See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/238371028

Evolution of the Sungei Buloh–Kranji mangrove coast,

Singapore

Article in Applied Geography · July 2004

DOI: 10.1016/j.apgeog.2004.04.002

CITATIONS READS

36 2,501

5 authors, including:

Michael I Bird Stephen Chua

James Cook University Nanyang Technological University
437 PUBLICATIONS 26,065 CITATIONS 16 PUBLICATIONS 177 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Michael I Bird on 11 January 2018.

The user has requested enhancement of the downloaded file.

 Applied Geography 24 (2004) 181–198
www.elsevier.com/locate/apgeog

Evolution of the Sungei Buloh–Kranji

mangrove coast, Singapore
Michael Bird a,, Stephen Chua a, L. Keith Fifield b,
Tiong Sa Teh a, Joseph Lai c
a
National Institute of Education, Nanyang Technological University, 1 Nanyang Walk, Singapore 637616
b
Research School of Physical Sciences and Engineering, Australian National University, Canberra,
ACT 0200, Australia
c
Sungei Buloh Wetland Reserve, 301 Neo Tiew Crescent, Singapore 718925

Abstract

The mangroves from Sungei Buloh Wetland Reserve to Kranji Dam represent the largest
intact mangrove forest left on mainland Singapore. Mangroves colonized the area around
6820 BP, as sea level rose following the last glacial maximum and a variable thickness of
Holocene sands, muds and peats (generally ~1 to >3.8 m thick) were deposited over the pre-
transgression land surface.
An analysis of a time series of photographs covering the period from 1946 to 2001 has
revealed major changes in the distribution of mangroves in the area resulting from develop-
ment-induced changes in the local hydrodynamic regime and clearance for aquaculture.
Mangroves covered 117.3 ha in the study area in 1946 and were actively advancing over the
coastal mudflats until 1980. Despite the addition of 6.24 ha from mangrove colonization, the
total area covered by mangroves was reduced by ~50% by 1980 due to clearance for aqua-
culture. Following 1980, a reduction in sediment supply possibly due to the construction of
the Kranji Dam, immediately east of the study area, led to the initiation of erosion along
much of the coastline, with the mangrove fringe having retreated by up to 50 m in 2001.
Establishment of the wetland reserve in 1992 enabled the partial regeneration of mangroves
in the area to 86.8 ha, 25% less than in 1946.
Three areas of undisturbed old growth mangroves >55 years in age have been identified
and are considered to be of high conservation value. Two of these areas are within the cur-
rent boundaries of the Sungei Buloh Wetland Reserve, but are located along coastal areas
that are undergoing severe erosion. The third area is located in the south of the study area,
protected from coastal erosion, but outside the current nature reserve boundary and hence is


Corresponding author. Present address: School of Geography and Geosciences, University of St.
Andrews, St. Andrews, Fife, KY16 9AJ, UK.
E-mail address: [email protected] (M. Bird).

0143-6228/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apgeog.2004.04.002
182 M. Bird et al. / Applied Geography 24 (2004) 181–198

susceptible to loss as a result of future development. This third area is possibly the oldest
undisturbed fragment of mangrove forest on mainland Singapore.
# 2004 Elsevier Ltd. All rights reserved.

Keywords: Mangroves; Remote sensing; Singapore; Holocene; Environmental management

Introduction

Singapore has undergone a period of rapid economic growth since the Second
World War. Increasing prosperity, along with increasing population, has resulted in
large changes to the physical environment. From the 1960s onwards, Singapore
entered a phase of large-scale land conversion for housing, industry and water storage
(Chua & Edwards, 1992). In addition, the area of the country has grown from 580 km2
prior to 1960, to 645 km2 currently, as a result of reclamation of shallow coastal water
areas. Reclamation has entailed the complete destruction of large stretches of natural
coastline and the destruction of the littoral ecosystems once associated with these
coasts. As a result of land conversion and coastal reclamation, the mangrove areas
that originally covered 13% of Singapore, now cover 0.5% (Ng & Sivasothi, 1999).
The mangroves that remain in Singapore are located in isolated patches on the
sheltered northern coast of the main island, and on the offshore islands of Pulau
Tekong and Pulau Ubin to the northeast, and Pulau Pawai and Senang to the
south (Fig. 1). Most of these remaining areas have been subject to some disturb-
ance in the last 30–40 years, especially to make way for the development of fish
and prawn farms. Upon abandonment of these enterprises, mangrove communities
have been able to re-establish themselves naturally to various degrees in some
areas. In recent years, there have also been efforts to actively promote the growth
of mangrove communities on reclaimed land at Pasir Ris in the northeast of the
mainland and on Pulau Semakau, an island that also serves as a major refuse dis-
posal site to the south of the mainland.
The number of mangrove areas that enjoy some measure of legislative protection
from development has decreased over the last few decades. In the Singapore Green
Plan 1992, mangrove areas at Mandai and Khatib Bongsu on the northern coast of
the main island were listed as conservation sites. In the Singapore Green Plan
2012, released in 2002, the status of both these areas was reduced to warranting
conservation ‘for as long as possible’, with the likelihood that both will be con-
verted for housing and water storage uses in the future. Such development would
mean that the mangroves of Sungei Buloh and Pulau Ubin will be the only sub-
stantial, publicly accessible natural or semi-natural areas of this ecosystem left in
Singapore. A positive development at Sungei Buloh was the announcement in late
2002 that a nature trail will be developed between Kranji Dam and the Sungei
Buloh Wetland Reserve (SBWR; Fig. 1), with the prospect that mangrove areas
along this trail may be incorporated into the existing reserve.
This study examines the geomorphic history and vegetation dynamics of the man-
grove dominated coast from Sungei Kranji, along the route of the new nature trail
from Kranji Dam, to the western end of the SBWR. As one of the few remaining
 M. Bird et al. / Applied Geography 24 (2004) 181–198 183

