12. CULTURAL HERITAGE IMPACT

Archaeological Survey and Assessment

12.1 Introduction

12.1.1 Commissioned by the Antiquities and Monuments Office (AMO) of the Leisure and Cultural Services Department, HKSAR, the Hong Kong Institute of Archaeology (HKIA) conducted a terrestrial archaeological survey at the proposed landing point (including different options) of Shenzhen Western Corridor in December 2001. The project of Shenzhen Western Corridor Landing Point (SWCLP) is related to Deep Bay Link Investigation and Preliminary Design (Agreement No. CE.109/98), and the objective of the archaeological survey is to assess the potential impact on the cultural heritage caused by the proposed landing point construction.

12.1.2 The 2001 archaeological survey at the landing point options walked over the entire works area, drilled 31 auger holes, made seven face cuts and excavated one test pit. During the survey, the archaeologist identified one beach site dating to the Song dynasty of the 12th - 13th century and possibly as early as the late Bronze Age of the first millennium B.C. in the area of Landing Point Option AB at Ngau Hom Shek.

12.1.3 Participants of the 2001 archaeological survey at SWCLP from the HKIA include LIU Mao, WANG Wenjian, WANG Fei, YIU Kam Lung, Nip Leung Hung and Nip Kwok Hung.

12.2 Geographic Setting

12.2.1 The works area of SWCLP is stretched along the coastline to the north of the Deep Bay Road in western Yuen Long, and the three options of the landing point, Option AB, Option C, and Option D, are located at Ngau Hom Shek, Ngau Hom Sha and Sheung Pak Nai, respectively (Figure 12.1). This coastal area is zero to 3 meters above the sea level and composed of wet lands with mangrove trees, sand beaches, and flat hillfoot sediments (Plate 1). At present, this area is largely occupied by numerous ponds, agricultural fields, animal farms and residential houses.

12.3 Previous Archaeological Works

12.3.1 The coastal area in western Yuen Long is rich with archaeological remains and the works area is almost entirely located within the boundary of the Ngau Hom Shek Archaeological Site. This site was first identified by Dr. Bard in 1978 and further investigated in 1994 (1), 1997 (2), 1999 (3) and 2000 (4). Remains from excavations date the site to the Neolithic period and late Bronze Age of the first millenium B.C. The distribution of the prehistoric deposit, however, is restricted to the lower hill slopes at Ngau Hom Shek and this area has been named as the Fu Tei Au Hill Site. Along the beaches at Ngau Hom Shek, archaeological remains of historical period have also been discovered but the beach site has been largely destroyed.

12.3.2 To the west and south of the Ngau Hom Shek Archaeological Site along the western Yuen Long coast, a series of Neolithic and Bronze Age sites have been discovered in the past. At Sheung Pak Nai, the location of SWCLP Option D, Neolithic remains have been retrieved from the third-tier sand dune deposit back to the hillfoot (5).

12.3.3 Previous archaeological investigations indicate that western Yuen Long coast is an area of high archaeological potential and archaeological deposits are originally distributed along the beach and hill foot. Comparatively, the current coastal line has extended much farther into the ocean in most part of the works area of SWCLP, and the focus of field survey should be placed on the locations with relatively less topographic change.

12.4 Assessment on Various Options

Ngau Hom Shek Area (SWCLP OPTION AB)

12.4.1 Ngau Hom Shek area is the location of SWCLP Option AB. This area is the focus of the field survey because it is within the boundary of the Ngau Hom Shek Archaeological Site, and also because the coastal line is very close to the hillfoot and therefore most likely has maintained its original shape from the past. In this area the archaeologist drilled 22 auger hole, made one face cut, and excavated one test pit (Table 12.1, Figure 12.2).

Table 12.1 Auger Record

Auger No.

Grid Coordinates

Context No.

Depth (cm)

Deposit Description

Note

 

N

E

 

 

 

 

K1

C1

40

Grayish clay with rubbles

Drilling at the bottom of FC1

K2

 

 

C2

100

Yellowish clay

Drilling at the bottom of FC2

 

 

 

C3

20

Yellowish sand

 

 

 

 

C4

30

Blackish mud

 

K3

 

 

C5

30

Yellowish clay

 

K4

834,612

814,137

C6

50

Grayish sand with shells

 

 

 

 

C7

90

Yellowish sand

 

K5

834,704

814,240

C8

50

Surface layer

Tile sherd

 

 

 

C9

30

Brownish clay

 

 

 

 

C10

20

Grayish sand with shells

 

K6

 

 

C73

120

Sandy deposit with shells

 

 

 

 

C74

40

Light grayish sand

 

K7

834,673

814,311

C11

20

Surface layer

Reached water level

 

 

 

C12

50

Grayish sand with shells

 

 

 

 

C13

50

Yellowish sand

 

K8

 

 

C75

20

Surface layer

 

 

 

 

C76

60

Yellowish sand with shells

Rock underneath

K9

834,698

814,243

C14

5

Surface layer-cement

 

 

 

 

C15

60

Light brownish coarse sand

 

 

 

 

C16

55

Light brownish fine sand

Reached water level

K10

835,029

815,227

C17

40

Surface layer-sand

 

 

 

 

C18

40

Light brownish coarse sand

Reached water level

K11

835,090

815,198

C19

20

Surface layer

 

 

 

 

C20

30

Light brownish coarse sand

 

 

 

 

C21

30

Light grayish sand

Reached water level

K12

835,029

815,170

C23

30

Surface layer

 

