Journal ofCoastal Research 101-124 West Palm Beach,Florida Winter 2003 ~

Fluctuating Holocene Sea Levels in Eastern andSoutheastern Brazil: Evidence from Multiple Fossil andGeometric Indicators

Louis Martin], Jose M.L. Dominguezt, and Abilio C.S.P. Bittencourtj

tCNPqflRD (ORSTOM)Laborat6rio de Estudos

CosteirosCentro de Pesquisas em

Geoffsica e GeologiaUniversidade Federal da

BahiaRua Caetano Moura 12aSalvador, BA 40210-340,

Brazil

:j:Laborat6rio de EstudosCosteiros

Centro de Pesquisas emGeofisica e Geologia

Universidade Federal daBahia

Rua Caetano Moura 123Salvador, BA 40210-340,

Brazil

.ttl!!II!I:.~~.

-: " «'...

ABSTRACT•••••••••••••••••••••••••••••

MARTIN, L; DOMINGUEZ, J.M.L, and BITTENCOURT, A.C.S.P., 2003. Fluctuating holocene sea levels in easternand southeastern Brazil: evidence from multiple fossil and geometric indicators. Journal of Coastal Research, 19(1),101-124. West Palm Beach (Florida), ISSN 0749-0208.

Relative sea-level curves have been delineated for several sectors of the Brazilian coast. In order to have homogeneousand consistent curves, very short segments of the coastline with the same framework are considered. To minimizesystematic errors tied to the particular nature of an indicator, we use the maximum number of different indicatorsavailable from former positions of relative sea level. These sea-level curves show that, during the last 7000 years, thecentral Brazilian coast has been subjected to a submergence phase which lasted until 5100 "C yr BP (5600 cal yrBP), followed by a drop in sea level. This drop in sea level was not continuous but was interrupted by two high-frequency oscillations. The Salvador curve, the most detailed, can be used as a reference for the central portion of theBrazilian coastline. It is now presented with corrections for isotopic fractionation and reservoir effects as well ascalibrations for astronomical ages. Additional supporting evidence for the existence of these high-frequencyoscillationsis provided.

ADDITIONAL INDEX WORDS: Sea-level curves, sea-level indicators, high-frequency sea-level oscillations, coastal evo-lution.

INTRODUCTION

During the last 25 years, research carried out on the cen-tral portion of the Brazilian coastline has provided a goodunderstanding of the Holocene coastal evolution for this re-gion (IJOMINCUEZ et al., 1987, 1990, 1992; MARTl.' et al.,1987a, 1996; Sucuro et al., 1985). As a result of these studies,a number of relative sea-level curves have been constructed(MARTIN et aI., 1979a, 1987a; PlRAZOLLI, 1991) (Figure 1).These sea-level curves show that during the last 7000 yearsthe central Brazilian coast has been subjected to a submer-gence phase which lasted until 5100 l·C yr. BP (5600 cal yrBP) followed by a drop in sea level. This drop in sea level wasnot continuous but interrupted by high-frequency oscilla-tions, during which sea-level dropped slightly below the pre-sent level and then rose afterwards to reach approximatelyits former position. These high-frequency oscillations had am-plitudes of 2 3 meters and durations of 400-500 years. Un-fortunately. most of the data pertaining to these sea-level

99071 received 15 April 1999, accepted in revision 15 April 2002.

curves have been published in low-circulation journals andsymposium proceedings never reaching a broader audience.

The major purpose of this paper is to present the Braziliansea-level curves to a broader audience, with a major focus onthe Salvador curve. Since an important aspect of these sea-level curves is the presence of high-frequency oscillations, ofthe same magnitude as is predicted for the future undergreenhouse conditions, additional supporting evidence for theexistence of these oscillations will be provided. Usually, theyhave been largely ignored when reconstructing coastal evo-lution and more recently their very existence has even beenquestioned (ANGULO and LESSA, 1997). The data providedhere will help us to better understand how the coastal sys-tems will respond to these sea-level changes. '

RECONSTRUCTION OF PAST SEA LEVELS

To reconstruct an ancient position of relative sea-level, itis necessary to fix a past sea-level record in space and time(MARTT!' 1'1 al., 1986a). To locate this record in space it isnecessary to know its present altitude in relation to its orig-inal position in relation to the sea-level at the time it was

102 Martin, Dominguez and Bittencourt

1 0

"c Yr BP x 1000

, 0"c Yr BP x 1000

1 0

14C Yr BP x 1000

, 0"c Yr BP x 1000

, 0"c Yr BP x 1000

7 6 5 4 3

}

SALVAOOR (BA).~._ _. _ '

m~c . - .~~~.7 .6 5 _ 4 3 2 1 0

_ ". , , ,.c Yr BP )( 1000

m511~HE~S (BA) ~ ~C' r==-----=-~76:; 43210

-5 ·"CYr BP x 1000

}

CARAVELAS (BA) - -' _ _m ~ r- -c _

51 ANGRAOOSm REIS (RJ)

7 6

-5

m5ISAN~-L-7/ 6 5 4

-5

51 CANANEIA-mIGUAP~-r= ~__

77~' 4 3 2 ~-=---":;:'~~---;O:--5 "c Yr BP x 1000

m5

] PARANAGUA_(P~ - _ _ r c-7 6 5 \4 r 3

-5JITAJAI- 'm LAGUN,:-{sg~_-- -, -~- C _

7; 6 5 4 3 2

FOGO RIVER

,

RJ PARAiBA DO SUL RIVER

\ - ,

30Q 1,\

(-'\,

11ES DOCE RIVER

1/ VITORIA

ANGRA DOS REISBERTIOGA

SANTOSPRAIA GRANDE

CANANEIAGUARATUilABAY PARANAGuA BAY

ITAJAILAGUNACARNIÇA

Figure L Holocene relative sea-level curves constructed for several sectors of the east and southeast coast of Brazil (modified from MARTI'l ct al., 1987a),Map shows location of places described in text (RN = Rio Grande do Norte State, PB = Paralba State, PE = Pernambuco State, AL = Alagoas State,SE = Sergipe State, BA = Bahia State, ES = Espirito Santo State, RJ = Rio de Janeiro State, SP ~ Sào Paulo State, PR = Paranà State, SC = SantaCatarina State),

formed, To define the record in time it is necessary to datethe time of formation, Once properly defined in time andspace the record will provide the position of an ancient sea-level at a certain date, To minimize systematic errors tied tothe particular nature of an indicator, it is advisable to con-front the information provided by a given indicator with thoseoriginating from other sources. Even if an indicator providesonly partial information, such as the trend of the sea-levelbehavior (rise or fall) and not an accurate reconstruction, thisinformation should not be discarded. In many cases, a con-vergence of data using different indicators is preferable to anisolated "precise" reconstruction, although the latter mayseem intrinsically better. On the other hand, one should notforget that there are nUmerous sources of error in the defi-nition, in space and time, of past sea-level positions. Thus,each reconstruction should be plotted along with error barsfor the estimated altitude and the measured age of the sea-level indicator used. Therefore, each reconstruction shouldplot not as a point but as a rectangle. This standard proce,dure has not always been observed (ANGULO and LESSA,1997; PlRAZzOLI, 1991). On the other hand, a "C age, in fact,i,< not an absolute age. Corrections for isotopie fractionatianand reservoir effects are necessary before ils significance canbe adequately assessed. Furthermore, the same HC age BP

may correspond to one or more caiendar ages (STUIVER andREIMER, 1993). Finally, one should not forget that, despiteexercising the utmost care in the sampJing and analysis,sorne dates might exhibit totally incoherent results. lt is thusnecessary, in arder to outline a curve of variations of relativemean sea-level, to have numerous ancient sea-level recon-structions, weil distributed in time. Obviously, a reliablecurve can only be derived from samples collected from smallseetors of the coastal zone, in order to minimize possible ef-fects of extraneous factors such as differential subsidence oruplift.

Several kinds of sea-level indicators have been used to re-construct the sea-level history along the central Braziliancoast. These indicators are listed below.

Sedimentary Indicators

Along almost the entire Brazilian coastlinc, marine, lagoonand mangrove swamp deposits situated above the presentsea-level are unquestionable evidence of ancient higher sea-levels.

LiHor,l Sand Deposits

Sands deposited in the littoral ZOote have many diagnosticsedimentary structures both of organic and inorganic origin,

Journa.! of Coastal Research, VoL 19, No. 1, 2003

Brazilian Coa~t.al Zone Evolution 103

Figure 2. Sedimentary indicators of higher than present sea-Ievel positions. A-perched beach deposits lying on top of an abrasion terrace. B-

104 Martin, :D6ininguez and Bittencourt

Figure 3. Sedimentary indicators of higher th an present sea-Ieve1 positions. A-paleolagoonal deposits outcropping along a river bank. B-detail of theoutcrop shawn in photo A. Oyster shells collected in these deposits provided an age of 6853(6494)6287 cal yr BP.

1975; FLEXOR and MARTIN, 1979; OLIVEIRA et al., 1990, BE-ZERRA et al., 1998). Trough cross-bedding is the dominantsedimentary structure, indicating they have been originallydeposited in the subtidal zone (upper shoreface) between thebreaker zone and the beach, under the influence of longshorecLirrents. Today these sandstones are positioned weil abovethe low-tide level (Figure 2E). Littoral mollusc shells, collect-ed from within these sandstones, were used in radiocarbondating.

Major uncertainties when using littoral deposits for recon-structing past sea levels are: i) reworkiug of shell materialmuch older than the time of final deposition and ii) recon-structions in space will be probably biased in the sense thatthey will tend to provide higher levels, as a resul t of prefer-entiai preservation of high-energy episodic phenomena suchas storm surges and abnormal wave heights.

Lagoonal Deposits

Int~gr'lr;0n of geologic-geomorphologic 'Ilapping, vibrocoring and "C dating hdS allowed the identification of paleola-

goon deposits in the Brazilian Quaternary coastal plains,whose tops are in many instances located higher than presentmean sea-Ievel (Figure 3). These filled lagoons along the cen-tral portion of the Brazilian coast, are thus testimonies ofperiods of higher-than-present sea levels (BITTENCOURT etal., 1979, 1983; BARBOSA et al., 1986; DOMINGUEZ et al., 1990;MARTIN and DOMINGUEZ, 1994; MARTIN and SUGUlO, 1989;MARTIN et al., 1980; SUGUIO and 1VlARTIN, 1978a, b; SUGUIOet al., 1987). These deposits usually incorporate shell andwood fragments viable for radiocarbon dating. Sea-Ievel re-constructions using this type of indicator will not be as pre-cise as others discussed herein, except for the case in whichmangTove swamp deposits can be identified 'in these lagoonalsequences. Because mangrove forests occupy the upper inter-tidal zone, reconstructions of past sea levels using this typeof indicator will be much more reliable_ However, care mustbe exercised, since organic-rich, water-saturated tine-gIaineddeposits a,-e much prone to compaction, due to dewateringand oxidation of the organic matter content. The', post-de-positional modiftcations can conceivably introduce e1'1'U,.s

Journal of Coasta] Research, Vol. 19, No_ 1, 2003

Brazilian Coastal Zone' Evolution 105

Figure 4. Biologiea! indieators of higher than present sea·level positions. A-abrasion terraee eovered by eolian sediments. B-in the abrasion terraeeshown on photo A vermetid gastropod incrustations are found indicating a sea level 3 meters above the present one (3680(3475)3375 cal yr BP). C-abrasion terraee eut into the Tertiary sediments and laler recovered by beach deposits. D-on lhe surface of lhis terrace, Sùkraslrea slellata coral headsare found in growth position (44l5(4140, 4095)3860 cal yr BP).

when estimating past sea levels by systematically shiftingthe reconstructions towards lower levels.