Fig. 1. Location of the Sungei Buloh Kranji coast.

mangrove areas in Singapore that enjoys a measure of protection as a conservation

area, an understanding of the environmental history of the area can serve as a guide in
management decisions aimed at ensuring that the area does not undergo further degra-
dation. In particular, an understanding of the environmental history can serve to ident-
ify less disturbed areas of high conservation value and areas that may be in danger of
loss through coastal erosion of vulnerable unconsolidated sediments. In addition, the
study can serve as a benchmark against which future changes can be assessed.
While there are several studies of mangrove change on decadal timescales based
on time series of maps and remotely sensed images (Dahoud-Guebas, 2002;
Manson, Loneragan & Phinn, 2003; Sulong, Mohd-Lokman, Mohd-Tarmizi &
Ismail, 2002; Panapitukkul, Duarte, Thampanya, Kheovongsri, Srichai, Geertz-
Hansen, Terados & Boromthanarath, 1998; Vasconcelos, Mussa Biai, Araujo &
Diniz, 2002; Verhheyden, Dahdouh-Guebas, Thomaes, de Genst, Hettiarachchi &
Koedam, 2002), few studies have examined environmental change both on millen-
nial timescales and decadal timescales.

Study area

Natural environment
The climate of Singapore is dominated by the northeast monsoon from
November to March and the southwest monsoon from June to September. The
184 M. Bird et al. / Applied Geography 24 (2004) 181–198

Fig. 2. Map of the sample area from aerial photography in 2001 and DGPS mapping in 2003.

v
mean annual temperature is 26.9 C with little monthly variability, and the mean
annual precipitation is 2300 mm.
The study area is located on the northwest coast of the main island of
Singapore, facing the Straits of Johor that separate Singapore from peninsular
Malaysia (Fig. 2). Wave fetch is generally short and the directions of maximum
fetch rarely coincide with those of the strongest winds, resulting in an average
breaker height of less than 20 cm (Chia, Khan & Chou, 1988). Tidal energy is low,
partly due to the construction, by 1913, of a causeway between Malaysia and Sin-
gapore, 4 km to the east of the study area, which restricted water movement
through the Straits of Johor (Lim, 1983; Lye, Lim, Sieh & Midun, 1991). The tides
are semidiurnal with a mean spring tide range of 2.4 m and a mean neap tide range
of 1 m (Wong, 1992).
The entire study area is underlain by ‘Old Alluvium’, a partly lithified and heav-
ily weathered sequence of fluvial sands and clays, probably dating from early Pleis-
tocene or Pliocene times (Gupta, 1986; Gupta et al., 1987; Public Works
Department, 1976). The Old Alluvium locally forms low hillocks up to 5–10 m
above the current intertidal zone inland from the coast.
During the mid-Holocene sea level ‘highstand’ (Geyh, Kudrass & Streif, 1979;
Hesp, Chang, Hilton, Ming & Turner, 1998), muddy, organic-rich sediments of the
transitional member of the Kallang Formation were deposited over the Old Alluv-
ium up to 4–5 m above modern sea level and these deposits have continued to
 M. Bird et al. / Applied Geography 24 (2004) 181–198 185

accumulate in the intertidal zone to the present day. Prior to development of the
area in the 1960s, the entire coastline in the study was fringed by a mangrove zone
generally a few hundred metres wide, fronted by a well-developed low-tide plat-
form of mud and peat to seaward, a few tens of metres wide. A network of large
and small tidal channels drained from the mangroves inland to the coast. The
major creek, Sungei Buloh, was flanked for several kilometres upstream by a fringe
of mangrove forest generally less than 100 m wide.

