Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

Traces of Ancient Sungai Batu Flow and Trading Complex from Geophysical Surveys and Archaeological Excavation Works

Journal: Songklanakarin Journal of Science and Technology ManuscriptFor ID SJST-2019-0195.R2 Review Only Manuscript Type: Review Article

Date Submitted by the 16-Sep-2019 Author:

Complete List of Authors: ROSLI, NAJMIAH; Universiti Sains , School of Physics A Rahman, Muhammad; Universiti Sains Malaysia, School of Physics SAIDIN, MOKHTAR; Universiti Sains Malaysia, Centre for Global Archaeological Research Saad, Rosli

Keyword: Sungai Batu, ancient river, trading complex, entrepot

For Proof Read only Page 1 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 TRACES OF ANCIENT SUNGAI BATU FLOW AND TRADING COMPLEX 5 6 7 FROM GEOPHYSICAL SURVEYS AND ARCHAEOLOGICAL EXCAVATION 8 9 WORKS 10 11 12 13 14 Review Article 15 16 Traces of Ancient Sungai Batu Flow and Trading Complex from Geophysical Surveys 17 18 and Archaeological Excavation Works 19 20 Najmiah Rosli1, Nazrin Rahman1, Mokhtar Saidin2 and Rosli Saad1* 21 For Review Only 22 1 23 Geophysics, School of Physics, Universiti Sains Malaysia, USM, 11800, Malaysia 24 25 2Centre for Global Archaeological Research, Universiti Sains Malaysia, USM, 11800, 26 27 Malaysia 28 29 30 * Corresponding author, Email address: [email protected] 31 32 33 34 Acknowledgment 35 36 37 The authors would like to thank Centre for Global Archaeological Research (CGAR) for 38 39 funding this research under 1002/PARKEO/910328 account. Gratitude is further 40 41 extended to USM geophysics staff and postgraduate students for their time and effort 42 43 during data acquisition. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 2 of 29

1 2 3 4 TRACES OF ANCIENT SUNGAI BATU FLOW AND TRADING COMPLEX 5 6 7 FROM GEOPHYSICAL SURVEYS AND ARCHAEOLOGICAL EXCAVATION 8 9 WORKS 10 11 12 13 14 Review Article 15 16 Traces of Ancient Sungai Batu Flow and Trading Complex from Geophysical Surveys 17 18 and Archaeological Excavation Works 19 20 21 For Review Only 22 23 Abstract 24 25 Excavations in Sungai Batu have revealed numerous archaeological artefacts that 26 27 link Sungai Batu to a life heavily associated with trading. The discovered artefacts 28 29 st 30 originated back to the 1 century; therefore, was established as one of the oldest known 31 32 civilisation in Southeast Asia. This encouraged researchers to dig deeper on Sungai 33 34 Batu’s archaeological events; however, many of the studies were relatively superficial 35 36 37 and did not portray the dynamics between Sungai Batu trading complex and its ancient 38 39 river as a whole but explained in localized sections. This paper seeks to provide a 40 41 conclusion of the past events in the area based on geological, geophysical and 42 43 geotechnical studies. The discussions include tracing the ancient Sungai Batu flows and 44 45 46 the trading complex for better illustration of their associations so that future 47 48 archaeological endeavours could discover more trade and religious artefacts aside from 49 50 jetty structural remains, iron smelting sites, potteries and stupa at Sungai Batu. 51 52 53 54 55 Keywords: Sungai Batu; ancient river; trading complex; entrepot 56 57 58 59 60

For Proof Read only Page 3 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 1. Introduction 5 6 7 As the richest archaeological zone in Malaysia, Lembah (Valley) Bujang covers 8 9 an extensive area of roughly 224 km2. The region consists of three archaeological 10 11 complexes; Sungai Batu, Sungai Mas and Pengkalan Bujang. Sungai Batu was named 12 13 14 after a tributary river that flows into the area. Civilisations at Sungai Mas and Pengkalan 15 16 Bujang emerged only after Sungai Batu complex started to decline economically; hence, 17 18 are chronologically younger (Murphy, 2018). Sungai Batu complex possessed abundant 19 20 archaeological structures and artefacts that could unravel the history of human activities 21 For Review Only 22 23 (primarily trading activities), the geographical changes of Sungai Batu ancient river and 24 25 ancient coastline. 26 27 With numerous studies that began more than a century ago, this brought about 28 29 30 mismatched and inorganized interpretations on the matter. Seeking for strong legitimacy, 31 32 this paper summarized a great number of studies regarding Sungai Batu trading complex 33 34 and its ancient river in order to align the archaeological events accordingly for future 35 36 37 references. 2D resistivity method was employed as the initial step in understanding 38 39 Sungai Batu’s geology with the assistance of physical samplings from borehole data and 40 41 auguring. Traces of ancient Sungai Batu flow was also investigated using magnetic and 42 43 3D resistivity method in conjunction with physical samplings (Nordiana et al., 2013; 44 45 46 Saad, Saidin, Fauzi & Tarmizi, 2015; Yusoh et al., 2018). 47 48 49 50 2. Location of Lembah Bujang 51 52 53 Lembah Bujang is situated in Kuala Muda, (Southern ) and covers 54 55 an area bordered by Mount Jerai in the north, Muda River in the south, in 56 57 the east and Straits of Malacca in the west (Saad et al., 2015; Roslan, Noor, Abdullah & 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 4 of 29

