer, C.E., 2011. IntCal09 and Marine09 radiocar-bon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51 (4), 1111–1150.
Sallenger, A.H., Stockdon, H.F., Fauver, L., Hansen, M., Thompson, D., Wright, C.W.,Lillycrop, J., 2006. Hurricanes 2004: an overview of their characteristics and coastalchange. Estuar. Coasts 29 (6), 880–888.
Smith, W.E.T., 1966. Earthquakes in Eastern Canada and Adjacent Areas 1928–1959, 32.Publication of the Dominion Observatory, Ottawa, pp. 87–121.
Sommerfield, C.K., Wong, K.-C., 2011. Mechanisms of sediment flux and turbidity mainte-nance in the Delaware Estuary. J. Geophys. Res. Oceans 116 (C1), C01005.
Stefanon, L., Carniello, L., D'Alpaos, A., Lanzoni, S., 2010. Experimental analysis of tidalnetwork growth and development. Cont. Shelf Res. 30 (8), 950–962.
Stewart, J.Q., 1962. The great Atlantic Coast Tides of 5–8March 1962.Weatherwise (June)117–120.
Stuckey, I.H., Gould, L.L., 2000. Coastal Plants from Cape Cod to Cape Canaveral. Universityof North Carolina Press, Chapel Hill (305 pp.).
Stuiver, M., Reimer, P.J., 1993. Extended 14C database and revised Calib 3.0 14C agecalibration program. Radiocarbon 35 (1), 215–230.
Stumpf, R.P., 1983. The process of sedimentation on the surface of a salt marsh. Estuar.Coast. Shelf Sci. 17 (5), 495–508.
Tiner, R.W., 1985. Wetlands of New Jersey. U.S. Fish and Wildlife Service, Newton Corner,MA.
Törnqvist, T.E., Wallace, D.J., Storms, J.E.A.,Wallinga, J., van Dam, R.L., Blaauw, M., Derksen,M.S., Klerks, C.J.W., Meijneken, C., Snijders, E.M.A., 2008. Mississippi Delta subsidenceprimarily caused by compaction of Holocene strata. Nat. Geosci. 1 (3), 173–176.
Tröels-Smith, J., 1955. Characterization of unconsolidated sediments. DanmarksGeologiske Undersagelse, Series IV 3, 38–73.
Turner, I.L., Coates, B.P., Acworth, R.I., 1997. Tides, waves and the super-elevation ofgroundwater at the coast. J. Coast. Res. 13 (1), 46–60.
Tuttle, M.P., Ruffman, A., Anderson, T., Jeter, H., 2004. Distinguishing tsunami from stormdeposits in Eastern North America: the 1929 Grand Banks tsunami versus the 1991Halloween storm. Seimological Research Letters 75 (1), 117–131.
van de Plassche, O., Wright, A.J., van der Borg, K., de Jong, A.F.M., 2004. On the erosive trailof a 14th and 15th century hurricane in Connecticut (USA) salt marshes. Radiocarbon46 (2), 775–784.
van de Plassche, O., Erkens, G., van Vliet, F., Brandsma, J., van der Borg, R.J., de Jong, A.F.M.,2006. Salt-marsh erosion associatedwith hurricane landfall in southern New Englandin the fifteenth and seventeenth centuries. Geology 34 (10), 829–832.
van Huissteden, J., van de Plassche, O., 1998. Sulphate reduction as a geomorphologicalagent in tidal marshes (‘Great Marshes’ at Barnstable, Cape Cod, USA). Earth Surf.Process. Landf. 23 (3), 223–236.
Williams, H.F.L., Flanagan, W.M., 2009. Contribution of Hurricane Rita storm surge depo-sition to long-term sedimentation in Louisiana coastal woodlands and marshes.J. Coast. Res. 56, 1671–1675.
Wilson, K., Kelley, J.T., Croitoru, A., Dionne, M., Belknap, D.F., Steneck, R., 2009. Strati-graphic and ecophysical characterizations of salt pools: dynamic landforms of theWebhannet salt marsh, Wells, ME, USA. Estuar. Coasts 32 (5), 855–870.
Woodruff, J.D., Donnelly, J.P., Emanuel, K., Lane, P., 2008. Assessing sedimentary records ofpaleohurricane activity using modeled hurricane climatology. Geochem. Geophys.Geosyst. 9 (9), Q09V10.