Sungei Buloh Wetland Reserve

The Sungei Buloh area was first gazetted as a 457 ha, dominantly mangrove, for-
est reserve in April 1890 (Tan et al., 1997). Though the area ceased to be a forest
reserve in April 1938, the Japanese occupation of Singapore from 1942 precluded
any significant development in the area until after the cessation of hostilities in
1945. While large areas above tidal influence were cleared for smallholder agri-
culture and coconut plantations prior to World War II, the pace of development in
the mangroves was slow until the 1970s when large areas were cleared and enclosed
by earth bunds for aquaculture, leaving only a narrow vegetation fringe to seaward
of the bunds. Drainage from the former mangrove flats was thereafter confined to
sluice gates.
A further major development was the damming of the Sungei Kranji to form the
Kranji Reservoir, completed in 1975. As part of this construction the mangrove
areas flanking Sungei Kranji were reclaimed and areas upstream of the dam wall
were inundated. The dam is likely to have changed the local hydrodynamic regime,
limiting freshwater inflow to short intense intervals when excess water is released
from the dam and also limiting sediment supply to the coast.
The Singapore Government took over the Sungei Buloh area, including the island
of Pulau Buloh and former prawn farming areas in 1989, and the SBWR, compris-
ing 87 ha of ponds, mangroves and terrestrial forest was opened in December 1993.
A series of nature trails now follow the line of the former bunds of the prawn ponds
and the formerly cleared prawn ponds are being actively reafforested, though clear-
ing of back mangrove areas for aquaculture is still occurring in areas outside the
SBWR boundary.
The reserve is home to about 250 species of vascular plants with the major man-
grove tree species being Avicennia alba, A. officinalis, A. rumphiana, Bruguiera
cylindrica, B. gymnorhiza, Rhizophora apiculate, R. mucronata, and Sonneratia alba
(Tan et al., 1997). Due to the long disturbance history, there is usually little evi-
dence of a zonation of species with respect to inundation frequency. In December
2002, the reserve was inducted into the East Asian–Australasian Shorebird Site
network by Wetlands International, in recognition of the fact that approximately
50,000 birds use the SBWR as a stopover in their annual migration. Approximately
80,000 people visit the reserve annually.
186 M. Bird et al. / Applied Geography 24 (2004) 181–198

Research methods

The positions of the modern mangrove fringe, dead mangrove patches, core
locations, trails, roads, survey benchmarks, sluice gates and other georeference
points broadly distributed across the study area were obtained using a Trimble
Pathfinder Pro XR differential GPS receiver and a TSC1 Data-logger. This instru-
ment has a horizontal accuracy of approximately 0.5 m when in locations cap-
able of receiving the differential GPS signal broadcast from Raffles lighthouse.
Where elevation data were required, or tree cover precluded the use of GPS,
manual levelling was carried out using an autolevel and 5 m staff, with reference to
Precise Level Bench Mark No. 439 (4.100 m above mean sea level) maintained by
the Singapore Land Authority of Singapore on the eastern end of Kranji Dam
(Fig. 2).
Reconnaissance augering of the Holocene sediments was carried out to 3.5 m
depth and coring was carried out using a modified Livingstone piston corer with a
5 cm diameter bore. The augering sites were chosen to maximize the chance of pen-
etrating the full thickness of the Holocene sequence on Pulau Buloh, the coring
sites were chosen to form a transect from the seaward low-tide mud platform,
through the mangroves to the back mangrove area in order to ascertain the thick-
ness of Holocene sediments underlying the modern mangroves. Cores were
collected in 1 m lengths and extruded at the collection site into half-sections of
PVC tube. The cores were wrapped to prevent moisture loss and transported to the
laboratory where they were cut into 10 cm lengths and weighed to enable the cal-
culation of bulk density. Each 10 cm length was then cut lengthways to facilitate
description of the colour, texture and sedimentary structures prior to sub-sampling
for further analysis.
A representative aliquot of each core sample was weighed and dried overnight in
v
a furnace at 60 C to obtain water content and allow the calculation of dry bulk
density. The dried samples were crushed to a powder and analysed for total carbon
content and total inorganic carbon content using a Shimadzu 5000A carbon analy-
ser. The total organic carbon (TOC) content of each sample was obtained by sub-
tracting the measured TIC from measured TC contents. Total inorganic carbon
content was very small for all samples analysed. A separate aliquot of each sample
was weighed and then wet sieved at 63 l, with the >63 lm fraction dried and
reweighed to obtain information on sediment texture.
Samples of macroscopic woody material were hand-picked and given an acid–
base–acid pretreatment before analysis by accelerator mass spectrometry using the
14UD accelerator in the Department of Nuclear Physics at the Australian National
University (Table 1). Radiocarbon ages (years BP) were converted to calendar ages
(cal BP) using Calib 4.4 (Stuiver, Reimer & Braziunas, 1998).
Remote sensing