1 2 3 4 Ali, 2016). The 1217 m high Mount Jerai, the highest point in the area, was particularly 5 6 7 useful during active trading period in Lembah Bujang as it served as a landmark for 8 9 traders and mariners (Zakaria, Saidin & Abdullah, 2011). Principal rivers that flow into 10 11 the valley include Sungai Merbok, Sungai Muda and Sungai Bujang which drains into the 12 13 14 straits (Figure 1a); therefore, were prime areas for archaeological investigations as they 15 16 held attractive features for living settlement due to close proximity to food sources and 17 18 fertile land (Washburn & Lancaster, 1968; Yesner, 1987; Haughey, 2009). It is clear that 19 20 most of the archaeological sites were found at very close distances to the rivers and coastal 21 For Review Only 22 23 line, as supported by the 86 sites found in Lembah Bujang (Allen, 1991). However, 24 25 changes in physical environment (river course or sea level rise) at some point of the 26 27 history caused some sites to be located further inland as seen at Sungai Batu, Gunung 28 29 30 Jerai and Tikam Batu sites. 31 32 Key interests in Lembah Bujang archaeological sites include Sungai Batu, 33 34 Pengkalan Bujang and Sungai Mas archaeological complexes. However, this paper 35 36 37 focuses on Sungai Batu archaeological site where a tributary river (Sungai Batu) that 38 39 branches off from Sungai Merbok played an important role in archaeological perspective 40 41 (Figure 1b). 42 43 44 45 46 Figure 1: Known archaeological sites in Kedah where a) depicts the sites’ locations 47 48 relative to neighbouring rivers while b) shows Sungai Batu archaeological site relative 49 50 to Sungai Bujang on the west and Sungai Merbok on the east 51 52 53 54 55 56 57 58 59 60

For Proof Read only Page 5 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 3. Geography and sea level changes around Sungai Batu 5 6 7 Geographically speaking, surface observations at Sungai Batu (Figure 1b) depict 8 9 that the region is largely covered by palm and rubber trees with localized swamps. Two 10 11 rivers; Sungai Merbok and Sungai Bujang flank each side of the archaeological complex. 12 13 14 The elevation surrounding the complex is relatively flat with less than 25 m height (Figure 15 16 2); hence, could be considered as a flood plain zone due to close proximity to rivers. 17 18 19 Figure 2: Topography surrounding Sungai Batu Archaeological Complex and 20 21 For Review Only 22 Sungai Merbok which illustrates that the majority of the area is a flood plain zone 23 24 25 In terms of shoreline, a large part of Lembah Bujang was engulfed by the sea 26 27 during the 1st – 2nd centuries due to high sea level at the time (Saidin, 2011). This explains 28 29 30 the interior localities of archaeological sites depicted by Figure 1A as a result of limited 31 32 land at the time. In the 9th – 13th centuries, the shorelines were more or less similar during 33 34 the high-water level except for the changed in Sungai Muda flow course (Figure 3). 35 36 th th 37 Drastic changed of shoreline could only be seen in the 14 – 15 centuries where the 38 39 receding sea level caused an abundant of sediments brought by sea and river to be 40 41 deposited onto the shore. This event encouraged coastal land development and 42 43 44 subsequently contributed to the prominent sandy-clay soil deposited in the area 45 46 (Wheatley, 1961; Allen, 1990). 47 48 49 50 51 th th 52 Figure 3: Shoreline regression throughout the 9 until 15 centuries relative to 53 54 Sungai Batu site that caused the exposure of landmass (modified from Allen, 2000) 55 56 57 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 6 of 29

1 2 3 4 4. Geology of Sungai Batu 5 6 7 A large part of Lembah Bujang overlies a broad band of Quaternary beach ridges 8 9 (Matang Gelugor Member) that extends up to 8 km inland (Bosch, 1988). On the contrary, 10 11 older Mahang Formation dated from Middle Ordovician to Early Devonian underlies the 12 13 14 eastern and southern flanks of Sungai Merbok (Habibah, Wan Fuad & Mohamad, 2004). 15 16 The formation composes of shale, red slate, grey slate and black slate. However, the 17 18 shallow subsurface of Sungai Batu complex is primarily made of fluvial sandy clay soil. 19 20 A detail mapping of Sungai Batu’s shallow subsurface was conducted in the last 21 For Review Only 22 23 15 years using not only archaeological methods, but also utilized geophysical methods 24 25 such as 2-Dimensional (2-D) Resistivity Imaging, magnetic and remote sensing methods. 26 27 The location of several studies conducted in the Sungai Batu complex are depicted in 28 29 30 Figure 4. Alashloo et al. (2011) investigated the type of soil in Sungai Batu site by 31 32 implementing 2D resistivity imaging with maximum penetration depth of 31 m. The 33 34 results depicted that the subsurface could be categorized into three layers; unsaturated 35 36 37 alluvial soil (sand and clay with occasional boulders) upper layer that have maximum 38 39 thickness of 10 m, followed by saturated alluvial soil with thickness of up to 20 m and 40 41 lastly underlain by hard bedrock. The first layer has relatively higher resistivity values 42 43 (50 – 1000 Ωm) compared to the second layer (< 50 Ωm) due to difference in moisture 44 45 46 content. The bedrock could be easily recognized due to its high resistivity values (> 1000 47 48 Ωm). 49 50 51 52 53 Figure 4: Location of Sungai Batu sites done by several researches which were all 54 55 within 500 m radius 56 57 58 59 60