Biological Indicators

Biological indicators of former sea-Ievels include vermetidgastropod incrustations, oysters, corals and coralline algae(Figure 4). These sessile organisms have their vertical distri-bution intimately controlled by mean sea level and as suchcan be used as indicators of former levels.

Vermetid Incrustations

The vertical distribution of these gastropods along rockyshorelines extends for approximately 50 cm and as such theyprovide very precise reconstructions of past sea levels (LA-BOREL, 1969, 1979, 1986). The altitude of a fossil vermetidincrustation should be measured in relation to the upper lim-it of its living counterparts, or in the absence of these, theupper limit of the infralittoral populations (LABOREL, 1986).In some situations the vermetid incrustations have associ-ated faunistic elements which allow one to establish if thesegastropods were originally positioned in the upper or lowerlimits of their growth zone. For exampJe the presence of Te·traclitas indicates that the vermetids were living close to the

upper limit of their e-rowth zone. Conversely, the occurrenceof coralline algae along with sessile forams and boring bi-valves inclicate the lower limit of the vermetid growth zone(LABOREL, 1986). It is worthwhile to observe that vermetidincrustations will hardly be preserved in a situation of risingsea level because once suhmerged, these incrustations will berapidly destroyed by boring organisms.

Coral Reefs

On the east-northeastern Brazilian coastline there aremany occurrences of coral reefs, which exhibit a truncatedtop, exposed at low tides (Figure 5). The top of these reefsare covered by dead coral and coralline algae .indicating anancient sea-Ievel highstand (LEÀO et al., 1985). AJthoughscleractinian corals can live in a wide range of water depths(down to 40 ml, their upper growth zone is determined byemersion which is normally close to mean low spring tide(HOPLEY, 1986). The truncated tops of the Brazilian coralreefs thus provide a lower constraint to the position of sealevel since at the time they were alive, the sea could not havestood below the level to which they had grown.

Journal of Coastal Research, Vol. 19, No. l, 2.003

106 Martin, Dorhinguez and Bittencourt

Figure 5. Biological indicators of higher than present sea levels. A-Exposed top of a coral reef showing truncated coral heads (Photos BandCl. These features are clear indications of the existence of former higherthan present sea levels: the coral head of photo C provided an age of3205(3075)2975 cal yr BP.

Coralline Algae

In tropical waters, coralline bioherms (algal ridges andtrottoir) develop in close association with sea-level and canbe indicators of that level (AOEY, 1986). Aclditionally thesecOl'alline bioherms develop at a level that is lower than theone occupied by the vel'metid gastl'opods (LABOREL, 1986). lnthe same \Vay as the exposed top of the coral reefs, the cor-alline algae bioherms can be used lo put constraints in theposition of past sea levels since at the time they were alive

Figure 6. Archaeological indicators of higher than present sea levels.These two shellmounds are made up of oyster shells daling3695(3475)3965 cal yr BP (Photo A) and 5640(5450, 5330)5050 cal yr BP(Photo BI. Bath shellmounds are located several kilometers away fromareas where mangrove swamps presently OCCUf.

the sea could not have stood below the level to which theyhad grown.

Other Sea-Level Indicators

Dnder this topic we discuss other indicators of sea-Ievelposition fOllnd along the coastal zone, that although not al·lowing precise sea-level reconstructions can still be used asconfirming evidence of trends in past sea-level behavior (riseor fa11), originally determined through the use of other indi-cators.

Archaeological Indicators

The only archaeological indicators present along the Bra-zilian coastal zone that can be used to draw inferences aboutthe past sea-level history are shellmounds ·("sambaquis")(Figure 6). Between the States of Bahia and Santa Catarina(Figure 1), numerous shellmounds have been built by the pre-historie peoples who inhabited the margins of lagoons andestuaries (MARTIN and SUGUIO, 1976; MARTIN et al., 1986b,1987b; SUGUIO el al., 1992). People who ouilt these moundsrell:ained thought of as technologically "primitive", in spiteof 2', expressive variety of bone artifacts sLrongly s'.,ggestingan .11icient fis' ing technology, as weil as some relatively

Journal of Coastal Research, Vol. 19, No. 1, 2003

Brazilian Caast'al Zone Evolution 107

1 Table L ... Variaûan..'af.[Y'C pnB· as a lunetwh.· afpOsitian. .af the samples· 'Ii,',' ';Table 2. Variation.::pf. [Y': G'PDB· as··a· ftln.çtion.;'b'f:àg'e':wiihin. the ·samewithin. a palealagoon.,.Can.aniia eoa.stal plain. (Stale ofSàa Paulo) (M.

108 Martin, Dominguez and Bittericourt

.Table 3. Radiocarbon ages of rè6ent liermè/id 'sampléS no! maf-ked by tM::'·.éffect of :h~r;"onuet~àl:.~omb eXplosions (MiJlTIN et at., 1996). .'

Radiocirbon ' Laboratory Position ofSampi. Age (yr BP) Numher Sea Level

B.l71 400 ± 80 Bah. 623 +0.5 ± 0.5 IDA.313 '. 390 ± 90 Bah. 488 +0.4 ± 0.5 IDLabJV 380 ± 90 . Gif. 1061 +0.5 ± 0.5 ID·B.22 . 365 ± 110 . :.Bah. 507 +0.3 ± 0.5mA.304 325 ± 100 Bah. 480 +0.4 ± 0.5 m

there has been no indication of such a scenario for the variousshellmounds studied in Brazil (GASPAR, 2000). These two ba-sic assumptions, i.e., proximity of the collecting zone and con-struction above local HWST level, enabled us to establish arather close relationship between geographic position ofshellmounds and ancient sea-Ievels. Shellmounds located far.: .....ther inland were probably associated with a period of maxi-mum lagoonal extent, correlated with a sea-Ievel maximum.Shellmounds whose substrate is located .beneath the presenthigh-tide level suggest an episode of lower sea level. Valuesof 313CPDB for carbonate shells dated from the shellmoundsgive us additional information about high· or low·sea-Ievelpositions, as weil as about the trend of sea-Ievel behavior.

Isotopie Indieators

3'SCPDB measurements of carbonate shells at coastal re-gions can exhibit an ample spectrum of values which are con-trolled by the degree of input of carbon from continentalsources. Shells from lagoonal organisms show 313CPDB valuesranging from about -13%0 (freshwater conditions) to about0%0 (marine conditions) (KEITH et al., 1964). The 313CPOB val·ue is a function of the geographic position of the organismswithin the lagoon (FLEXOR et al., 1979).

Shells of organisms living in t"8 outer zone of a lagoon(near the open-sea edge) show only slightly negative 3'3CPDBvalues, while those living in the inner zones of the lagoon

{dose t6the mainlarid shorelirie)aÇe:(:r~aI1Yèlïdracthized:'b~;;more riegativevalues reflecÜng larger inputs of organic car-'bon from continental sources (Tab,le 1). As time progressesand the lagoon increases o~decreases in size, shell~ from anidéntical geographic location b,ut with different ages will ex-hibit distinct 313CPDB values. Tlie~efore, at a given geographiclocation, .. Iagoonal shells living. during a period' of marinetransgression will exhibit éOgTesslyêly lessnegative values.On the other himd,a dropiiJ.s~ki~velor a ~ëg.ession willresult in an increase of contiriental c~rbon influe;'èe, Wi.th th~lagoonal shells exhibiting more negative 3'SCPDB values. Bythe same token 313CPDB values of shells collected in shell-mounds whose occupation persisted for extended periods oftime can be used as a yardstick ta estimate variations in la-goon area and thus indirectly of sea-Ievel fluctuations (MAR-TIN et al., 1986b). Conversely 313CPDB values of shells collectedin several shellmounds of different ages located around asmall common lagoonal area will conceivably register varia-tions in lagoon area as related to fluctuations in sea level. Amajor assumption intrinsic to the reasoning presented aboveis that the evolution of these coastal lagoons is controlledessentially by sea·level variations, i.e. there should be no ma-jor input of large rivers entering into it, in order to minimizeeffects such as reduction in lagoon size due to sediment de-position. This is the case of the paleolagoons studied in south-ern Sào Paulo state, since the most important rivers in thearea flow toward the interior of the continent, to the Paranariver basin, instead of doing it directly to the sea. Table 2shows variations in &'3CPDB values, as a function of age inshells collected in 6 different shellmounds distributed alongthe margin of a small sector of the 19uape-Cananeia paleo-lagoon (State of Sào Paulo-Figure 1).

Although the archaeological and isotopie records do notprovide precise reconstructions of past sea levels they canprovide constraints or supporting evidence for informationfrom other 30urces, particularly times of maximum and min-imum expansion of coastal lagoons.

Qj 5>.!!'

'"'" 20.0

'"lc:.9

'"üJ0

+ +

6 5 4 3 2 1 014C yr BP x 1000

Fi~:re 8. Relative sea-lp\'el curve constructed by ANGULO aE.' LESSA (1997). This curve. W~:;; con~~,ructed ~oleJy on radiocarhon data from vermetidgastropods.

Journal of Coastal Research, Vol. 19, No. l, 2003

m5

Brazilian Coasta! Zone Evolutidn" 109

4

3

2

A

~ -""Juruvauva Il=

Present sea-Ievel 5

)uruvauva 1

uruvauva III

4Cal yr BP x 1000

B

5

'tapa!! Il

4

Cal yr BP x 1000

Figure 9, Comparison of the Salvador sea-level curve (A) and the Sl3Cpa• values curve (E) of molluscan shells callected in several sh~llmounds locatedaround the same paleolagoonal area for the period 5800--4200 cal yr BP (madified fram MARTIN et al., 1986b),

RELATIVE SEA-LEVEL CURVES FOR THE LAST7000 YEARS

Using the various types of indicators described above, ithas been possible to delineate curves or sketches of curves ofrelative sea-Ievel fluctuations for the last 7000 14C years forseveral sectors of the central part of the Brazilian coast(DOMINGUEZ et al., 1987, 1990, 1992; MARTIN and SUGUIO,1976; MARTIN et aL, 1979a, 1987a, 1987b, 1996; SUGUIO etal., 1980, 1985) (Figure 1). In order to have relatively homog-enous and consistent curves, only very short segments of thecoastline (60 to 80 km), with the same geologic frameworkand numerous data, were considered. Although these curvesare very similar to each other they show notièeable differ-ences in vertical amplitudes (MARTIN et al., 1987a).