Four aerial photographs for the periods 1946, 1969, 1980 and 2001 were selected
for the analysis of mangrove change, with a time interval between successive
 M. Bird et al. / Applied Geography 24 (2004) 181–198 187

Table 1
Radiocarbon dates on samples from cores SBU-1 and SBU-3
Lab code Sample Elevation d13C 14C age Error Calibrated age Probability
(ANUA-) name (m rsl) (%) (years BP) (years) (years cal BP)
22729 SBU 1-1-3 0.9 28.8 6400 190 7480–7160 0.86
22728 SBU 1-1-7 1.33 28.3 6350 190 7430–7140 0.75
22727 SBU 1-2-1 1.68 27.7 6820 200 7840–7550 0.86
22726 SBU 1-2-8 2.48 25 6820 190 7840–7550 0.89
22730 SBU 3-1-6 2.2 25 2240 200 2470–1990 0.93

images ranging from 11 to 23 years (Table 2). Due to the large changes accom-
panying development over this interval, it was not possible to select a single set of
georeferencing points applicable to all photographs.
Twelve control points spaced across the study area for which differential GPS
data were obtained, were used to register the 2001 photograph in MAPInfo Pro-
fessional 7.0. A new set of control points identifiable on both the 1980 and 2001
photograph were then located and used to register the 1980 photograph. This pro-
cess was repeated with successively older photographs. Once registered, mangrove
areas and other features were digitised to polygons on-screen.
The position error for individual features identifiable in the 2001 photograph is
better than 5 m. The georeferencing errors for successively older photographs are
cumulative, but it is estimated that the errors associated with the mapping of all
boundaries across the study area are unlikely to exceed 15 m in any photographs,
except locally, in areas of heavy shade when the photograph was taken while the
sun was close to the horizon.
The mangrove polygons derived from successive time intervals were overlain to
obtain estimates of mangrove advance or retreat and mangrove stand age across
the study area. Terrestrial forest above the intertidal zone was not considered in
this analysis, although it is possible that small areas of terrestrial forest, growing

Table 2
Details of aerial photographs used in this study
Date Number Source
1946 Accession 10 photo 3019 NAS
1946 Accession 10 photo 3020 NAS
1946 Accession 10 photo 4019 NAS
1969 106 SN 68 NIE
1980 1980/R2/0815 (BG03) MINDEF
1990 1990/R1/5296 (FJ59) MINDEF
2001 S2001/R15/18 (AG28) MINDEF
NAS, National Archives of Singapore; NIE, National Institute of Education; MINDEF, Ministry of
Defence.
188 M. Bird et al. / Applied Geography 24 (2004) 181–198

on small isolated protrusions of Old Alluvium surrounded by mangrove forest,

may have been included.

