Indonesian Journal on Geoscience Vol. 5 No. 3 December 2018: 221-234

INDONESIAN JOURNAL ON GEOSCIENCE Geological Agency Ministry of Energy and Mineral Resources

Journal homepage: hp://ijog.geologi.esdm.go.id ISSN 2355-9314, e-ISSN 2355-9306

The Magnetostratigraphy and the Age of So’a Basin Fossil-Bearing Sequence, Flores, Indonesia

Dida Yurnaldi, Ruly Setiawan, and Emma Yan Patriani

Centre for Geological Survey, Geological Agency Jln. Diponegoro no 57 Bandung - 40122

Corresponding author: [email protected] Manuscript received: November, 07, 2017; revised: January, 08, 2018; approved: June, 05, 2018; available online: November, 19, 2018

Abstract - Three fossil-bearing intervals have been recognized in the Pleistocene So’a Basin, with the upper one holding important evidence of hominin fossils. The sequence also contains numerous in situ stone artifacts and fossils of other vertebrate taxa. Therefore, multiple dating techniques are crucial to secure the age of the fossil and artifact- bearing layers, especially the one with the hominin remains. This paper deals with the palaeomagnetic dating of the So’a Basin sequence to assist other dating methods that have been applied, and to refine the chronostratigraphy of the area. Palaeomagnetic sampling was conducted in four sections along a west to east transect. Four magnetozones can be recognized, consisting of two reverse and two normal polarity zones. By using the available radiometric ages as a guide and comparing the So’a Basin magnetostratigraphy with the Standard Geomagnetic Polarity Time Scale (GPTS), it became clear that both reverse magnetozones are part of the Matuyama Chron, while the normal magnetozones are the Jaramillo subchron and the Brunhes Chron. These palaeomagnetic dating results support the available radiometric dates and refine the age of the fossil-bearing deposits of the So’a Basin. Keywords: Palaeomagnetic, Homo floresiensis,hominin, dating, polarity, rock magnetics © IJOG - 2018. All right reserved

How to cite this article: Yurnaldi, D., Setiawan, R., and Patriyani, E.Y., 2018. The Magnetostratigraphy and the Age of So’a Basin Fossil-Bearing Sequence, Flores, Indonesia. Indonesian Journal on Geoscience, 5 (3), p.221-234. DOI: 10.17014/ijog.5.3.221-234

Introduction fossils, which are estimated to be c. 700,000 years old (Brumm et al., 2016). In order to refine the Recently, a mandible fragment and six hominin chronostratigraphy of the So’a Basin sequence teeth were recovered from one of the fossil-bearing and to allow basin-wide cross correlation, palaeo- strata in the So’a Basin, Flores, Indonesia. These magnetic sampling and analysis were carried out specimens are relatively similar to the holo-type on four stratigraphic sections. The combination of Homo floresiensisIJOG from Liang Bua, except the of palaeomagnetic dating and other radiometric size which is slightly smaller and the mandibular dating techniques has widely been used and first molar that has more primitive characteristics proven to be successful for dating archaeo- and than the Liang Bua specimens (van den Bergh et palaeontological sites around the world, for al., 2016a; see also Brown and Maeda, 2009). example: the Koobi Fora site (Lepre and Kent, Multiple dating methods were applied to 2010), Sterkfontein Cave (Herries et al., 2010), determine the age of these important hominin and the Melka Kunture site (Tamrat et al., 2014)

Accredited by: - LIPI, valid August 2016 - August 2021 Indexed by: Scopus - RISTEKDIKTI, valid May 2016 - May 2021 221 Indonesian Journal on Geoscience, Vol. 5 No. 3 December 2018: 221-234 in Africa, the Ceprano- Fontana Ranuccio site up to 5° to the south (Hartono, 1961; O’Sullivan (Muttoni et al., 2009) and the Poiana Cireşului et al., 2001; Suminto et al., 2009). The Ola Bula site in Europe (Zeeden et al., 2009), and the Ni- Formation unconformably overlies the Ola Kile hewan Basin (Zuo et al., 2011; Ao et al., 2013) Formation and fills much of the basin with an up in Asia. In Indonesia, palaeomagnetic dating was to 100 m thick sequence. It consists of relatively used to date the Sangiran site in Java (Hyodo et undisturbed and horizontally bedded volcanic al., 1993; 2011) and The Talipu site in Sulawesi and sedimentary deposits, which started to in- (van den Bergh et al., 2016b). Magnetostratigra- fill the basin during the Late Pliocene or Early phy was applied to better constrain of the numeric Pleistocene (Murouka et al., 2002; Suminto et ages and to complement other dating methods, al., 2009). The Ola Bula Formation is divided into including Fission-Track, Argon-Argon, Electron three lithological members, from old to young: Spin Resonance, and optical dating techniques. the Tuff Member, the Sandstone Member, and the Limestone Member (Hartono, 1961; van den Geological, Palaeontological, and Geochrono- Bergh, 1999; and Suminto et al., 2009 - Figure 2). logical Context of The So’a Basin Three main fossil-bearing intervals have been The So’a Basin is located in the central recognized within the Ola Bula sequence. The part of Flores, almost entirely surrounded by oldest fossiliferous layers are developed within mountains and active volcanoes (Figure 1). It the Tuff Member, which has its most complete is drained by the Ae Sissa River, which empties development near the Tangi Talo site in the the basin through a deeply incised gorge to the central and deepest part of the basin. The fossil northeast, and forms a delta plain on the north fauna from this level consists of a dwarfed extinct coast (O’Sullivan et al., 2001; Suminto et al., elephant (Stegodon sondaari), a giant tortoise, 2009). The basin basement consists of the Ola and Varanuskomodoensis, but the site has not Kile Formation comprising massive and resis- yielded any stone artifact (Sondaar et al., 1994; tant andesitic breccias interbedded with minor Aziz et al., 2009). Two fossil intervals higher up tuffaceous siltstones, sandstones, and lava flows. in the sequence lie within the Sandstone Member It had been tilted by tectonic activity which dips and can be found at the Mata Menge site. Both

120o53’51” E 121o22’36” E So’a Basin < 400 m 32’58” S 32’58” S o o

8 8 Keli Esu s a is S e Welas Caldera 2 A 400 - 1000 m 4 Kelilambo

So’a < < < < 1000 - 2000 m

1 3 > 2000 m Bajawa Boawae Kelindora

Ambulobo Active volcano

Inactive volcano

Inirie Town

Sawu Sea < Excavation/Trench 1 = Mata Menge 3 = Tangi Talo 0 km 10 20 2 = Wolo Sege 4 = Gero 58’56” S 58’56” S IJOG N o o 8 8 120o53’51” E 121o22’36” E INDONESIA Flores

Komodo 8 River Liang Bua Ruteng Bajawa Ende Maumere

19 Soa Basin Boundary of 0 km 100 So’a Basin sequence 120 122 124 10

Figure 1. The four site locations of the So’a Basin that were sampled for this study (map modified after van den Bergh, 2010).