 

 

 

C24

20

Light grayish sand

 

 

 

 

C25

50

Light brownish coarse sand

Reached water level

K13

834,998

815,112

C26

40

Surface layer

 

 

 

 

C27

20

Light grayish coarse sand

 

 

 

 

C28

55

Brownish coarse sand

Reached water level

K14

835,060

815,141

C29

60

Surface layer-garbage

 

 

 

 

C30

50

Dark brownish coarse sand

Reached water level

K15

834,968

815,055

C31

20

Surface layer

 

 

 

 

C32

45

Yellowish sand with shells

 

K16

834,937

815,055

C33

10

Surface layer

 

 

 

 

C34

20

Brownish fine sand

 

 

 

 

C35

50

Blackish mud

Reached water level

K17

834,998

815,112

C36

20

Surface layer

 

 

 

 

C37

50

Dark brownish coarse sand

 

 

 

 

C38

30

Light grayish sand

 

 

 

 

C39

20

Brownish fine sand

 

K18

834,937

815,170

C40

20

Surface layer

 

 

 

 

C41

40

Brownish clay

 

 

 

 

C42

30

Grayish sand with shells

sherd

 

 

 

C43

30

Light grayish fine sand

 

 

 

 

C44

20

Brownish fine sand

 

K19

834,937

815,198

C45

20

Surface layer

 

 

 

 

C46

50

Brownish clay

sherd

 

 

 

C47

30

Brownish fine sand

Reached water level

K20

834,952

815,152

C48

40

Surface layer

 

 

 

 

C49

30

Brownish clay

 

 

 

 

C50

40

light grayish fine sand

Reached water level

K21

835,020

814,884

C51

30

Surface layer

Rock underneath

K22

 

 

C52

40

Surface layer

Rock underneath

K23

835,069

815,278

C53

50

Surface layer

Reached water level

K24

835,078

815,301

C54

20

Surface layer

 

 

 

 

C55

40

Grayish sand

Celadon sherd

 

 

 

C56

20

light grayish fine sand

 

K25

 

 

C57

30

Surface layer

Circular flake

 

 

 

C58

30

Grayish sand

White porcelain sherd

 

 

 

C59

30

light grayish fine sand

 

 

 

 

C60

10

Brownish fine sand

 

K26

 

 

C61

20

Surface layer

 

 

 

 

C62

80

Light brownish clay

Celadon sherd

K27

 

 

C63

30

Surface layer

Water pipe underneath

K28

 

 

C64

40

Surface layer

Rock underneath

K29

835,053

815,304

C65

20

Surface layer

 

 

 

 

C66

80

Light brownish clay

Celadon sherds

 

 

 

C67

40

Brownish fine sand

 

K30

835,025

815,418

C68

25

Surface layer

 

 

 

 

C69

70

Light brownish clay

Celadon and glazed sherds

 

 

 

C70

40

Dark brownish fine sand

Celadon sherd

K31

834,977

815,164

C71

40

Surface layer

 

 

 

 

C72

60

Grayish sand

Reached water level

12.4.2 This area can be divided into three sections. In the middle is a small path running through several groups of residential houses and gardens (Plate 2 - 1). To the west are patches of agricultural fields and ponds (Plate 2 - 2), and the east section is largely occupied by a piggery just above the beach (Plate 2 - 3).

12.4.3 The survey in the western section revealed no original archaeological deposit. Surface walking collected a few pieces of blue-and-white porcelain and most auger holes did not find any clue of archaeological deposit. But, the auger K18 and K19 between two large ponds pulled out ceramic sherds, and therefore one test pit of 1 x 1 meters was opened in this location.

12.4.4 Test Pit T1 uncovered two layers of deposit with cultural remains underneath the surface layer (Table 12.2, Figure 12.3, Plate 3 - 1). The remains from Layer 2 (C102) and Layer 3 (C103) indicates, however, the deposit at this location is only secondary with relics of different periods mixed together. Layer 2 yielded a piece of coarse corded pottery sherd probably dated to the Bronze Age (Plate 3 - 2) and a bronze hairpin (Plate 3 - 3, but in Layer 3 underneath, ceramic remains of the Ming and Qing periods were unearthed (Plate 3 - 4, 5).


Table 12.2 Record of Test Pit Excavation and Slope Face Cutting

T & FC No.

Grid Coordinates

Acreage (m)

Context

Depth (cm)

Deposit Type

Remain in Context

T1

N834,977 E835,164

1 x 1

C101

25

Surface layer (South wall profile)

Celadon sherd

C102

28

Light brownish sand

Flake, cord-marked coarse pottery sherds, celadon pieces, gray stoneware, bronze hairpin

C103

30

Brownish coarse sand

Green glazed and brown glazed sherds

C104

>25

Brownish fine sand, reached water level

FC1

N834,518 E813,869

1

C301

10

Surface layer

 

 

 

C302

20

Grayish clay with rubbles

 

 

 

C303

20

Yellowish clay

 

 

 

C304

>50

Grayish clay with rubbles

FC2

 

1

C305

30

Surface layer

 

 

 

C306

50

Light brownish clay

 

 

 

C307

10

Light brownish clay

FC3

N834,527 E813,648

1

C308

15

Surface layer

 

 

 

C309

25

Light brownish silty clay

 

 

 

C310

40

Grayish clay with rubbles

 

 

 

C311

20

Blackish mud

FC4

N834403 E813,874

1

C312

20

Surface layer

 