Results and discussion

Holocene development
Prior to flooding of the area by post-glacial sea level rise, Sungei Buloh was an
incised creek draining a hilly terrain composed of Old Alluvium. The creek drained
north into a major river flowing south from Malaysia into the (terrestrial) Straits
of Johor, then west around the main landmass of Singapore and on into what is
now the Straits of Singapore.
Coring and augering in the mangrove forests of SBWR (Fig. 2) demonstrates
that the glacial age terrestrial land surface is now overlain along the entire seaward
fringe and modern intertidal zone up to several hundred metres inland by 1 to
>3.5 m of unconsolidated sands, muds and peats of Holocene age. It was conceiv-
able that Pulau Buloh included a core of consolidated material (a former hill in the
pre-transgression landscape), but augering has indicated that the entire island is
composed of unconsolidated Holocene sediments at least down to below the low-
tide line. This is significant because the seaward coast of Pulau Buloh is suffering
from coastal erosion (discussed in the next section), and therefore erosion has been,
and will be, facilitated by the fact that there is no core of more resistant material
present on the island.
Coring on a transect from the back mangrove area, flooded only during times of
highest tides, to the mudflat fronting the mangroves in the east of the study area
(Fig. 2) suggests that the pre-transgression land surface slopes gently to seaward
and is blanketed by 2–3 m of unconsolidated Holocene mangrove sediments
(Fig. 3). Cores SBU-1 and SBU-3 penetrated the entire Holocene sequence and
bottomed in the underlying deeply weathered Old Alluvium at 3.3 and 0.5 m rsl
(relative to mean sea level), respectively. The entire Holocene sequence was not
penetrated at SBU-2 but at least 2.8 m of Holocene sediments overlie basement at
this location, which lies below 1.8 m rsl. It was not possible to determine the
inland extent of the Holocene mangrove sequence due to major aquaculture devel-
opment SW of SBU-3.
Fig. 4 shows the sedimentary characteristics of the Holocene sequences pene-
trated by SBU-1 and SBU-3. The lowermost section of both cores is composed of
very consolidated sandy clays of the Old Alluvium, with dry bulk densities gener-
ally above 1.5 g/cm3, with low TOC of <0.5%. In both cores, the Old Alluvium is
overlain by a transgressive sequence of peaty sands and clays becoming progress-
ively richer in organic matter up the core. Bulk densities in the Holocene sequence
are generally lower than in the Old Alluvium. Although clastic-rich samples can
have densities up to ~1.5 g/cm3, the organic-rich peats generally have bulk den-
sities of ~0.5 g/cm3 or lower.
 M. Bird et al. / Applied Geography 24 (2004) 181–198 189

Fig. 3. Surface and sub-surface morphometry from levelling and coring on the eastern coast of the area
(see Fig. 2 for transect location). Also shown are tide levels as follows: HAT—highest astronomical tide;
MHWS—mean high water spring tide; MSL—mean sea level; MLWS—mean low water spring tide.

Radiocarbon dating of organic material close to the base of the Holocene

sequence at SBU-1 indicates that mean high water spring tide level reached 2.5 m
by ~6800 years BP (Table 1) consistent with Holocene sea level curves from Singa-
pore (Hesp et al., 1998; after correction of carbonate dates for a marine reservoir
effect) and the Straits of Malacca (Geyh, et al., 1979).
Thereafter mangrove sediments began to accrete very rapidly, with ~2 m of
sediment accumulated in ~400 years (~5 mm/year). The top of core SBU-1, corre-
sponding to the surface of the modern low-tide mud platform fronting the man-
groves is a fibrous organic-rich peat of mid-Holocene age (~6400 BP). Sediments
younger than mid-Holocene have been eroded during formation of the
modern low-tide platform.
The sequence in core SBU-3 is more complicated than in core SBU-1. The trans-
gressive mangrove sequence is overlain by a regressive mangrove sequence
deposited as sea level fell following the mid-Holocene highstand. In addition, the
upper part of the core was formed very recently during the period when the area
was enclosed by bunds to serve as a prawn pond. The former tidal channel at the
SBU-3 location was rapidly filled by very fine brown clay washed in from the
disturbed areas inland from the site. Following the cessation of prawn farming,
mangroves re-established in the area and the mud became mixed with organic
detritus to form a loose low-density uppermost sediment layer. The age at which
the sea transgressed the elevation corresponding to the interface between the Old
Alluvium and the Holocene sediments was not determined, but the sediments
immediately beneath the recent prawn farm muds were deposited around 2240 BP,
confirming that the upper mangrove sequence in the core was deposited during the
period of sea level regression following the mid-Holocene highstand.
190 M. Bird et al. / Applied Geography 24 (2004) 181–198

Fig. 4. Sediment characteristics of cores SBU-1 and SBU-3, including chronology provided by radio-
carbon dating of wood fragments.

Recent change

The aerial photographs suggest that the course of the major creek, Sungei Buloh,
changed several times prior to 1946. The creek originally discharged to the west of
what is now the island of Pulau Buloh, but coastal erosion and/or channel
migration led to breaching of a former meander loop at the eastern end of Pulau
Buloh, and the current outlet of Sungei Buloh developed (locations 1 and 2 on
Fig. 4). The island of Pulau Buloh is clearly identifiable on the 1886 map of Singa-
pore (National Archives of Singapore) and thus the formation of the current outlet
is not a recent phenomenon. The former channel south of Pulau Buloh, (called
‘Sungei Bilabong’) is now largely inactive and filled to low-tide level with muddy
sediments.
Fig. 5 shows that the main channel of Sungei Buloh in the centre of the study
area had changed at least once due to natural channel migration which cut through
a large meander loop prior to 1946, straightening the last kilometre of the
Sungei Buloh main channel (locations 3 and 4 on Fig. 5). A small amount of
 M. Bird et al. / Applied Geography 24 (2004) 181–198 191