222 The Magnetostratigraphy and the Age of So’a Basin Fossil-Bearing Sequence, Flores, Indonesia (D. Yurnaldi et al.)

Stratigra- Vertebrate fauna Age Lithology phic Unit & stone tools ime (Ma) T Recent Holocene volcanics & Tuffaceous sands and silts; lava flows alluvium gravels, sands, and silts. Hiatus 0.65 Thin-bedded micritic ftreshwater limestone and tuffaceous silts; minor tuffaceous sands. Member Limestone

Tuffaceous sands and silts; debris flows; sheet flow deposit; conglomerates; Stegodon florensis minor white tuffs; Varanus komodoensis increasingly fine-grained in upper part. Hooijeromys nusatenggara

Ola Bula Fm Small crocodile Sandstone Member 0.80 Large crocodile Stone artefacts 0.88 Early to Middle Pleistocene

White and pink pumice tuffs and Stegodon sondari tuffaceous silts and sands; Varanus komodoensis Giant tortoise f Member minor fluvial channels.

uf Small crocodile

0.94 T Hiatus 1.8 Andesitic breccia + minor tuffaceous

Fm. silts and sands. Ola Kile Late Pliocene

Figure 2. Generalized lithostratigraphy of the So’a Basin (after Suminto et al., 1999). levels contain the remains of a larger Stegodon recognized throughout the basin, including at the (Stegodon florensis), a giant rat (Hooijeromys Tangi Talo site, where it occurs at ~20 m above nusatenggara), Komodo dragon (Varanus komo- the Tangi Talo fossil bearing layer. Therefore, the doensis), unidentified crocodilian remains, and Tangi Talo fossil bearing layer could be much stone artifacts (Sondaar et al., 1994; van den Bergh older than ~0.9 Ma. 1999; van den Bergh et al., 2009; Brumm et al., In 2016, the Mata Menge two fossil-bearing 2016). In addition, the upper level has yielded the layers have been dated as well. Both fossil intervals remains of a small-sized hominin, and thought to occur above the WSI, indicating that they must be be the ancestral form of Homo floresiensis (van younger than ~1.02 Ma. Ages of the lower fossil den Bergh et al., 2016). interval lie between ~0.88 Ma and ~0.8 Ma, based Dates have come from multiple methods. Pre- on the fission-track dating of two sampled levels, viously, the Tangi TaloIJOG fossil assemblage was es- one directly below and one associated with the timated to be 0.9 Ma old based on palaeomagnetic fossil bearing layer (O’Sullivan et al., 2001). The and fission track dating (Sondaar et al., 1994; upper fossil-bearing interval, containing hominin Morwood et al., 1998). More recently, 40Ar/39Ar remains, has an age estimate based on its strati- dating by Brumm et al. (2010) of a widespread graphic position, because no direct dating associ- ignimbrite marker bed yielded an age of ~1.02 ated with this sedimentary layer. The hominin layer ma. This Wolo Sege ignimbrite (WSI) can be occurs 12.5 m above the level that yielded the F-T

223 Indonesian Journal on Geoscience, Vol. 5 No. 3 December 2018: 221-234 date of ~0.8 Ma and 13.5 m below a tephra layer a magnetic stirrer for SEM-EDX purposes. SEM- dated at ~0.65 Ma by means of 40Ar/39Ar dating, EDX analysis was performed using a JEOL-JSM and another primary tephra layer occurring at the 636OLA at the Geological Laboratory of the top of the Mata Menge section has been 40Ar/39Ar Centre of Geological Survey of Indonesia. dated at ~0.51 Ma (Brumm et al., 2016). In addi- Magnetic mineral domains were determined tion, there are also U-series ages of a hominin tooth using Dunlop plot (Dunlop et al., 2002). The Char- root fragment and combined U-series and electron acteristic Remanent Magnetization (ChRM) was spin resonance (ESR) dates of two S. florensis analyzed using Zijderveld diagrams (Zijderveld, molars that were found in situ in the same layer 1967) and Principal Component Analysis (PCA; as the hominin fossils. The U-series dating yield Kirschvink, 1980) with Puffin Plot software (Lur- a minimum age of at least 0.55 Ma, whereas the cock and Wilson, 2012). The direction of ChRMs combined U-series/ESR dating indicates minimum was determined from orthogonal plots in at least ages around 0.36 Ma and 0.69 Ma, respectively four to five successive measurement steps with (Brumm et al., 2016). Based on those dates above, the maximum angular deviation (MAD) setting Brumm et al. (2016) concluded that the hominin at <15º. The group mean direction was analyzed fossil layer is estimated to be ~0.7 Ma old. using Fisher statistics (Fisher, 1953) in IAPD 2000 software. For reversal tests, the method of McFad- den & McElhinney (1990) and Butler (1998) was Materials and Methods followed using PMAGTOOL v 4.2 by Hounslow (2006). One issue that must be taken into account At least six to eight hand specimens were taken is that low latitude areas have a relatively weak from each of twenty-nine clay and/or silt layers geomagnetic force as compared to middle and from the baulks of four step trenches in the So’a high latitudes, especially with regard to the verti- Basin. Sampling was done by carving fresh sedi- cal component (Shimizu et al., 1985). Kobayashi ment surfaces to fit into 8 cm3 acrylic cubes. Ori- et al. (1971) have found that near the equator, the entation of the samples followed the procedure of inclination (vertical component) has large fluctua- Tauxe (2003), with a little modification where the tions, so that it is not suitable to use it for tracing laboratory arrow is the same as the specimen strike geomagnetic reversals. On the other hand, it is also arrow. Additional non-oriented samples were also noted that the declination (horizontal component) taken from each layer for rock magnetic analysis. shows sharp shifts at the reversals. Natural Remanent Magnetization (NRM) was measured using 2G a three-axis cryogenic mag- netometer from William S. Goore, Inc. (WSGI). Result and Disscussion Specimens were stepwise demagnetized using an alternating field (AF) with increment intervals of The Carrier of the Magnetization 0.5 - 2.5 mT up to 100 mT. A duplicate thermal A Dunlop plot (see Figure 3) shows that demagnetization (TD) was also used on each the magnetization carriers consist of a mix of ‘twin’ specimen. However, most of the TD decay single- and multidomain (SD-MD) grains, with pattern was found to be erratic and hard to inter- a dominance of MD grains of up to 60 – 90 %. pret. Rock magnetic parameters were measured Some of the grain plots show a shift to the SD using a Princeton MicromagIJOG 3900 series Vibrat- and superparamagnetic (SP) mix curve, which ing Sample Magnetometer (VSM). All analyses indicates that SP components also occur in some were performed in a magnetic field-free room at samples. During demagnetization most of the the Palaeomagnetic Laboratory of the Australian intensities are reduced to 10 % from the total in- National University (ANU) Canberra. tensity prior to demagnetization, at relatively low Magnetic minerals were extracted from 10 g AF peaks of 30-50 mT (Figure 4). Above 50 mT, samples dispersed in 80 ml distilled water using most specimens are demagnetized completely.