The more detailed ofthese curves was produced for a sectorof approximately 50 km situated to the north of Salvador city(State of Bahia-Figure 1) (MARTIN et al" 1979a). Sixty twodeterminations of former relative sea-Ievel posltions, coveringthe last 7000 years (Appendix 1) were used to construct thiscurve, which is certainly one of the most detailed producedworldwide. Indicators used to construct the curve were ver-metid incrustations, calcareous algae crusts and cor;ls col-.lected above their present living zone. Also used were radio-carbon dates of shells collected in cemented littoral zone de-posits. In this case vertical reconstruction of past sea levelwas done based on the sedimentary structures present inthese deposits. On the sea-Ievel curve, reconstructions basedon vermetid and littoral deposits are plotted as a rectangle,whose width and height represent respectively, the age and

the altitude uncertainty associated with the reconstruction.Sea-level reconstructions based on corals and calcareous al-gae are plotted as a point and age error bar. The plot depictsa minimum sea-Ievel position, the arrow pointing upward in-dicates that sea level was probably higher than the positionpresently occupied by the indicator. Reconstructions based ondatings of sLells collected in shell mounds, are also plottedas a point with an age error bar, depicting a maximum po-sition of sea leve!. The arrow pointing downward indicatesthat sea level was probably lower than the base of the shell-mound. Finally, radiocarbon datings of shells and wood col-lected in marine and lagoonal sediments outcropping abovepresent sea level were also used for construction of the sea·level curve. In this case the reconstructions are plotted as apoint and associated age error bar. The arrow pointing up-wards indicates that by the time these sediments were de-posited sea level was higher than the al titude of these de-posits.

AlI reconstructions used to construct the Salvador sea-Ievelcurve are listed in the table presented in the appendix 1.Figure 7 shows this curve plotted both as uncalibrated (Fig-ure 7A) and calibrated "C years (Figure 78). We have optedfor this dual display for the reasons presented below.

The calculation of the radiocarbon age of a sampIe assumesthat the specifie activity of the 14C in the atmospheric CO.has been constant. However, this is not true. The l'C activityin the atmosphere and other reservoirs, and thus the initialactivity of the samples dated has varied over time (STUIVERand BRAZlUNAS, 1993). A calibration dataset is necessary to

Journal of Coastal Research, Vol. 19, No. 1,2003

110 Martin, Dominguez and Bitlencourt

....... ~

•Monte PascoalIsland

,.,.. ...:-:-..~.~.....

23'475'. '. . .;,,>. ,46'02 5'.c=:3 HOloce"e - beach':dep~sii '.,lIIIiIÎ ~~'Prkca~b~~m b~~~~e~~ti~')~' ~ .~. à~ach rid ~s: -,~,;;,: .

.:.j' -"~

23'50'

o 2km-'

m 5 Wm;IIIIilllQ

4

3

Shell- rlch bed~.

iTrnnfrn~m:m-T':"';:'~ 5980 (5905) 5740 cal yr BP

2\---------------- HWL

~-=----nO'4m~ LWLQ"--- --=-.Y..=.'-

m 54

3

Maximumsea-Ievel4.6! DAm

2

Present sea·levelBah. 609

67 5.5cal yr BP x 1000

Figure 10. Location map of the MOITO da Enseada outcrop in the Bertioga coastal plain (Sr) (modified from MARTIN el al., 1979b,c). The sea·levelposition reconstructed from this site has been plotted on the Salvador sea·level curve. See text for delails.

convert conventional radiocarbon ages ("C j'T) into calibratedyears (cal j'T). We have used, the program CALIB 3.0, writtenand distributed by the Quaternary Isotope Laboratory, Uni-versity of Washington (8TUfVER and REIMER, 1993). It isclear that this calibration will only have real meaning if theoriginal 14C ages are first eorreeted for the reservoir and iso-topie fractionation effects. During the seventies when most ofthe dates reported herein were performed (MARTIN el al.,1979a), no reservoir correction was applied, although the iso-topie fractionation eITects were systematically corrected.Measured reservoir effect values are not available for the een-b-al part of Brazi!. We could either apply a systematic-400years correction to ail our samples ofmarine carbonates (cor-al, eoralline algae, vennetid incrustations, elc.) or ',:,je an in-direct method to estimate this effE;ct. In doing this \',: selected

vermetid samples that provides the more reeent ages, not yetmarked by the eITect of thermonuclear bomb explosions, andwere in a vertical position very close to the present level. Theage of these vermetids wouId thus be largely eontrolled bythe reservoir effect (see Table 3).

Based on these assumptions an average value of350 :!:: 100years can be attributed to the coastal oeeanic water of thecentral part of the Brazilian coast. This value' is slightly lessthan the 400 years usually used.

The calibrated curve of Salvador thus shows that (Figure 7B):

• the present zero (mean Jevel) was exceeded for the firsttime in the Holocene at ~bout 7800 cal yr BP;

• about 5600 cal yr BP, tl'c' relative sea-level Wc: • throl:,-ha first maximum of 4.7 :'.. 05 m above the present leveJ;

Journal of Coasta! Research, Vol. 19, No. l, '2003

Brazilian COBstal Zone Evolution III

.after~his;~aximurir;:there was a rapid regression unti\"5300 cal yr BP, slowing down up to 4400 cal,yr BP, andspeeding up again. at 4200 cal yr BP. At about this time.the sea level passed through a minimum, pr~bably belowthe present level; ,

• after 3700 cal yr BP, a rapid transgression occurred, andat about 3500 cal yr BP, the relative sea-Ievel passedthrough. a second maximum of 3.5 ± 0.5 m above the pre-··sent level; ,.....

• between 3500 and 2800 yr BP, the relative sea-Ievel fellslowly and regularly. Starting at 2800 cal yrs BP, the de-cline became very rapid again. At 2600 cal yr BP the ma-rine level passed through a second minimum, probably be-low the present level;

• after 2300 cal yr BP, the relative sea-level rose very rap-idly, passing through a third maximum of 2.5 ± 0.5 mabove the present level at about 2100 cal yr BP;

• since 2100 cal yr BP, the relative sea-level has fallen reg-ularly to its present position.

In summary, it is possible to recognize along the centralpart of the Brazilian coast during the last 7800 years, threemain events of submergence (7800-5600, 3700-3500 and2300-2100 cal yr BP) alternating with three main events ofemergence (5600-4200, 3500-2800 and after 2100 cal yr BP).

COMPLEMENTARY SUPPORTING EVIDENCEFOR THE EXISTENCE OF THE

HIGH-FREQUENCY OSCILLATIONS

The Salvador curve shows the existence of two distinctivehigh-frequency oscillations at 4300-3500 cal yr BP and 2700-2100 cal yr BP. Also the interval between 5900 and 5300 calyr BP was characterized by a rapid rise followed by an equal-Iy rapid fall. Recently ANGULO and LESSA (1997) using sea-level reconstructions based exclusively on I·C dating of ver-metid incrustations have published a new sea-Ievel curve fol'Brazil (Figure 8) in which they questioned the existence ofthese oscillations and the maximum of the Holocene trans-gression reported above. These authors, based on the as-sumption that "...vermetids are widely recognized as an ac,curate paleo sea-level indicator. .." considered that, when theinformation provided by the vel'metids, conflicts with infor-mation provided by other indicators, the former should bepreferred. At first this might scem a controversy of local in-terest. However, wc all have during the last decade becomeaware of the fundamental role exerted by sea-level changesnot only in controlling evolution of coastal areas but also thedeposition of entire sedimentary packages at the continentalmargins(VAlL et al., 1977; VAN WAGONER et al., 1990) (thesequence stratigraphy paradigm). Additionally the high fre-quency sea-Ievel oscillations delineated in the Salvador curveare of a magnitude and frequency of those predicted for thenext century as a result of global warming. It is thus neces-sary to discuss these high-frequency oscillations a bit further.

Reconstructions of past sea-levels for eastern and south-eastern Brazil indicate that at 4200-3700 and 2600-2300 calyr BP, relative sea level was positioned very close to the pres-P''', level, rising rapid!y afterwH",ls. Thus any vermcticl in-crustations produced at these law s8a level periods would be

'.- subsequentlycovéred by li risind'~e~lé'Y~:i'a~ddèstroyèd byboring organisms~'This might bé' the reason why in the, Sal-

, vador curve there is a gap in information from vermetid gas-tropods for these two time intervals. The sea-Ievel positionsplotted in the curve for these two periods are based on recon-structions using littoral sandstoncs. The addi tional evidencethat we provide hérein supportive of the existence of thesehigh~frequencyoscillations cornes from the study of the evo-lution of the coastal plains located along the eastern andsoutheastern Brazil coast (DOMINGUEZ et al., 1987; MARTINand SUGUIO, 1992; MARTIN et al., 1987a, 1993). These studiesshow that during these two periods important environmentalchanges affected these plains and the pre-historie peoples liv-ing there.

The Maximum: of the Holocene Transgression

In the Salvador curve the maximum of the Holocene trans-gression (4.7 ± 0.5m above present sea level around5660(5590)5495 cal yr BP) is based in one reconstruction(MARTIN et al., 1979a), which could at first be a source ofcontroversy. However, information from other sectors of theBrazilian coast corroborates the timing and/or altitude of themaximum of the Holocene Transgression in the Salvadorcurve.

ln the Cananéia region (Sao Paulo State-Figure 1) the1113CPDB values of molluscan shells collected in shellmoundslocated around the same paleolagoonal area (MARTIN et al.,1986b) provide interesting data (Table 2). These data showclearly that at the Cananéia paleolagoon the lowest influenceof continental carbon (i.e., the less negative 1113CPDB value)occurred between 5650 and 5450 cal yr BP (5245 and 5110HC yr BP), which probably corresponds to the greatest ex-pansion of the lagoon, suggesting that the maximum of theHolocene trans,,'Tession has been reached at that time. Com-parison of the Salvador sea-level curve and the 1113CPDB valuescurve shows a good match for the period 5800-4200 cal yrBP (Figure 9), in close agreement with the time of the max-imum of the Holocene transgression.