 

 

C313

40

Yellowish clay

 

 

 

C314

80

Reddish weathered clay

 

 

 

C315

>40

Yellowish weathered clay

FC5

N834,394 E813,919

1

C316

20

Surface layer

 

 

 

C318

>50

Yellowish weathered clay

FC6

N834,382 E814,034

1

C319

10

Surface layer

 

 

 

C320

80

Yellowish weathered clay

 

 

 

C321

10

Brownish weathered clay

FC7

N E

1

C322

10

Surface layer

 

 

 

C323

80

Modern fill

Porcelain sherds, lithic pieces, and plastic cord

12.4.5 The survey in the eastern section turned out to be quite fruitful. Many pieces of lithic and ceramic remains were spotted on the beach (Plate 4 - 1). A face cut (FC7) also yielded lithic and ceramic remains from the deposit just underneath the platform of the piggery (Plate 4 - 2). The lithic remains collected from this location almost all belong to a jue-ring making workshop, including pounding tools like hammer (Plate 4 - 3) and circular flakes as the waste or unfinished products from the process of ring making (Plate 4 - 4, 6 - 1).

12.4.6 The ceramic remains include both pottery ware with geometric patterns and the celadon stoneware of the Song dynasty. The types of surface treatment or decoration include coarse zigzag pattern (Plate 4 - 5), fine zigzag pattern (Plate 5 - 1), small trellis pattern (Plate 5 - 3), large trellis pattern (Plate 5 - 4), impressed square and raised dot pattern (Plate 5 - 5), incised square pattern (Plate 5 - 6), and the coarse cord-marked pattern (Plate 5 - 2).

12.4.7 Based on the characteristics of the geometric pattern, the stone and pottery remains can be dated to the late Bronze Age, or the Late Western Zhou and the Spring-and-Autumn period, from the 8th to the 5th century B.C. (6), (7).

12.4.8 The landform of the piggery is almost entirely covered by cement. Augering at some opening spots, however, yielded Song celadon pieces from almost every hole (Table 12.1, Plate 6 - 2, 3, 4, 5, 6). Therefore, this location can be identified as a site of cultural heritage with archaeological significance and named as "Ngau Hom Shek Beach Site". This site contains deposits of not only the Song Dynasty (A.D. 960-1279) but also the late Bronze Age.

Ngau Hom Sha Area (SWCLP OPTION C)

12.4.9 The location of Option C at Ngau Hom Sha area is mainly a small village surrounded by beach, ponds, and a river course (Plate 7 - 1, 2). No archaeological remains were found from the field walking and augering in this area.

Sheung Pak Nai Area (SWCLP OPTION D)

12.4.10 The location of Option D at Sheung Pak Nai is basically an unoccupied open area, composed of beaches, deserted fields, ponds, and hillfoot slopes (Plate 7 - 3). Six face cuts were made in this area but no archaeological remains were found from the underground deposits (Plate 7 - 4). The only significant antiquity from this location is a celadon ware of the Song dynasty collected from the beach (Plate 7 - 5).

Endorsed Alignment Option

12.4.11 As presented in Section 4, this EIA study is based on the s curve alignment endorsed by the Engineering Working Group in October 2001, which is Option AB as stated in para. 12.2.6

12.5 Mitigation Measures

12.5.1 Two sites of cultural heritage with archaeological significance have been identified at the locations within or very close to the direct impact zone of the proposed SWC alignment, including the selected landing point of SWC. These sites are the Ngau Hom Shek Beach Site and Ngau Hom Shek Hill Site. The SWC especially the location of pilecaps, should be designed to minimize the impact to the two sites of cultural heritage.

12.5.2 The western part of the Ngau Hom Shek Beach Site is located within the landing point of SWC. A rescue excavation therefore should be carried out in the project area with earthworks and building works in this area by the project proponent.

12.5.3 The site is now almost entirely covered by piggery and no further archaeological work can be conducted unless the piggery is moved. The total size of the rescue excavation at the Ngau Hom Shek Beach Site is tentatively estimated 100 square metres (pending the specific design of earthworks and building works). The average depth of cultural deposit is around 1.5 metres according to the result of augering on the site and estimated time length of field work for the rescue excavation is one month (based on an assumption that one qualified archaeologist, four experienced field assistants and eight labour workers are participated in the excavation; the same assumption is also taken for the estimation of the time length of field work on other sites).

12.5.4 The coastal area at Ngau Hom Shek to the immediate west to the proposed SWC landing point is not accessible for archaeological survey at this stage but the possibility of identifying significant archaeological remains can-not be completely excluded, although the archaeological potential of this area is low. It is recommended therefore to conduct a further archaeological survey and, if necessary, to carry out a rescue excavation in the project area with earthworks and building works by the project proponent after land resumption and before the commencement of the construction works of the site. The areas where archaeological survey, rescue excavation and protection from indirect impact are required in coastal Ngau Hom Shek are indicated by Figure 12.4.

12.5.5 The Ngau Hom Shek Hill Site of the prehistoric period is located slightly outside and above the impact zone of the proposed SWC landing point, the construction therefore will not cause direct adverse impact to this site (Figure 12.5). It should be emphasized, however, the site area should be protected from indirect impact potentially caused by the construction, and facilities and activities related to the construction should be kept away from this site during the entire process of construction.
12.5.6 For the area subject to landscape works in the Ngau Hom Shek Beach Site (including the west of the site if necessary) should be filled up to a level that the adverse impact of the proposed tree roots to the existing ground can be avoided. The remaining part of the site will not be directly affected by the construction but it should be protected from potential indirect impact with a layer of soil (30 cm minimum) covered on the top. See Figure 12.5A.