For Proof Read only Page 7 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 Samuel, Saad, Muztaza, Ismail and Saidin (2016) had further divided the upper 5 6 7 layer into two more layers based on resistivity, magnetic and auguring methods at Sungai 8 9 Batu (Figure 5) in the 14 m penetrated depth. Conducted just opposite of Alashloo’s study 10 11 area, this study shows that the topsoil (0 – 3 m depth) in the first layer has lower resistivity 12 13 14 values of < 200 Ωm due to sticky clayey material with decayed organic materials in the 15 16 shallow subsurface. Muscovite was also found in the soil sample, suggesting a close 17 18 proximity to sediments source considering that muscovite is a soft mineral; thus, is easily 19 20 weathered. Going further in depth until 5 m, the upper layer changes in grain size with 21 For Review Only 22 23 presence of sandy clay having values ranging from 73 – 152 Ωm. Furthermore, Ismail, 24 25 Nordiana, Saidin, Masnan and Abir (2018) selected SB1 and SB2 sites in Sungai Batu to 26 27 run resistivity survey on top a mound with exposed baked-clay bricks. Within the 10 m 28 29 30 depth from ground surface, two distinct layers could be distinguished; top brick layer with 31 32 high resistivity values (≥ 3000 Ωm) and lower alluvial soil layer with low resistivity 33 34 values (50 – 100 Ωm). The bricks, which were identified to be a floor structure, are highly 35 36 37 resistive to current flow due to the intense ceramic bonds form at high temperatures (500 38 39 – 800 °C). A schematic diagram of the subsurface is demonstrated in Figure 5 based on 40 41 these data. 42 43 44 45 46 Figure 5: Schematic diagram of Sungai Batu's lithology and the relative position of the 47 48 archaeological floor structure in the form of a brick mound 49 50 51 52 53 5. Ancient river flow at Sungai Batu 54 55 Believing that Sungai Batu used to flow in a more extensive area, Nordiana et al. 56 57 (2013) conducted a comprehensive study to predict the ancient river flow in the 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 8 of 29

1 2 3 4 archaeological complex through physical samplings (two boreholes) in the area (Figure 5 6 7 4). In less than 2 m depth, a hard layer of iron pans was detected in boreholes ADH1 and 8 9 ADH2. This signifies an occurrence of iron-rich fluid oxidation and silica loss that 10 11 typically transpired due to surface exposure or at very shallow subsurface (Saidin, 2009). 12 13 14 At depths of < 5 m, the boreholes exhibit stiff silty-clay soil layer in the area with SPT 15 16 N-values of more than 30. Muscovite was also found in the alluvial deposits here. 17 18 Presence of iron pan and muscovite at these sites suggests the possibility that they were 19 20 previously situated on a river bank. 21 For Review Only 22 23 Based on magnetic residual results in Sungai Batu by Saad et al. (2015), two zones 24 25 were categorized where low magnetic residual values of < 0 nT is believed to be an 26 27 ancient river (alluvium) whereas the high magnetic values of > 0 nT is unlikely to be one. 28 29 30 While most of magnetic surveys used in archaeology were used in mapping fire-related 31 32 artefacts (burning causes soil susceptibility to increase), occurrence of naturally high 33 34 susceptibility soil in the absence of burning does exist due to the presence of fine-grained 35 36 37 soil such as clay and silt which contain a large amount of iron (Crowther, 2003). Using 38 39 the produced magnetic residual map, the ancient river and present Sungai Merbok were 40 41 mapped onto the present elevation map (Figure 6). The existing Sungai Batu has an 42 43 average width of only 35 m as it is a tributary river from Sungai Merbok. However, the 44 45 46 suspected ancient tributary river used to have a longer route and flowed further into the 47 48 land from the primary Sungai Merbok. Traces of the ancient river extend up to 1 km in 49 50 width which flowed roughly from northwest to southeast direction. 51 52 53 54 55 Figure 6: Ancient and present Sungai Batu flows based on magnetic data overlaid on 56 57 top of elevation map of Sungai Batu archaeological complex 58 59 60

For Proof Read only Page 9 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 Another detail survey was done by Yusoh et al. (2018) which covered almost the entire 5 6 7 Sungai Batu archaeological complex in an attempt to map the ancient river using 3- 8 9 Dimensional Resistivity method. 11 resistivity lines were conducted to generate a dense 10 11 data for good subsurface visualization. In contrast to the location suggested by Saad et al. 12 13 14 (2015), the ancient river identified from resistivity method was at the eastern zone of the 15 16 complex. Figure 7 depicts the suspected ancient Sungai Batu with resistivity values of 0 17 18 – 45 Ωm where the 3-D model were sliced at two different depths; 0 – 6 m and 12 m. The 19 20 shallow depth slice (Figure 7a) has similar resistivity values up to 6 m depth that extends 21 For Review Only 22 23 up to ± 600 m in width. This indicates that the area composes of both the ancient river 24 25 and its floodplains. The shallow depth slice also shows two paths of river flows; western 26 27 and eastern flow-paths that were separated by roughly 200 m distance but became 28 29 30 interconnected down-stream. In the 12 m depth slice (Figure 7b), traces of the eastern 31 32 ancient river start to diminish whereas the western river’s traces are still clear until 21 m 33 34 depth, thus indicating that the western river was not only flowing deeper, but is older too. 35 36 37 Any traces of the ancient river courses throughout the centuries completely diminishes at 38 39 33 m depth. 40 41 42 43 Figure 7: Ancient and present Sungai Batu flows based on resistivity data where a) is 44 45 46 depth slice of 3D resistivity model at 6 m while b) at 12 m depth. 47 48 49 50 With regards to previous studies discussed beforehand, the ancient rivers mapped 51 52 53 by Saad (2015) and Yusoh et al. (2018) tallied with borehole results by Nordiana et al. 54 55 (2013), seeing that both ADH1 and ADH2 were situated on the riverbanks of the proposed 56 57 ancient river (Figure 8). 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 10 of 29