224 The Magnetostratigraphy and the Age of So’a Basin Fossil-Bearing Sequence, Flores, Indonesia (D. Yurnaldi et al.)

0.5 SD Specimen Mrs/Ms 10% Synthetic Magnetite

0.025 m SP Superparamagnetic 20% 0.4 SD Singledomain 20% MD Multidomain 30% Mrs Saturation Remanent Magnetization 1 m 30% Ms Saturation Magnetization SP + SD mixing curve Hcr Coercivity of Remanent 0.3 40% Hc Coercive Force 0.010 m 0.12 m 0.19 m SP

60% 0.2 60%

70%

3

0.1 1 2 SD + MD 90% mixing curve 90%

MD

0 1 2 3 4 5 Hcr/Hc

Figure 3. Hysteresis data for samples from So’a Basin, compared to theoretical model curves by Dunlop (2002). Percentages along mixing curve are the proportion of grain.

6 2 a 5 b 1.5 4 MTNT 400-1 MTXII 505BR-1 3 1

2 0.5

Magnetization (A/m) x 10E-5 1 Magnetization (A/m) x 10E-5

0 0 20 40 60 80 100 20 40 60 80 3-axis AF strength (mT) 1 3-axis AF strength (mT) 3 0.8 c d 2.5

0.6 2 WLSG 95ALT-3 WLSG 401-4 0.4 1.5

1 0.2 Magnetization (A/m) x 10E-4

Magnetization (A/m) x 10E-5 0.5 0 20 40 60 80 0 20 40 60 80 3-axis AF strength (mT) 3-axis AF strength (mT) 10 2

8 e f 1.5

6 TT FAEX-1 1 GR 3 - 4 4 IJOG 0.5 2 Magnetization (A/m) x 10E-4 Magnetization (A/m) x 10E-6

0 0 10 20 30 40 50 10 20 30 40 50 3-axis AF strength (mT) 3-axis AF strength (mT)

Figure 4. Typical of progressive demagnetization curves of samples from Mata Menge (a, b), Wolo Sege (c, d), Tangi Talo (e), and Gero (f).

225 Indonesian Journal on Geoscience, Vol. 5 No. 3 December 2018: 221-234

This indicates that the carriers of the magnetiza- plain why the thermal demagnetization (TD) did tion are dominated by low coercivity minerals, not work, because it is not suitable for the magneti- such as magnetite and/or titanomagnetite. The zation with magnetite carrier (McElhinney, 2000). typical tall and thin hysteresis loops also reflect the occurrence of magnetite and/or titanomagne- The Characteristics Remanent Magnetization tite (Figure 5), although such profiles can also rise (ChRM) from the combination of magnetic minerals with Most magnetization vectors obtained from all contrasting coercivities (Roberts et al., 1995). specimens show two to three separated compo- However, SEM-EDX was able to confirm that nents of NRM on the orthogonal planes (Figure the carriers of the magnetization likely represent 7). This means that these specimens were af- MD titanomagnetite (Figure 6). This would ex- fected by a secondary magnetization. However,

+200E-3 +20E-3 m (emu) m (emu) 11/29/2012 17:27 02/23/2012 11:16 HC 44.82 Oe HC 67.93 Oe Mr 5.645 memu Mr 1.086 memu Ms 114.6 memu Ms 10.58 memu initial slope N/A Initial slope N/A Squareness ratio 0.0492 Squareness ratio 0.1026 S* 0.0310 S* 0.0409 Slope corr. -2.244 μemu/Oe Slope corr. -1.172 μemu/Oe Field increment 40.0 Oe Field increment 40.0 Oe 0 0 Avg. time 200.0 ms Avg. time 200.0 ms Orientation 0.0 deg Orientation 0.0 deg

MTXII-18 NT-230cm

H (Oe) H (Oe) -200E-3 -20E-3 -5E+3 0 +5E+3 -5E+3 0 +5E+3 Description: [Not assigned] Description: [Not assigned] File: mtxii-18.hyscorr File: NT230cm.hyscorr

+50E-3 +20E-3 m (emu) m (emu) 02/21/2012 10:12 02/23/2012 14:16 HC 33.65 Oe HC 84.71 Oe Mr 1.341 memu Mr 1.417 memu Ms 31.44 memu Ms 12.45 memu Initial slope N/A Initial slope N/A Squareness ratio 0.0426 Squareness ratio 0.1138 S* 0.0000 S* 0.0348 Slope corr. -595.0 μemu/Oe Slope corr. -474.4 μemu/Oe Field increment 40.0 Oe Field increment 40.0 Oe 0 0 Avg. time 200.0 ms Avg. time 200.0 ms Orientation 0.0 deg Orientation 0.0 deg

WLSG-95 WLSG-15cmB-IG

H (Oe) H (Oe) -50E-3 -20E-3 -5E+3 0 +5E+3 -5E+3 0 +5E+3 Description: [Not assigned] Description: [Not assigned] File: WLSG95.hyscorr File: WG15cmB-IG.hyscorr

+20E-3 +20E-3 m (emu) m (emu)

02/23/2012 14:40 02/23/2012 15:43 HC 92.99 Oe HC 72.86 Oe Mr 1.918 memu Mr 1.543 memu Ms 15.28 memu Ms 13.51 memu Initial slope N/A Initial slope N/A Squareness ratio 0.1255 Squareness ratio 0.1142 S* 0.0423 S* 0.0461 Slope corr. -640.8 μemu/Oe Slope corr. -1.131 μemu/Oe Field increment 40.0 Oe Field increment 40.0 Oe 0 0 Avg. time 200.0 ms Avg. time 200.0 ms Orientation 0.0 deg Orientation 0.0 deg IJOGTT-250cm GR-3mWL

H (Oe) H (Oe) -20E-3 -20E-3 -5E+3 0 +5E+3 -5E+3 0 +5E+3 Description: [Not assigned] Description: [Not assigned] File: TT250cm.hyscorr File: GR3mWL.hyscorr

Figure 5. Typical hysteresis loop curves from Mata Menge (top), Wolo Sege (middle), Tangi Talo (lower left), and Gero (lower right). The wasp-waisted curves represent typical hysteresis loops for mixed magnetic mineral assemblages domi- nated by magnetite.