Fig. 5. Geomorphology and mangrove distribution in 1946, with features discussed in the text marked
by numbers. The low-tide line in 2001 is shown for reference purposes.

encroachment of back mangrove area by coconut plantations had already occurred

in the east of the study area by 1946. The mangrove forest established over most of
the area and particularly in the seaward areas was comparatively young and the
canopy of the forest was not closed, even on Pulau Buloh, suggesting either that
some logging within the forest had occurred prior to 1946 or that the mangroves
were actively establishing on what had been bare tidal mudflat areas.
The mangrove fringe in the eastern part of the study area was substantially thin-
ner than it is today, with 100–200 m of bare mudflat, dotted by occasional individ-
ual trees, exposed between the mangroves and the modern low-tide line. A total of
117.3 ha of the study area was covered by mangroves in 1946.
Fig. 6 shows the changes that occurred in the distribution of mangroves between
1946 and 1969. During this period, the existing mangrove forests thickened sub-
stantially, with the forest canopy over much of the area becoming closed. The
mangroves also advanced substantially, infilling what had been tidal inlets across
the area, and advancing over areas that had been bare mudflat in 1946. While the
exact magnitude of the advance is uncertain due to the likely magnitude of errors
192 M. Bird et al. / Applied Geography 24 (2004) 181–198

Fig. 6. Changes in geomorphology and mangrove distribution between 1946 and 1967. The back man-
grove line for 1946 and the low-tide line for 2001 are shown for reference.

associated with the digitising process, direct comparison of the photographs, cou-
pled with GIS assessment suggests that mangroves may have advanced by ~100 m
in places between 1946 and 1969. It may have been that the construction of the
causeway to Johor (circa 1913), with attendant changes to tidal flows (Lim, 1983)
may have caused the mangrove advance evident by 1969 and possibly already
underway prior to 1946.
Natural loss of mangroves through coastal erosion occurred as a result of the
gradual widening of the mouth of Sungei Buloh and small-scale clearance for agri-
culture (in the northwest) and aquaculture (in the south and east) had begun. An
area of 29.6 ha of new mangroves was established between 1946 and 1969, while
18.4 ha was lost, giving a net increase in mangrove area of 9.5% to 128.8 ha.
Fig. 7 shows the changes in distribution that occurred between 1969 and 1980.
Mangroves continued to advance over the mudflats on the northeast coast and also
along the southern coast of Pulau Buloh, but this period marked the rapid loss of
mangroves to aquaculture. Most of the western mainland part of the study area
was cleared at this time, leaving remnant patches enclosed by bunds and a
 M. Bird et al. / Applied Geography 24 (2004) 181–198 193

Fig. 7. Changes in geomorphology and mangrove distribution between 1967 and 1980. The back man-
grove line for 1946 and the low-tide line for 2001 are shown for reference.

mangrove fringe to seaward of the bunds. At the eastern end of the study area,
construction of the Kranji Dam, with attendant reclamation to seaward of the dam
wall led to considerable loss of mangrove area. Gain in mangrove area, sometimes
over previously cleared but abandoned areas amounted to 12.7 ha, while mangrove
loss amounted to 73.7 ha, giving a net loss of 47.4% between 1969 and 1980, with
67.3 ha covered by mangroves in 1980.
The changes in mangrove distribution that occurred between 1980 and 2001 are
shown in Fig. 8. This period encompasses the end of aquaculture and development
of SBWR. Substantial areas of mangrove have re-established in the area of SBWR
and also along Sungei Buloh south of the reserve. These increases have been offset
somewhat by a change from mangrove advance to mangrove retreat along most of
the coastline, including the northern coast of Pulau Buloh. There was a gain in
new mangrove area during this period of 36.2 ha, offset by a loss of existing man-
groves of 15.0 ha, yielding a net increase of 31.5% to 86.8 ha.
194 M. Bird et al. / Applied Geography 24 (2004) 181–198

Fig. 8. Changes in geomorphology and mangrove distribution between 1980 and 2001. The back man-
grove line for 1946 and the low-tide line for 2001 are shown for reference.