In the Bertioga coastal plain (Sao Paulo state-Figure 1)there is an ancient littoral deposit (Figure 10) cemented bycalcium carbonate, which accumulated in a small protectedreentrant in a rocky shoreline, The study of the sedimentarystructures in this deposit indicates that it has resulted fromslow in situ vertical accretion oflow-tide, shell-rich sedimentsaccompanying a rise in sea level (MARTIN et al., 1979b). Com-parison of sedimentary structures in this littoral deposit withtheir modern counterparts suggests that the top of this de-posit accumulated when sea level was positioned 4.5 ± 0.5 mabove the present leve!. On the other hand, shell fragmentscollected in sediment positioned 2.0 m above '.their moderncounterparts provided an age of 5980(5905)5740 cal yr BP(Bah.609). The information derived from this outcrop indi-cates that sea level rose to a level 4.5 ± 0.5 m above thepresent level after that time. This reconstruction agrees withthe altitude of 4.7 ± 0,5 m indicated by vermetids in theSalvador CUfve. Unfortunately it was not possible to preciselyd" t.e the top ofthe outcrop because of shell contamination bymore recent carbon (MARTIN et al., 1979c).

Journal of Coastal Research, Vol. 19, No, l, 2003

112

.-'ie.-!-::",-.~:.~-.

• Precambrian

• Pleistocene - alluvial depositso Pleis!ocene - beach deposi!s

o Holocene - lagoonal - mangrove depositsD Holoeene - beaeh depositsCS] Oulerop location

Beach sands

Lagoona! mud

Tree trunks in life position7240 (7170) 7000 cal yr BP (Gij. 3845)7430 (7380, 7340) 7230 cal yr BP (Bah. 327)

2.0m

Top of present day 0.5mmangrove O.Sm

Bah.327

Present

0.5m Equivalent present day0.501 depositlevel

Maximumsea-Ievel4.5:+c 0.4m m

5

4

3

2

6 cal yr BP

Figure Il. Location map of the Praia Grande region (SP) outcrop (modified from MA.RTJN el al., 1979b). The sea-Ievel position reconstrucled from thissile has been plolted on the Salvador sea-Ieve! curve. See text for details.

A somewhat similar situation (MARTIN et al., 1979b) isfound at the Praia Grande region (Sao Paulo state-Figure1). lri this region a littoral deposit, comprised by lagoonalfacies in its lower portion is ovedain by sediments indkdliveof beach-face dep-'sition (Fj re 11). Comparison of sedimen-tary structures in these d'''posits with their modern Counter-parts indicates that the beach-face sedim!'nts accumulated ata time when relative sea level was 4.5 ~ 0.5 m above the

present level. In the lagoonal sediments, trunks of mangrovetrees in life position indicate a sea level 2.0 := 0.5 m abovethe level where modern mangrove trees grow. Radiocarbondating of these trunks provided ages of 7430(7380,7340)7230cal yrs BP (Bah.327) and 7240(7170)7000 cal yr BP (Gif.3845).'n·:" arrangement of facies indicates that the 10\\·... portionof the deposit accumulated dur; .'é': a tra:,gressive· phase,whereas the upper porlion accumulalt·, much latter and in

Journal of Coastal Research, Vol. 19, No. 1; 2003

m5

4

3

2

l

.Brazilian Coastarz;,neEvo!ution

Present sea·level,

Laborel

113

6 5 cal yr BP x 1000

Sam pie Age (cal yr BP) Material

Curray 585') (5595)5320 Oyster shells(LJ.1364 )B 190 5855(5640,5620)5475 Vermetid gastropod

ES 44 5930 (5880, 5825, 5775)5655 Vermetid gastropod

ES 17 5565(5280,5150,5090)4865 Vermetid gastropod

Laborel 4860(4800,4770,4590,4575)4410 Vermetid gastropod

Figure 12. Sea-!evel reconstructions from different sites in southern Brazit, for the period 6000-4500 cal yr BP (MARTIN el al., 1996), when plotted onthe Salvador curve, show remarkable agreement.

association with a higher sea leveL This occurrence indicatesthat the maximum of the Holocene transgression reached 4.5:t 0.5 m above the present sea-Ievel, sometime after 7000 cal'fT BP. This reconstruction also agrees with the 4.7 :t 0.5 msea-Ievel position inferred from vermetids in the Salvadorcurve.

Several other sea-level reconstructions from southern Bra-zil (MARTIN et al., 1996), when plotted on the Salvador curve,display a remarkable agreement, helping to pinpoint the tim-ing of the maximum transgression in space and time (Figure12).

The information provided above thus corrobora tes the po-sition in space and time of the maximum of the Holocenetransgression in the Salvador curve (4.7 ::': 0.5 m above pre-sent level at 5660(5590)5495 cal YI' BP)

The High-Frequency Oscillation of4200-3700 cal yr BP

The Rio do Fogo Mangrove Swamp Deposit

In the Rio do Fogo region (Rio Grande do Norte State-Figure 1), a mangrove swamp deposit a few decimeters thickcontaining wood and moluscan shells (Lucina peetinata) CI'OpSoufat the beach-face on a sector of the shoreline experiencingerosional retreat today (Figure 13). The top of this mangrovedeposit is positioned slightly below the upper surface of mod-ern mangrove swamps. Radiocarbon dating of a piece of woodcollected in these sediments provided an age of4260(4225,4190,415.5)4095 cal yr BP (pa.1670) (o"Cpou =-25,2';100). This date provides additional evidence that around4200 cal 'fT BP, relative sea level was positioned slightly be-

low the present level. An eventual compaction of this depositcan not significantly arTect this reconstruction.

The Maratua Shellmound

The Maratua shellmound (Santos region-Sâo PauloState-Figure 1) provides interesting indirect information onsea-Ievel position. The base of this shellmound is located be-low present sea level. Considering the modern ideas of thesocial and cultural meaning of the shellmound (DE BLASIS etal., 1998; GASPAR, 2000) and its very location right in themiddle of lagoonal-mangrove deposits, it is reasonable to as-sume that this mound was constructed during a period oflower than present sea level. Il would be very improbable tosuppose that prehistoric people would pile sueh an impressiveamount of shells in the middle of a lagoon, at a point locatedat a minimum distance of 2 kilometers from any high ground.On the other hand the fact that the base of the shellmoundis located beneath present sea level cannot be attributed tosubstrate subsidence due to loading, because the shellmoundlayers are not deformed in any way. As previously concludedby EMPERAlRE and LAAlING-EMPERAlRE (1956), if the mid-den had sunk under its own weight, the center would be ex-pected to have lowered farther than the margins and the in-ternaI layers would have been deformed. Two samples fromthe submerged part ofthis shellmound had been dated at thebeginning of the radiocarbon investigations. The results,7330 :t 1300 "c 'fT BP (Gif.15) and 7800 :t 1300 14C 'fT BP(Gif.l6) were initially accepted because the dates did coincidewith a period when relative sea-Ievel was definitely lowerthan today (LA1I1JNG-EMPERAIRE, 1968). However, archaeo-

Journal of Coastal Research, Vol. 19, No. 1, 2003

114

........ ,."

Martin, Doriiiilgt,ez and BittencoUit

Dune . -: -:' ../ Erosion

HWL

Formermangrove

4260 (4225, 4190,4155) 4035 cal yr BP (Pa. 1670)

ŒWLD:::'./_LWL

m5

4

3-!---------- L~ Present sea-Ievel~~-=t~

5Pa (1670)

4 cal yr BP

Figure 13. Mangrove s\Vamp deposit olltcropping at the beach face in the Rio do Fogo region (RN). The sea-Ievel reconstruction from this site, ",henplotted on the Salvador curve coincides with a period of sea level very close to the present one, associated with the first high-frequency oscillations ofthe Salvador curve. See text for details.

logical remains within the Maratua shellmound suggestedmuch later dates. Redating of EMPERAlRE and LAMING'Ssamples (MARTIN et al., 1986b) produced ages of4,080(3850)3650 cal yr BP (Bah382) and 3895(3825,3750)3640 cal yr BP (la 9185), which are more in keepingwith the artifactual evidence. These new dates thus coincideswith the lower sea level event between 4300 and 3500 cal yrBP depicted in the Salvador curve (Figure 14).

The Rio Boguaçu ShellmolU1ds

Two other shellmounds (PR09 and f' .: 0) from Rio Rog-\ JÇU (Figure 15) (Baia de Guaratuba, Stnt\', of Parana- Fig-l're 1) a 50 provide additional p., ,.lence of the 4200-3700 calyr B.P. oscillation (MARTIN et al, 1987b) The shellmoundnamed PR09 rests on an island made up of lagoonal sedi-

ments. Nowadays its substrate is located below present sealevel. Mollusc shells collected at this midden provided agesof 4130(3840)3625 cal yr BP (Bah 1272), The 813CPDB value ofthese shells is -3,560/00. In the same area, another shell-mound named PR10,located 1200 m from shellmound PR09,rests on top oflagoonal sandy deposits positioned 1.5 m abovev,'sent high tide level. Mollusc shells sampled from the lowerperipheral portion of this mound provided ages of3355(3075)2820 cal yr BP (Bah.1273), and 8 J3CPDB value was-0.67f

Brazilian Coastal Zone Evolution 115

• PrecambriamIII Pleistocene - Aluvial depositso Pleistocene - Beach deposits

o Holocene· Lagoonal - Mangrove deposits[J Holocene - Beach deposits54 © - Maratua Shell midden

MaratuaShell midden

4,5

Present sea-Ievel

m43 -1-------__.