226 The Magnetostratigraphy and the Age of So’a Basin Fossil-Bearing Sequence, Flores, Indonesia (D. Yurnaldi et al.)

11000

MTXII-505 BR a 10000

9000

8000

7000

6000 Carts 5000

4000

3000

2000

1000

0 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 ke V Element (keV) mass% Error% At% Compound mass% Cation K O 28.22 Ti K 4.508 20.08 5.06 31.17 TiO2 33.49 5.70 29.1484 Fe K 6.398 51.70 7.47 68.83 FeO 66.51 12.59 70.8516 Total 100.00 100.00 100.00 18.30

WLSG 35cm b 8000 7200

6400

5600

4800 Carts 4000

3200

2400

1600

800

0 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 ke V Element (keV) mass% Error% At% Compound mass% Cation K O 27.32 Si K 1.739 3.81 3.58 9.73 SiO2 8.15 1.91 3.4118 Ti K 4.508 8.50 4.41 12.73 TiO2 14.18 2.50 12.3949 Fe K 6.398 60.37 6.42 77.54 FeO 77.67 15.20 84.1933 Total 100.00 100.00 100.00 19.60

2000 TT 2.5 m c 1800 1600

1400

1200

Carts 1000

800

600

400

200

0 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 ke V

Element (keV) mass% Error% At% Compound mass% Cation K O 30.11 AIK 1.486 2.05 0.47 2.76 AI203 3.87 0.97 1.5525 Si K 1.739 4.64 0.41 12.04 SiO2 9.93 2.11 4.5219 Ti K 4.508 12.82 0.52 19.49 TiO2 21.38 3.41 19.7141 Fe K 6.398 50.39 0.75 65.71 FeO 64.82 11.51 74.2115 Total 100.00 100.00 100.00 18.00

Figure 6. SEM images of representative samples from Mata Menge (a), Wolo Sege (b), and Tangi Talo (c). The euhedral crystal shapes correspond with titanomagnetite of volcanic origin. as mentinued before,IJOG the secondary magnetization or can be defined as the mean vectors that are was easily removed at an AF demagnetization stable in the intensity upon a higher AF demag- between 5 to 20 mT, while Characteristics Rema- netization. nent Magnetizations (ChRMs) could be isolated Fisher statistics for group mean directions for at peaks of 30 - 50 mT. each sampled layer are provided in Table 1 and All the ChRMs are either progressing to the equal area projection is shown in Figure 8. the origin of the orthogonal vector projections Although some points are scattered, two groups

227 Indonesian Journal on Geoscience, Vol. 5 No. 3 December 2018: 221-234

a b N,U 1 MTNT 400-1 MTXII 505BR-1

N,U 15 W,S E,N fit from 5 1 dec = 15.56 NRM

NRM inc = -33.39 20 fit from MAD dec/inc : 17.74/19.63 dec = 241.8 J 0 = 1.71e-05 inc = 33.05 10 10 MAD dec/inc : 18.04/16.47 J 0 = 5.43e-05 20 20 10 NRM

W,S E,N 15 15

NRM

5 4 S,D S,D

N,U c 8 d NRM N, U WLSG 95ALT-3 5 WLSG 401-4

20

40 W,S E,N 25

fit from 20 dec = 59.48 10 40 inc = -23.53 10 fit from MAD dec/inc : 11.51/6.03 dec = 240.15 20 inc = 25.82 20 NRM J = 9.66e-05 0 MAD dec/inc : 15.48/6.19 NRM 40 J 0 = 3.55e-05 40 W,S E,N NRM 2 8 15 S,D

2 S,D

N,U 6 e NRM f TT FAEX-1 GR 3 - 4

N,U fit from 15 10 dec = 245.97 NRM fit from inc = -21.13 dec = 27.64 20 MAD dec/inc : 12.72/15.17 inc = -19.81 MAD dec/inc : 19.17/7.42 J 0 = 8.34e-06 J 0 = 1.66e-04 40 W,S E,N 6 40 1 10 20 10

20 NRM W,S 40 10 20 40 E,N 15 10

5 S,D

NRM

6 S,D

Figure 7. Representative progressive Zijderveld demagnetization diagrams (Orthogonal planes) of samples from Mata Menge (a, b), Wolo Sege (c, d), Tangi Talo (e), and Gero (f). of opposite polarities can be easily recognized, The grouping of the points, both normal and and they all passed the reversal test as marked reverse, are better after demagnetization and by the overlap of the normal α95 circle with the the normal polarity group has slightly moved antipode of the reverseIJOG α95 circle (grey ghost away from the present magnetic direction after circle) (Butler, 1998). demagnetization. One possibility is that some The McFadden & McElhinney reversal test samples were strongly affected by the secondary is also passed, where the observed angle (Ɣo) is magnetic field. not exceeding the critical angle (Ɣc) (16.41º to N represents the number of specimens ana- 18.69º) and is classified as a “C” class according lyzed per layer, although no firm criteria exist to McFadden & McElhinney (1990). for acceptability of palaeomagnetic data . Within

228 The Magnetostratigraphy and the Age of So’a Basin Fossil-Bearing Sequence, Flores, Indonesia (D. Yurnaldi et al.)

Tabel 1. Magnetic intensities and Characteristic Remanent Magnetization Directions (ChRMs) for sampled layers in the So’a Basin. N represents the number of specimens analyzed per layer, R represents resultant vector, κ represents the dispersion of the direction population and α95 represents the confidence limit. Although no firm criteria exist for acceptability of palaeomagnetic data. Within-site κ > 30 and α95 values of <15 are generally considered to indicate reliable directions (Butler, 1998)