1 2 3 4 5 6 7 Figure 8: Integration of suspected ancient river flows mapped by Saad et al. (2015) and 8 9 Yusoh et al. (2018) using resistivity and magnetic surveys 10 11 12 13 14 6. Trades in Sungai Batu, Lembah Bujang 15 16 Lembah Bujang is situated at the northern part of Straits of Malacca which was 17 18 one of the key factors that boost the region into life, subsequently became a strategic port 19 20 for trading and industry due to its prime location. With close proximity to the straits, 21 For Review Only 22 23 Lembah Bujang was the pioneer of Kedah Tua Kingdom’s civilization (Jacq- 24 25 Hergoualc’h, 1992, Bellwood 1997). Up to present, abundance of relics, ancient 26 27 structures and iron foundries were discovered in Sungai Batu archaeological complex. 28 29 nd 30 These brought to light that the complex was founded around 110 AD (2 century) based 31 32 on laboratories results of these artefacts such as jetty remains, iron smelting sites and a 33 34 clay brick monument (Nordiana et al., 2013). Two structures that have come to light, a 35 36 st nd 37 possible stupa and a jetty, may date as early as the 1 – 2 centuries CE, while an iron 38 39 smelting site has been dated to the 3rd – 6th centuries CE. These results now indicate that 40 41 Sungai Batu is chronologically older than it was previously thought (Nordiana, 2014). 42 43 Accommodated with fertile land and nearby water resource, the land flourished 44 45 th 46 around the 4 century and continued to expand economically via trading surge. Lembah 47 48 Bujang is believed to be not only the earliest entrepot but also a religious centre in Kedah 49 50 based on the discoveries of Hindu-Buddhist temple, statues, porcelain, beads and many 51 52 53 other artefacts (Wales, 1970; Rahman, 2008). Presence of traders brought about a variety 54 55 of artefacts in Lembah Bujang such as seen by the discovery of a brick structure in Sungai 56 57 Batu that appears to be associated to either Buddhist or Hindu (Allen, 1991). On top of 58 59 60

For Proof Read only Page 11 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 that, a Buddhist stone inscription was also recovered that was dated to be from the 5th – 5 6 th 7 6 centuries (Zakaria et al., 2011). With more than 90 archaeological mounds recognized 8 9 in Sungai Batu with high artefacts potential, SB2B and SB2D (Figure 4) are among those 10 11 that have been excavated to reveal an ancient river connected to the Sungai Batu fluvial 12 13 14 system (Zakaria et al., 2011). Structures made of brick, roof tiles and timber-log pillar 15 16 highly suggest that they were previously a river jetty, in conjunction with their strategized 17 18 location and the structures’ condition (water-damaged) as depicted in Figure 9. Jetty is an 19 20 important means to facilitate water and landmass’ related activities such as trading (Giri, 21 For Review Only 22 23 2004). Along with the structures came the relics such as potsherds, beads, stone tools and 24 25 iron slags. Charcoal sample taken here indicated that it was from the 5th century via 26 27 radiocarbon dating. Together with previously dated artefacts, these results illustrate that 28 29 th 30 the jetties were associated with iron smelting activities not just during the 5 century but 31 32 even back in the 1st century. The jetties were primarily useful for loading and unloading 33 34 of trade commodities that was also inclusive of iron ores, hence indirectly endorsing the 35 36 37 existence of an upriver-downriver economical interaction and the occurrence of iron- 38 39 based trading in Sungai Batu while concurrently acted as a religious centre. Gradually, 40 41 Lembah Bujang evolved to a collecting centre for trading goods from all over Malay 42 43 Peninsula before it was established as an entrepot in the 7th century AD (Omar, 44 45 46 Mohammed, Nordin, Johari & Ibrahim, 2010). 47 48 49 50 Figure 9: Jetty remains consist of floor, steps and lens-like wall structures found near 51 52 53 ancient Sungai Batu channel (modified from Zakaria et al., 2011) 54 55 56 57 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 12 of 29

1 2 3 4 A turn of events transpired when Lembah Bujang fell under the influence of 5 6 7 Srivijaya in 670 AD, resulting the influential region to be under Srivijaya’s control for 8 9 more than 400 years before Lembah Bujang regained its power at the end of the 11th 10 11 century due to the crippling of the Srivijaya Empire. In the 14th century, Merbok Estuary 12 13 14 continued to evolve as an influential port and trading centre that catered both local and 15 16 international traders including China, India and the Middle East as it became an extensive 17 18 entrepot on the west coast of Thai-Malay peninsula (Khoo, 1996). Carnelian beads and 19 20 Guangdong ceramic were among the discovered trading products in Sungai Batu site that 21 For Review Only 22 23 proved the international relationship between Lembah Bujang and foreign countries 24 25 (Zakaria et al., 2011). Although Sungai Batu was blessed with favourable bay and estuary, 26 27 changes of monsoonal seasons were among the favourable factors that force ships to 28 29 30 layover for months to wait for the winds to change direction (Murphy, 2018). 31 32 Sungai Batu ancient river’s headstream came from Mount Jerai, known as the 33 34 landmark of the . Allen (1988) did not characterized Sungai Batu Complex 35 36 37 as one of the trading sites, although its river played a vital role in connecting remote areas 38 39 to Sungai Merbok that flows into the Strait of Malacca due to its strategized location; on 40 41 the banks of ancient rivers. Despite that, iron smelting site discovered in Sungai Batu 42 43 (SB2B and SB2D sites) suggested otherwise as this discovery proved that Sungai Batu 44 45 46 was once a site for iron accumulation before being exported (Mohd, 2009). Sungai Batu 47 48 port was used in transporting trading commodities to areas along the river as iron ores 49 50 were found at both Sungai Batu and Mount Jerai neighbourhood (Bean & Hill, 1969; 51 52 53 Bradford, 1972; Almashoor, 1974). Other than iron, forest products were also one of the 54 55 traded commodities here as the area was rich in forest resources (Dunn, 1975; 56 57 Kathirithamby-Wells & Villiers, 1990). 58 59 60