2

1

3.5

cal yr BP x 1000

Figure 14. Location map of the Maratua shellmound in the Santos coastai plaill ISUGUlü and MARTIN, 1978a). Ages of two shell samples From the lowerpart of the midden when plotted in the Salvador sea-Ievel curve coincides with a s'-'gment of that curve characterized by a 1cJwer than present sea level.See text for detai\;,

Maratua shellmound, and ii, that when shellmound PRI0was built (between 3850 and 3100 cal yr BP) sea Jevel washigher than the present !evel due to the 013C PDa values as-sociated wiLh this mound (Figure 15), Please note that onFigure 15 radiocarbon datings for shellmound PRI0 plots be-low the Salvador sea-Ievel curve, This is due to the fact thatthe Paranagua sea level curve as shown on Figure 1, exhibitslower amplitudes jf compared to the Salvador curve,

Holocene Evolution of the Doce River Strandplain

Detailed reconstruction of the Doce river strandplainevo-!ution has shown that shoreline progradation during the Ho-locene was interrupted by several erosional episodes (SUGUIOet al., 1982; DOMINGUEZ, 1987; DOMINGUEZ et al., 1987; MAR-TIN and SUGUIO, 1992). Most of these episodes were relatedto changes in longshore dlift direction (MARTIN et al., 1993).However there have been two episodes of severe erosion,which because of its widespread nature and association wiLhpartial inundation of the strandplain, as evidenced by " ,0presence of lagoonal sediments have been interpreLed aô aresult of a rise in sea Jevel. The Holocene evolution of the Rio

Doce coasta! plain began with the formation of a barrier is-landllagoonal system (Figure 16A). Soon af1erwards the RioDoce started construction ofa large intralagoonal delta. Afterthe maximum of the Holocene transgression, progradation ofthe shoreline produced extensive littoral sand deposits form-ing beach ridges. In the central part of the Doce River coastalplain, in front of the 'intralagoonal paleodelta, it is possibleto recognize, from the geometry of these beach-lidges, thepositions occupied by five former river mOllths. Truncationrelationships show that these paleo-liver mouths \Vere aIl ac-tive during the same period of time, possibly favored by thedrop in sea level af1er 5600 cal yr B.P. (Figure 16B), A woodfragment sampled froru beach-ridges related to one of theseriver mouths has been dated as 4560(4410)4150 cal yr BP(Bah. 964), which coincides with a period of sea-Ievel drop inthe Salvador curve. After the depositional phase following theexit of the distributaries, the coasta! zone was subjected tosignificant changes ma~ked by severe erosion affecting al-most the entire co"sL!ine. At the same time the sea invadedparts of the strandplain as docume::LPd by DOMINGUF:Z(1987) and MARTIN and SUGuro (1992) (Figure 16C). The five

Journal of CoastaJ Research, Vol. 19, No, 1,2003

116 . Martio,:Doming"ez and Billencoun

m

• Precamboam basement

• Pleistocene - alluvial deposits

o Pleislocene - beach deposits

o Holoeene - lagoonal - fiuvial depositso Holoeene - beach depositsJo. PR. 09 . PR. 10, Shell middens

PR.l0 Jol' Il! l! Il! 1 III ~I 1 ! 1 ! Il! 'll-4

6" C (POB) = • 0.6%Present sea-level PR.09

1~!III"!I!II!ill!IIIIU~

6" C (POB) = - 3.5%3

cal yr BP x 1000

Figure 15. Location map of the RIO Boguaçu (bala de Guaratuba, PR) PR-10 and PR-09 shellmounds (1V1ARTIN el al., 1988). Ages and 5 13CPD8 values ofmolluscan shells collected in the lower part of these two middens, when plotted in lhe Salvador cur"e, show a remarkable agreemenl between the historyof occupation of Ibis sile an'd the sea-Ievel history. See text for delails.

river mouths were drowned and subsequently abandoned.Afterwards the Doce river concentrated its fJow in just oneriver mouth (Figure 16Dl. Radiocarbon dating of lagoonalsediments irilllling low-lying areas between sets of beachridges provided ages around 3400 cal yrs B.P. Additionallythe existence of shellmounds built on top of these beach ridg-es right at the margin of the paleolagoon, and dating3695(3475)3365 cal yr BP corroborates that t!lis importantenvironmental change was related to a rise in sea level. Thise':olutionary history thus provides support for a sea-level be-havior ch.:acterized by a fal! in sea leve] hefore4fil,0(411 »4150 cal yr DP foL·.ved by a lise before363,,(3465)3280 cal yr BP (Figure 16), which is in agreem~ntwith the nrst high-frequency osciUation depicted on the Sal-vador curve.

The High-frequency Oscillation of 2700:-2100 cal yr BP

The Shellinound of Pedra Oca

The Pedra Oca Shellmound is located at the margin of To-dos os Santos Bay (Bahia State-Figure 1) Its base rests ontop of a veneer of littoral sands (Figure 17) which overlie anabrasion terrace carved by wa ves into sandstones and mud-stones of Cretaceous age (Figure 17). The top of the abrasionterrace is located 0.7 m above the present day high waterlevel or 3.5 m above mean low tide, Vermetid incrustationsat this abrasion terrace, indicating a past sea level 3.0 ::': 0.5-rn 2bove the present level have been datee! providing aL ageof 2870(2765)27;:;;, cal yr BP (lhi-h 542). Archeteological stud-ies of Ihis site by CALDERON (1964) has discJo~.:d the r'èm-nants of hearth-stones on the perchee) transitlonal sands at

Journal of Coastal Research, Vol. 19, No. 1,2003

Brazilian Coastal Zone Evoiution

C:J Teriary deposils Holocene - Beach deposilsPleislocene - Beach deposils E:J Holocene-Ouvial deposils (de Ilaie) ... Shell-midden

117

,.,,:.

Figure 16. Holocene evolution of the Doce river coastal plain (ES). A-barrier/island lagoonal system. B-contemporaneous exit of five distributaries ofthe intralagoonal delta. C-erosion of the entire eoastline and inception of the second lagoonal phase. D-abandonment of the five river mouths whichwere replaced by a single mou th (modified from MARTIN and Sueulo, 1992). The timing of the contemporaneous exit orthe Doce river distributaries andof the second lagoonal phase, when ploLted on the Salvador curve coincides respectively with a period of ses·level drop and sea·Jevel rise.

the base of the shellmound, contemporaneous to the begin-nings of midden accumulation. Shells collected at the base ofthe midden were dated as 2745(2705,2490)2350 cal yr BP(Si.470) and as 2355(2325)2145 cal yr BP (Gif.878) (MARTINet al, 1986b).

From the data presented above it is possible to derive thefollowing sea leve\ history:

• a'round 2800 cal yrBP relative sea level was positioned 3.0:!: 0,5 m above the present level as indicated by the ver-metid incrustations on top of the abrasion terrace;

• the accumulation of the perched transitional sand depos-its recovering the abrasion terrace occurred soon after-wards;

• construction of the shellmound only began after sea levelhas dropped. This inference is based on the presence of thehearth-stone found at the base of the shellmound. Al-though people today sometimes build beach tires below thehigh water line, if we consider the modern interpretationof these shellmounds as discussed before, we think it isperfcctly reasonable to suppose that the hearth-stone pres-enl in this site was positioned in a supra-tidal level whenthe pre-historie people arrived al the site.

Jequitinhonha Stransplain

Paleogeographical reconstructions for the JequitinhonhaRiver strandplain (Bahia State-Figure 1) (DOMINGUEZ,1983, 1987; DOMINGUEZ et al., 1982, 1983, 1987) has shownthe existence of three paleo-mouths associated with three dis-tinct generations of littoral sand deposits (Figure 18). It isqui te clear examining these three river roouths and associ-ated beach-ridge systems, that changes in river-mouth posi-tion occurred very abruptly and not as a result of shifts con-trolled by littoral processes such as longshore sediment trans-port. Simple meandering processes can be discarded in thiscase because of the large scale of the changes involved. Sub-sidence is not a factor also to be considered because topo-graphie levelling shows that beach-ridge altitude regularlydecreases seaward (A.C.S.P. BrITENcouRT, unpublisheddata). However a rapid rise in sea level would have the sameefTect that subsidence does in controlling river avulsion. Thisinterprelation corroborated by other Jjnes of evidence is fa-vored herein. Our attention is to·.be focused on the secondand third positions occupied by the Jequitinhonha rivermou th. Wood debris collected in the most external portion ofthe littoral sands deposited in association with the second

Journal of Coastal ReseArch, Vol. 19, No. 1, '2003

Ï18 Martin, Domïn'guez and Bittencoü'rt'

..;1.'...1IIIÎIII.....llliililliIIiIIiI_...._ ........_IIIiIïIII....~......iiIIII:l__ÏllliillIIIIIIIII__

A

B

Presentsea·level

/ Vermelid incrustallon

cal yr BP x 1000

Figure 17. Location map of the Pedra Oca shellmound at the margin of the Todos os Santos Bay (BA) (CALDEROI', 1964). Ages of vennetid gastrùpodincrustations and of two shell samples of the lower part of the shellmound when plotted on the Salvador curve, show a remarkable agreement betweenthe history of occupation of this sile and the sea·level history as depicled i1l the Salvador curve. See text for more delails.

river mouth were dated al 2765(2740)2480 cal yr BP(Eb) 432). On the other hand, wood deblis collected in theinnumost portion of the littoral sands associated with thel!lin] river mouth was dated as 2345(2310,2230,2210)213Ccalyr HP (Bah.325J. These dHles sho," that the Jequitinhonh"river mouth shifted [rom position 2 to position 3 after2765(2740)2480 cal yr BP and before 2345(2310,2230,2210)

2130 cal yr BP. This shift is possibly associated with a lisein sea level since the abandoned river mouth (river mouth 2)exhibit~ in aerial photographs and satellite imab~;Y' mor-phologîcal [catures indicative of d"owning, the cen' "al portionof the littoral S8:iès being covered with fine-grained serli·ments (DOMfNC1Jl::Z, 1983, 1987) (Fl~. :~e 18). V/hen thesedates are placed on the Salvador curve (Figure 18) they are

Journal of Coastal Research, Vol. 19, No. 1, 2003

Brazilian Coastal Zone Evolution 119

.... _ l.,\-

39"JOW 38'~OW

Tertiary deposits

Paleo·channel . Jequitinhonha river

Former moulhs . systems 1, 2 &. 3

Fluvial deposils

Plelslocene· beach depasils

H·o·locene· beach deposils

-- Truncalions in beach·ridge alignment...... .-_. . " .

15'40'S

'" 15'50'SN

aNN

ci'"NN

ci

'"~'"v'"N

Figure 18, Simplified geology-geomorphology map of lhe Jequitinhonha River coaslal plain (BA) showing posilions occupied by the Jequilinhonha Rivermouth and assoeiated beach-ridge syslems during lhe Holocene evolulion M lhe plain (modilied from DOMINGUEZ, 1983). Radiocarbon dales of beach-ridge systems 2 ,nd 3, when plotled on the Salvador relative se"level curvc" indicate thal river·moulh shilling was probably a consequence of a rapidsea·level rise associ,ted wilh lhe terminal phase of the second high·frcquency oscillation of the Salvador cW'\'e, Sec text for details.

coincident with the 2700-2100 cal yr BP high-frequency os-cillation, with the river mouth shifting occurring during therising stage of that oscillation, On the other hand truncationsin beach-ridge alignment observed within each of the threemapped generations of littoral sand deposits have also beenidentified which does presenta clear relationship with sedi-ment dynamics, These truncations are better explained asepisodes of severe erosion rel"ted to continuous/interrnittentlateral migration of river mouths and changes in longshoredrift direction,

The Caravelas Strandplain

The northern portion of the Caravelas strandplain (south-em Bahia State--Figure 1) was characterized during the Ho-locene by extensive progradation of the shoreline (MARTIN etal,. 1993), This progradation was not continuous but was in-ten'upted by erosional episodes as indicated by truncationsin beach-ridge alignment (Figure 19), These truncations oc-cur in different scales and hierarchics reAecting different de-grees in sevelity of the erosional episodes, The Caravelas

strandplain is not related ta a major river course, 50 erosivephases reftected in these beach-ridge truncations can not berelated to changes in river mouth dynamics, On the otherhand many of these beach-ridge truncations are of small scaleand clearly related to changes in coastal dynamics, Howeverthere is one extensive truncation that affects the entirenorthern part of the plain (Figure 19), We think that becauseof its extension this truncation is better explained as the re-suIt of a rapid rise in sea level. This truncation is widespreadand affects the entire northern portion of the strandplain,This truncation is also associated with a narrow low-lyingarea filled with lagoonal sediments, Wood debris and mollus-can shells collected in these lagoonal sediments provided agesof 2340(2320)2150 cal yr BP (Beta 104 779) and2355(2345)2330 cal yr BP (Ly.7959) respectively, '\'hen plot-ted on the Salvador sea-level curve, these dates coincide withthe rising sea-level stage of the 2700-2100 cal yr BP high-frequency oscillation (Figure 19), thus suggesting a geneticrelaLionship between the two phenornefla, i.e" the \\idespreaderosion was a consequence of a rapid rise in sea le\'e1.