Intensity (A/m) Remanent Direction No Samples ID Before After Mean Mean N R κ α95 Demagnetization Demagnetization Declination Inclination 1 MTXII 90 2.27E-05 2.35E-07 50.2 -9.7 3 2.96 46.95 18.2 2 MTXII 260 1.06E-05 2.04E-07 77.5 34.7 5 4.76 16.93 19.1 3 MTXII 60 BR 6.92E-04 7.09E-06 21.2 0.5 4 3.9 31.31 16.7 4 MTXII 120 BR 7.32E-04 6.67E-06 8.9 -5.9 4 3.98 180.7 6.9 5 MTXII 265 BR 1.60E-04 4.98E-06 13.7 -4.1 5 4.53 8.52 27.8 6 MTXII 340 BR 9.82E-05 4.29E-06 31.7 -15.5 5 4.82 21.91 16.7 7 MTXII 505 BR 1.37E-05 1.06E-06 24 -15.9 5 4.86 29.59 14.3 8 MTXII 1318 1.06E-05 6.35E-07 24.4 -6.7 4 3.92 272.2 5.6 9 Mata 180 2.01E-04 6.87E-06 41.1 -13.3 4 3.95 57.06 12.3 10 Mata 220 4.21E-05 4.07E-06 22.7 -26.7 4 3.84 18.34 22 11 MTNT 230 4.76E-05 1.36E-06 59.3 -31.2 3 2.76 2.72 97.3 12 MTNT 270 6.52E-06 4.24E-07 57.5 5.3 4 3.9 31.31 16.7 13 MTNT 400 9.13E-05 3.82E-06 277.7 27.6 5 4.84 24.61 15.7 14 MTNT 500 1.08E-04 7.43E-06 206.1 23.9 4 3.74 11.76 28 15 MTNT 7C 2.29E-04 3.40E-05 29.9 -32.6 5 4.08 12.55 22.5 16 MTNT IGN 73 7.81E-04 9.59E-06 63.6 -21.5 5 4.36 6.21 33.4 17 MTNT MUD 2.70E-03 9.37E-05 210.8 32.9 4 3.96 77.8 10.5 18 WLSG 328 1.51E-04 6.53E-06 70.3 7 4 3.99 272.2 5.6 19 WLSG 401 3.38E-05 1.76E-06 232 28.3 4 3.97 111.9 8.7 20 WLSG 95 ALT 1.12E-04 6.74E-06 59.5 -23.4 4 3.99 559 3.9 21 WLSG 745 1.92E-04 1.42E-06 39.3 -22 5 4.96 105.8 7.5 22 TT 2.5 1.56E-04 1.02E-06 30.8 -31.3 4 3.89 26.33 18.2 23 TT 50a2010 3.44E-04 2.61E-06 228.2 1.8 5 4.79 19.18 17.9 24 TT 2000 2.79E-05 3.57E-06 249.9 38.8 3 2.94 31.04 22.5 25 TT 2116 6.99E-05 5.01E-06 235.7 53.7 4 3.8 14.94 24.6 26 TT FA excav 6.86E-06 4.53E-07 248.3 -6.3 4 3.71 10.28 30.1 27 TT 10 BFA 2.13E-06 5.36E-08 260.8 1.5 5 4.8 19.99 17.5 28 TT 2406 IJOG1.91E-05 5.54E-06 235 -12.1 5 4.96 110.1 7.3 29 GR 3 1.98E-04 4.19E-06 25.9 -28.1 5 4.83 23.94 16 site к >30 and α 95 values of <15 are generally The Magnetostratigraphy of the So’a Basin considered to indicate reliable directions (Butler, Figure 9 shows the sampled stratigraphic 1998). sections from the So’a Basin together with the

229 Indonesian Journal on Geoscience, Vol. 5 No. 3 December 2018: 221-234

D : 19.2 I : -24.4 D : 38.5 N : 20 I : -12.3 R : 14.32 N : 20 k : 3.34 R : 18.14 95 : 21.2 k : 10.22 95 : 10.7

D : 233.7 D : 225 I : 19.8 I :26.1 N : 10 N : 10 R : 9.04 R : 7.82 k : 9.39 k : 4.13 95 : 16.6 95 : 27.1

Before demagnetization After demagnetization

Figure 8. The equal area projection of individual mean directions obtained from the sampled levels of the So’a Basin. Open and solid squares in the equal area projections represent the upper and lower hemisphere, respectively. The black circles with centred crosses represent the mean magnetization directions (α95 circle) of normal (northern direction) and reverse (southern direction) polarities. Grey ghost circles represent the antipodes of the reverse α95 circles. A red cross represents the present-day magnetization direction. magnetic polarities. Each section is plotted ac- Furthermore, N2 at Mata Menge coincides with cording to their relative topographic elevations, the sequence that is dated at ~0.8 – 0.5 Ma, and with Gero section being the highest and Tangi therefore correspond with the normal polarity of Talo the lowest. There are also the polarity cor- the Brunhes Chron. As the event/subchron of the relations between sections and the composite N1 and N2 magnetozones is known, R1 and R2 litho- and magneto- stratigraphy as compared can be inferred to correspond with the reverse with the Standard Geomagnetic Polarity Time polarity of Matuyama Chron (Figure 9). Scale (GPTS). The fossil bearing level at Tangi Talo (F3 in A total of four polarity magnetozones have Figure 9) is situated below the lower boundary of been recognized in the So’a Basin. At the Mata the N1 magnetozone (Jaramillo) and it resulted Menge excavation section, all four magnetozones in the relative age of > 1.07 Ma. It is difficult to were recovered, which consist of two reverses (R1 determine the maximum age of this fossil layer, and R2) and two normal (N1 and N2) zones. At because the other subchrons are below the Jara- the Wolo Sege trench section, there are two nor- millo, such as the Cobb Mountain Event (1.24 mals (N1 and N2) and one reverse (R2) zone. At – 1.22 Ma) or the Gilsa Event (1.68 Ma) were the Tangi Talo trench section, two magnetozones not recognized, which is not surprising since can be recognized, which consist of one Reverse the sampling density in this interval is very low. (R1) and one Normal (N1)IJOG zone, whereas at the Denser sampling in the future may reveal these much shorter Gero trench section, only one mag- events and allow a higher resolution in the oldest netozone was found (N2). part of the So’a Basin sequence. However, this N1 is associated with the WSI. The numeri- fossil layer probably will not exceed the bottom cal 40Ar/39Ar date of 1.02 Ma ± 0.2 for the WSI boundary of the Olduvai subchron (1.95 Ma), (Brumm et al., 2010) indicates that N1 corre- because a sample from the Ola Kile Formation, sponds with the Jaramillo subchron on the GPTS. the top of which lies at ~2 m below the lowest

230 The Magnetostratigraphy and the Age of So’a Basin Fossil-Bearing Sequence, Flores, Indonesia (D. Yurnaldi et al.)

Age Subchrons, ALO (Ma) Excursion events T Specimen (m) Polarity GERO Formation Keybed Site Age (Ma) Member

Specimen ...... Polarity MENGE

Member 0 (m) ...... A 0 MTXII 91 ...... T

...... LS ...... ANGI ...... 0.6-0.5 ...... 0.12 Blake member 2 ...... Limestone T ...... MTXII 260 member . . WOLO SEGE ...... Limestone ...... 2 . . . MA ......