For Proof Read only Page 13 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 Sea level and river course changes also have significant impact on the location of 5 6 st th 7 the Lembah Bujang’s entrepot. During the 1 – 6 centuries, Sungai Batu was among the 8 9 few places on Lembah Bujang’s limited land that was not submerged during high sea 10 11 level (Saidin, 2011). Considering the close proximity of Sungai Batu to the shoreline 12 13 14 connecting to Straits of Malacca combined in conjunction with the presence of Sungai 15 16 Batu tributary channel for trading pathway, trading initially began here. As shoreline 17 18 shifted with time (sea level receded), new passageways opened which provided better 19 20 paths for traders and mariners into Lembah Bujang, thus promoted other entrepots to took 21 For Review Only 22 23 over Sungai Batu such as Pengkalan Bujang and Sungai Mas as illustrated in Figure 10 24 25 (Saidin, Abdullah, Osman & Abdullah, 2011; Murphy, 2018). Sungai Mas and Sungai 26 27 Bujang were either submerged or were too swampy to be inhabited in the 1st – 3rd 28 29 30 centuries, but they later were more viable for access from Straits of Malacca compared to 31 32 Sungai Batu as sea level and river courses changed over time. These eventuated the rivers 33 34 and streams to become clogged with silt and clay as the coastal plain continued to grow 35 36 37 before finally succumbed to permanent landlocked and ceased as a viable location for 38 39 maritime trade. 40 41 42 43 Figure 10: The shifting of trading ports throughout the centuries as an impact from sea 44 45 46 level and river course changes 47 48 49 50 7. Conclusion 51 52 53 A wide range of studies were executed at Sungai Batu archaeological complex, 54 55 yet most provided little satisfaction to literate on. This paper presented the overall review 56 57 in an organized manner based on the studies conducted. Geological, geophysical and 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 14 of 29

1 2 3 4 geotechnical results complimented each other which gave rise to the discovery of 5 6 7 potential locations of ancient river flow in Sungai Batu. Series of studies, excavation and 8 9 dating tests done revealed that Sungai Batu was a popular trading complex based on 10 11 remaining jetty structures, iron smelting site and clay brick monuments. Bricks, roof tiles 12 13 14 and timber-log pillars attest to the existence of jetty that was once situated here. Trade- 15 16 related artefacts excavated were situated on the river banks of the predicted ancient 17 18 Sungai Batu, therefore provided a strong support on the matter. Sungai Batu complex was 19 20 not only a trading centre but also served as a religious centre. All these attributes were 21 For Review Only 22 23 due to the strategic location of Sungai Batu during the long period of high sea level which 24 25 was above sea level and near the straits, thus permitting the complex to hold its stature as 26 27 an entrepot for several centuries. 28 29 30 31 32 References 33 34 Alashloo, S. M., Saad, R., Nawawi, M. N. M., Saidin, M., & Alashloo, M. M. (2011). 35 36 37 Magnetic and 2D electrical imaging methods to investigate an archaeological 38 39 site at Sungai Batu, Kedah, Malaysia. In 3rd International Conference on 40 41 Chemical, Biological and Environmental Engineering, IPCBEE (Vol. 20). 42 43 Allen, J. (1988). Trade, transportation, and tributaries: exchange, agriculture, and 44 45 46 settlement distribution in early historic-period Kedah, Malaysia (Doctoral 47 48 dissertation, University of Hawai'i, Manoa, United States of America). Retrieved 49 50 from 51 52 53 https://scholarspace.manoa.hawaii.edu/bitstream/10125/9329/uhm_phd_89121 54 55 29_r.pdf 56 57 58 59 60

For Proof Read only Page 15 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 Allen, J. (1991). Trade and site distribution in early historic-period Kedah: 5 6 7 Geoarchaeological, historic, and locational evidence. Bulletin of the indo-pacific 8 9 prehistory Association, 10(307-19). 10 11 Allen, J. (2000). In support of trade: Coastal site location and environmental 12 13 14 transformation in early historical-period Malaysia and Thailand. Bulletin Indo- 15 16 Pacific Prehistory Association, 20:62-78. 17 18 Almashoor, S. S. (1974). Geology of Gunung Jerai, Kedah. (Unpublished M. Sc. thesis, 19 20 National University of Malaysia, Bangi, Malaysia). 21 For Review Only 22 23 Bean, J. H., & Hill, J. H. (1969). The iron-ore deposits of West Malaysia (Vol. 2). 24 25 Geological Survey, West Malaysia. 26 27 Bellwood, P. (1997). Ancient seafarers. Archaeology, 50(2), 20-22. 28 29 30 Bosch, J. H. A. (1988). The Quaternary deposits in the central plains of Peninsular 31 32 Malaysia. Geological Survey of Malaysia Report, Quaternary Geology 33 34 Section, 87 pp. 35 36 37 Bradford, E. F. (1972). Geology and mineral resources of the Gunung Jerai area. Kedah 38 39 Geological Survey District Memoir, 13. 40 41 Crowther, J. (2003). Potential magnetic susceptibility and fractional conversion studies 42 43 of archaeological soils and sediments. Archaeometry, 45(4), 685-701. 44 45 46 https://doi.org/10.1046/j.1475-4754.2003.00137.x 47 48 Dunn, F. L. (1975). Rain-forest collectors and traders: a study of resource utilization in 49 50 modern and ancient Malaya (No. 5). Monographs of the Malaysian Branch of 51 52 53 the Royal Asiatic Society. 54 55 56 57 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 16 of 29

