Holocene deposits on the coast of north-central : radiocarbon ages and implications for coastal changes

Yoko Ota Departmenl 01 Geography, Yokohama Nalional UnlversHy, Tokiwadai, Hodogaya-ku , Yokohama 240, Japan

Roland Paskoff Departmenl 01 Geography, Lumlere Unlverslty, Lyon, France

ABSTRACT

Eight radiocarbon-dated shell samples collected from the emerged marine deposits on Bahra Herradura nearCoquimbo and Bahra Tongoy areas, north-central Chile, provide new constraints on the rate of coastal changa during the Holocene. The Holocene transgression culminated ca. 6,000 yr B.P. based on a shell age from esluarine deposits in Estero Tongoy. In Bahía Tongoy, the spatial separation of parallel beach ridgas of ca. 910 yr B.P. and ca. 2,300-2,400 yr S.P. suggests a progradation rate of ca. 0.14 m/yr. Assuming a constant progradation rate, the oldest Holocene beach ridga is ca. 5,400 yr B.P. The Holocene uplift is very small, probably 2-3 m in most cases, and 4-5 m in the most rapidly uplifting area (Altos de Talinay). The low Holocene uplift rates are similar to the ones obtained from the elevation above present sea level ofthe last interglacial terrace in this region .

Key words : Holocene marine deposlts, Beach rldge, Uplift rate, Radlocarbon age, North-central Chile.

RESUMEN

Depósitos holocenos en la costa del norte-centro de Chile: edades radiométricas Cl. e implicaciones para los cambios costeros. Muestras de conchas recolectadas en la terraza marina más baja de las bahfas de , Herradura y Tongoy en el Norte Chico de Chile (30°$) informan sobre la evolución de la costa durante el Holoceno. La transgresión posglacial terminó, probablemente, hace 6.000 años, según lo indicado por la edad de depósitos esluarinos en el estero de Tongoy. En la bahfa Tongoy, una comparación entre la posición de la playa hace 900 años y hace 2.300-2.400 años AP., indica una velocidad de progradación del orden de 0,14 m por año. La edad de la playa holocena más antigua seria de 5.400 años A.P. si la velocidad de progradación fuera constante. En los alrededores de Coquimbo y Tongoy, el solevantamiento de la costa durante el Holoceno ha sido de poca importancia, probablemente menos de 3 m y menos de 5 m en el sector más inestable de los Altos de Talinay. Estas estimaciones concuerdan con las que se pueden hacer para la terraza marina del último perfodo interglacial en esta región.

Palabras claves: Depósitos marinos ho/ocenos, Cordón de playa, Velocidad de solevantamiento, Dataclones radlocarbónlcas, Norte Chico, Chile.

INTRODUCTlON

The study area is the coast along Bahía Tongoy coastal line of this area is fringed by Holocene (Tongoy Bay) and Bahía Herradura (Herradura Bay) emergent terraces or beach ridges. Two radiocarbon near 30°5. north-central Chile (Fig. 1). Most of the ages for Holocene shells were reported from Estero

RfJII1shJ Geol6gloa d. Chile, Vol. 20, No. 1, p. 25-32, 6 FIgs., 1 Table, JuIy 1993. 26 HOLOCENE DEPOSITS ON THE COAST OF NORTH-CENTRAL CHILE: RADIOCARBON AGES AND IMPlICATIONS ...

80 60 La Serena

(, \ \ \ (, o ~ N

FIG.1 . Mapshowinglhelocalionofthesludy area (a and b). Conlours: 200 m inlerval afler la Serena 1:500,000 o ! lopographic map. Insel shows the I location offigu 191 . Puerto Aldea Faull limils the Allos de Talinay lo the wesl and lowland lo Ihe east.