The extent of recent retreat (determined by the existence of dead and fallen trees
beyond the living mangrove fringe) was quantified by DGPS for the eastern part of
the study area (Fig. 2) and ranges between 0 and 10 m. The retreat of the northern
coast of Pulau Buloh was not quantified by DGPS but analysis of the aerial photo-
graphs suggests retreat amounting to 20–50 m since 1980 along most of the coast
of the island. The comparatively recent change from mangrove advance to man-
grove retreat may have been initiated by the construction of Kranji Dam, east of
the study area, which may have changed the local hydrodynamic environment and
reduced the amount of sediment arriving at the coast. In addition, mangrove
retreat that has occurred in areas in front of old bunds may have been partly due
to the initiation of more vigorous bottom scouring due to the reflection of wave
energy by the bunds. While ‘natural’ wave energy in the Straits of Johor is low,
many vessels use the Straits and generate waves exceeding 20 cm in height at the
Sungei Buloh coast.
 M. Bird et al. / Applied Geography 24 (2004) 181–198 195

Fig. 9. Distribution of mangrove stands of 34–55 years and >55 years in the study area, areas of man-
groves <34 years not shown. The back mangrove line for 1946 and the low-tide line for 2001 are shown
for reference. Letters indicate locations referred to in the text.

The sequence of photographs analysed for this study allows the identification of
forest within the study area that has survived the large disturbances that character-
ized the 1960–1990 period. Fig. 9 shows the distribution of mangrove areas with
apparent ages 34–55 years and >55 years in 2001. Many of the smaller areas may
be an artefact of digitising errors or are too small to warrant special consideration.
Other larger areas, particularly in the western mainland area and the eastern end of
the study area may have been cleared and regrown between photos, and hence not
as old as suggested by this analysis. Two areas, areas D and E in Fig. 9 certainly
were cleared and regrown between photographs, so though they appear as old
growth areas by the definitions imposed during the digitizing process, the trees in
these areas are actually much younger. Despite these complications, three areas
stand out as having special conservation importance as areas of undisturbed old
growth forest.
In 1946, Pulau Buloh (Area A; Fig. 9) was covered with relatively immature
mangrove forest consisting mostly of isolated trees and clumps of trees. Since that
196 M. Bird et al. / Applied Geography 24 (2004) 181–198

time the island has not been disturbed and the forest has matured into a dense
closed canopy forest mostly older than 55 years in age in 2001. Likewise, Area B
(Fig. 9), east of Sungei Buloh for about 400 m has not been disturbed, though
much of the forest only began colonizing the mudflats between 1946 and 1969.
Further east remnants of mature forest remain, but the area is partly enclosed by
bunds, possibly constructed as part of the Kranji dam project, and field investi-
gation indicates that much of this area may have been cleared and regrown
between 1969 and 1980.
Area C (Fig. 9), an isolated 1.3 ha fragment of mangrove forest in the south of
the study area on the western bank of Sungei Buloh was relatively mature man-
grove forest by 1946 and has escaped disturbance since that time. Field investi-
gation has confirmed that this patch of forest is composed dominantly of old
growth Rhizophora spp. and therefore appears to be the oldest surviving forest
remnant in the study area. Other such remnants may exist in the SBWR area, but
these have not been confirmed by field visits.

Conclusions

The dissected and deeply weathered land surface of the last ice age in the study
area was flooded by post-glacial sea level rise by 6820 BP. Thereafter a sequence of
variably sandy to muddy unconsolidated mangrove sediments and peats were
deposited to a depth of several metres over the low-lying parts of the area. Follow-
ing a mid-Holocene highstand, sea level gradually receded, eroding some of the
sediments deposited during the earlier transgression, and depositing a younger ser-
ies of mangrove sediments. After the sea reached its current level, erosion of the
Holocene sediments continued, and a low-tide platform 10–50 m wide was cut
shorewards into the Holocene sediments along most of the seaward fringe of the
area.
Analysis of a time series of aerial photographs has allowed the elucidation of
massive changes that have occurred in many parts of the study area in more recent
time (since 1946). The effects of a major phase of clearance between 1969 and 1980
(or later), which reduced the area of mangrove forest by almost 50%, have been
partly offset by reafforestation in more recent times and an addition of 6.2 ha from
seaward advance of the mangrove fringe since 1946. Nevertheless, the total area
colonized by mangroves has shrunk by over 25% from 117.3 ha in 1946 to 86.8 ha
in 2001.
From 1946 to 1980, mangroves were actively colonizing bare mudflats on the
seaward fringe of the area, possibly due to changes in local hydrodynamic regime
associated with construction of the causeway between Singapore and Johor in the
early 20th century. Since 1980, the seaward fringe of the mangroves has been
retreating due to coastal erosion, possibly the result of construction of the Kranji
Dam. Three areas of high conservation value, old growth mangroves have been
identified in the study area. Of particular importance to managers, two of these
areas, Pulau Buloh and the coastal fringe east of Sungei Buloh, are currently
 M. Bird et al. / Applied Geography 24 (2004) 181–198 197

undergoing significant coastal erosion of up to 50 m since 1980. Augering on Pulau

Buloh in particular has demonstrated that there is no core of consolidated material
at shallow depth that might serve to slow the rate of coastal erosion on the island
in the future. A third area of high conservation potential, in the south of the study
area, lies outside the current SBWR boundaries and consideration should be given
to extending the park to provide protection to what is the oldest undisturbed man-
grove forest in SBWR and therefore possibly on Mainland Singapore.