Polarity ...... (m) GR 3 ...... Specimen

Formation Keybed ...... Age (Ma) Member ...... 0.2 ...... 4 ...... 0 ...... 4 ...... 0.260 CR0 ...... MTXII 91 ...... 6 ...... LS ......

member ...... 0.6-0.5 2 ...... 0.319 CR1 ...... Limestone 6 ...... MTXII 260 ...... GR 7 N2 ...... MTXII 60BR ...... 8 ...... MTXII120BR ...... 4 ...... 8 ......

Specimen ...... Polarity ......

Member ...... MTXII 265BR ...... (m) ...... 10 ...... c s s pgr ...... MTXII 340BR 6 ...... 0 ...... 10 ...... 12 . . MTXII 505BR ...... 0.543 Big Lost/CR2 ...... N2 Brunshes . . MTXII 60BR ...... 8 ...... 2 ...... MTXII 120BR ...... 12 ...... MTXII 1318 ...... 0.593 CR3 ...... 0.6 ...... Sandstone Member ...... 14 ...... MTXII 265BR ...... 10 . . . . . 4 ...... 14 ...... MTXII 340BR

...... 16 ...... 12 . . MTXII 505BR 6 . . . . . 16 ...... MTXII 1318 ...... 18 ...... 0.781 ...... 0.7 ...... F3 14 . . . . . 8

...... Sandstone Member ......

Mata Menge 20 ...... N2 . . . . . 16 10 ...... N2 ...... 22 ...... 0.7 ......

F3 18 12 ...... 0.99 ...... 1.0 ......

Sandstone Member Jaramillo ...... 24 ...... Sandstone Member ...... 14 ...... 20 ...... 1.07 ...... 26 ...... 16 ...... 22 ...... MATA 180 ...... MATA 220 ...... Ola Kile Formation ...... 1.2 28 ...... 0.8 ...... F2 ...... 1.22 ...... Cobb 24 ...... 18 ...... 1.24Mountain ...... MTNT 230 ...... MTNT 270 ...... 30 ...... 26 ...... 20 ...... MTNT 400 ...... Ola Kile Formation ...... R2 ...... WLSG 328 ...... MATA 180 . . . . . MTNT 500 ...... 32 ...... MATA 220 ...... MTNT 7c ...... WLSG 401 0.8 28 . . . . . 22

F2 ...... 1.4 ...... olosege WSI . . R2 ...... N1 . . . . MTNT ign 7.3 W Ignimbrite . . 1.02 ...... WLSG 95alt ...... MTNT 230 ...... ? ...... 34 ...... MTNT 270 ...... 30 ...... 24 ...... MTNT 400 . . . . . N1 ...... R2 uff Member ...... WSI WLSG 745 ...... 36 TT 50a2010 ...... MTNT 500 ......

. . T ...... 32 ......

......

26

......

? ...... MTNT 7c ...... (m) ...... c s s p gr . . . WSI N1 ...... 1.02 MTNT ign 73 ? ......

WOLOSEGE IGNIMBRITE . . . 38 0 ...... 34 ...... 28 ...... 1.68 Gilsa ...... 2 Matuyama . . uff Member ...... 40 ...... TT 2.5 ...... 36 ...... T N1 30 . . . . . R1 ...... 1.77 ...... 4 . . . MTNT mud . . 42 . . . . . WSI . . . 38 . . 1.8 uff Member alo T 6 T 44 Olduvai

...... Remarks : ...... TT 50a2010 ......

...... angi R1 ...... 8 ......

...... 46

...... T

...... 1.95 ......

...... 2.0 ...... TT 2000 ...... Normal Polarity ...... N 10 ...... 48 ...... R Reverse Polarity ...... TT 2116 ...... 12 ...... 50 ...... 2.14 Reunion ...... TT faexcav 0.9 WSI F1 ...... Wolo Sege Ignimbrite (1.2 Ma) ...... TT 10bfa 14 ...... 52 ...... 2.19 R1 ...... 2.2 ...... TT 2406 ......

uff member ...... Tangi Talo Fossil layer . . . . . F1 ......

T 16 . . . . . 54 . . . . . Silty Clay F2 Mata Menge Fossil layer (0.8 Ma) . . . 2.33 18 . . . . . Sand 56 ...... 2.39 . . . TT 2000 2.4 ...... F3 . . . Hominid Fossil layer (0.7 Ma) ...... Conglomerate ...... 20 ...... 2.44 ‘X’ ...... TT 2116 ...... Palaeosol ...... Lake sequence (0.6-0.5 Ma) ...... LS ...... 22 ...... F1 ...... TT faexcav ...... Micritic Limestone ...... TT 10bfa Composite stratigraphy ...... 2.588 ...... 1.2 Correct age 24 ...... TT 2406 ...... Ola Kille Breccia ...... 26 Fossil Remain 0.9 Erroneous age Gauss Stone Artefact 28 GPTS (ICS, 2010)

Figure 9. Magnetostratigrahy of the So’a Basin compared with radiometric dating results and the Standard Geomagnetic Polarity Time Scale [Geomagnetic Polarity Time Scale (GPTS), 2016]. Declination and inclination are the averages of the higher coercivity stable magnetization (ChRM). Black squares represent Normal Polarities and white squares represent Reverse Polarities.

Reverse sample in the Tangi Talo section, was which means that both of them are younger than dated at 1.8 Ma based on FT dating (O’Sullivan ~0.781 Ma. et al., 2001). These palaeomagnetic dating results have The two fossiliferous intervals in the Mata implications for the previous age estimates of Menge section (F2IJOG and F3 in Figure 9) are fossil assemblages from the So’a Basin. located above the boundary between R2 and Firstly, palaeomagnetic dating of the Tangi N2. The lower (F2) and the upper (F3) are situ- Talo section combined with the 40Ar/39Ar dates ated about 2.5 m and 12 m above the R2-N2 given by Brumm et al. (2010) indicates that the boundary, respectively. This means that those Tangi Talo fossil-bearing level is older than previ- fossil layers are located above the boundary of ously thought. Instead of 0.9 Ma it now appears to Matuyama event (R2) and Brunhes event (N2) be older than 1.07 Ma. Secondly, palaeomagnetic