1 2 3 4 Giri, S. (2004). Flow, turbulence and erosion induced by river structures (Doctoral 5 6 7 dissertation, Hokkaido University, Japan). Retrieved from 8 9 https://ci.nii.ac.jp/naid/500000282039/ 10 11 Habibah, J., Wan Fuad, W. H., & Mohamad, M. T. (2004). Pengaruh jenis batuan sekitar 12 13 14 terhadap taburan Pb dalam sedimen muara Sungai Merbok, Kedah. Geological 15 16 Society of Malaysia Bulletin 48, 7-11. Retrieved from 17 18 http://archives.datapages.com/data/geological-society-of- 19 20 malaysia/bulletins/048/048001/pdfs/7.pdf 21 For Review Only 22 23 Haughey F. M. (2009). People and Water: A study of the relationship between humans 24 25 and rivers in the Mesolithic and Neolithic with particular reference to that within 26 27 the Theme Basin. (Doctoral thesis, University College London, United 28 29 30 Kingdom). Retrieved from 31 32 http://discovery.ucl.ac.uk/1460514/1/Fiona_Haughey_vol_1.pdf 33 34 Ismail, N., Nordiana, M. M., Saidin, M., Masnan, S. S. K., & Abir, I. A. (2018, April). 35 36 37 Detection of Shallow Buried Archaeological Remains Structure using 2-D 38 39 Resistivity Method at Sungai Batu, Lembah Bujang, Kedah. Journal of Physics: 40 41 Conference Series (Vol. 995, No. 1, p. 012086). IOP Publishing. Retrieved from 42 43 https://iopscience.iop.org/article/10.1088/1742-6596/995/1/012086/pdf 44 45 46 Jacq-Hergoualc'h, M. (1992). La civilisation de ports-entrepôts du Sud Kedah 47 48 (Malaysia): Ve-XIVe siècle. Editions L'Harmattan. 49 50 Kathirithamby-Wells, J., & Villiers, J. (Eds.). (1990). The Southeast Asian port and 51 52 53 polity: rise and demise. Singapore University Press, National University of 54 55 Singapore. 56 57 58 59 60

For Proof Read only Page 17 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 Khoo, T. T. (1996). Geomorphological evolution of the Merbok estuary area and its 5 6 7 impact on the early state of Kedah, northwest peninsular Malaysia. Journal of 8 9 Southeast Asian Earth Sciences, 13(3-5), 347-371. 10 11 https://doi.org/10.1016/0743-9547(96)00042-6 12 13 14 Mohd, M. N. A. (2009). The Discovery of Iron Smelting in Sg. Batu, Lembah Bujang, 15 16 Kedah. Indo-Pacific Prehistory Association 2009 Program and Abstracts. 17 18 Murphy, S. A. (2018). Revisiting the Bujang Valley: A Southeast Asian entrepôt complex 19 20 on the maritime trade route. Journal of the Royal Asiatic Society, 28(2), 355- 21 For Review Only 22 23 389. https://doi.org/10.1017/S1356186317000505 24 25 Nordiana, M. M., Saad, R., Saidin, M. M., & Kamaruddin, N. A. (2014). Archeomagnetic 26 27 Studies of Anomaly at Sungai Batu, Lembah Bujang, Kedah (Malaysia). 28 29 30 Electronic Journal of Geotechnical Engineering, 19, 2315-2323. Retrieved from 31 32 http://www.ejge.com/2014/Ppr2014.217mar.pdf 33 34 Nordiana, M. M., Saad, R., Saidin, M., Shyeh, S. K., Kiu, Y. C., Ismail, N. A., & El 35 36 37 Hidayah Ismail, N. (2013). Mapping Alluvial Landscape Using 2-D Resistivity 38 39 with EHR technique at Lembah Bujang, Kedah (Malaysia). Electronic Journal 40 41 of Geotechnical Engineering, 18. Retrieved from 42 43 http://www.ejge.com/2013/Ppr2013.098alr.pdf 44 45 46 Omar, M. W., Mohammed, S., Nordin, R., Johari, A., & Ibrahim, S. (2010). Hidden Tales 47 48 of the Bujang Valley. Asian Culture and History, 2(2), 221. 49 50 10.5539/ach.v2n2P221 51 52 53 Rahman, H. S. B. N. A. (2008). Lembah Bujang: dari perspektif arkeologi dan 54 55 pelancongan. Institut Alam dan Tamadun Melayu, Universiti Kebangsaan 56 57 Malaysia. 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 18 of 29

1 2 3 4 Roslan, S. A., Noor, N. M., Abdullah, A., & Ali, Z. M. (2016). Analysis of shrines 5 6 7 properties using remote sensing approach: case study of Lembah 8 9 Bujang. Planning Malaysia Journal, 14(4). 10 11 http://dx.doi.org/10.21837/pmjournal.v14.i4.167 12 13 14 Saad, R., Saidin, M., Fauzi, A., & Tarmizi, H. (2015). Identify Sungai Batu Ancient River 15 16 by magnetic method. Electronic Journal of Geotechnical Engineering, 20(18), 17 18 11143-11148. Retrieved from 19 20 https://www.researchgate.net/profile/Fauzi_Andika/publication/285429633_In 21 For Review Only 22 23 dentify_sungai_batu_ancient_river_by_magnetic_method/links/58aa7f9ba6fdc 24 25 c0e079831ac/Indentify-sungai-batu-ancient-river-by-magnetic-method.pdf 26 27 Saidin, M. (2009). Report for borehole record at Sungai Batu area, Lembah Bujang, 28 29 30 Saidin, M. (2011). Dari Zaman Batu ke Tamadun Awal di Malaysia: Pemerkasaan Jati 31 32 Diri Bangsa (Penerbit USM). Retrieved from 33 34 https://books.google.com/books?hl=en&lr=&id=VRatCQAAQBAJ&oi=fnd&p 35 36 37 g=PA13&ots=9LgNX_Ywio&sig=lDUYOAmguOO3_PUgbBY41JKyJR0 38 39 Saidin, M., Abdullah, J., Osman, A. J., & Abdullah, A. (2011). Issues and problems of 40 41 previous studies in the Bujang Valley and the discovery of Sungai Batu. Bujang 42 43 valley and early civilizations in Southeast Asia, 15-36. Retrieved from 44 45 46 https://www.researchgate.net/publication/288890645_Issues_and_Problems_of 47 48 _Previous_Studies_in_The_Bujang_Valley_and_The_Discovery_of_Sungai_B 49 50 atu 51 52 53 Samuel, Y. M., Saad, R., Muztaza, N. M., Ismail, N. A., & Saidin, M. M. (2016). 54 55 Magnetics, 2-d resistivity and geotechnical studies for shallow subsurface soil 56 57 58 59 60