Acknowledgements

We wish to thank Sungei Buloh Wetland Reserve for permission to undertake

this study.

References
Chia, L. S., Khan, H., & Chou, L. M. (1988). The Coastal Environmental Profile of Singapore. ASEAN/
US Coastal Resources Management Project. Technical Publications, Series 3.
Chua, B. H., & Edwards, N. (1992). Public space: Design, use and management (pp. 240). Singapore:
Singapore University Press.
Dahdouh-Guebas, F. (2002). The use of remote sensing and GIS in the sustainable management of tropi-
cal coastal ecosystems. Environment, Development and Sustainability, 4, 93–112.
Geyh, M. A., Kudrass, H. R., & Streif, H. (1979). Sea-level changes during the late Pleistocene and
Holocene in the Straits of Malacca. Nature, 278, 441–443.
Gupta, A. (1986). Quaternary geology in Singapore: A review. Proceedings of the CCOP Symposium on
Progress in Quaternary Geology of east and Southeast Asia (pp. 263–269). Thailand: Bangkok.
Gupta, A., et al. (1987). The Old Alluvium of Singapore and the extinct drainage system to the South
China Sea. Earth Surface Processes and Landforms, 12, 259–275.
Hesp, P. A., Chang, C. H., Hilton, M., Ming, C. L., & Turner, I. M. (1998). A first tentative Holocene
sea-level curve for Singapore. Journal of Coastal Research, 14, 308–314.
Lim, L. C. (1983). Current movement in the East Johor Straits and its adjacent waters. Singapore Jour-
nal of Primary Industries, 11, 83–97.
Lye, K. H., Lim, P. E., Sieh, K. C., & Midun, Z. (1991). Simulation of hydrodynamic changes in Johor
Strait, Malaysia. In Chou et al. (Eds.), Towards an integrated management of tropical coastal resour-
ces (pp. 103–107). ICLARM conference Proceedings 22, National University of Singapore.
Manson, F. J., Loneragan, N. R., & Phinn, S. R. (2003). Spatial and temporal variation in distribution
of mangroves in Moreton Bay, sub-tropical Australia: a comparison of pattern metrics and change
detection analyses based on aerial photographs. Estuarine, coastal and Shelf Science, 56, 1–14.
Ng, K. L., & Sivasothi, N. (1999). A guide to the Mangroves of Singapore. Singapore: Singapore Science
Centre.
Panapitukkul, N., Duarte, C. M., Thampanya, U., Kheovongsri, P., Srichai, N., Geertz-Hansen, O.,
Terados, J., & Boromthanarath, S. (1998). Mangrove colonization: mangrove progression over the
growing Pak Phanang (SE Thailand) mudflat. Estuarine, coastal and Shelf Science, 47, 51–61.
Public Works Department. (1976). Geology of the Republic of Singapore (pp. 79). Public Works Depart-
ment.
Stuiver, M., Reimer, P. J., & Braziunas, T. F. (1998). High-precision radiocarbon age calibration for ter-
restrial and marine samples. Radiocarbon, 40, 1127–1151 (Data available on the Internet at http://
www.calib.org/).
Sulong, I., Mohd-Lokman, H., Mohd-Tarmizi, K., & Ismail, A. (2002). Mangrove mapping using Land-
sat imagery and aerial photographs: Kemaman District, Terengganu, Malaysia. Environment, Devel-
opment and Sustainability, 4, 135–152.
 198 M. Bird et al. / Applied Geography 24 (2004) 181–198

Tan (1997). A botanical survey of Sungei Buloh Nature Park, Singapore. Gardens Bulletin Singapore, 49,
15–35.
Vasconcelos, M. J. P., Mussa Biai, J. C., Araujo, A., & Diniz, M. A. (2002). Land cover change in
Guinea Bissau (1956–1988). Applied Geography, 22, 139–156.
Verhheyden, A., Dahdouh-Guebas, F., Thomaes, K., de Genst, W., Hettiarachchi, S., & Koedam, N.
(2002). High resolution vegetation data for mangrove research as obtained from aerial photography.
Environment, Development and Sustainability, 4, 113–133.
Wong, P. P. (1992). Impact of a sea level rise on the coasts of Singapore: preliminary observations. Jour-
nal of Southeast Asian Earth Sciences, 7, 65–70.

View publication stats