12.5.7 The archaeological survey and excavation at Ngau Hom Shek should be conducted by a qualified archaeologist who should apply for a licence under the provision of the A & MO (Cap. 53). The entire process of licence approval takes minimum two months after submitting the application. The programme and details of the archaeological works should be agreed with the AMO.

12.6 References

  1. Chiu S.T., "Major Achievements of Archaeological Rescue Projects Conducted in Hong Kong Since 1990", in C. T. Yeung and W. L. Li ed., Archaeology in Southeast Asia (Hong Kong: The University Museum and Art Gallery of the University of Hong Kong, 1995), pp. 397 – 428.
  2. 1997 ( ) ( 1997)
  3. 西 ( 1999)
  4. W. Meacham, Deep Bay Link Archaeological Impact Assessment (File in the Antiquities and Monuments Office, Hong Kong, 2000).
  5. 〈香 1999 6 26 – 41
  6. P. R. Rogers, N. W. Leininger, S. Mirchandani, J. V. D. Bergh and E. A. Widdowson, Tung Wan Tsai: A Bronze Age and Han Period Coastal Site, Antiquities and Monuments Office Occasional Paper No. 3 (Government of Hong Kong, 1995).
  7. :〈 〉,《 1998 7 17 – 30

Marine Archaeological Investigation

12.7 Introduction

12.7.1 In accordance with the 1998 EIA Ordinance (Cap. 499, S16), the Antiquities and Monuments Office (AMO) requested a Marine Archaeological Investigation (MAI) for the Shenzhen Western Corridor in Deep Bay. It is proposed to construct a bridge to link Ngau Hom Shek of HKSAR and Shekou of the PRC across Deep Bay. This study only covers the section of the route which will be in HK territorial waters. The aim of the MAI was to locate and access underwater archaeological resources within the designated study area.

12.8 Methodology

12.8.1 This study was undertaken using standard MAI techniques described below which follow the Guidelines issued by the Antiquities and Monuments Office (AMO).

Study Area

12.8.2 Figure 12.6 shows the location and boundaries of the study area. The survey area lies within Deep Bay, to the north west of Yuen Long. Oyster farming has taken place in this area for at least 200 years. It generally uses the bottom culture method with spat collected by laying old shells, rocks or concrete piles as clutches on the mud flat.

Baseline Review

12.8.3 A comprehensive review was carried out to determine the archaeological potential of the study area. This included examination of AMO records and archaeological and historical academic publications. A review was also made of geological publications.

Archive Search

12.8.4 All archives holding information on shipwrecks in Hong Kong waters were explored for relevant data.
Geophysical Survey

12.8.5 Geophysical survey is a standard and internationally accepted method for marine archaeological prospection (United States Department of the Interior, Minerals Management Services, 1991). It provides a fast and reliable method of assessing the seabed and sub-surface sediments for archaeological material. Combined with accurate position fixing, geophysical surveying facilitates detailed mapping of submerged cultural resources.

12.8.6 The archaeological objectives of the survey were:

· To map shipwrecks and anomalous seabed features.
· To determine the underlying significant geological horizons to assess the preservation potential of the seabed sediments.
· To map iron and/or steel debris and artefacts on the seabed.

12.8.7 The survey was divided into two phases. The first phase comprised the marine magnetic survey carried out between January 7th and 9th 2002. Due to the presence of an unaccepted level of persistent electro-magnetic interference by an unidentified transmitted signal at all times except between the hours of 01.30 and 05.30 the survey had to be carried out during that specified four hour period. The second phase comprised echo sounding, seismic profiling and side scan sonar surveys was carried out on the 14th, 16th and 18th January 2002.

Equipment

12.8.8 The following equipment was mobilized and used for this survey:

The EGS seismic profiler (high resolution boomer)
EGS hyrdophone &EG&E seismic power pack
Waverley thermal recorder
EGS signal processor
360 Octopus processor

Klein Side Scan Sonar System 2000

Marine Magnetics, Seaspy Marine Overhauser Magnetometer
Computer for magnetic data logging
GEM, GSM-19 Overhauser magnetometer

Sercel NR103 10-channel DGPS receiver
Sercel DGPS NDS200 HF transmitter
The EGS computerised navigation system

Class III commercially licensed survey vessel
Locally licensed sampan

Horizontal Location Control

12.8.9 The survey vessel was located by a Different Global Positioning System (DGPS) unit using different corrections broadcast from the government RTCM transmitter. The GPS requires a minimum of four satellites to calculate the position of the survey vessel. However, the US military has injected an error in the pseudo range of each satellite. The differential signal was used to correct the pseudo ranges of the satellites received on each survey boat.

Computerised Navigation

12.8.10 The EGS Computerised navigation system was added to the positioning system to control the steering of the boat along the traverse specified, and to log all horizontal and vertical control data.

12.8.11 This system provides the user with a dynamic analogue and digital screen display on which the following area continuously updated:

· Skewed grid set parallel to the desired line direction
· Boat position
· Water depth in metres
· Date and Time
· DGPS diagnostics enabling quality control

12.8.12 Other information displayed for the assistance of the hydrographic surveyor includes, course, speed, fixing status and co-ordinates on the Hong Kong Metric Grid as well as a number of other user defined options for marine surveying. The system includes a multiple instrument "fixing" unit, and permits the operation of a printer, a plotter and a heave compensator for removing the effects of waves from echo sounding records.