For Proof Read only Page 19 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 characterization in Lembah Bujang, Kedah. Jurnal teknologi, 78(7-3), 123-127. 5 6 7 https://doi.org/10.11113/jt.v78.9504 8 9 Tieng, F. S. (2010). Hoabinhian Rocks: An Examination of Guar Kepah Artifacts from 10 11 the Heritage Conservation Centre in Jurong (Doctoral dissertation, National 12 13 14 University of Singapore, Lower Kent Ridge Road, Singapore). Retrieved from 15 16 http://scholarbank.nus.edu.sg/handle/10635/20404 17 18 Wales, H. Q. (1970). Malayan Archaeology of the" Hindu period": some 19 20 reconsiderations. Journal of the Malaysian Branch of the Royal Asiatic 21 For Review Only 22 23 Society, 43(1 (217), 1-34. Retrieved from 24 25 http://myrepositori.pnm.gov.my/bitstream/1/185/1/JB0039_MAHP.pdf 26 27 Washburn, S. L. & C. S. Lancaster. (1968). The evolution of hunting. Man the Hunter, 28 29 30 Chicago, Aldine:293–303 31 32 Wheatley, P. (1961). The Golden Khersonese: Studies in the historical geography of the 33 34 Malay Peninsula before AD 1500. Retrieved from 35 36 37 http://myrepositori.pnm.gov.my/bitstream/1/209/1/JB0047_TGKh.pdf 38 39 Yesner, D. R. (1987). Life in the Garden of Eden: Causes and consequences of the 40 41 adoption of marine diets by human societies. Food and evolution: Toward a 42 43 theory of human food habits, 285-310. 44 45 46 Yusoh, R., Saad, R., Saidin, M., Muhammad, S. B., Anda, S. T., Ashraf, M. M., & 47 48 Hazreek, Z. A. M. (2018, April). Visualizing Sungai Batu Ancient River, 49 50 Lembah Bujang Archeology Site, Kedah–Malaysia using 3-D Resistivity 51 52 53 Imaging. Journal of Physics: Conference Series 995(1), 012117. 10.1088/1742- 54 55 6596/995/1/012117 56 57 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 20 of 29

1 2 3 4 Zakaria, I. I., Saidin, M., & Abdullah, J. (2011). Ancient Jetty at Sungai Batu Complex, 5 6 7 Bujang Valley, Kedah. Postgraduate Student Forum. Retrieved from 8 9 http://www.cuhk.edu.hk/ant/PostgraduateForum2011/Arch/IklilIzzatiZAKARI 10 11 A.pdf 12 13 14 15 16 17 18 19 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Page 21 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 FIGURE FILE 5 6 7 8 Figure 1: Known archaeological sites in Kedah where a) depicts the sites’ locations 9 10 relative to neighbouring rivers while b) shows Sungai Batu archaeological site relative to 11 12 Sungai Bujang on the west and Sungai Merbok on the east 13 14 15 Figure 2: Topography surrounding Sungai Batu Archaeological Complex and Sungai 16 17 18 Merbok which illustrates that the majority of the area is a flood plain zone 19 20 21 Figure 3: Shoreline regressionFor Reviewthroughout the 9th Onlyuntil 15th centuries relative to Sungai 22 23 Batu site that caused the exposure of landmass (modified from Allen, 2000) 24 25 26 Figure 4: Location of Sungai Batu sites done by several researches which were in 500 m 27 28 in radius 29 30 31 32 Figure 5: Schematic diagram of Sungai Batu's lithology and the relative position of the 33 34 archaeological floor structure in the form of a brick mound 35 36 37 Figure 6: Ancient and present Sungai Batu flows based on magnetic data overlaid on top 38 39 of elevation map of Sungai Batu archaeological complex 40 41 42 Figure 7: Ancient and present Sungai Batu flows based on resistivity data where a) is 43 44 45 depth slice of 3D resistivity model at 6 m while b) at 12 m depth. 46 47 48 Figure 8: Ancient rivers mapped by Saad et al. (2015) and Yusoh et al. (2018) 49 50 51 Figure 9: Jetty remains consist of floor, steps and lens-like wall structures found near 52 53 ancient Sungai Batu channel (modified from Zakaria et al., 2011) 54 55 56 Figure 10: The shifting of trading ports throughout the centuries as an impact from sea 57 58 59 level and river course changes adapted from Murphy (2018) 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 22 of 29

1 2 3 4 5 6 7 8 Gunung 9 a Jerai Bukit 10 Choras 11 Gurun 12 13 14 15 Merbok 16 17 18 19 20 Sungai Petani 21 For Review Only 22 23 24 Kota Kuala 25 Muda 26 Tikam 27 Batu 28 29 30 31 b 32 33 34 Sungai Batu 35 Archaeological site 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 Figure 1 Known archaeological sites in Kedah where a) depicts the sites’ locations 54 relative to neighbouring rivers while b) shows Sungai Batu archaeological site 55 relative to Sungai Bujang on the west and Sungai Merbok on the east 56 57 58 59 60

For Proof Read only Page 23 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 5 6 5.705 40 7 5.703 8 35 9 5.701 10 5.699 11 Sungai Batu 30 5.697 Archaeological Complex