• 14C sampling site

1m-- Holocene lowfand area with beach ~ ridges Q :;' Boh/o Tongoy Q ...... Crest of well-delined beach ridge '<

:::": Quebrada (wash) deposits

~ Aockysea c!iff

Abandoned Holocene sea cliff ~ and Pleistocene marine terrace

~ Fauij (geoiogically and geomor- phoiogically defined). N ~ FIG.2. Marphology andsamplingsiles o 2km in !he Sahfa Tongoy al9a. b==*==' Location is shown in figul9 1. Y. ata and R. Paskoff 27

Location 2 Location 3 Location 6 Surface height Surface height ca. 3.4 m ca. 3.5 m m m O , " ,1, weak top soil shells (3·2)', weak soil (slope wash deposits, _:..: LUl 2,510i100yrB.P. thickness unknown) ~.:.~;~ ::. ~' .• ;o;:;.. .• .t: fine sand (fluvial?) C. Ql O Cl gravel, max. 60 cm diameter with shells, 2,370 i 100 yr B.P. sand and gravel ca. 3.5 m a.m.s.1. with sparse shells silt with shells in growth posilion 6,380 i 120 yr B.P. bedrock stratified sand and gravel 2 slream level

gravel wilh shells (3·1)', 3,140i 90 yr B.P. abrasion plalform sandslone wilh gravel 25 cm diameler

'Sample numbers corresponding lo Table 1.

FIG. 3. Section of Holocene deposits. Locations are shown in figure 2.

Tongoy (Location Ain figure 2) and Bahía Herradura coastal areas (Figs. 2, 5) . The paleontology (Herm, (Location B in figure 5) by Paskoff (1973). No previous 1969) and morphostratigraphy (Paskoff, 1970) of the study, however, was done on Holocene sea level marine terraces, particularly in the Bahía Coquimbo history and Holocene coastal evolution in the study area (Fig. 5), have been studied. The Herradura 11 area. terrace at Coquimbo, about 15-20 m in altitude, is the The purpose of this paper is to report eight new lowest of five major Pleistocene terraces in this area radiocarbon ages from Holocene deposits and to (Paskoff, 1970); it is probably of last Interglacial age discuss the Holocene transgression, coastal based on amino acid ratio (Hsu et al., 1989) and ESR progradation rate and vertical movement based on analysis (Radtke, 1989) of shells. The uplift rate at these materials. Coquimbo is thus very low, about 0.1-0.2 mlkyr In the weste rn part ofthe study area, a conspicuous (Leonard and Wehmiller, 1992). normal fault (Puerto Aldea Fault; Paskoff, 1970) A low marine terrace below Herradura 11, at less forms the eastern margin of north-south trending than 7 m around the bays of Coquimbo and Tongoy, Altos de Talinay highlands (Fig. 1), which are com­ was reported as early as the middle of last century by posed of sedimentary and igneous metamorphosed Darwin (1846) and Domeyko (1848), and later by Paleozoic rocks, and intrusive Mesozoic rocks (Gana, Brüggen (1929). These authors attributed this low 1991). The coastal area to the west of this fault is terraceto recent continental uplift. Radiocarbon ages mostly underlain by Miocene-Pliocene sedimentary from the terrace in Bahía Herradura (3,700±120 yr rocks, but Mesozoic rocks are exposed in headland B.P.) and Bahía Tongoy (2,400±240 yr S.P.) confirm areas (Thomas, 1967). a Holocene age for this terrace (Paskoff, 1973). Well-preserved marine terraces characterize the