Journal of CoastaJ Research, Vol. 19, No, l, 2003

120 Martin, Domiri"guez and Bittericourt

., .......... .

17'40'S

tl\l)iIl\VlWIA~~Ponta da Baleia

Ponta do Catoeiro

o Tertiary deposits Cl Holocene - beach depositsPleistocene - beach deposits cg Holocene - lagoonal - mangrove deposils

~ Beach ridges

3

Present sea-Ievel cal yr BP x 1000L+--+-+------t--r--t---r---t-4-+------t~f~F___r___,_r_t-

2

Figure 19. Simplified geology-geomorphology map of the Caravelas strandplain (BA) (modified from MARTIN el al., 1993). Radiocarbon dating oflagoonaldeposits infilling the low-Iying area 3ssociated with an important truncation in beach-ridge aligllment, when plotted on the Salvador relative sea-levelcurve, coincides with the rapid sea-Ievel rise associated with the terminal phase of the second high-frequency oscillation of that curve.

The Paraîba do SLÙ Strandplain

In the northern part of the Paraîba do Sul River coastalplain (Rio de Janeiro State-Figure 1) three generations ofHolocene littoral sand deposits have been identined (MARTINet ai, 1987a, 199.3) (Figure 20C) As it happened in the Doceliver and Caravelas strandplo""ô, the prograddtion of thecoastline was not continuou, hUé interrupted by erosionaJ ep-isodes as indicatd by truncatio,~ in beach-ridge alignments.These truncations occur in different scales and hierarchies

rcf!ecting different types of erosional episodes. Natural var-iations in the discharge of the Paraîba do Sul river havecaused episodes of localized erosion (NIARTLN et al_, 1984),perfectly identified as truncations in beach-ridge sets in thevicinity of the river mouth. Other causes ofbeach ridge trun-cation are relatcd t0 changes' in coastal dyna r , '::;s such aoinversions in longshore drift direction (MM' rrN" ri SUGurü,] 992). Of interest here is the f"ct that the secoLd and thirdgenerations of Holocene littoral sand deposits are separated

Journal of Coastal Research, Vol. 19, No. 1,2003

Br~zilian Coastal Zone Evolution li1"

El Tertiary deposits_ Pleistocene - Beach deposits

mIIIIIIIIl Paleo-channel

o Holocene - Fluvial depos;ts.-- Holocene - Beach depositsL- 1",2~ and 3" Genera'l;ons

cal YI' ElP x 10002

Bah,1098Bah,1261

Start of the constructionof the third generalion ofHolocene beach deposifs

3

Present sea-Ievel

D

Figure 20, Sirnplined Holocene evolution of the norlhern part or the Paralba do Sul slrandplain (RJ) (modined from MARTIN et al., 1997), This evolu-lionary scheme was prepared bascd on lhe integration of patterns of beach·ridge tru ncation and radiocarbon dati ng. A-deposition of tlie second generalionof the littoral sand tleposits. B-erosion of the external portion of the second generation of the littoral sand deposits. C-beginning of the de position ofthe third generation of the littoral sand deposits. Radioearbon dating ofsamples collected in the low·lying area separabng the second and third generationsof littoral sand deposits, wh en plotted on the Salvador relative sea-Ievel curve, coincides with the rise in sea level that characterizes the final stage ofthe second high-frequency oscillation that curve. See text for details.

by a low-lying area filled with lagoonal sediments (MARTINet al., 1987a, 1993). The formation of the lagoon was precededby an episode of severe erosion of the externaJ portion of thesecond generation of littoral sand deposits as indicated bytruncation in beach-ridge alignment (Fig-ures 20A and B). Onthe other hand the lagoon could only form aCter the beginningof deposition of the third generation of littoral sands (Figure20C). Shells from the lagoonal sediments provided ages of2350(2145)l945 cal. yr BP (Bah.1098) and 2300(1985)1815 yrBP (Bah.1261) (MARTIN et al., 1993). When plotted on theSalvador curve, these ages coincide with the rise in sea levelthat characterized the final stage of 2700-2100 cal yr BPhigh-frequency oscillation of that curve (Figure 20D).

CONCLUSIONS

Relative sea-Ievel curves constructed for the last se\'eralthousand years provide an important reference against which

the evolution of the coastaJ zone can be investigated. The rel-ative sea-level curve for Salvador can be used for a referenceto the central portion of the Brazilian coastline. Since it is verydetailed it is now preseoted herein with corrections for isotopicfractionation and reservoir effects as weU as calibrations forastronomical ages. The information presented herein from oth-er parts of the Brazilian coastal zone provides additional sup-porting evidence for the time when the maximum of the Ho-locene transgression was reached (araund 5600 cal YT BP) andfor the existence of the high-frequency oscillations depicted inthe sea-level CUl"Ve. Contrary to recent work by At'

MartiD, DO;;1Ïri'~ez and Bittencourt

mi)-riy cases; a_conv.ergent data sebis preferable tb.isolated ifl,~,·~1formation, although the latter may intrinsically seem bettêr,j.Dàta convergenCe' in favor of the existence of at least two high::;'frequency oscillatio~s after 5600 cal yr BP is too great fontsomeone to be in doubt. As Dr N. A. Morner sagaciously statedduring a scientific meeting: "There are more than mere coin',..cidences." (MARTIN et al., 1998).

ACKNOWLEDGMENTS

This paper was greatly improved by the critical reviews ofD.F. Belknap (University of Maine) and D.L. Forbes (BedfordInstitute ofOceanography) to whom the authors are sincerelygreatful. J.M.L. Dominguez and A.C.S.P. Bittencourt wouldalso like to thank CNPq (Brazilian Research Counci\) for pro-viding 2 Research Fellowship.

LITERATURE CITED

AnEY, W.R, 1986. Coralline algae as indicators. In: VAN DE PLAS-SCRE, O. (ed.) Sea·level Research: A Manual for the Collection andEvaluation of Data. Norwich: Geo Books, 9, 229-280.

ANGULO, R J. and LESSA, G.C., 1997. The Brazilian sea-levelcurves: a critical review with emphasis on the curves from Para-nagua and Cananéia regions. Marine Geology, 140, 141-166.

BARBOSA, L.; BITTENCOURT, A.C.S.P.; DOMINGUEZ, J.M.L., andMARTIN, L., 1986. The Quaternary coastal deposits of the State ofAlagoas: influence of the relative sea-level changes. Quaternary ofSouth America and Antartic Peninsula, 4, 269-290.

BEZERRA, RR; LIMA-FILRO, F.P.; AMAJ'AL, RF.; CALDAS, L.RO.,and COSTA-NETO, LX.,1998. Holocene coastal tectonics in NE Bra-zi!. In: Coastal Tectonics, Special Publication, Geological Society,London, 146, 279-293.

BIGARELLA, J.J., 1975. Reefs sandstone from northeastern Brazi!.Anais Academia Brasileira Ciências, 47, 395-409.

BITTENCOURT, A.C.S.P.; MARTIN, 1,.; DOMINGUEZ, J.M.L., and FER-REIRA, YA., 1983. Evoluçao paleogeografica quaternaria da costado Estado de Sergipe e sul da costa do Estado de Alagoas. Revista13rasilcira Geociências, 13, 2, 93--97.

Brr'TENCOURT, AC.S.P.; MARTIN, L.; VILAS-BoAS. G.S., and FLExoR,J-M., 1979. Quaternary marine formations of the coast of the Stateof Bahia, Brazi!. International Symposium on Coastal Evolution inthe Quaternary. Sao Paulo: Proceedings, 232-253.

BRANNER, J.C., 1904. The stone reefs of Brazil, their geological andgeographical relations. Bull. of the Museum of Comparative Zool-ogy. Geological series, 44, 7, 207-275. .

CALDERON, V., 1964. 0 sambaqui de Pedra Oca. Instituto de CiênciasSociais, Universidade Federal da Bahia, 2, 1-89.

DARWIN, C.R, 1841.0n a remarkable bar of sandstone on the Per-nambuco coast of Brazi!. Philosophical Magazine and Journal ofSciences, 19: 257-260.

DE BLASIS, P.; FISH, S.K; GASPAR, M.D., and FISH, P.R., 1988. Jour·nal of American Archaeology, 15, 75-105.

DOMINGUEZ, J.M.L, 1983. Evoluçao quaternaria da planicie coste iraassociadaà foz do Rio Jequitinhonha (BA): influência das variaçoesdo nivel do mar e da deriva litorânea de sedimentos.Curso de Pés-Graduaçao em Geologia, 1nstituto de Geociências, UniversidadeFederal da Bahia, M.S. Thesis, 73p.

DOMINGUEZ, J.M.L., 1987. Quaternary sea-level changes and the de-position al architecture of beach-ridge strandplains along the eastcoast of Brazi\', Rosenstiel School of Marine and Atmospheric Sci-ence, University of Miami, Florida, USA, PhD TI··' •.,is, 288p.

DOMINGUEZ, J.M.L.; BrTTENcouRT, A.C.S.P.; 1".',0 Z.M.AN., anrlAVEDo. A.E.G., 199(j Geologia do Quaternario costeiro do [stad.,de Pernarnbuco. Rev! ,1" Brasileira Geociêncios, 20, 208·-215

DOMINGUEZ, J.M.L.; Br'iTENcOURT, AC.S.P., and MARTI"', L., 1983.o papel da deriva litorânea de sedimentos arcnosos na construçaodas planicies costeiras associadas a desembocaduras dos rios Sao

'iFfancisco{SE/~)Î',J.e.\l'1t~iJP.,~pf};f!.(~-N,.P~"3~g;;$}'ie"g~rafb,a;;(!'o'sul (RJ). Revlsta BrasLlelra Geoclênc.as, 13, .2, 93-105. .