231 Indonesian Journal on Geoscience, Vol. 5 No. 3 December 2018: 221-234 dating has also confirmed the estimated age of J. Morwood and Adam Brumm. The authors the hominin remains, which were assumed to be would like to acknowledge Gerrit van den Bergh ~0.7 Ma old (Brumm et al., 2016). This estima- and Brad Pillans for assessment and academic tion matches the palaeomagnetic dating result discussion of this project. The authors are also of Mata Menge section, which indicates that the grateful to the former Heads of the Geological hominin layer (F3) is younger than ~0.781 Ma. Agency, Surono and Ego Syahrial, the succes- In addition, the palaeomagnetic results also show sive Directors of Centre for Geological Survey, that previous age estimates for the lower fossil- A. D. Jumarma, Agung Pribadi, and Muhammad bearing interval at Mata Menge (0.8 - 0.88 Ma; Wafid for permission to undertake this project. see O’Sullivan et al., 2001) were slightly too old, In addition, the acknowledgements are due to and should be considered as minimal ~0.781 Ma. Iwan Kurniawan, Erick Setyabudi, and Wiji for their technical support during fieldwork, David Heslop for a support in the ANU rock magnetic Conclusion laboratory, and Juju Jumbawan in the CGS SEM laboratory. Palaeomagnetic dating in the So’a Basin has recovered four polarity magnetozones. These magnetozones, from old to young, are as follows: References R1 corresponds with the Matuyama Chron below the Jaramillo subchron; N1 with the Jaramillo Ao, H., An, Z., Dekkers, M.J., Li, Y., Xiao, G., subchron; R2 with the Matuyama Chron between Zhao, H., and Qiang, X., 2013. Pleistocene the Jaramillo subchron, and the Brunhes Chron; magnetochronology of the fauna and Paleo- and N2 with the Brunhes Chron. lithic sites in Nihewan Basin: Significance These palaeomagnetic dating results support for environmental and hominin evolution in and refine the previously published radiometric North China. Quaternary Geochronology, 18, ages related to the palaeontological and/or ar- p.78-92. DOI: 10.1016/j.quageo.2013.06.004 chaeological sites of the So’a Basin. The Tangi Aziz, F., van den Bergh, G.D., Morwood, M.J., Talo fossil layer is situated below the bottom Hobbs, D.R., Kurniawan, I., Collins, J., and boundary of the Jaramillo. This provides a relative Jatmiko. 2009. Excavation at Tangi Talo, cen- minimum age of > 1.07 Ma for the Tangi Talo fos- tral Flores, Indonesia. In: Aziz, F., Morwood, sil fauna, which is older than previously thought. M.J., and Bergh, G. D. van (eds.), Palaeontol- Furthermore, two fossil intervals occurring in ogy and archaeology of the So’a Basin, central the Mata Menge section are located above the Flores, Indonesia. Indonesian Geological Matuyama-Brunhes boundary, which indicates Survey Institute, Bandung, p.1-18. that they are younger than 0.781 Ma. Brown, P. and Maeda, T., 2009. Liang Bua Homo However, it is necessary to propose additional floresiensis mandibles and mandibular teeth: studies with high resolution sampling interval in a contribution to the comparative morphology the studied area, in order to obtain a complete of a new hominin species. Journal of Human and the more appropriate stratigraphic age suc- Evolution 57, p.571-596. DOI: 10.1016/j. cession data. jhevol.2009.06.002 IJOGBrumm, A., Jensen, G.M., van den Bergh, G.D., Morwood, M.J., Kurniawan, I., Aziz, F., and Acknowledgements Storey, M., 2010. Hominins on Flores, Indo- nesia, by one million years ago. Nature, 464, This research project was financially supported p.748-752. DOI:10.1038/nature08844. by the Australian Research Council (ARC) Dis- Brumm, A., van den Bergh, G. D., Storey, M., covery grant (DP1093342) awarded to Michael. Kurniawan, I., Alloway, B. V., Setiawan,

232 The Magnetostratigraphy and the Age of So’a Basin Fossil-Bearing Sequence, Flores, Indonesia (D. Yurnaldi et al.)

R., Setiyabudi, E., Grün, R., Moore, M. W., transition constrains the age of Javanese Yurnaldi, D., Puspaningrum, M. R., Wibowo, Homo erectus in the Sangiran dome, Indo- U. P., Insani, H., Sutisna, I., Westgate, J. A., nesia. Proceedings of the National Academy Pearce, N. J. G., Duval, M., Meijer, H. J. M., of Sciences of the United States of America, Aziz, F., Sutikna, T., van der Kaars, S., Flude, 108 (49), p.19563-19568. DOI: 10.1073/ S., and Morwood, M. J., 2016. Age and con- pnas.1113106108 text of the oldest known hominin fossils from Kirschvink, J. L., 1980. The least-squares line and Flores. Nature, 534, p.249-253. DOI: 10.1038/ plane and the analysis of the palaeomagnetic nature17663 data. Geophysical Journal International, 62 Butler, R. F., 1998. Paleomagnetism: magnetic (3), p.699-718. domains to geologic terrains. Blackwell Sci- Kobayashi, K.,Kitazawa, K., Kanaya, T., and entific Publications, Boston. 238pp. SAKAI, T., 1971. Magnetic and micropaleon- Dunlop, J. D., 2002. Theory and application of the tological study of deep-sea sediments from the

Day plot (Mrs/Msversus Hcr/Hc) 2. Application west-central equatorial Pacific.Deep-Sea Re- to data for rocks, sediments, and soils. Journal search, 18. p.1045-1057. DOI: 10.1016/0011- of Geophysical Research, 107 (B3), p.1-22. 7471(71)90093-3 DOI: 10.1029/2001jb000487 Lepre, C.J. and Kent, D.V., 2010. New magne- Fisher, R.A., 1953. Dispersion on a sphere. Pro- tostratigraphy for the Olduvai Subchron in ceedings of the Royal Society of London. Series the Koobi Fora Formation, northwest Kenya, A, 217, p.295-305. with implications for early Homo. Earth and Geomagnetic Polarity Time Scale (GPTS), 2016. Planetary Science Letters, 290, p.362-374. International Commission on Stratigraphy. DOI: 10.1016/j.epsl.2009.12.032 http://www.stratigraphy.org/index.php/ics- Lurcock, P. C. and Wilson, G.S., 2012. Puffin- chart-timescale. Accessed October 29, 2018. Plot: A versatile, user-friendly program for Hartono, H. M. S., 1961. Geological investiga- paleomagnetic analysis. Geochemistry, Geo- tion at Olabula, Flores. Djawatan Geologi, physics, Geosystems, p. 13 (6) p.1-6. DOI: Bandung, 40pp. 10.1029/2012gc004098 Herries, A.I.R., Hopley, P., Adams, J., Curnoe, McFadden, P. L. and McElhinny, M.W., 1990. D., and Maslin, M., 2010. Geochronology and A new fold test for palaeomagnetic studies. palaeoenvironments of the South African early Geophysical Journal International, 279 (103), hominin bearing sites. American Journal of p.163-169. Physical Anthroplogy, 143, p.640-646. DOI: McElhinny, W. and McFadden, P.L., 2000. Paleo- 10.1002/ajpa.21389 magnetism: Continent and Ocean. Academic Hounslow, M.W., 2006. PMagTools version Press, San Diego, 386pp. 4.2- a tool for analysis of 2-D and 3-D di- Morwood, M. J., O’Sullivan, P.B., Aziz, F., and rectional data. http://dx.doi.org/10.13140/ Raza, A., 1998. Fission-track ages of stone RG.2.2.19872.58880. tools and fossils on the east Indonesian is- Hyodo, M., Watanabe, N., Sunata, W., Susanto, E. land of Flores. Nature, 392, p.173-176. DOI: E., and Wahyono, H., 1993. Magnetostratig- 10.1038/32401 raphy of hominid fossil bearing formations Murouka, H., Nasution, A., Urai, M., Takashi, M., in Sangiran and IJOGMojokerto, Java. Anthropo- Takashima, I., Simanjuntak, J., Sundhoro, H., logical Sciences, 101 (2), p.157-186. DOI: Aswin, D., Nanlohy, F., Sitorus, K., Takahashi, 10.1537/ase.101.157 H., and Kosek, T., 2002. Tectonic, volcanic, Hyodo, M., Matsu’ura, S., Kamishima, Y., and stratigraphic geology of the Bajawa Takeshita, Y., Kitaba, I., Danhara, T., Aziz, F., geothermal field, central Flores, Indonesia. Kurniawan, I., and Kumai, H., 2011. High- Bulletin of Geological Survey of Japan, 53, resolution record of the Matuyama-Brunhes p.109-138. DOI: 10.9795/bullgsj.53.109