DESCRIPTION OF RADIOCARBON LOCALlTIES

BAHIA TONGOY or on very narrow abrasion platforms. Along the lowlands, to the east of the Puerto Aldea Fault, a well­ The Altos de Talinay highlands are fringed by developed beach extends for 15 km. Two samples cliffs along a rocky coast. Holocene deposits occur (Locations 1 and 2 in figure 2, table 1 ) were obtained only locally as beach deposits filling emerged notches from the rocky eastern coast of the Altos de Talinay, 28 HOLOCENE OEPOSITS ON THE COAST OF NORTH-GENTRAL CHILE: RAOIOCARBON AGES ANO IMPLICATIONS ••• four samples (Locations 3, 4, and 5 in figure 2, Table sand and gravel, with abundant well-preserved shells 1) from beach ridges located west of Quebrada (Fig. 3). The radiocarbon age for a lower sample (3- Pachingo, and one sample from the lower reach of 1) at 1.1 m a.s.l., immediately aboye the abrasion the Estero Tongoy (Location 6 in figure 2, Table 1). platform, is 3, 140±90 yr S.P. An upper safTltlle (3-2) Selow, sites from the southwestern part of the are a to at 3.2 m a.s.1. yields an age of 2,51 0±1 00 yr S. P. the northeastern part are described. Location 4 is situated on the extension of the ridge Location 1 is an elevated marine notch, which is on which Location 3 is, and Location 5 is on the outer­ 5.1 m a.s.1. at its retreat point and is covered with most fossil beach ridge (Fig. 2). At these localities, beach sand with shell fragments, exposed in a quarry beach deposits are mainly composed of rounded about 600 m southeast of the Punta Lengua de Vaca gravels with little fine-grained matrix, but including (Fig. 2). Radiocarbon age for these shell fragments is abundant shells and shell fragments. The radiocarbon 3,220±80 yr S.P. age is 2,330±80 yr S.P. (ca. 4.1 m a.s.l.) for Location Location 2 is in beach deposits composed of well­ 4 and 91 0±90 yr S.P. (ca. 2.7 m a.s.l.) for Location 5. rounded boulders up to 60 cm in diameter with shell Location 6 is a section on the right bank of Estero fragments, resting on an abras ion platform at ca. 3.5 Tongoy (Figs. 2-4). This is the only locality where m a.s.l., overlain by slope wash deposits (Figs. 2,3). Holocene estuarine deposits were identified in the These shell fragments have an age of 2,370±1 00 yr study area. At Location 6, an estuarine silt bed, which S.P. has abundant fossil shells with paired valves, is Location 3 is a quarry about 100 m seaward from exposed. The silt bed is overlain by fine sand, probably the abandoned Holocene sea cliff and just north of a of fluvial origino The upper limit of the estuarine main track headingto Puerto Aldea. Of all the sampling deposits was not confirmed at this locality. Fossil sites in Holocene deposits, this site is the closest to shells in estuarine deposits at ca.1.8 m a.s.1. yield an the abandoned Holocene sea cliff, but is actually age of 6,380±120 yr S.P. located on the fourth beach ridge fromthe cliff (Fig. 2) . Shell fragments with an age of 2,400±240 yr S. P. Seach deposits rest unconformably on an abrasion (Location A; Paskoff, 1973) were collected from platform, about 1.0 m in altitude, which truncates beach deposits at about 1 km downstream from Miocene sandstone. They are composed of stratified Location 6 (Fig. 2).

FIG. 4. A view of Location 6 on the right bank of Estero Tongoy. Holocene estua­ rine deposits which haveshells ngrowth position, are ex­ posad just above the stream level. Y. ata and R. Paskoff 29

TABLE 1. RADlOCARBON AGES FOR SHELLS FROM HOLOCENE SEDIMENTS OF THE BAHIA TONGOY ANO BAHIA HERRA­ OURAAREAS

Location' Sample number2 14C age Species Environment altitude (yr B.P.)' (m a.s.I.)5

1 GaK-15002 3,220± 80 Protothaca sp. beach 5.1 2 GaK-14998 2,370±100 Protothaca sp. beach 3.5 3-1 GaK-14999 3,140± 90 Mactra sp. beach 1.1 3-2 GaK-15003 2,510±100 Mactra sp. beach 3.2 4 GaK-15000 2,330± 80 Mactra sp. beach 4.1 5 GaK-15001 910± 90 Mactra sp. beach 2.7 6 GaK-14997 6,380±120 Protothaca sp.' estuary 1.8 A 2,400±240 shell beach ca. 2.0 7 GaK-14996 4,520±110 Argopecten beach 2.0 purpura tus B 3,700±120 shell beach 2.0

Notes: 'locations 1·7 correspond to this study. locations A and B are alter Paskoff (1973). 2 Dated at Gakushuin Univers~y. Japan, except lor locations A and B .... hich .... ere dated at the Inst~ute 01 Geology, Universily 01 Bordeaux. 3 Using libby' s ha~ lije (5,570 yr) .... ~h 2 standard devlations. • Collected in growth positlon. SHeights were measured by handlevel and hand·heJd barometer. referring to the present sea level and corrected to the artitude aboye mean sea leve l. using tide table. Errors 01 height measurement are estimated to be ca. ±O.5 m, up to ca. ± 1 m at Location 6 which is lar Irom the coast.