12 (° N) 13 5.695 25 14 5.693 15 15 16 Northing 5.691 17 5.689 10 18 5.687 19 5 20 5.685 Current 21 For Review SungaiOnly Merbok -2 22 100.44 100.446 100.452 100.458 100.464 100.47 23 Elevation (m) 24 Easting (° E) 25 26 Figure 2 Topography surrounding Sungai Batu Archaeological Complex and Sungai 27 Merbok which illustrates that the majority of the area is a flood plain zone 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 24 of 29

1 2 3

4 5.74 5.74 5 6 5.72 5.72 Sungai Batu Sungai Batu 7 site site 5.7 5.7 8 STRAITS STRAITS 9 OF OF 5.68 5.68 MALACCA 10 MALACCA 11 5.66 5.66 12 13 5.64 5.64

14 5.62 5.62 15 16 5.6 5.6 17 5.58 18 5.58 Sungai Muda 19 5.56 5.56 Sungai Muda 20 9th century shoreline For Review10-11th Only century shoreline 21 100.3 100.34 100.38 100.42 100.46 100.3 100.34 100.38 100.42 100.46 22 5.74 5.74

23 5.72 5.72 24 Sungai Batu Sungai Batu site site 25 5.7 STRAITS 5.7 STRAITS 26 OF OF MALACCA 27 5.68 MALACCA 5.68

28 5.66 5.66 29 30 5.64 5.64 31 32 5.62 5.62

33 5.6 5.6 34 35 5.58 5.58

36 Sungai Muda Sungai Muda 37 5.56 5.56 38 12-13th century shoreline 14-15th century shoreline 39 100.3 100.34 100.38 100.42 100.46 100.3 100.34 100.38 100.42 100.46 40 Figure 3 Shoreline regression throughout the 9th until 15th centuries relative to 41 Sungai Batu site that caused the exposure of landmass (modified from Allen, 2000) 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Page 25 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Figure 4 Location of Sungai Batu sites done by several researches which were all 39 within 500 m radius 40 41 Mound with 42 exposed bricks 43 Depth 44 Organic clay Unsaturated ± 3 m > 45 alluvial Sandy-clay 46 soil < 10 m Bricks in alluvial 47 Soil soil at depth 0-1.5 m (>3000 ohm.m) 48 > boundary 49 Saturated 50 alluvial soil 51 52 Alluvial soil 53 underneath < 30 m (50-100 ohm.m) 54 Hard rock 55 56 57 Figure 5 Schematic diagram of Sungai Batu's lithology and the relative position of 58 the archaeological floor structure in the form of a brick mound 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 26 of 29

1 2 3 4 5 6 7 8 9 10 nT 11 12 13 14 15 16 17 18 19 20 21 For Review Only 22 23 24 25 26 27 Current Sungai Merbok 28 5.705 29 30 Ancient Sungai Merbok (from magnetic survey) 31 5.7 32 Sungai Batu 33 Archaeological Complex

(° N) ADH1 5.695 ADH2 34 Elevation contour (m) 35

36 Northing 37 5.69 38 39 40 5.685 41 42 100.44 100.446 100.452 100.458 100.464 100.47 43 Easting (° E) 44 45 Figure 6 Ancient and present Sungai Batu flows based on magnetic data overlaid on 46 top of elevation map of Sungai Batu archaeological complex 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Page 27 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 5 Resistivity 6 (Ohm.m) 7 8 9 10 11 12 13 14 15 16 17 18 a b Easting (° E) 5.705 19 5.705 20 5.7 21 For Review5.7 Only

(° N) ADH1 ADH1 22 ADH2 ADH2 23 5.695 5.695

24 Northing 25 5.69 5.69 26 5.685 5.685 27 0-6 m depth slice 12 m depth slice 28 100.44 100.446 100.452 100.458 100.464 100.47 100.44 100.446 100.452 100.458 100.464 100.47 29 Easting (° E) Easting (° E) Ancient Sungai Merbok 30 Current Sungai Merbok Elevation contour (m) 31 (from resistivity survey) 32 Figure 7 Ancient and present Sungai Batu flows based on resistivity data where a) 33 is depth slice of 3D resistivity model at 6 m while b) at 12 m depth 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI Page 28 of 29

1 2 3 4 5 6 5.705 7 8 9 10 5.7 11 Sungai Batu Archaeological Sungai Batu Complex 12 Archaeological Site (° N) ADH1 13 5.695 ADH2 14 15 16 Northing 5.69 17 18 19 20 5.685 21 For Review Only 22 100.44 100.446 100.452 100.458 100.464 100.47 23 Easting (° E) 24 Current Sungai Merbok Ancient Sungai Merbok Elevation contour (m) 25 (from resistivity survey) 26 27 Ancient Sungai Merbok (from magnetic survey) 28 29 Figure 8 Integration of suspected ancient river flows mapped by Saad et al. (2015) 30 and Yusoh et al. (2018) using resistivity and magnetic surveys 31 32 33 34 35 Steps 36 37 Floor 38 39 40 41 Wall 42 43 44 45 46 47 48 49 50 51 52 53 54 Figure 9 Jetty remains consist of floor, steps and lens-like wall structures found near 55 ancient Sungai Batu channel (modified from Zakaria et al., 2011) 56 57 58 59 60

For Proof Read only Page 29 of 29 Songklanakarin Journal of Science and Technology SJST-2019-0195.R2 ROSLI

1 2 3 4 Gunung 5 Jerai 6

7 Pengkalan Bujang 8 (7th-9th centuries)

9 Pengkalan Bujang (11th-14th centuries) Sungai Batu 10 (3rd-6th centuries) 11 12 13 14 15 16 Sungai Petani 17 18 19 20 For KotaReview Kuala Only 21 Muda 22 23 24 Sungai Mas 25 (10th-15th centuries) 26 27 Figure 10 The shifting of trading ports throughout the centuries as an impact from 28 sea level and river course changes adapted from Murphy (2018) 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Proof Read only