Shore Control

12.8.13 The government DGPS transmitter was located at Kau Yi Chau which is shown on Figure 12.6.
Calibration, Accuracy and Quality Assurance

12.8.14 The positioning system was calibrated by checking the co-ordinates displayed by the navigation system at a previously co-ordinated point. An accuracy of ± 2m or better was ensured by carrying out the above quality assurance checking procedure at the beginning of each survey day.

Echo Sounder Calibration

12.8.15 A temporary pressure tide gauge was set up next to the pier at Ash Lagoon and levelled into Principal Datum Hong Kong (PD) from a previously co-ordinated and levelled control point. The temporary bench mark (TBM) of the tide gauge had a level of 5.720m above PD. All tide data was reduced to Principal Datum, Hong Kong (PD).

12.8.16 The datum at use or implied in HK are as follows:

------------------------------------------------------ MSL

1.27m

(An admiralty value)

------------------------------------------------------ Principal Datum (PD)

0.146

(SMO definition)

------------------------------------------------------ Chart Datum (CD)

0.2m

Astronomic component

------------------------------------------------------ LAT

0.25m rms

Meterological component (EGS value)

------------------------------------------------------ (-0.45m CD, -0.68m PD)

 

SMO = Survey and Mapping Office, Hong Kong Government

 

Survey Vessels

12.8.17 A Class III commercially licensed survey vessel was used to carry out the offshore survey. As the water depth is very shallow over the inshore part of the survey area, a motorised sampan was used for the survey. During the survey period carried out by this motorised sampan, the Class III commercially licensed survey vessel was in attendance to comply with Marine Department regulations.

Coverage

12.8.18 Survey lines were run as illustrated on the Hydrophone Track Plots. The presence of an Oyster bed, approximately from 814210E 836650N to 814455E 836325N within the survey corridor prevented complete adherence to the survey plan.

Echo Sounding, Seismic Profiling and Side Scan Sonar

12.8.19 The seismic profiling and side scan sonar surveys were carried out simultaneously with the echo sounding survey. A swell filter was used to minimise the effects of surface waves from the seismic records. The side scan sonar was towed from the stern of the survey waves at a depth of 5 to 15m below the sea surface. Data was recorded digitally throughout the seismic survey. The echo sounder was calibrated using the 'bar check' method at the start and end of every survey period.

Land Based Magnetic Survey

12.8.20 The land based, GSM-19 magnetometer was set up at Chung Hom Kok prior to the commencement of the survey. The magnetic field base readings from 15:46:32 on the 7th January 2002 to 11:11:41 on the 11th January 2002 were averaged to generate a background value of 44850.2nT. The base and survey magnetometer crystal controlled clocks were synchronised with reference to the GPS time at the commencement of the survey and compared for relative drifts at the completion.

Marine Magnetic Survey

12.8.21 The depth sensor inside the marine magnetometer was calibrated at the commencement of the survey. The recording parameters for the survey were listed as follows:

Vessel Speed : 2.0 - 3.0 m/sec
Fix Interval : 10 seconds
Fish Depth : 2m
Sampling Rate : 1 reading per second (Marine Magnetometer)
Sampling Rate : 1 reading per 3 seconds (Chung Hom Kok Base station)

Site Safety

12.8.22 Safety was generally in accordance with the 'Marine Geophysical Operations Safety Manual' (International Association of Geophysical Contractors, Eighth Edition, 1996).

12.9 Results - Baseline Review

Physical Setting

12.9.1 Located in the northwest New Territories near the market town of Yuen Long, the shores of Deep Bay were historically one of the major agricultural areas of the north-western New Territories. The shallow soils surrounding Deep Bay in the salt-marches close to the sea produced a form of brackish-water paddy crop in the late summer. This rice was considered somewhat inferior to better quality rice produced from more fertile inland valleys such as Lam Tsuen and Shatin (Ng, 1983).

12.9.2 In Deep Bay, the coastal area comprises low-lying mud flats and brackish back swamps. Mangroves fringe the shoreline and the mud flats are a refuge for a wide variety of crustaceans and other fauna. This is the most important mangrove habitat in the HKSAR (Pitman & Peck, 1979). Figure 12.7 presents a photograph of the Study Area. Accretion of sediment around the roots of the mangroves had led to a seaward extension of the mud flats and consequently of the coastline. To the south and around the southern coast of the western New Territories, the granite of the Castle Peak area forms a shoreline of low relief.

12.9.3 The English name, Deep Bay, is a complete misnomer as the water depth nowhere exceeds 5m. The bay is a sheltered inlet on the eastern margin of the Pearl River estuary with a relatively large catchment. Inter-tidal channels occur along the eastern margin of the bay where the Sham Chun and Shan Pui rivers discharge fresh water into the bay. Both the northern and southern shores of Deep Bay comprise mud flats with mangroves. Prior to reclamation and the creation of ponds in the last century, the mudflats were more extensive and tidal influences extended further inland.

Fisheries and Oyster Production

12.9.4 Fishing and oyster gathering were major local activities since the mid-eighteenth centuries. Both were undertaken mainly by boat-dwelling Tanka and Hoklo people. Later there was considerable Punti participation in the oyster industry. Land-dwellers were also involved in coastal fisheries, either as part-time stake-net operators along the coastline or full-time as brackish or fresh-water pond-fish farmers. There were two thousand six hundred and ninety eight mu of fishponds registered in what was then Sun On County in 1898; many of them were in the Deep Bay area of the county (Ng, 1983). Fish-ponds were a major source of revenue in the Deep Bay area until the late 1980s, when many were filled in to provide building land.