BAHIA HERRADURA Barnacles are attached to some rounded gravels. Shells from this site at about 2.0 m a.s.1. are dated At Location 7, beach deposits including at 4,520±11 Oyr B.P. This age is much olderthan the abundant shell fragments are exposed in the right 3,700±120 yr B.P. previously reported by Paskoff bank of Quebrada Los Chines (Figs. 5, 6; Table 1). (1973).

HOLOCENE SEA LEVEL HISTORV ANO COASTAL EVOLUTION

THE HOLOCENE MAXIMUM TRANSGRESSION. Bahía Tongoy area described in section 2. BAHIA TONGOY However, it is difficult to determine the position of sea level at the time of the culmination of the One sample from the estuarine silt at Location transgression and the amount of vertical movement 6 in Bahía Tongoy providesthe oldest radiocarbon during the Holocene. This is because the upper limit of age (6,380±120yrB.P.)forHolocenetransgressive estuarine deposits is not determined exactly at Location deposits in north-central Chile. The Holocene 6, and no Holocene sea level curve is available from transgression reached at least about 3 km inland north-central Chile at present. from the present coast, resulting in the formation of The height of sea level at the culmination of the a narrow drownedvalley along Estero Tongoy, and transgression estimated from Estero Tongoy sample is probably culminated slightly later than the above­ significantly lower than the height of ca. 30 m in Isla mentioned ages. It is tentatively assumed that the Mocha, central Chile (Kaizuka et al., 1973) and the sea level reached a slightly higher position than the height of more than 20 m in southern Chile (Hervé, F. sample horizon, probably about 3 m a.s.1. at maxi­ and Ota, Y.; in preparation). The uplift rate of the Bahía mum ca. 6,000yr B.P., considering the elevation of Tongoy area is evidently low. the abrasion platform and beach ridges in the 30 HOLOCENE OEPOSITS ON THE COAST OF NORTH-GENTRAL CHILE: RAOIOCARBON AGES ANO IMPUCATIONS •••

N Bahtá Coquimbo ! • 14 e sampling site Holocene k>w marine terraoe

Ephemeral stream (quebrada)

Abandoned Holocene sea clitt

Pleistocene marine terrace

@-

O Ikm b====:d FiG. 5. Morphology and sampling sites in the Bahía Coquimbo area.

HOLOCENE UPUFT inner margin of the Holocene terrace, and also BAHIA TONGOY ANO BAHIA HERRADURA because the abras ion platform at Location 3 is underlain by soft Tertiary rocks. However, a slight emergence during the Holocene can be postulated. In addition to the emerged estuarine deposit at No evidence is available for Holocene activity of the Estero Tongoy, there is evidence which indicates a Puerto Aldea Fault. slow uplift in the Bahía Tongoy and Bahía Herradura Around Bahía Herradura, it is not possible to give areas. a precise altitude of sea level at the time of the On the eastern coast of the Altos de Talinay, two culmination of the Holocenetransgression. However, radiocarbon ages indicate that beach deposits, 1i11ing it cannot be higher than a few meters according to the a notch or restin;¡ on an abrasion platform cutting the topographic profile (Fig. 6). hard Paleozoic-Mesozoic rocks, are Holocene in age The amount of Holocene uplift in the study area is (3,220±80 yr B.P. at ca. 5 m a.s.1. at Location 1, and very small, judged by the above-mentioned data; <5 2,370±100 yr B.P. at ca. 3.5 m at Location 2). No m on the eastern coast of the Altos de Talinay positive evidence is available for establishing the sea highlands and <3 m in the other areas. This estimate level position 01 the culmination 01 the Holocene sea is consistent with the low uplift rate since the last level rise 01 these localities. However, the erosional interglacial (ca. 0.1-0.2 mlkyr, Hsu et al., 1989). No topography beneath the Holocene deposits probably evidence of an epi so die uplift during the Holocene is represents that position, which was not higher than available from any locality in the study area.The ca. 5 m. Radiocarbon ages youngerthan ca. 6,OOO'yr Bahía Tongoy and Bahía Herradura areas have B. P. at these locations are probably due to depositlon remained relatively stable during the late Quaternary. 01 younger deposits which reoccupied a previous sea This may indicate a different nature of the subduction level position, as suggested by Paskoff (1970). zone on this area along the Perú-Chile Trench from At Location 3, the abras ion platform beneath the that of the central and southern part of Chile, where Holocene deposits is 1.1 m a.s.I., slightly lower than a typical Chilean-type subduction (e.g. Uyeda, 1982) Locations 1 and 2. with a shallow dipping Benioff Zone associated with It is difficult to compare this altitude directly with strong compression and resultant high rate of uplift those at Locations 1 and 2, beca use Location 3 is is known. situated in the e:nbayment and is seaward 1rom the Y. Ola and R. Paskoff 31