DOMINGUEZ, J,M.L.; BIT'TENCOURT, AC.S.P., and ~1ARTIN,L., 1992..Controlson Quaternary coastal evolution of the east-northeasterncoast of Brazi!: roles ofsea-levels history, trade ~inds and climate.Sedimentary Geology, 80, 213--232.

DOMINGUEZ, J.M.L.; MARTIN, L., and BITTENCOCR. A.C.S.P., 1982,Evoluçao paleogeografica do "delta" do Rio Jequitinhonha (BA)durante 0 Quaternario: influência das variaçoes do nivel do maL

.Alas do N Simp6sio do Quaternario no Brasil: Rio de Janeiro, 62-92, .

DOMINGUEZ, J.M.L.; MARTIN, L.; and BIT'TENCOL'RT, A.C.S.P., 1987.Sea-level history and Quaternary evolution of river mouth-asso-ciated beach-ridge plains along the east-southeast Brazilian coast:a summary. In: Nm-IMEDAL, D.; PILKEY, O.H. ,and HOWARD. J.D(eds.) Seo·kve! fluctuation and coastal evolution, Special Publica-tion of the Society of Economie Paleontologists and Mineralogists,41, 115-127.

EMPERAIRE, J. and 1AJv[lNG, A., 1956. Les sambaquis de la côte mêr-idionale du Brêsi!. Campagne de Fouilles (1954-1956). Journal So-ciété des Américanistes, 45, 5-163.

FAIRBRIDGE. R.W., '1976. Shellfish-eating preceramic indians in coast-al Brazi\. Science, 191, 353-359.

FLEXOR, J-M. and MARTIN, 1,., 1979. Sur l'utilisation des grès co-quilliers de la région de Salvador (Brésil) dans la recônstructiondes lignes de rivage holocènes. International Symposium on Coast-al Evolution in the QucLternary. Sao Paulo: Proceedings, 343-355.

FLEXOR, J-M.; MARTIN, L., and SUGUIO, K, 1979. Utilisation durapport 13C/12C comme indicateur d'oscillations lagunaires. Inter-national Symposium on Coastal Evolution in the Q1Iaternary. SaoPaulo: Proceedings, 346-357. .

GASPAR, M.D. 2000. Sambaqui: arqueologia do litoral brasileiro. Jor-ge Zahar ed., 89p.

HOPLEY, D., 1986. Corals and reefs as indicators of paleo-sea levelswith special reference to the Graet Barrier reef. In: \'A.~ DE PLAS-SCHE, O. (ed.) Sea-level Research: a Manual for the Collection andEvaluation of Data. Norwich: Geo Books, 8, 195-228.

KEITH. M.L.; ANDERSON. G.M., and EICHELER. R., 1964. Carbon andoxygen isotopic composition of mollusk shells from marine andfreswater environments. Geochim. Cosmocim. Acta, 28, 1787.

LABOREL, J., 1969. Les peuplements de Madréporaires des côtestropicales du Bréei!. /Innahs de l'Université d.'A,è:djan Série E, 2(3), 1-260.

LABOREL, J., 1979. Fixed marine orga nisms as biological indicatorsfor the study of recent sea level and climatic variations along thebrazilian tropical coast. International Symposium on Coastal Evo·lution in the Quaternary (Sao Paulo) Proceedings, pp. 193-211.

LABOREL, J., 1986. Vermetid gastropods as sea-level indicators. In:VAN DE PLASSCHE. O., (ed.) Sea-level Research: a Manual for theCollection and Evaluation of Data. Norwich: Geo Books, 10, 281-310.

LAMING-EMPERAJRE, A., 1968. Missions archéologiques françaises auChili Austral et au Brêsil Méridional: datations de quelques sitespar le radiocarbone. Journal Société Américanistes, 57, 77-99.

LEAo, Z.M.A.N.: BITTENCOURT, A.C.S.P.; DOMIXGUEZ, J.ML.; NO-LASCO, M.C., and MARTIN, L., 1985. The effects of Holocene sea-level fluctuations on the morphology of the Brazilian coral reefs.Revista Brasileira de Geociências, 15(2),154-157.

MARTIN, L.; BITTENCOURT, AC.S.P.; DOMINGL"EZ, J.M.L.; FLEXOR,J-M., and SUGUIO, K 1998. Oscillations or not oscillations, thatis the question: Comment on ANGULO, RJ. and LEsSA, G.C. TheBrazilian sea·level curves: a critical review with emphasis on thecurves from the Paranagua and Cananéia regioll5. Marine Geology,150(1-4), 179-187.

MARTIN. L.; BITTENCOURT, AC.S.P.; VILAs-Bo.-\S. G.S., and FLEXOR,J-M., 1980. Mapo geo/6gico do quaternârio costeira do Estado daBahia. 2 folhas eo':ala 1/250000, Texto explicativo, Secrctaria deMinas e Energi8, . ;"lvador (Brasi!).

M.ARTIN, L. and ;)(!\lINGUEZ, J.M.L., 199,1. Geological ,.,story ofcoastal lagoom wn.h particular referer,ce ta exa:nples j',-orn' theeast coast of Brazi!. In: KJERVE, B. (ed.) Coaslal Lagoon Process,Elsevier Oceanography Series, 60, 41-68.

Journal of Coastal Research, Vo!. 19, No. 1,2003

Brazilian Coastal Zone Evolution '123

MARTIN, ,L,.; F;l,EXQR, J.M., and SUGUIO, K 1984 Enregistœment des ,,;périodes de fortes et faihles énergies à l'embouchure d'un fleuve.

, Le tas du .Panoa do Sul (Brésil). Implications paléoclimatiques.. ,C.R. Ac. Sc. Patis, série il, 299,.66J..,664.,;:r· .

·MAfmN,''L.;FI'.ExoR, iNI.;'VÏi:":S.'BoAS, ·G.S.;· BIT~ENCOURT,AC.S.P., and GUIMÀRAES, M.M.M., 1979 a. Courbe de variation duniveau t'elaÙ( .dela·mer au cours des 7 000 dernières années sur :',un secteilr'hom~gènedu littàral brésilien (nord de Salvador). In-tern~'tional Symposium on Coastal Evolution in the Quaternary.Sâo .Paulo: Proceedings, 264-274. .

MARTiN, L.; MORNER, N.A.; FLEXOR, J-M. and SUGUIO, K,1986 a.: F!1,ndamentos eTeconstruçao de antigos niveis marinhos do Qua-. ternario.BOletim IGJ USP,4, 1-16l.MARTIN, L. 'and SUGUIO, K, 1976. Les variations du niveau relatif

de la mer au Quaternaire récent dans le sud de l'Eat de Sao Paulo.Utilisation des "sambaquis" dans la détermination des ancienneslignes de rivage holocènes. Actes du XLII Congrès des Américan-istes, IX: 73-83.

MARTIN, L. and SUGUIO K, 1989. Excursion route along the Brazil-ian coast between Santos (State of Sao Paulo) and Campos (Northof State of Rio de Janeiro). International Symposium on GlobalChanges in South America during the Quaternary. Sao Paulo: Spe-cial Publication, 2, 1-136.

MARTIN, L. and SUGUlO, K, 1992. Variation of coastal dynamicsduring the last 7000 years recorded in beach-ridge plains associ-ated with river mouths: example from the central Brazilian coast.Palaegeography Palaeoclimatology Palaeoecology, 99, 119-140.

MARTIN, L.; SUGUIO, K; DOMINGUEZ, J.M.L., and FLEXOR, J-M.,1997. Geologia do Quaternario costeiro do litoral norte do Rio deJaneiro e do Espirito Santo. 2 folhas escala 1/250 000, Texto ex-plicativo 104 p, CPRM-Serviço Geolôgico do Brasil.

MARTIN, L.; SUGUIO, K, and FLEXOR, J-M., 1979 b. Le Quaternairemarin du littoral brésilien entre Cananeia (SP) et Barra de Guar-atiba. International Symposium on Coastal Evolution in the Qua-ternary. Sao Paulo: Proceedings, 296-331.

MARTIN, L.; SUGUIO, K, aild FLEXOR, J-M., 1979c. Précautions surl'utilisation d'âges HC dans la reconstruction des anciens niveauxmarins. Additif et correctif à: Le Quaternaire marin du littoralbrésilien entre Cananeia et Barra de Guaratiba. InternationalSymposium On Coastal Evolution in the Quaternary (Sao Paulo)Proceedings, pp. 332-342.

MARTIN, L.; SUGUlO, K, and FLEXOR, J-M., 1986 b. Shell middensas a source for additional information in Holocene shoreline andsea-Ievel reconstruction: examples from the coast ofBrazil. In: VANDE PLASSCHE, O. (ed.) Sea-level Research: A Manual for the Collec-tion and Evaluation of Data. Norwich: Geo Books, 18, 503-521.

MARTIN, L.; SUGUlO, K, and FLEXOR, J-M., 1987 b. Relative sea-level reconstruction during the last 7000 years along the states ofParana and Santa Catarina coastal plains; additional infonna-tions derived from shell-middens. Quaternary of South Americaand Antartic Peninsula, 4, 210-236. .

MARTIN, L.; ABSY, M.L.; FLEXOR, J.M.; FOURNlER, M.; MOURGUIART,PH.; SIIFEDDlNE, A, and TURcQ, B. 1993. Southern oscillation sig-nal in South American paléoclimatic data of the last 7000 years.Quaternary Research, 33, 338-346

MARTIN, L.; SUGUlO, K, and FLEXOR, J-M., 1993. As flutuaçoes donivel do mal' durante 0 Quaternario superior e a evoluçao geolô-gica de "deltas" brasileiros. Boletim IG/ US?, 15, 1-186.

MARTIN, L.; SUGUlO, K; FLEXOR, J-M., and AzEVEDO A.EG., 1988.Mapa Geol6gico do Quaternario costeira dos Estados do Parana eSanta Catarina. 2 folhas escala 1/250 000, Texto explicativo 40 p.Departamento Nacional da Produçao Mineral, Série Geologia 28,Seçao Geologia Basica 18.

MARTIN, L.; SUGUIO, K; FLEXOR, J-M.; DOMINGUEZ, J.M.L., and

'BI'ITENCOURT, AC.S.P., 1987 a. Quaternary evolution of the cen-tral part of the Brazilian coast. The l'ole of relative sea-Ievel var-iati?n a~d of shorelin~, d'1ft; In:, Qu~te'(L.a~~:Çoaftql Qep!9gy. pt

:'. West·Afnca and Soutli'Amenca, T;JNESCq'Report in' Marine Sci-enc'e, 43, 97-145. '.'