Rural Reclamation Work

12.9.5 Gei-wai or salt-paddy reclamation is the process of reclaiming small piecemeal areas of land from shallow tidal areas by means of small weirs. These are flushed out over a period of several years with fresh water and then planted with varieties of salt-resistant paddy. The land gradually silts up and raises slightly above sea level. This process was undertaken in slow phases by local farmers for generations along the shores of Deep Bay and work was done on a more extensive basis in the 1920s and 1930s.

12.9.6 The shoreline of Deep Bay has, as a direct result of gei-wai reclamation, moved steadily northwards over the past century. Yuen Long's Tin Hau Temple, which dates in parts from the mid-eighteenth century, was once located on or very near the Deep Bay coastlines. It is now over 2 kilometres inland. From this, we can infer that any archaeological resources that may have once been in its vicinity are both deep underground and very far from the sea.

European Contacts in the Vicinity of Deep Bay

12.9.7 Reference has been made in numerous works to the early Portuguese presence at Lintin and Tuen Mun, both in the immediate vicinity of Deep Bay. Castle Peak and the Deep Bay are clearly marked in O Livro de Francisso Rodrigues, (Translated by Armando Cortesão as The Suma Oriental of Tomé Pires and The Book of Francisco Rodrigues, originally written in 1514). Rodrigues was a very early Portuguese pilot, cartographer and captain who sailed in these waters in the early sixteenth century. Rodrigues was one of the commanders of Alvares' flotilla during the voyage from Malacca to China in 1517-18 (Braga, 1955). His book contains the earliest maps of the region drawn by a European based on actual observation and by men who were familiar with the places depicted. Maps tended to be approximations drawn from the oral accounts of professional pilots, usually Malay or Arab.

12.9.8 The first European navigator known to have reached the China coast, a Portuguese named Jorge Alvares, made his landfall in 1513. Alvares commenced his mission in Malacca, now in Malaysia, that the Portuguese had captured in 1511. Merchandise brought from China, especially porcelain, fetched extremely high prices, with good quality ceramics fetching twice their own weight in silver when re-sold at Goa. Instead of relying on Chinese traders, the Portuguese intended to establish a sea-route to China, and purchase for themselves at source.

12.9.9 Most of Alvares' time was spent at Lintin, the island known as "Solitary Nail" in Chinese due to its sharply-pointed shape. The Portuguese named it Tamão - an approximation of the Cantonese pronunciation of Tune Mun - and the name appears on many old maps of the Pearl River Delta.

12.9.10 The local mandarinate at Tuen Mun anchorage, headquartered at nearby Nam Tau, received the Portuguese in a friendly manner and trade commenced. A padrão or stone carved with the Portuguese cross and crest was erected by Alvares at Lintin, though nothing of it now survives. These stones functioned more a marker of passage for later seafares than as a territorial claim, and were erected wherever the Portuguese mariners sailed, from Mombasa and Ormuz to western India and the Moluccas. They are still to be seen in some of these places today. Alvares' young son accompanied him on the voyage from Malacca, but he died at Lintin and was buried at the base of the padrão erected by his father.

12.9.11 Alvares' flotilla remained at Tuen Mun for ten months, only returning to Malacca when the south-west moonsoon winds permitted them to sail. While not given much freedom of movement, traders visiting Lin Tin were not as closely confined by the Chinese authorises as in later centuries, and while it is not recorded, it is highly likely that Alvares and his men visited the nearby mainland.

12.9.12 Jorge Alvares made two more voyages to China, in 1519 and again in 1521. On what was his last voyage to the China coast, he died on 8th July 1521 and was buried beneath the padrão in the same spot at his son (Braga, 1955). No trace reamins today of the padrão. Ming Dynasty gazetteers record that in 1516 the Portuguese came again, but the Chinese under Naval Commander Wang Hong defeated them (Ng, 1983).

12.9.13 Later the Portuguese navigator Simao Peres de Andrade built a fort in the neighbourhood of Deep Bay without receiving permission from the Chinese to do so; in 1521 a Chinese naval force attacked the Portuguese in the vicinity and defeated them (Ng; 1983).

Shipwreck Data

12.9.14 Practically nothing is known about the archaeological potential of the seabed deposits in Hong Kong. The only marine archaeological discovery was that of a later Sung/early Ming Dynasty boat exposed during construction of the High Island Reservoir, near Sai Kung (Frost, 1974). Since then, no other historic shipwreck has been found. However, this is probably because there were no dedicated marine archaeological surveys until the introduction of the 1998 EIA Ordinance. Marine archaeology is therefore a new area of study in Hong Kong with very little baseline data to draw upon.

12.9.15 Formation of archaeological sites underwater is mainly due to shipwrecks (Muckelroy, 1978). Since these are random and haphazard events it is difficult to predict their exact location if no written references survive. A combination of geophysical and diver survey is the most efficient method for accurately assessing the archaeological potential of the seabed.

12.10 Results - Archieve Research

12.10.1 The UK Hydrographic Office (UKHO) holds a database of surveyed shipwrecks in Hong Kong, including those not shown on Admiralty Chart. There are no records of shipwrecks within the study area.
12.10.2 The UKHO holds navigation charts of the study area dating from 1899 (Figure 12.9) and 1966 (Figure 12.10). These charts are particularly useful as they may show wrecks which have been subsequently buried or broken up. A 1905-1909 British War Office Map (Figure 12.11) is also of interest as it shows the lack of development in the study area.