BEACH RIDGE DEVELOPMENT. ridge was formed after the emergenee of estuarine BAHIA TONGOY deposits.

A series of beaeh ridges around Sahía Tongoy HOLOCENE EMERGEN CE ANO HUMAN OCCUPATlON. has resulted from both a fall of relative sea level and BAHIA HERRADURA from progradation of the shoreline due to high sediment diseharge by several ephemeral streams. Prior to radioearbon datings :he low marine Sueeessive shoreline positions are marked by indi­ terraee below the last interglaeial terraee on Sahía vidual beaeh ridges. The radiocarbon age of the Herradura (Herradura II terraee) was thought to be of youngest fossil beaeh ridge is 91 0±90 yr S.P. Holoeene age on areheologieal grounds (Paskoff, Similar radioearbon ages from Loeation 3 and 1970). Near Location 7 (Fig. 5), remnants of the Loeation 4 (2,51 0±1 00 yr S.P. at Loeation 3 and preeeramie 'Anzuelo de Conehaeulture' (shell middens 2,330±80 yr S.P. at Loeation 4) eonfirm that both at ca. 6,000 yr S.P.) are never fOLnd on this low sites are loeated on the same beaeh ridge. terraee, but only at the top of the abandoned sea eliff The shoreline prograded about 210m between whieh forms the outer edge of Herradura 11 terraee. ca. 2,300-2,500 yr S.P. and ca. 900yr S.P., giving an Artifaets and shell middens of the ycunger 'El Molle average rate of 0.14 m/yr. The ealeulated eulture' (about 2,000 yr S.P.) are oeeasionally loeated progradation rate between ca. 900 yr S.P. and the on the lowterraee (Fig. 6, after Monta,é,1964). Thus, present (distanee is ca. 120 m) is similar: 0.13 mlyr. this low terraee was still submerged at ca. 6,000 yr If the same rate applies to the earlier beaeh ridges, S.P., but had emerged by ca. 2,000 y' S. P. when the the oldest beaeh ridge, whieh is about 420 m inland El Molle people settled on it. Two radioearbon ages fromthe ca. 2,300-2,500 yr S.P. beaeh ridge, is ca. from Loeation 7 (4,520±11 O yr S.P.) and Loeation S 5,400 yr S.P. This estimated age is slightly younger (same loeality as Loeation 7, but different sample: than the inferred age of the eulmination of Holoeene 3,700±120 yr S.P.) eonfirm this estimate, although we transgression (ca. 6,000 yr S.P). The age seems to eannot interpret an age differenee of these two samples be reasonable, eonsidering that the oldest beaeh at presento

"""O -g~ § ~ a "E.Q b .DI"' o _Holocene Terrace _..: _ Last Interglacial Terrace (Herradura 11) - m Preceramic shell middens t- 20 Pacific Storm Dune I ndian shell middens (Anzuelo de Concha) Ocean beach (El Molle) ca. 6,O~0 yr B.P. ca. 2,000 yr B.P. ~, r 10 1 1 1 ...... :.~.--____-'7-/ . . . Herradura 11 deposits ----.;<. ~----~ O O 100 \. 200 300 m Holocene beach deposits Shell m,ddens (4.520 ± 110 yr B.P .(7) , 3.700 + 120 yr B.P. (B)

FiG.6. Prolile 01 matine terraces near Location 7, Bahía Herradura, modilied Irom Paskoff (1970). Topographical prolile is alter Montané (1964). 32 HOLOCENE OEPOSITS ON THE COAST OF NORTH-CENTRAL CHILE: RAOIOCARBON AG ES ANO IMPLlCATlONS •••

SUMMARV ANO CONCLUSIONS

Postglacial -elative sea level rise in the study The amount of Holocene tectonic uplift appears area probably culminated at ca. 6,000 yr B.P. at a to be very small and has been consistent with a low slightly higher position (Iess than ca. 5 m on the rocky uplift rate since the last Interglacial. The Holocene coast of the Altos de Talinay and less than 3 m in the progradation rate of the beach ridge system in the other areas) than the present sea level. Bahía Tongoy area is ca. 0.14-0.13 mlyr.