233 Indonesian Journal on Geoscience, Vol. 5 No. 3 December 2018: 221-234

Muttoni, G., Scardia, G., Kent, D. V., Swisher, C. Tauxe, L., 2003. Paleomagnetic principles and C., and Manzi, G., 2009. Pleistocene magneto- practice. Kluwer academic publisher. New chronology of early hominin sites at Ceprano Jersey, USA, 314pp. and Fontana Ranuccio, Italy. Earth and Plan- van den Bergh, G. D., 1999. The Late Neogene etary Science Letters, 286, p.255-268. DOI: elephantoid-bearing faunas of Indonesia. 10.1016/j.epsl.2009.06.032 Scripta Geologica, 117, 388pp. O’Sullivan, P., Morwood, M.J., Hobbs, D., van den Bergh, G.D, Kurniawan, I., Morwood, Aziz, F., Suminto, and Situmorang, M. 2001. M.J., Lentfer, C.J., Suyono., Setiawan, R., Archaeological implications of the geology Aziz, F. 2009. Environmental reconstruction and chronology of the So’a Basin, Flores, of the Middle Pleistocene archaeological/pal- Indonesia.Geology, 29 (79), p.607-610. DOI: aeontological site Mata Menge, Central Flores, 10.1130/0091-7613(2001)029%3C0607:aiot Indonesia. In: Pleistocene Geology, paleontol- ga%3E2.0.co;2 ogy and archaeology of the So’a basin, Central Roberts, A. P., Cui, Y., and Verosub, K. L., 1995. Flores, Indonesia. Special Publication Centre Wasp-waisted hysteresis loops: Mineral mag- for Geological Survey, Indonesia. p.59-81 netic characteristics and discrimination of van den Bergh, G.D. 2010. 2010 So’a Basin components in mixed magnetic systems. Jour- field report. University of Wollongong. Un- nal of Geophysical Research, 100, p.17,909- published. 17,924. DOI: 10.1029/95jb00672 van den Bergh, G. D., Kaifu, Y., Kurniawan, Shimizu,Y., Mubroto, B., Siagian, H., and Un- I., Kono, R.T., Brumm, A., Setiabudi, E., tung, M., 1985. A paleomagnetic study in the Aziz, F., and Morwood, M.J., 2016a. Homo Sangiran area. Quaternary Geology of the floresiensis-like fossils from the early Middle Hominid Fossil Bearing Formations in Java. Pleistocene of Flores. Nature, 534, p.245-248. In: Watanabe, N. and Kadar, D. (eds.), Geo- DOI: 10.1038/nature17999 logical Research and Development Centre, van den Bergh, G. D., Li, B., Brumm, A., Grün, Special Publication, 4, p.275-308. R., Yurnaldi, D., Moore, M. W., Kurniawan, I., Sondaar, P. Y., van den Bergh, G.D, Mubroto, Setiawan, R., Aziz, F., Roberts, R. G., Suyono., B., Aziz, F., de Vos, J., and Batu, U.L., 1994. Storey, M., Setiabudi, E., and Morwood, M. J. Middle Pleistocene faunal turnover and colo- 2016b. Earliest occupation of Sulawesi, Indo- nization of Flores, Indonesia by Homo erectus. nesia. Nature, 529, p.208-211. DOI: 10.1038/ Human Paleontology, p.1255-1262. nature16448 Suminto, Morwood, M. J., Kurniawan, I., Aziz, Zeeden, C., Hambach, U., Steguweit, L., and F., van den Bergh, G. D., and Hobbs, D. R., Anghelinu, M., 2009. Loess stratigraphy using 2009. Geology and fossil sites of the So’a palaeomagnetism: Application to the Poiana Basin, central Flores, Indonesia. In: Aziz, F., Cireşului archaeological site, Romania. Qua- Morwood, M.J., and Bergh, G. D. van (eds), ternary International, 240, (1-2), p.100-107. Palaeontology and archaeology of the So’a DOI: 10.1016/j.quaint.2010.08.018 Basin, central Flores, Indonesia. Indonesian Zijderveld, J. D. A., 1967. “A. C. Demagnetization Geological Survey Institute, Bandung. p.19- of Rocks: Analysis of Results.” In: Collinson, 40. DOI: 10.1130/0091-7613(2001)029%3C0 D. W., Creer, K. M., and Runcorn, S. K. (eds.), 607:aiotga%3E2.0.co;2IJOGMethods in Palaeomagnetism, Elsevier, Am- Tamrat, E., Thouveny, N., Taieb, M., and Brugal, sterdam, New York, p.254-286. J.P. 2014. Magnetostratigraphic study of the Zuo, T.W., Cheng, H.J., Liu, P., Xie, F., and Deng, Melka Kunture archaeological site (Ethiopia) C.L., 2011. Magnetostratigraphic dating of the and its chronological implications. Quartenary Hougou Paleolithic site in the Nihewan Basin, International, 343. p.5-16. DOI: 10.1016/j. North China. Science in China, 54. p.1643- quaint.2013.11.030 1650. DOI: 10.1007/s11430-011-4221-2

234