12.11 Geophysical Survey

Sounding Data

12.11.1 The influence of wave action was corrected in real time by the motion compensator. The smoothed sounding data was then digitized and reduced to levels below Principal Datum using the measured tidal data.
Interpretation of Side Scan Sonar Records

12.11.2 Processing was carried out as follows by EGS:

· The features in the side scan sonar data were interpreted and marked on the original records.
· The marked features were digitised and plotted.
· The plotted seabed features were checked by reconciling them with the original raw side scan records, and making final adjustments accordingly.

12.11.3 The side scan sonar records collected over this area are characterised by the following features:
· Extensive areas of oyster beds (Figure 12.12)
· Man made navigation channel
· Trawl marks (Figure 12.13)
· 5 Unidentified Features (Figures 12.14 to 12.17)

12.11.4 All of the features listed above are drawn onto the Seabed Features Charts. Figure 12.18 shows the distribution of the 5 Unidentified Features. There are no magnetic anomalies associated with the Unidentified Targest.

Interpretation of Seismic Profiler Records

12.11.5 Record quality was slightly affected over parts of the survey area by masking within and close to the base of the Marine Deposits: such masking is probably caused by the anaerobic decomposition of organic plant material trapped during a recent inundation. The gas bubbles so generated absorb the seismic energy, thereby preventing reflections from deeper horizons (Permchitt et al, 1992). In this case, such masking was both patchy and variable, enabling the identification of deeper horizons on th seismic records in many of the affected areas.

12.11.6 Processing was carried out as follows by EGS:

· Seismic horizons were selected for interpretation and highlighted on copies of the seismic records during preliminary interpretation.
· All traverse crossing points were then calculated and marked on the seismic records.
· The records were then physically correlated at all of these points.
· The correlated horizons were digitized, plotted and contoured.
Seabed Stratigraphy

12.11.7 Geophysical survey data enables (1) establishment of the marine geology and (2) the archaeological preservation potential of the marine sediments. The survey data will be discussed in terms of its relevance to the marine archaeological investigation; it is not appropriate to provide full geological analysis of the study area.

12.11.8 The following general succession applied to this area:

-------------------------------------------------------------------- Sea Bed

RECENT DEPOSITS

(soft silty CLAY or clayer SILT)

-------------------------------------------------------------------- Base of Recent Deposits

MARINE DEPOSITS (Hang Hau)

(Soft Silty CLAY, or clayer SILT, sometimes

Sandy at the base and close to the shorelines)

-------------------------------------------------------------------- Base of Marine Deposits

ALLUVIUM (Chek Lap Kok)

Silty, Sandy CLAY or fine to medium

Silty SANG

--------------------------------------------------------------------- Top of Grade V Rock

DECOMPOSED ROCK

Silty fine/coarse SAND, fine sandy SILT

--------------------------------------------------------------------- Top of Grades III Rock

GRADE III ROCK

Pink, grey coarse grained granite

---------------------------------------------------------------------

12.11.9 Based on data from numerous borehole records and seismic surveys of the territory, the Hong Kong Geological Survey has mapped the offshore surficial deposits (e.g. Strange & Shaw, 1986; Langford et al, 1989; Strange et el, 1990). The subdivided the submarine deposits into two formations: the pleistocene Chek Lap Kok Formations (alluvial/estuarine deposits) and the overlying Holocene Hang Hau Formation (marine muds). Recently, the stratigraphy has been refined Fyfe et al., (1997; 1999), and the Sham Wat (~130-85 ky.) and/or Waglan (~85-10 ky.) Formations (both shallow marine units) are sometimes positioned between the Chek Lap Kok and Hang Hau Formations.

The Hang Hau Formation

12.11.10 From an archaeological perspective, the marine sediments assigned to the Hang Hau Formation have the greatest archaeological potential.

12.11.11 The formation consists of relatively homogenous very soft to soft, greenish grey silty clay (Fyfe et al., 1997) and has a high moisture content. Therefore, the Hang Hau Formation sediments potentially provide an excellent substrate for the preservation of archaeological material. Additionally, the soft nature of the sediments would make it possible for archaeological material to be buried within the formation, where it would have greater protection than if it were exposed on the seabed.

12.11.12 Throughout the study area the Hang Hau formation is more than 2m thick.

Sham Wat and Chek Lap Kok Formations
12.11.13 The (pre-Holocene) sedimentary formations beneath the Hang Hau Formation in the study area are considered to offer limited to negligible archaeological potential. The original land surface that would have existed at the top of these deposits would have been planned-off during the Holocene marine transgression (Fyfe, 1997). Thus, any evidence of late Pleistocene human land activity in the study area would almost certainly have been removed. In the area of Deep Bay the sedimentary character of the Chek Lap Kok Formation has been controlled by the outflow of the Shenzhen River, a tributary of what is now the Fanshi Channel of the Pearl River Estuary.

12.11.14 There was a cluster of diffraction hyperbole on the northern edge of the navigation channels. After cross referencing them with the Seismic Profiler data it was concluded that these were caused by the presence of coarser seabed sediments rather than buried objects.