ACKNOWLEOGEMENTS

This study was financially supported in 1989 by Geología y Minería de Chile} for their support during the InternationaJ Scientific Research Program of the the research. This paper was improved through Ministry of Education, Science and Culture of Japan critical comments by N.P. Psuty, C.E. Stearns, L. (project number 01041 040). We thank Takahiro Recart, G. Pringle, and A.A. Nelson (United States Miyauchi and Motoharu Koba, and other members of Geological Survey). We thank Yoshiaki Matsushima the project, for their collaboration in the field and for for identification of shell species and Junko Miyake their helpful discussion. We are also grateful to J. for drawing the figures. This is a contribution to IGCP Corvalán and J.A. Naranjo (Servicio Nacional de Project 274.

REFERENCES

Brüggen, J. 1929 Texto de Geología. Geología General. terrace reoccupation inferred from mollusk amino­ Primera Parte. Imprenta 'El Globo: 450 p. . stratigraphy. Coquimbo Bay area, Chile. Quaternary Darwin, C. 1846. Geologieal observations on South America. Research, Vol. 38, p. 246-259. Smith E/der 8?ld Co. , 279 p. London. Montané, J. 1964. Fechamiento tentativo de las ocupacio­ Domeyko, 1. 1848. Mémoire sur le terrain tertiaire et les nes humanas en dos terrazas a lo largo del litoral lignes d'ancien niveau de l'Océan du Sud, aux environs Chileno. Arqueología de Chile Central y áreas vecinas. de Coquimbo (Chili). Annales des Mines, Vol. 14, p. Congreso Internacional de Arqueología Chilena, No. 3, 153-162. p. 109-124. Santiago. Gana, P. 1991. Magmatismo bimodal del Triásico Superior­ Paskoff, R. 1970. Le Chili semi-aride. 420 p. Biseaye Jurásico Inferior, en la Cordillera de la Costa, Provincia Freres. Bordeaux. de Elqui y Umarr, Chile. Revista Geológica de Chile, Paskoff, R. 1973. Radiocarbon dating of marine shells Vol. 18, No. 1 p. 55-67. taken from the north and central coast of Chile. Inter­ Herm, D. 1969. Marines Pliozan und Pleistozan in Nord-und national Union forQuaternary Research.ln International Mittel-Chile unter besonderer Berücksichtigung der Geological Congress, No. 9, Proceedings, p.281-282. Entwicklung der Mollusken-Faunen. Zitteliana, Vol. 2, Christchurch, New Zealand. p. 1-159. Radtke, U. 1989. Marine Terrassen und Korrallenriffe. Das Hsu, J.T.; Leonard, E.M. ; Wehmiller, J.F. 1989. Aminostrati­ Problem der quartaren Meeresspiegelschwankungen graphy of Pe;uvian and Chilean Quaternary marine erlautert anFallstudien aus Chile, Argentinien und Bar­ terraces. Quaternary Science Review, Vol. 8, p. 255- bados. DOsseldolferGeographische Schriften, Vol. 27, 262. p.1-246. Kaizuka, S.; Matsuda,T.; Nogami, M.; Yonekura, N. 1973. Thomas, H. 1967. Geología de la Hoja , provincia de Quaternary lectonic and recent seismic crustal Coquimbo. Instituto de Investigaciones Geológicas, movements in the Arauco Peninsula and its environs, Boletín, No. 23, p. 1-58. central Chile. T~o Metropolitan University, Geographieal Uyeda, S. 1982. Subduction zones: an introduction to com­ Reports, Vol. 8, p.1-49. parative subductology. Tectonophysics, Vol. 81, p. 133- Leonard, E.M.; Wehmiller, J.F. 1992. Low uplift rates and 159.

Manuscript received: Ja,uary 23, 1992; accepted: March 30, 1992.