MARTIN,' L.; SUGUIO, K.; .FLExOR, J:M.; DOMINGUEZ, J.M.L. andBITIENCOURT, AC.S:P, 19.96. Quaternary Bea-Ievel history andvariàtion in Dynamics along the central Brazilian coaôt: conse-quenceson coastal plain construction. Anais Academia BrasileiraCiências, M,' 303:'354:

OLIVEIRA, M.I.; BAGNOLJ, E.; FARIAS, C.C.; NOGUEIRA, AM. B., anJSANTIAgoiM., 1990. Consideraçàes sobre a geometria, petrogralia,

. sedimeritologia, diàgênse e idades dos beachrocks do Rio grandedo Norte. Anais XXXVI Congresso Brasileirode Geologia, 2,621-634.

PIRAZOLLI, P., 1991. World Atlas of Holocene sea-level changes. NewYork: Elsevier, Oceanography Series, 58, 1-300.

STUIVER, M. and BRAZIUNAS, T. F., 1993. Modeling atmospheric "cinfluences and HC ages of marine samples back to 10,000 BC. Ra-diocarbon, 35, 137-189.

STUlVER, M. and REIMER, P.J., 1993. Radiocarbon Calibration Pro-gram REV 3.0.3c. Radiocarbon, 35, 215-230.

SUGUlO, K and MARTIN, L., 1976. Presença de tubos fosseis de Cal-lianassa nas formaçàes quaternarias do litoral paulista e sua util-izaçao na reconstruçao paleoambiental. BoUG1US?, 7, 17-26.

SUGlJ10, K and MARTIN, L., 1978 a. Mapa geol6gico do Quatemàriocosteiro do Estado de Sao Paulo. 8 folhas escala 11100000. Governodo estado de Sao Paulo, Secretaria de Obras e do meio Ambiente,Departamento de Àgua e Energia Eléctrica.

SUGUIO, K, and MARTIN, L., 1978 b. Quaternary marine formationsof the State of Sao Paulo and southern Rio de Janeiro. Interna-tional Symposium on Coastal Evolution in the Quaternary. SaoPaulo: Proceedings, 232-253.

SUGUlO, K; MARTIN, L., and DOMINGUEZ, J.M.L.,,1982. Evoluçaodo "delta" do Rio Doce (ES) durante 0 Quaternario: influência dasvariaçoes do nivel do mar. Atas .do IV Simp6sio do Quatemario noBrasil, Rio de Janeiro, 93-116.

SUGUIO, K; MARTIN, L.; DOl\lINGUEZ, J.M.L.; BITTENCOURT,AC.S.P., and FLEXOR, J-M., 1985. Flutuaçàes do nivel relativo domal' durante 0 Quaternario superior ao longo do litoral brasileiroe suas implicaçoes na sedimentaçao costeira. Revista BrasileiraGeociências, 15, 273-286.

SUGUIO, K; MARTIN, L., and FLEXOR, J-M., 1980. Sea-level fluctu-ations during the past 6 000 years along the coast of the State ofSao Paulo (Brazil). In: MORNER, NA, (ed.) Earth Rheology, Isos-tasy and Eustasy. New York, Wiley, pp. 471-486.

SUGUIO, K; MARTIN, L., and FLEXOR, J-M., 1992. Paleoshorelinesand the Sambaquis of Brazil. In: JOHNSON, L.L. (ed.) Paleoshore-lines and Prehistory. Boca Raton: CRC Press, pp. 83-99.

SUGUIO, K; MARTIN, L.; FLEXOR, J-M., and AzEVEDO, AE.G., 1987.The Quaternary sedimentary deposits in the States of Parana andSanta Catarina coastal plains. Quaternary of South America andAntartic Peninsula 4: 3-25.

VAlL, P.R.; MITCUM, R.M, and THOMSON, III S., 1977. Seismic stratig-raphy and global changes of sea level, part 3: relative changes ofsea level from coastal onlap. In.- PAYTON C.N. (ed.) Seismic Stra-tigraphy Applications ta hydrocarbon Exploration. Tulsa,Oklahoma: Amer. Association Petroleum Geologists, Memoir 26,pp. 63-97.

VAN WAGONER, J.C.; MITCHlThl, R.M.; CA..'\IPION, KM., and RAHMANIAN,V.D., 1990. Siliclastic sequence stratigraphy in welllogs, cores andoutcrops: concepts for high-resolution correlation of time and fa-cies. American Association vf Petroleum Geologists., Methods Ex-ploration, Series, 7, 1-55.

Journal of Coastal Research, Vol. 19, No. 1,2003

124 Martin, Domingilez and Bittencourt'

" .. ' . ," ;!',:~{--st,0.0 6630 :t 175 6280:t 175 . 7375 (7195) 6950B.25 Shen BR Bah.510 + 1.7 (:t0.5) 66]0 :t 180 6260:t 180 7365 (7175) 6900B.131 Shen BR Bah.601 +2.2 (:t0.5) 6515 :t 170 6261:t 170 7220 (7140, 7020) 6810B.19 Shen Bah.687 >0.0 6500 :t 175 6150:t 175 7215(7015)6795B.63 Shen Bah.557 >0.0 6440:t 170 6090:t 170 7175 (6910) 67408.116 Shen .Bah.576 >0.0 6300 :t 170 5950 :t 170 7000 (6780J 6565B.20 Shen Bah.505 >0.0 6240:t 155 5890:t 155 6885 (6730) 6500B.59 Shen Bah.553 >0.0 5970 :t 170 5620 :t 170 6630 (6410) 62808.54 Shen BR Bah.543 +2.0 (:t0.5) 5940 :t 155 5590 :t 155 6525 (6400, 6350) 6215R18 Shen Bah.504 >0.0 5675 :t 145 5325 :t 145 6285 (6lïO, 6100) 5930B.44 Shen BR Bah.524 +2.0 (:t0.5) 5470:t 160 5120 :t 160 6160 (5905) 56658.17 Shen Bah.675 >0.0 5265 :t 150 4915 :t 150 5880 (56-15) 5485B.I07 Verm. Bah.567 +4.7 (:t0.5) 5195 :t 110 4845:t 110 5660 (5590) 5465R119 Verm. Bah.579 +2.7 (:t0.5) 4970 :t 150 4620 :t 150 5570 (5310) 5045R267 Coral Bah.776 >+2.2 4960 :t 90 4610:t 90 5455 (5310) 5075B.43 Shen Bah.523 +2.7 (:t0.5) 4545 :t 120 4195 :t 120 5860(4820,4660)4535R49-A Verm. Bah.519 +2.5 (:t0.5) 4245:t 95 3895:t 95 4425 (4350,4305)4150B.123 Shen BR Bah. 589 0.0 (:t0.5) 4175:t 85 3825 :t 85 4405 (4230, 4160) 4090B.04 Shen BR Bah.497 0.0 (:t0.5) 3880 :t 120 3530:t 120 3970(3825,3740)3635T.C. Shen BR Gif.2150 0.0 (:t0.5) 3780 :t 130 3430 :t 130 3835 (3685,3650) 3475R178 Shen Bah.104 +3.5 (:t0.5) 3595:t 120 3245 :t 120 3625 (3465) 3355MD.l Shen Bah.269 +3.3 (:t0.5) 3550 :t 130 3200 :t 130 3565 (3390) 3265B.167 Coral Bah.621 >+1.5 3545 :t 105 3195 :t 105 3475 (3390) 3275lIlD.2 Shen Bay.270 +3.3 (:t0.5) 3450 :t 120 3100 :t 120 3455 (3340,3280)3155B.27 Coral Bah.512 >+1.6 3290 :t 55 2940 :t 55 3205 (3075) 2975R50 Shen Bah.539 +3.2 (:t0.5) 3260:t 100 29]0 :t 100 3210(3060,3015)2875B.51 Shen Bah.540 +3.1 (:t0.5) 3100 :t 120 2750 :t 120 2960(2845,2820)2755B.53 Verm. Bah.542 +3.1 (:t0.5) 3030 :t 120 2680:t 120 2870 (2765) 2735B.52 Shen Bah.541 +3.0 (:t0.5) 2990 :t 120 2640:t 120 2850 (2755) 2715B.I08 Verm. Bah.568 +3.0 (:t0.5) 2990 :t 135 2640 :t 135 2860 (2755) 2520PO.01 Sambaqui Si.470 +0.5 2250:t 80 1900 :t 80 1925 (1830) 17208.283 Verm. Bah. +2.0 (:t0.5) 1915 :t 100 1565 :t 100 1540 (1415) 1335B.268 Verm Bah.777 +1.5 (:t0.5) 1705 :t 120 1355 :t 120 1345 (J285) 1165B.174 Cale. Alg. Bah.626 >+0.8 1565 :t 120 1215 :t 120 1275 (1135, 1100) 995B.113 Verm. Bah.573 +0.7 830 :t 90 480 :t 90 545 (515) 470B.114 Verm. Bah.574 +0.8 (:t0 ..5) ~30 :t 90 480 :t 90 545 (515) 470B.170 Verm. Bah.622 +0.7 (:t0.5) 760 :t 80 410 :t 80 515 (490)320 -.B.24 Verm. Bah.509 +0.7 (:t0.5) 700 :t 110 350 :t 110 510 (435, 330) 290

.B.125 Cale. Alg. Bah.596 >+0.5 675 :t 125 325 :t 125 505 (425, 395) 0Bl24 Verm. Bah595 +0.7 (:t0.5) 660 :t 80 280 :t 80 440 (305) 0B.172 Verm. Bah.624 +0.5 (:t0.5) 610 :t 80 260 :t 80 430 (295) 0B.173 Verm. Bah.625 +0.5 (:t0.5) 570 :t 100 220 :t 100 420(285)0B.l71 Verm. Bah.623 +0.5 (:t0.5) 400 :t 80 50 :t 80 255 (0108.22 Verm. Bah.507 +0.3 (:t0.5) 365:t 115 15 :t 115 134 (01 0

Reconstruction in space and time of former relative sea-Ievel positions usee! ta eonstruct the Salvador curve ("C yr BP and cal yr BP 1. Verm.-vermetidgastropod, Cale. Alg.-calcareous algae, Shen BR-littoral sandstones (beach-rock), Sambaqui-shellmidden.

Journal of Coastal Research, Vol. 19, No. 1,2003

Martin Louis, Dominguez J.M.L., Bittencourt A.C.S.P. (2003)

Fluctuating holocene sea levels in eastern and southeastern

Brazil : evidence from multiple fossil and geometric

indicators

Journal of Coastal Research, 19 (1), 101-124

ISSN 0749-0208