Marine Magnetic Data

12.11.15 Corrections were made to the magnetic data to compensate for the influences of diurnal magnetic drift. The natural diurnal variations of the magnetic data were recorded continuously throughout the survey periods by a static magnetometer base station established at a quiet place in Chung Hum Kok.

12.11.16 Base station records were subjected to a de-spiking filter process to remove any spurious transients from the data sequence and the temporal variations of the earth's magnetic field were expressed relative to a base datum level of 44000nT. Those variations were subtracted from the survey recorded data a eliminate or minimize the component of diurnal variation in the survey data values.

Sounding Plan

12.11.17 The seabed is extremely shallow within the survey area with depths ranging from 0m PD and -2m PD. A small channel was mapped during the survey and it was detected by both the echo sounding and side scan sonar surveys.

12.12 Barging Point

12.12.1 A barging point is proposed at Lung Kwu Sheung Tan (see Figure 6.3) for batching plant and storage. It was discussed and agreed with AMO that the piece of land on reclamation area could be used for works activities involved piling work. Details of the piling works should be sent to AMO for comment prior to works commencement.

12.13 Conclusion

12.13.1 The study area is exceptionally shallow with water depths ranging from 0m - 2m PD. The geophysical survey revealed that significant areas of the seabed are covered with oyster beds. These serve to cover the natural seabed and prevented archaeological investigation. A navigation channel has been dredged across the northern section of the study area.

12.13.2 The side scan sonar data also revealed that the seabed within the study area has been extensively trawled. This is indicated by the presence of a large number of deeply incised linear scars on the seabed. The trawling activities may have served to destroy or redistribute archaeological material, if present, thereby reducing the archaeological potential of the study area.

12.13.3 There are five Unidentified Targets within the southern section of the study area and it is not possible to assess their archaeological significance until without a diver inspection. All the five targets are found to locate outside the 100m EIA Study zone (refer Figure 12.19) as stipulated in the EIA Study Brief Clause 3.4.8.3. The proposed works limit is at 50m from both edges of the SWC footpath and therefore there will be no physical disturbance due to the proposed marine works of the project to the seabed of the 5 sites. Therefore, diving inspection is not recommended for all the five unidentified objects.

12.13.4 Diver inspection is required to cover the area with no data due to extremely shallow water or oyster beds within the strip of land directly beneath SWC footprint, refer Figure 12.20. The visual inspection of the archaeological investigation area found no visible features or objects of archaeological interest. This may be result of the high sedimentation rates in the area and the extensive disturbance to the seabed from fishing and oyster cultivation or the absence of archaeological features. Due to the proximity of the archaeological investigation area to the coastal archaeological site, Ngau Hom Shek, the potential for such resources cannot be ruled out. However, based on the visual inspection and review of the site conditions, the archaeological potential is classified, as very low and further investigation is not recommended.

12.13.5 It was proposed that a batching plant may be required at Lung Kwu Sheung Tan barging point. The contractor shall submit the proposal to AMO for agreement prior to any works commencement.

12.14 References

Braga, J. M 1995. China Landfall 1513. Jorge Alvares Voyage to China. A complilation of some relevant material. Macao. Imprensa Nacional.
Cortesao, A. 1994. The Suma Oriental of Tome Pires and The Book of Francisco Rodrigues. London. Hakluyt Society.
Davis, D. W., Sewell, R. J. & Campbell, S. D. G. 1997. U-Pb dating of Mesozoic igneous rock from Hong Kong. Journal of the Geological Society of London, 154: 1067-1076.
Davis, S. G. 1949. Hong Kong in its Geographical Setting. London. Collins.
Frost, R. J. 1974. Sha Tsui, Hing Island. Journal of the Hong Kong Archaeological Society, 5: 23-30.
Fyfe, J. A., Shaw, R. 1997. The Offshore Geology of Hong Kong. Hong Kong Geological Survey Memoir. Geotechnical Engineering Office, Hong Kong.
Hong Kong Environmental Protection Department. 1997. Technical Memorandum on Environmental Impact Assessment Ordinance Process (EIA Ordinance, CAP. 499, S16). Hong Kong Government Press.
Landford, R. I., Lai, K. W., Arthurton, R. S. & Shaw, R. 1989. Geology of the Western New Territories, 1:20000 Sheets, 2, 5 & 6. Hong Kong Geological Survey Memoir No. 3. Hong Kong Government Printer.
Lo, H. L. 1963. Hong Kong and its External Communications Before 1942. Institute of Chinese Culture, Hong Kong.
Muckelroy, K. 1978. Maritime Archaeology. Cambridge University Press.
Ng, P. Y. L. 1983. New Peace County: A Chinese Gazetteer of the Hong Kong Region. Hong Kong. Hong Kong University Press.
Pickman, J. & Peck, R. 1979. The physical nature of the Hong Kong seashore. In: B. Morton (ed) The Future of the Hong Kong Seashore. Oxford University Press, Hong Kong.
Premchitt, J., Rad., N. S., To, P., Shaw, R. & James, J. W. C. 1992. A Study of Gas in Marine Sediments in Hong Kong. Continental Shelf Research, 12: 1251-1264.
Sayer, G. R. 1975. Hong Kong 1862-1919. University Press, Hong Kong.
Strange, P. J. & Shaw, R. 1986. Geology of Hong Kong Island and Kowloon. Hong Kong Geology Survey Memoir, No. 2. Geotechnical Control Office.
United States Department of the Interior, Minerals Management Service, Gulf of Mexico OCS Region. 1991. Requirements for Archaeological Resource Field Surveys.